stuff
 

See this on step 2 for the video demo:

    http://nitroproject.com/wiki/pages/Nitro+in+Action+one+step+2

A better method is welcomed.  *smile*

Bye,
     William

-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of James Britt
Sent: Friday, 7 July 2006 00:25
To: General discussion about Nitro
Subject: Re: [Nitro] [Oxy] pending questions 2006-07-03
Importance: Low

>>code-element rendering is suppose to create *inline* markup, which is 
>>what I wanted, not that block code with the unreadable color scheme.
> ....
> Define unreadable color scheme, white on black is unreadable? See a 
> previous question[1] on oxyliquit, where a color change request was 
> made. I will integrate almost anything, just tell me what.

....
Sorry; "unreadable" is quite subjective.  I prefer dark text on light
background and have a hard time reading it otherwise.  I generally use a 
Firefox script to turn off page colors in such cases.   The code colors 
on oxy seem to mix small white fonts with a very dark background, but
various keywords use colors that do not (to my eyes) offer sufficient
clarity.

(I took a stab at writing a Greasemonkey script to apply my preferred colors
but it doesn't quite work yet.)

I expect there's no good way to pick a single colour scheme to please
everyone.

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.9/382 - Release Date: 04-Jul-2006
 


From transfire at gmail.com  Thu Jul  6 11:33:30 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 6 Jul 2006 11:33:30 -0400
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: 
References: <3ff63f9b0606270721s1891de30k70053913522bd1f@mail.gmail.com>
	
	<4b6f054f0606301033n2bc9987dled56a73a8e6d0c88@mail.gmail.com>
	
	<4b6f054f0607011223w3210b68i12aaba57660b19ee@mail.gmail.com>
	
	<4b6f054f0607011618t36c6973co53f1c46d9df6359a@mail.gmail.com>
	<4b6f054f0607060728h6b449c91j64eca05e3be04a9@mail.gmail.com>
	
Message-ID: <4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>

On 7/6/06, George Moschovitis  wrote:
> > I added settings/configuration to facets. I renamed Configuration to
> > just Settings, no alias --hope that's cool.
>
> Please, add the alias too!

Okay. But why do you need both names?

T.

From george.moschovitis at gmail.com  Thu Jul  6 11:42:14 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 6 Jul 2006 18:42:14 +0300
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: <4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>
References: <3ff63f9b0606270721s1891de30k70053913522bd1f@mail.gmail.com>
	
	<4b6f054f0606301033n2bc9987dled56a73a8e6d0c88@mail.gmail.com>
	
	<4b6f054f0607011223w3210b68i12aaba57660b19ee@mail.gmail.com>
	
	<4b6f054f0607011618t36c6973co53f1c46d9df6359a@mail.gmail.com>
	<4b6f054f0607060728h6b449c91j64eca05e3be04a9@mail.gmail.com>
	
	<4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>
Message-ID: 

> Okay. But why do you need both names?

Different names make sense in different places. To make the source
code more readable.

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From james.britt at gmail.com  Thu Jul  6 12:29:13 2006
From: james.britt at gmail.com (James Britt)
Date: Thu, 06 Jul 2006 09:29:13 -0700
Subject: [Nitro] [Oxy] colour idea
In-Reply-To: <010701c6a111$24a077d0$84af07ca@ghostgum>
References: <010701c6a111$24a077d0$84af07ca@ghostgum>
Message-ID: <44AD3A59.8090701@gmail.com>

* William wrote:

> 
> My first reaction is why bother with Greasemonkey?  Can't you just override
> the Ruby-code-style with a USER.CSS file?  I believe it is possible, would I
> bother?  

I'll check.  I think that is for all sites, not just one or another.
This page has some info, though:

http://www.squarefree.com/userstyles/

> 
> More importantly, ... Would "mom and pop" user have a clue what I'm talking
> about?!  Would they know what Greasemonkey is?!!!!!  Let alone understand
> it.
> 
> For James and other technically inclined folks, you can use a greasemonkey
> toy called platypus (no relation), its job is to automate
> presentation-revisions as James mentions.
> 

Oh, nice.  Thanks!

> For "mom and pop" user; I was thinking about something in Nitro to return
> individual CSS stream per user.  That data-stream is a HTTP stream and could
> be templated via Ruby as something we can call an rCSS (likened to rXHTML).

This all interesting, but a perhaps simpler suggestion is to offer a few 
alternative style sheets and allow the browser to select one.

http://www.alistapart.com/stories/alternate/

-- 
James Britt

"Blanket statements are over-rated"

From james.britt at gmail.com  Thu Jul  6 12:43:02 2006
From: james.britt at gmail.com (James Britt)
Date: Thu, 06 Jul 2006 09:43:02 -0700
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: 
References: <3ff63f9b0606270721s1891de30k70053913522bd1f@mail.gmail.com>		<4b6f054f0606301033n2bc9987dled56a73a8e6d0c88@mail.gmail.com>		<4b6f054f0607011223w3210b68i12aaba57660b19ee@mail.gmail.com>		<4b6f054f0607011618t36c6973co53f1c46d9df6359a@mail.gmail.com>	<4b6f054f0607060728h6b449c91j64eca05e3be04a9@mail.gmail.com>		<4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>
	
Message-ID: <44AD3D96.50500@gmail.com>

George Moschovitis wrote:
>>Okay. But why do you need both names?
> 
> 
> Different names make sense in different places. To make the source
> code more readable.

Um ... maybe, but it makes deciphering code harder.

If I see different names, my first assumption is that they are different 
things.  Makes it harder to understand how to use things, because now 
you have to track when multiple names are referring to the same thing.

Honestly, is the difference between "Settings" and "Configuration" such 
that  they cannot be used interchangeably?  Are there that many cases 
(or even *one*) where using one name makes anything more clear than 
using the other?   Is there a case where anyone would get upset because 
they had to call it "Settings" instead of "Configurations"?

And if the concept of "settings" is indeed different from 
"configurations" then the names shouldn't refer to the same thing.

I'm not a big fan of There's Only One Way To Do It, but it (as Guido 
argues) it really does make certain things easier to learn.

TMTOWTDI can become EDID (Everyone Does It Differently) quite fast.


-- 
James Britt

"Blanket statements are over-rated"

From john at oxyliquit.de  Thu Jul  6 15:40:02 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Thu, 06 Jul 2006 21:40:02 +0200
Subject: [Nitro] [Oxy] colour idea
In-Reply-To: <010701c6a111$24a077d0$84af07ca@ghostgum>
References: <010701c6a111$24a077d0$84af07ca@ghostgum>
Message-ID: 

Hi,

> These comments gave me an idea.  Why can't a user on the www2 platform of
> the future have his/her own CSS choices?  Colour is a good start.

Put on todo. Thx for suggestion.

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From john at oxyliquit.de  Thu Jul  6 15:40:03 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Thu, 06 Jul 2006 21:40:03 +0200
Subject: [Nitro] [Oxy] pending questions 2006-07-03
In-Reply-To: <44AD1D53.5090906@gmail.com>
References: <005901c69f1d$732413d0$52af07ca@ghostgum>
	<44AC89C7.4090400@gmail.com> 
	<44AD1D53.5090906@gmail.com>
Message-ID: 

Hi,

> Sorry; "unreadable" is quite subjective.  I prefer dark text on light

See post to William.

> The forced-block code is the bigger issue, and thanks for looking after
> that.  (I really appreciate oxiwtf and hope I don't sound too thankless
> for complaining about the code formatting.)

Nono, everything helps! Nothing is worse than no feedback :)

> Screw the exams!
T_T

> No, seriously, best of luck.  Let us know how you do.

Will do, many thanks ^_^

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From nusgnaf at gmail.com  Thu Jul  6 16:31:24 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 6 Jul 2006 20:31:24 +0000
Subject: [Nitro] mongrel adapter vs. large request
Message-ID: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>

Hi, devs on the list, while running nitro that handle file uploads
with mongrel adapter, I hit the following error:
"Fri Jul 07 00:50:58 UTC 2006: ERROR: can't convert Tempfile into String"
A quick googling reveals this discussion on mongrel-user:
http://rubyforge.org/pipermail/mongrel-users/2006-April/000103.html
So, when mongrel pass request body to Nitro engine, the request body
maybe a StringIO or a Tempfile, for performance reason, this spots the
following bug: lib/nitro/adapter/mongrel.rb line128:
xxxxxxxxxxx
 StringIO.new(req.body || "")
xxxxxxxxxxx
My understanding is that we wrap http request around Nitro's context
class, this makes our code uniform and easy to maintain, but cause it
invole copy whole request body to memory, this make mongrel's
optimization much useless, so what's the better way to fix the problem
and gain better performance from the optimization?

From transfire at gmail.com  Thu Jul  6 17:06:08 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 6 Jul 2006 17:06:08 -0400
Subject: [Nitro] Annotations 2.0
Message-ID: <4b6f054f0607061406j1b8b9016u85b4872360bd2238@mail.gmail.com>

Okay, by some feat of magic I managed a much nicer implementation of
annotations. The major difference between the old and new:

1) Only one layer. Ie. No automatuic chaining, X.ann.a.b.c.d.e... etc.
One data layer is usually enough and in the rare case that you have to
have more layers, one can assign the OpenCascade themselves.

  X.ann :a, b => OpenCascade.new
  X.ann.a.b.c.d.e....

2) You no longer need the bang method to change an annotatin directly.
This is achived via a clever RWDelegator.

3) A "lookup" notation is supported, which I actually find more intuitive:

    X.ann(:foo).bar == X.ann.foo.bar

This notation also allows non-symbol annotation keys to be used.

(Q: Is that worth having or should all keys be symbols, period?)

5) Not so sure about this, but as it stands I am storing the
class/module and name of the annotation in the annotation object. This
will limit their reusability, but might be worth it. Problem arises
though if you wanted to do something like:

  X.ann.foo = X.ann.bar

(Q: Is that something that should be doable, or is it no big loss?)

6) I haven't dealt with the module case yet but it should be a simple
as using class_extension.

The new code requires Facets 1.5, now under devlopment, so you'll have
to wait until I get the repo up to test it out.

Jonas, I'm using Darcs, hope that's okay. I'm familiar enough with
Darcs now that I'd be happy to upload the repo to devlab if that's
cool with you?

T.

From william.full.moon at gmail.com  Thu Jul  6 22:32:33 2006
From: william.full.moon at gmail.com (* William)
Date: Fri, 7 Jul 2006 12:32:33 +1000
Subject: [Nitro] [Oxy] colour idea
In-Reply-To: <44AD3A59.8090701@gmail.com>
Message-ID: <013601c6a16d$a38816f0$84af07ca@ghostgum>

Hi 

The platypus link is:  https://addons.mozilla.org/firefox/737/

It is also good for debugging your CSS, it highlights the item and shows the
class and style info in the status bar.  I'd use it for just that sometimes.
:D

Yes, I believe the alternative style sheets idea is more suitable to "mum
and pop"

    http://www.alistapart.com/stories/alternate/

My comment was more on the lines of the mechanism.  A template based
mechanism for CSS, like the method Nitro used for XHTML (templates) pages
sounded interesting.  One could then allow the user to alter some aspects.
Example:

   %-CSS-
     body {
        background: #{user_bg};
        color:      #{user_fg};
     }
   -CSS-
 
The selection of alternates, provides people with a base choice.  The
Generation-Y mass market of the first half of this millennium will demand
personalization.  The funkier the better.  You can look at all the covers
for mobile phones and iPods to verify that forecast. *grin*

I think as a mechanism, you can do either, both, and more with a template
based CSS idea. 

More fun,
          Will.


-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of James Britt
Sent: Friday, 7 July 2006 02:29
To: General discussion about Nitro
Subject: Re: [Nitro] [Oxy] colour idea
Importance: Low

* William wrote:

> For James and other technically inclined folks, you can use a 
> greasemonkey toy called platypus (no relation), its job is to automate 
> presentation-revisions as James mentions.
> 

Oh, nice.  Thanks!

> For "mom and pop" user; I was thinking about something in Nitro to 
> return individual CSS stream per user.  That data-stream is a HTTP 
> stream and could be templated via Ruby as something we can call an rCSS
(likened to rXHTML).

This all interesting, but a perhaps simpler suggestion is to offer a few
alternative style sheets and allow the browser to select one.

http://www.alistapart.com/stories/alternate/

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.9/382 - Release Date: 04-Jul-2006
 


From william.full.moon at gmail.com  Thu Jul  6 22:49:29 2006
From: william.full.moon at gmail.com (* William)
Date: Fri, 7 Jul 2006 12:49:29 +1000
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: <44AD3D96.50500@gmail.com>
Message-ID: <013a01c6a16f$fafeeb50$84af07ca@ghostgum>

Hi ho all ...

>> Um ... maybe, but it makes deciphering code harder.

Unfortunately that is the truth.  It might be OK for writing and write-once
code.  It has no legs when it comes to fixing bugs and maintain code.  And
Ruby didn't invent it (nor small talk) it goes way back to FORTRAN and
assembler.

People used to define things in COMMON blocks, where a variable was called
"Temperature" in one place and "Degrees" on another line, in the same
function.

C programmers and macro assembler people can define "Temperature" and
"Degrees" to mean the same variable.  Or worse you can make 

    pTemp    = pDegrees;
    (*pTemp) = 50;

In the long run, things get broken because not everyone will check to see if
"Temperature" is the _same_thing_ as "Degrees".  It fact it makes life
EXTREMELY TIRESOME to double check every variable or methods for an "alias".

To me it is a practice best avoided.  Besides if you use up all the cool
words on this feature, there are less valid _potential_ words to use of the
next feature.

Cheers,
        Will.


-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of James Britt
Sent: Friday, 7 July 2006 02:43
To: General discussion about Nitro
Subject: Re: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
Importance: Low

George Moschovitis wrote:
>>Okay. But why do you need both names?
> 
> 
> Different names make sense in different places. To make the source 
> code more readable.

Um ... maybe, but it makes deciphering code harder.

If I see different names, my first assumption is that they are different
things.  Makes it harder to understand how to use things, because now you
have to track when multiple names are referring to the same thing.

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.9/382 - Release Date: 04-Jul-2006
 


From manveru at weez-int.com  Thu Jul  6 23:47:08 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Fri, 7 Jul 2006 12:47:08 +0900
Subject: [Nitro] [Oxy] colour idea
In-Reply-To: <013601c6a16d$a38816f0$84af07ca@ghostgum>
References: <013601c6a16d$a38816f0$84af07ca@ghostgum>
Message-ID: <200607071247.08262.manveru@weez-int.com>

Afaik we have a CSS-templating already...
should be in the announcements as well :)

On Friday 07 July 2006 11:32, * William wrote:
> Hi
>
> The platypus link is:  https://addons.mozilla.org/firefox/737/
>
> It is also good for debugging your CSS, it highlights the item and shows
> the class and style info in the status bar.  I'd use it for just that
> sometimes.
>
> :D
>
> Yes, I believe the alternative style sheets idea is more suitable to "mum
> and pop"
>
>     http://www.alistapart.com/stories/alternate/
>
> My comment was more on the lines of the mechanism.  A template based
> mechanism for CSS, like the method Nitro used for XHTML (templates) pages
> sounded interesting.  One could then allow the user to alter some aspects.
> Example:
>
>    %-CSS-
>      body {
>         background: #{user_bg};
>         color:      #{user_fg};
>      }
>    -CSS-
>
> The selection of alternates, provides people with a base choice.  The
> Generation-Y mass market of the first half of this millennium will demand
> personalization.  The funkier the better.  You can look at all the covers
> for mobile phones and iPods to verify that forecast. *grin*
>
> I think as a mechanism, you can do either, both, and more with a template
> based CSS idea.
>
> More fun,
>           Will.
>
>
> -----Original Message-----
> From: nitro-general-bounces at rubyforge.org
> [mailto:nitro-general-bounces at rubyforge.org] On Behalf Of James Britt
> Sent: Friday, 7 July 2006 02:29
> To: General discussion about Nitro
> Subject: Re: [Nitro] [Oxy] colour idea
> Importance: Low
>
> * William wrote:
> > For James and other technically inclined folks, you can use a
> > greasemonkey toy called platypus (no relation), its job is to automate
> > presentation-revisions as James mentions.
>
> Oh, nice.  Thanks!
>
> > For "mom and pop" user; I was thinking about something in Nitro to
> > return individual CSS stream per user.  That data-stream is a HTTP
> > stream and could be templated via Ruby as something we can call an rCSS
>
> (likened to rXHTML).
>
> This all interesting, but a perhaps simpler suggestion is to offer a few
> alternative style sheets and allow the browser to select one.
>
> http://www.alistapart.com/stories/alternate/

From zimba.tm at gmail.com  Fri Jul  7 03:15:13 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Fri, 7 Jul 2006 09:15:13 +0200
Subject: [Nitro] Annotations 2.0
In-Reply-To: <4b6f054f0607061406j1b8b9016u85b4872360bd2238@mail.gmail.com>
References: <4b6f054f0607061406j1b8b9016u85b4872360bd2238@mail.gmail.com>
Message-ID: <3ff63f9b0607070015n587e2b99xfa7e425e09dca2ad@mail.gmail.com>

On 06/07/06, TRANS  wrote:
> (Q: Is that worth having or should all keys be symbols, period?)

I prefer that annotation remains simple. So only with symbols. If
somebody needs something more advanced, he/she can still build his how
object like you showed with OpenCascade.

What would be nice altrough, is to have a way to extend annotations.
For example Glue::Configuration could use annotations if it supported
an extension mechanism.

> (Q: Is that something that should be doable, or is it no big loss?)

Not for me

> Jonas, I'm using Darcs, hope that's okay. I'm familiar enough with
> Darcs now that I'd be happy to upload the repo to devlab if that's
> cool with you?

definately. You can already scp the repo in your htdocs folder. It
will then be available here : http://devlab.oree.ch/~trans/

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From zimba.tm at gmail.com  Fri Jul  7 03:24:35 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Fri, 7 Jul 2006 09:24:35 +0200
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
Message-ID: <3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>

Hi Fang Sun,

On 06/07/06, Fang Sun  wrote:
[...]
> So, when mongrel pass request body to Nitro engine, the request body
> maybe a StringIO or a Tempfile, for performance reason, this spots the
> following bug: lib/nitro/adapter/mongrel.rb line128:
> xxxxxxxxxxx
>  StringIO.new(req.body || "")
> xxxxxxxxxxx

This code is bad. Once upon a time, Nitro supported http body
streaming. But I guess it was abandoned against super useless features
like Elements.

[...]
> so what's the better way to fix the problem
> and gain better performance from the optimization?

Mongrel supports the X-Sendfile extension if you installed the `sendfile` gem.
In your action, set the @headers['X-Sendfile'] = '/path/to/your/file'
and return no body. Mongrel will intercept that header and join the
file as a streamed body.

PS : This solution may only works on POSIX

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From transfire at gmail.com  Fri Jul  7 03:41:48 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 7 Jul 2006 03:41:48 -0400
Subject: [Nitro] Annotations 2.0
In-Reply-To: <3ff63f9b0607070015n587e2b99xfa7e425e09dca2ad@mail.gmail.com>
References: <4b6f054f0607061406j1b8b9016u85b4872360bd2238@mail.gmail.com>
	<3ff63f9b0607070015n587e2b99xfa7e425e09dca2ad@mail.gmail.com>
Message-ID: <4b6f054f0607070041x3286a805w3fa5e70f36a32b5f@mail.gmail.com>

On 7/7/06, Jonas Pfenniger  wrote:
> On 06/07/06, TRANS  wrote:

> What would be nice altrough, is to have a way to extend annotations.
> For example Glue::Configuration could use annotations if it supported
> an extension mechanism.

How do envision this?

> > (Q: Is that something that should be doable, or is it no big loss?)
>
> Not for me
>
> > Jonas, I'm using Darcs, hope that's okay. I'm familiar enough with
> > Darcs now that I'd be happy to upload the repo to devlab if that's
> > cool with you?
>
> definately. You can already scp the repo in your htdocs folder. It
> will then be available here : http://devlab.oree.ch/~trans/

?
  scp trans at devlab.oree.ch facets/trunk
?

T.

From zimba.tm at gmail.com  Fri Jul  7 03:49:53 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Fri, 7 Jul 2006 09:49:53 +0200
Subject: [Nitro] Annotations 2.0
In-Reply-To: <4b6f054f0607070041x3286a805w3fa5e70f36a32b5f@mail.gmail.com>
References: <4b6f054f0607061406j1b8b9016u85b4872360bd2238@mail.gmail.com>
	<3ff63f9b0607070015n587e2b99xfa7e425e09dca2ad@mail.gmail.com>
	<4b6f054f0607070041x3286a805w3fa5e70f36a32b5f@mail.gmail.com>
Message-ID: <3ff63f9b0607070049p1d265f7dkc1a6b994bff66872@mail.gmail.com>

>   scp trans at devlab.oree.ch facets/trunk

scp source destination

ex:
scp -r facets-repo trans at devlab.oree.ch:htdocs

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From aglarond at gmail.com  Fri Jul  7 03:54:07 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Fri, 7 Jul 2006 09:54:07 +0200
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: <3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
Message-ID: <55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>

Hey zimba,

On 7/7/06, Jonas Pfenniger  wrote:
>
>
> Mongrel supports the X-Sendfile extension if you installed the `sendfile`
> gem.
> In your action, set the @headers['X-Sendfile'] = '/path/to/your/file'
> and return no body. Mongrel will intercept that header and join the
> file as a streamed body.
>

So that's how that works?  Cool.  Aidan, did you read this?

The OP was concerned about handling the request.  This sendfile magic only
deals with the response though, right?  What would be the most effecient way
of handling a request?

- Dimitri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060707/1deba52d/attachment.html 

From zimba.tm at gmail.com  Fri Jul  7 04:20:48 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Fri, 7 Jul 2006 10:20:48 +0200
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: <55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
	<55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>
Message-ID: <3ff63f9b0607070120p2ceaacecw58257b8a2974adb2@mail.gmail.com>

On 07/07/06, Dimitri Aivaliotis  wrote:
> The OP was concerned about handling the request.  This sendfile magic only
> deals with the response though, right?  What would be the most effecient way
> of handling a request?

Gh I missed that point. I don't know really.

IMHO the easiest way is to build a new MongrelUploadAdapter that deals
with file uploads. Nitro is too complicated some times.

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From surrender_it at yahoo.it  Fri Jul  7 06:29:08 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Fri, 07 Jul 2006 12:29:08 +0200
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: <44AD3D96.50500@gmail.com>
References: <3ff63f9b0606270721s1891de30k70053913522bd1f@mail.gmail.com>		<4b6f054f0606301033n2bc9987dled56a73a8e6d0c88@mail.gmail.com>		<4b6f054f0607011223w3210b68i12aaba57660b19ee@mail.gmail.com>		<4b6f054f0607011618t36c6973co53f1c46d9df6359a@mail.gmail.com>	<4b6f054f0607060728h6b449c91j64eca05e3be04a9@mail.gmail.com>		<4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>	
	<44AD3D96.50500@gmail.com>
Message-ID: 

James Britt ha scritto:
> George Moschovitis wrote:
> 
>>>Okay. But why do you need both names?
>>
>>
>>Different names make sense in different places. To make the source
>>code more readable.
> 
> 
> Um ... maybe, but it makes deciphering code harder.

+1 from me, too, less aliasing is better in this case, imho.


-- 

blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From george.moschovitis at gmail.com  Fri Jul  7 08:39:01 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 7 Jul 2006 15:39:01 +0300
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: <3ff63f9b0607070120p2ceaacecw58257b8a2974adb2@mail.gmail.com>
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
	<55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>
	<3ff63f9b0607070120p2ceaacecw58257b8a2974adb2@mail.gmail.com>
Message-ID: 

> IMHO the easiest way is to build a new MongrelUploadAdapter that deals
> with file uploads. Nitro is too complicated some times.

I agree. I will have to write something like this for personal use real soon.

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From nusgnaf at gmail.com  Fri Jul  7 11:35:27 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Fri, 7 Jul 2006 15:35:27 +0000
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: <3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
Message-ID: <716700c90607070835q4d70914p226f3ccc6b121a01@mail.gmail.com>

On 7/7/06, Jonas Pfenniger  wrote:
> Hi Fang Sun,
>
> On 06/07/06, Fang Sun  wrote:
> [...]
> > So, when mongrel pass request body to Nitro engine, the request body
> > maybe a StringIO or a Tempfile, for performance reason, this spots the
> > following bug: lib/nitro/adapter/mongrel.rb line128:
> > xxxxxxxxxxx
> >  StringIO.new(req.body || "")
> > xxxxxxxxxxx
>
> This code is bad. Once upon a time, Nitro supported http body
> streaming. But I guess it was abandoned against super useless features
> like Elements.
>
> [...]
> > so what's the better way to fix the problem
> > and gain better performance from the optimization?
>
> Mongrel supports the X-Sendfile extension if you installed the `sendfile` gem.
> In your action, set the @headers['X-Sendfile'] = '/path/to/your/file'
> and return no body. Mongrel will intercept that header and join the
> file as a streamed body.
>
> PS : This solution may only works on POSIX
>
> --
> Cheers,
>   zimbatm
>
> http://zimbatm.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
The camping adapter code budled with mongrel(lib/mongrel/camping.rb)
show that X-Sendfile header is handled seperately. I saw Zed Shaw
mentioned in other emails that mongrel use sendfile to serve static
file so it's different thing to X-Sendfile header magic.

From george.moschovitis at gmail.com  Fri Jul  7 12:16:03 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 7 Jul 2006 19:16:03 +0300
Subject: [Nitro] Change in Og.
Message-ID: 

Dear devs,

I am considering the following change in Og. At the moment, the
polymorphic relation of Og works like this:

class Comment
  ...
  belongs_to :parent, Object  # --> polymorphic, we dont define parent class
end

class Article
  has_many :comments
end

automatically generates the Article::Comment class.

I am thinking about making Og generate the ArticleComment class
instead (ie, not a nested class). I am wondering what are your
opinions on this.

regards,
George.


-- 
http://www.gmosx.com
http://www.nitroproject.org

From nusgnaf at gmail.com  Fri Jul  7 14:00:16 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Fri, 7 Jul 2006 18:00:16 +0000
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: 
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
	<55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>
	<3ff63f9b0607070120p2ceaacecw58257b8a2974adb2@mail.gmail.com>
	
Message-ID: <716700c90607071100g29beeedbh64f9afe0d009fe17@mail.gmail.com>

Hi, My working ground is this:
modify lib/nitro/adapter/mongrel.rb line128:
=-=-=-=-=
 StringIO.new(req.body || "")
=-=-=-=-=
to
=======
 StringIO.new(req.body.read || "")
=======
 However, as others point out, this is not performance wise...

From george.moschovitis at gmail.com  Fri Jul  7 14:29:47 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 7 Jul 2006 21:29:47 +0300
Subject: [Nitro] Annotations 2.0
In-Reply-To: <4b6f054f0607061406j1b8b9016u85b4872360bd2238@mail.gmail.com>
References: <4b6f054f0607061406j1b8b9016u85b4872360bd2238@mail.gmail.com>
Message-ID: 

Hello,

> Okay, by some feat of magic I managed a much nicer implementation of
> annotations. The major difference between the old and new:

Argh, changes to annotations make me nervous. If the interface changes
I will have to make many many changes in nitro source code. Please be
careful.

> 1) Only one layer. Ie. No automatuic chaining, X.ann.a.b.c.d.e... etc.
>   X.ann.a.b.c.d.e....

hmm I think it is ok.

> 2) You no longer need the bang method to change an annotatin directly.
> This is achived via a clever RWDelegator.

thats great! (does the bang method still exist though?)


> 3) A "lookup" notation is supported, which I actually find more intuitive:
>
>     X.ann(:foo).bar == X.ann.foo.bar

didn't that exist already?

> 5) Not so sure about this, but as it stands I am storing the
> class/module and name of the annotation in the annotation object. This
> will limit their reusability, but might be worth it. Problem arises
> though if you wanted to do something like:
>
>   X.ann.foo = X.ann.bar

what kind of problem ?

-g.


-- 
http://www.gmosx.com
http://www.nitroproject.org

From transfire at gmail.com  Fri Jul  7 15:29:27 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 7 Jul 2006 15:29:27 -0400
Subject: [Nitro] Annotations 2.0
In-Reply-To: 
References: <4b6f054f0607061406j1b8b9016u85b4872360bd2238@mail.gmail.com>
	
Message-ID: <4b6f054f0607071229l6f606f71l41c33b08491c84d@mail.gmail.com>

On 7/7/06, George Moschovitis  wrote:
> Hello,
>
> > Okay, by some feat of magic I managed a much nicer implementation of
> > annotations. The major difference between the old and new:
>
> Argh, changes to annotations make me nervous. If the interface changes
> I will have to make many many changes in nitro source code. Please be
> careful.

Yes. You are right about that. This will have to be tested well.

> > 1) Only one layer. Ie. No automatuic chaining, X.ann.a.b.c.d.e... etc.
> >   X.ann.a.b.c.d.e....
>
> hmm I think it is ok.

Okay, cool.

> > 2) You no longer need the bang method to change an annotatin directly. This is achived via a clever RWDelegator.
>
> thats great! (does the bang method still exist though?)

Ah, only so so great. It's nice to have no bang but it is less
efficient this way.  And their are some possible issues related to
equality in the RWDelegator.
I think I stick with the bang for now and we can revisit this possibility later.

>
> > 3) A "lookup" notation is supported, which I actually find more intuitive:
> >
> >     X.ann(:foo).bar == X.ann.foo.bar
>
> didn't that exist already?

Yes. But the implementation is now much cleaner.

> > 5) Not so sure about this, but as it stands I am storing the
> > class/module and name of the annotation in the annotation object. This
> > will limit their reusability, but might be worth it. Problem arises
> > though if you wanted to do something like:
> >
> >   X.ann.foo = X.ann.bar
>
> what kind of problem ?

If then name (and owner class) for that matter is stored in the
annotation itself then you can't reuse the same one for different
classes and methods. With the above:

  X.ann.foo = X.ann.bar
  X.ann.foo.annotation_name #=> :bar  #WRONG!

But maybe it's not important to be able to share them? Hmmm, maybe I
can add a catch to dup the annotation on the fly? I'll let you know.

T.

From transfire at gmail.com  Fri Jul  7 15:35:56 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 7 Jul 2006 15:35:56 -0400
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: 
References: <3ff63f9b0606270721s1891de30k70053913522bd1f@mail.gmail.com>
	<4b6f054f0607011223w3210b68i12aaba57660b19ee@mail.gmail.com>
	
	<4b6f054f0607011618t36c6973co53f1c46d9df6359a@mail.gmail.com>
	<4b6f054f0607060728h6b449c91j64eca05e3be04a9@mail.gmail.com>
	
	<4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>
	
	<44AD3D96.50500@gmail.com> 
Message-ID: <4b6f054f0607071235w6fa1f760w33b986989c27ed7a@mail.gmail.com>

George,

Pick the one you use most/like best, Settings or Configuration. I will
use that in Facets. If you want to use both in Nitro then you can just
add to Nitro:

  Configuration = Settings  (or vice-versa)

Cool?

T.

From george.moschovitis at gmail.com  Fri Jul  7 16:28:39 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 7 Jul 2006 23:28:39 +0300
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: <4b6f054f0607071235w6fa1f760w33b986989c27ed7a@mail.gmail.com>
References: <3ff63f9b0606270721s1891de30k70053913522bd1f@mail.gmail.com>
	
	<4b6f054f0607011618t36c6973co53f1c46d9df6359a@mail.gmail.com>
	<4b6f054f0607060728h6b449c91j64eca05e3be04a9@mail.gmail.com>
	
	<4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>
	
	<44AD3D96.50500@gmail.com> 
	<4b6f054f0607071235w6fa1f760w33b986989c27ed7a@mail.gmail.com>
Message-ID: 

Ok, please use Settings and I will add the alias in Nitro.

-g.

On 7/7/06, TRANS  wrote:
> George,
>
> Pick the one you use most/like best, Settings or Configuration. I will
> use that in Facets. If you want to use both in Nitro then you can just
> add to Nitro:
>
>   Configuration = Settings  (or vice-versa)
>
> Cool?
>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Fri Jul  7 16:29:28 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 7 Jul 2006 23:29:28 +0300
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: 
References: <3ff63f9b0606270721s1891de30k70053913522bd1f@mail.gmail.com>
	<4b6f054f0607011618t36c6973co53f1c46d9df6359a@mail.gmail.com>
	<4b6f054f0607060728h6b449c91j64eca05e3be04a9@mail.gmail.com>
	
	<4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>
	
	<44AD3D96.50500@gmail.com> 
	<4b6f054f0607071235w6fa1f760w33b986989c27ed7a@mail.gmail.com>
	
Message-ID: 

Or maybe I will drop the alias altogether and use Settings everywhere.

-g.

On 7/7/06, George Moschovitis  wrote:
> Ok, please use Settings and I will add the alias in Nitro.
>
> -g.
>
> On 7/7/06, TRANS  wrote:
> > George,
> >
> > Pick the one you use most/like best, Settings or Configuration. I will
> > use that in Facets. If you want to use both in Nitro then you can just
> > add to Nitro:
> >
> >   Configuration = Settings  (or vice-versa)
> >
> > Cool?
> >
> > T.
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> --
> http://www.gmosx.com
> http://www.nitroproject.org
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From nusgnaf at gmail.com  Fri Jul  7 17:24:09 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Fri, 7 Jul 2006 21:24:09 +0000
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: 
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
	<55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>
	<3ff63f9b0607070120p2ceaacecw58257b8a2974adb2@mail.gmail.com>
	
Message-ID: <716700c90607071424g77671e38p7d3c3f74ebf45a7f@mail.gmail.com>

Hi, George
I didn't see the reason to write specific mongrel adapter to handle
file upload yet. Rightnow mongrel can save large request as tempfile
then passit to further application framework, In the nitro cgi code,
we also use tempfile to multipart/form-data, so this should not be the
bottleneck, in my opinion the problem is that we convert every request
to StringIO and pass it to nitro, this consume too much memory and
make it impossible to utilize clever hacks such as use tempfile to
store large request.
What's your opinion on this?
On 7/7/06, George Moschovitis  wrote:
> > IMHO the easiest way is to build a new MongrelUploadAdapter that deals
> > with file uploads. Nitro is too complicated some times.
>
> I agree. I will have to write something like this for personal use real soon.
>
> -g.
>
> --
> http://www.gmosx.com
> http://www.nitroproject.org
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>

From james.britt at gmail.com  Fri Jul  7 17:55:54 2006
From: james.britt at gmail.com (James Britt)
Date: Fri, 07 Jul 2006 14:55:54 -0700
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: 
References: <3ff63f9b0606270721s1891de30k70053913522bd1f@mail.gmail.com>	<4b6f054f0607011618t36c6973co53f1c46d9df6359a@mail.gmail.com>	<4b6f054f0607060728h6b449c91j64eca05e3be04a9@mail.gmail.com>		<4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>		<44AD3D96.50500@gmail.com>
		<4b6f054f0607071235w6fa1f760w33b986989c27ed7a@mail.gmail.com>	
	
Message-ID: <44AED86A.9090107@gmail.com>

George Moschovitis wrote:
> Or maybe I will drop the alias altogether and use Settings everywhere.

Yes, please.




Thanks,

James

From surrender_it at yahoo.it  Fri Jul  7 18:38:35 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sat, 08 Jul 2006 00:38:35 +0200
Subject: [Nitro] Publishable#path to allow easy access to
	Publishable.mount_path ?
Message-ID: 

Hi everyone,

I always thought that instances of Nitro::Controller and other 
Publishables had an instance variable @mount_path, possibly because the 
documentation says so.. but the documentation was obviously referring to 
a _class_ instance variable, and thus to access that information in a 
view it is necessary to write

  self.class.mount_path
or
  ThisController.mount_path #losing modularity

Would it be possible to apply this micro patch that adds a "path" method 
to Publishable objects? Simple test attached (both for 
Controller.mount_path and Controller#path, a two-liner :)




-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: publishable_path_tests.patch.gz
Type: application/octet-stream
Size: 3844 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060708/8a5b7956/attachment.obj 

From neokolor at gmx.de  Fri Jul  7 19:36:50 2006
From: neokolor at gmx.de (jfwittmann)
Date: Sat, 08 Jul 2006 01:36:50 +0200
Subject: [Nitro] [PATCH] table-helper and pager-helper
Message-ID: <44AEF012.8040707@gmx.de>

Hello,

here are three patches.

- the first two output-fix patches fix problems with the table and pager 
helper if they get no results.
- the third patch add a new option to the pager helper to set the 
navigation titles.

additional, I add a simple test app to show the changes.

cu,

Felix




 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: table_helper_test.zip
Type: application/x-zip-compressed
Size: 2422 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060708/6cca83af/attachment-0001.bin 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: table_helper_ouput_fix
Url: http://rubyforge.org/pipermail/nitro-general/attachments/20060708/6cca83af/attachment-0003.pl 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: pager_helper_add_option
Url: http://rubyforge.org/pipermail/nitro-general/attachments/20060708/6cca83af/attachment-0004.pl 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: pager_helper_ouput_fix
Url: http://rubyforge.org/pipermail/nitro-general/attachments/20060708/6cca83af/attachment-0005.pl 

From zedshaw at zedshaw.com  Fri Jul  7 21:50:37 2006
From: zedshaw at zedshaw.com (Zed Shaw)
Date: Fri, 07 Jul 2006 21:50:37 -0400
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: 
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
	<55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>
	<3ff63f9b0607070120p2ceaacecw58257b8a2974adb2@mail.gmail.com>
	
Message-ID: <1152323437.6667.13.camel@localhost.localdomain>

On Fri, 2006-07-07 at 15:39 +0300, George Moschovitis wrote:
> > IMHO the easiest way is to build a new MongrelUploadAdapter that deals
> > with file uploads. Nitro is too complicated some times.
> 
> I agree. I will have to write something like this for personal use real soon.

A couple of quick points:

* request.body is always an IO object.  Further wrapping it in another
StringIO object is kind of pointless.
* It's also rewound to the beginning as if it was a recently opened
file.
* There's support for Notifiers in 0.3.13.3 which has even better
support in the 0.4.  The rails guys whipped up a quick
mongrel_progress_upload plugin that's very framework agnostic so you
could just use that to base the nitro one on (or just use that and write
the nitro part).  This upload progress uses notifiers.
* File upload/send is very efficient in Mongrel now.  It will read
chunks to a tempfile when it gets too large and uses sendfile or other
tricks as well as conditional compilation.  Nitro should be doing less
of this work and simply read requests, process bodies, write requests
and go.  There's even a sendfile API (check out the DirHandler).

Feel free to hit me up on IRC for help with these.  The notifier stuff
changes pretty dramatically in 0.4 but you can grab a pre-release of 0.4
at:

  gem install mongrel --source=http://mongrel.rubyforge.org/releases/


-- 
Zed A. Shaw
http://www.zedshaw.com/
http://mongrel.rubyforge.org/
http://www.railsmachine.com/ -- Need Mongrel support?


From zedshaw at zedshaw.com  Fri Jul  7 21:53:10 2006
From: zedshaw at zedshaw.com (Zed Shaw)
Date: Fri, 07 Jul 2006 21:53:10 -0400
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: <716700c90607071100g29beeedbh64f9afe0d009fe17@mail.gmail.com>
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
	<55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>
	<3ff63f9b0607070120p2ceaacecw58257b8a2974adb2@mail.gmail.com>
	
	<716700c90607071100g29beeedbh64f9afe0d009fe17@mail.gmail.com>
Message-ID: <1152323590.6667.17.camel@localhost.localdomain>

On Fri, 2006-07-07 at 18:00 +0000, Fang Sun wrote:
> Hi, My working ground is this:
> modify lib/nitro/adapter/mongrel.rb line128:
> =-=-=-=-=
>  StringIO.new(req.body || "")
> =-=-=-=-=
> to
> =======
>  StringIO.new(req.body.read || "")
> =======
>  However, as others point out, this is not performance wise...

Also, req.body is already guaranteed to be a rewound IO object.  I think
you guys are probably trying to be too clever :-)

I mean, what your new code above basically does is say:

 "take the request.body and read it into a string and put it back into
an IO object"

Additionally, the || "" is pointless since you obviously just
called .body on req and then .read which doesn't return nil objects.
The || "" never gets called (unless you're cracking open IO to make it
do evil).

Basically, the code could just be that you use req.body.


-- 
Zed A. Shaw
http://www.zedshaw.com/
http://mongrel.rubyforge.org/
http://www.railsmachine.com/ -- Need Mongrel support?


From william.full.moon at gmail.com  Fri Jul  7 23:16:26 2006
From: william.full.moon at gmail.com (* William)
Date: Sat, 8 Jul 2006 13:16:26 +1000
Subject: [Nitro] Change in Og.
In-Reply-To: 
Message-ID: <006001c6a23c$ec5d1ae0$85af07ca@ghostgum>

Hi George

Reading your example, it seems that Comment.parent can be any object.  I'd
definitely want to re-think that myself.  Also, I prefer the KISS principle
(keep it simple software).  I'd vote to keep Og simple.  And add a
OO-persistence on-top of Og

An effort to add more sophisticated OO features to Og is going to remove Og
essential simplicity, clarity, and effectiveness.  Better to wrap Og with
fancy stuff and avoid being 'too clever'.  Returning to the specific example
...


I see that 'parent' could be polymorphic, as can *any* thing that is base
type 'Object'.  To be useful in an OO modelling sense you need to be more
restrictive -- Or in truth, more focused.

This is an example of containment/ownership.  For example, using the
ubiquitous drawing Shape example:

   class Comment
      ...
     belongs_to :item, Shape    # --> polymorphic, item's class 
                                #     must descend from Shape
   end  #Comment

   class DesignNote  < Comment
      : :
   end  #DesignNote

   class Shape
     has_many :comments
   end  #Shape


With a generic object type as 'parent' in the first example, any program bug
could store a 'Canoe' as a 'Comment'.  (Where: 'Canoe' is-not-a 'Shape').
Where the desirable behaviour is that 'DesignNote' is-a 'Comment' and thus
valid.

I accept that if 'Canoe' passes a 'Shape Duck-test', that is acceptable --
The semantics for "has_many" *must* clearly identify the 'category' (type)
of objects that "belong to" Shape.

When that is not done the honest expression of what is happening is to
declare comments as so.

   class Shape
     has_many :comments,  (void *)
   end  #Shape

That alternative is not so much OO-persistence or polymorphism as, suicide
so far.  However I'm not dead yet so I live in hope. ;-)


In making this distinction we can be clear about the following questions?

   * Does has_many auto-mantically identify valid parent classes 
     for 'comments'.

   * In the containment example the polymorphism is resident in the
     owner, 'Shape' class in the second example and the 'Article' class
     in the first example.  (See next point)

   * In the database sense there is NO need for 'Comment' to restrict
     its content to things that can be owned-by 'Shape'.  The desirable
     morphology for 'Comment' is any 'Comment' class and its children 
     can be validly stored as 'Comment'-table elements.

     -- So INSERT 'DesignNote' IN 'Comment' ;  # is OK
           INSERT 'Canoe'      IN 'Comment';   # Fails

     -- Because a 'Canoe' is-not-a 'Comment' and 'DesignNote' is-a 'Comment'



The second part of the post sounds good if I have it right.

>> I am thinking about making Og generate the ArticleComment class 
>> instead (i.e., not a nested class). I am wondering what are your 
>> opinions on this.

This would mean connecting Article to Comment by a join/foreign key table?
Is that so?  That definitely sound preferable to a nested anything in a
database.  If my programming philosophy is KISS, my database philosophy is
to KIFF (keep it files flat, as possible) *lol*

This sounds really interesting.

Aloha,
       Will.

-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of George Moschovitis
Sent: Saturday, 8 July 2006 02:16
To: General discussion about Nitro
Subject: [Nitro] Change in Og.
Importance: Low

Dear devs,

I am considering the following change in Og. At the moment, the polymorphic
relation of Og works like this:

class Comment
  ...
  belongs_to :parent, Object  # --> polymorphic, we dont define parent class
end

class Article
  has_many :comments
end

automatically generates the Article::Comment class.

I am thinking about making Og generate the ArticleComment class instead (ie,
not a nested class). I am wondering what are your opinions on this.

regards,
George.


--
http://www.gmosx.com
http://www.nitroproject.org
_______________________________________________
Nitro-general mailing list
Nitro-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/383 - Release Date: 07-Jul-2006
 


From george.moschovitis at gmail.com  Sat Jul  8 02:54:52 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 8 Jul 2006 09:54:52 +0300
Subject: [Nitro] [PATCH][nitroproject] Some small fixes to Og and Glue
In-Reply-To: <44AED86A.9090107@gmail.com>
References: <3ff63f9b0606270721s1891de30k70053913522bd1f@mail.gmail.com>
	
	<4b6f054f0607060833r25a83c37k8864fcac39b1a13c@mail.gmail.com>
	
	<44AD3D96.50500@gmail.com> 
	<4b6f054f0607071235w6fa1f760w33b986989c27ed7a@mail.gmail.com>
	
	
	<44AED86A.9090107@gmail.com>
Message-ID: 

> > Or maybe I will drop the alias altogether and use Settings everywhere.
>
> Yes, please.


Ok.

-g.


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sat Jul  8 02:56:46 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 8 Jul 2006 09:56:46 +0300
Subject: [Nitro] [PATCH] table-helper and pager-helper
In-Reply-To: <44AEF012.8040707@gmx.de>
References: <44AEF012.8040707@gmx.de>
Message-ID: 

Please,  resend the patches in tgz or zip format because Gmail messes
up with the plain text files. Thanks!

-g.

On 7/8/06, jfwittmann  wrote:
> Hello,
>
> here are three patches.
>
> - the first two output-fix patches fix problems with the table and pager
> helper if they get no results.
> - the third patch add a new option to the pager helper to set the
> navigation titles.
>
> additional, I add a simple test app to show the changes.
>
> cu,
>
> Felix
>
>
>
>
>
>
>
>
>
>
> New patches:
>
> [helper/table.rb: fix problem with table output
> neokolor at gmx.de**20060707232548] {
> hunk ./nitro/lib/nitro/helper/table.rb 237
> -    options[:values][0][0].respond_to?(:to_ary)
> +    options[:values][0].respond_to?(:to_ary) && options[:values][0][0].respond_to?(:to_ary)
> }
>
> Context:
>
> [dispatcher-add_rule-patch
> bryan.a.soto at gmail.com**20060630000135
>  Duplicate of George's add_route to add_rule patch as his repo was unavailable.
> ]
> [og-tc-sti-find
> bryan.a.soto at gmail.com**20060629233146
>  Adds a test case for manveru's sti fix.
> ]
> [og-entity-from-string
> bryan.a.soto at gmail.com**20060629232926
>  Modifies the definition of Og::Entity::entity_from_string slightly. It assumed only one manager would ever exist. Not a valid assumption in the test suite.
> ]
> [og-thread-safe-options
> bryan.a.soto at gmail.com**20060629221018
>  Adds the ability to specify :thread_safe as an option to Og.setup with the default begin true. Without this patch, thread_safe was true and it couldn't be modified without modifying the Og code.
> ]
> [og-aggregate-fix
> bryan.a.soto at gmail.com**20060629220858
>  Accomodates the fact that the Postgres adapter doesn't return an Array like the other stores by modifying the aggregation code.
> ]
> [fix for schema_inheritance queries
> manveru at weez-int.com**20060627085447]
> [localization-support for table-helper
> m.fellinger at gmail.com**20060420100859]
> [update-facets-1.4.2
> bryan.a.soto at gmail.com**20060622000622
>  Updates to Facets-1.4.2. Uses facets/core/module/class_extension instead of classinherit. No other changes needed to pass tests.
> ]
> [fix problem with :plural_name in admin controller
> neokolor at gmx.de**20060618220838]
> [hack-enh-fcgi-rewind
> bryan.a.soto at gmail.com**20060609220620
>  A small hack to ensure that you can access the raw_body of a request, by converting the incoming stream to a StringIO if it doesn't respond to :rewind.
> ]
> [nitro-sendfile-support
> bryan.a.soto at gmail.com**20060608201940
>  Adds sendfile supported. Submitted by Jan A. Lambert . Tested on IE 6.0 and FireFox 1.5.
> ]
> [nitro-dispatcher-router-strip_path
> bryan.a.soto at gmail.com**20060608020204
>  Makes the dispatcher honour the Router.strip_path setting. This, basically, makes Nitro ignore a portion of the url that could represent a directory name.
>
>  Basically, without the setting:
>  /my_nitro_apps_directory/nitro-app => looks for a controller mapped to / or /my_nitro_apps_directory.
>
>  With Router.strip_path = '/my_nitro_apps_directory'
>  /my_nitro_apps_directory/nitro-app => looks for a controller mapped to / or /nitro-app.
>
>  Many thanks to Vagabond for the patch.
>
> ]
> [bugfix-mongrel-adapter-request-uri
> bryan.a.soto at gmail.com**20060605211152
>  Removes REQUEST_URI header manipulation and accomodates an upcoming change in Mongrel 0.3.13. An ugly class check... Lovely.
> ]
> [fixup-guilles-patch
> bryan.a.soto at gmail.com**20060518004717
>  Brings Guille's patch up to date by resolving a conflict.
> ]
> [og/lib/glue/taggable.rb refactor + support for passing :condition parameter
> Guillaume Pierronnet **20060512094822
>  SqlStore#prepare_statement moved to public scope
>  test case on taggable runs fine
> ]
> [test-fix-og-tc_ez
> bryan.a.soto at gmail.com**20060518001436
>  Fixes up the ez testcase Manveru submitted to make it run and pass.
> ]
> [ez-fix for trueclass and testcase for it
> m.fellinger at gmail.com**20060513111040]
> [error-page-fix for source-preview
> m.fellinger at gmail.com**20060513110945]
> [to_rexml and to_xml
> fabian at fabian-buch.de**20060517200153
>
>  to_rexml returns a REXML::Element object
>  to_xml returns an XML-String
>
> ]
> [introduction of OgObject.to_xml method
> fabian at fabian-buch.de**20060517143323
>
>  to_xml returns a REXML object of an Og object
>
>  example usage:
>    User[1].to_xml
>
>  returns a REXML object, so use to_s to get it as String containing XML
>
> ]
> [test-fix-nitro-tc_render
> bryan.a.soto at gmail.com**20060517223259
>  Makes the TestController subclass Nitro::Controller so it get's the #encode_url method.
> ]
> [adds-some-blank-lines-to-og-tc_setup
> bryan.a.soto at gmail.com**20060517222452]
> [gabrielle-renzis-linefeeds
> bryan.a.soto at gmail.com**20060517221725
>  Switches line endings from Dos to Unix.
> ]
> [crash-early Og.setup
> surrender_it at yahoo.it**20060505123051
>  This patch removed the Exception handling routine from Og.setup.
>  With the current code if an exception is raised during the execution of
>  the method there is no crash, but the crash will happen every time someone references Og.manager, cause that variable won't be initialized .
>  This patch makes Og.setup crash as soon as possible, thus allowing faster diagnostics.
>
>  The patch also adds a simple test case for Og.setup, which seem not tested anywhere, but the test case may need some love  for proper setup (it relies on glycerin as of now, cause loading CONFIG.rb seem to raise a strange exception related to mixed use of facets 1.3 and 1.0 on my box)
>
> ]
> [better test logging
> surrender_it at yahoo.it**20060505010206
>  As of now tests for Og swallow every error ingormation while not in $DBG mode,
>  this one-line patch allows visualization of logged informations at WARN level and more, which has little impact on the visualization of the tests and still allows quick diagnostics while staying out of $DBG
> ]
> [george-glue-tag-fixup
> bryan.a.soto at gmail.com**20060516203704]
> [Minor.
> George Moschovitis **20060516150705]
> [Moved examples, spark, flare to a separate repository.
> George Moschovitis **20060516073246]
> [Removed wee helper and example. Not really useful in Nitro, better use wee standalone + Og if you like Wee components (or wait for a Nitro solution).
> George Moschovitis **20060516072823]
> [Removed controller_name from Render, user controller.mount_path instead.
> George Moschovitis **20060516072350]
> [Fixed: dont overwrite already defined methods in the new scaffold code.
> George Moschovitis **20060516071615]
> [Improved R (encode_url) handles strings and entities. Redirect implicitly uses R. Form builder implicitly uses R for form actions.
> George Moschovitis **20060516071508]
> [Fixed name_to_jsfile (absolute url).
> George Moschovitis **20060515065439]
> [Reenabled format in Logger.
> George Moschovitis **20060513042251]
> [Added some :nodoc: all to cleanup generated RDocs.
> George Moschovitis **20060513034929]
> [Bumbed version number in many files.
> George Moschovitis **20060512154849]
> [Wrote a detailed parts howto in RDoc form in parts.rb as requested by Jonas. To be improved after community feedback.
> George Moschovitis **20060512153754]
> [Initial code for the new scaffolding system, *very* under construction, ignore for the moment.
> George Moschovitis **20060511101456]
> [Added a model macro to controllers that links related classes. At the moment it is used in the sweeper do delete cached pages from the linked controller base directory.
> George Moschovitis **20060510102836]
> [add-ostruct-require
> bryan.a.soto at gmail.com**20060512060511
>  Adds missing ostruct require to testcase.rb.
> ]
> [more-mongrel-adapter-fixes
> bryan.a.soto at gmail.com**20060510222051
>  Fixes up static file serving.
> ]
> [nitro-proto-page
> bryan.a.soto at gmail.com**20060510060546
>  Updates the intro page with a link to the rubyforge page for examples to be downloaded as suggested by James Britt.
> ]
> [mongrel-fixes
> bryan.a.soto at gmail.com**20060510060439
>  Gives the Mongrel adapter some need updating.
> ]
> [test-suite-fixup
> bryan.a.soto at gmail.com**20060509063643
>  Fixes the full test suite so that .\script\test.rb completes.
> ]
> [Fixed Session.current and Controller.current
> George Moschovitis **20060508075645]
> [Store current context in a thread local variable, accesible through Context.current
> George Moschovitis **20060507160715]
> [Removed glue/attribute, use Facet's cattr instead. Updated nitro/og source to make compatible.
> George Moschovitis **20060507085703]
> [Added script adapter file.
> George Moschovitis **20060507062034]
> [Removed flexob and references to it, use Facet's OpenObject instead.
> George Moschovitis **20060507061555]
> [Renamed to ScriptAdapter, keep ConsoleAdapter alias.
> George Moschovitis **20060506103921]
> [Yeah, the ConsoleAdapter kinda works now ;-)
> George Moschovitis **20060506102010]
> [Introduced new nitro command/runner that will replace the current spaghetti code. Based on facetes Console::Command. Also introduced the ConsoleAdapter.
> George Moschovitis **20060506095503]
> [Removed accumulate (in facets).
> George Moschovitis **20060506080052]
> [Some small updates.
> George Moschovitis **20060506075526]
> [TAG 0.30.0
> George Moschovitis **20060506075053]
> Patch bundle hash:
> ab7c5788477df6c255f17446a093347e8e1b2223
>
>
>
> New patches:
>
> [helper/pager.rb: add option to set nav link titles
> neokolor at gmx.de**20060707231808] {
> hunk ./nitro/lib/nitro/helper/pager.rb 27
> +
> hunk ./nitro/lib/nitro/helper/pager.rb 44
> -  def initialize(request, per_page, total_count, key = Pager.key)
> +  def initialize(request, per_page, total_count, key = Pager.key, link_titles = Pager.link_titles)
> hunk ./nitro/lib/nitro/helper/pager.rb 47
> -    @request, @key = request, key
> +    @request, @key, @link_titles = request, key, link_titles
> hunk ./nitro/lib/nitro/helper/pager.rb 194
> -        First
> -        Previous
> +        #{CGI.escapeHTML(@link_titles[:first])}
> +        #{CGI.escapeHTML(@link_titles[:previous])}
> hunk ./nitro/lib/nitro/helper/pager.rb 201
> -        Last
> -        Next
> +        #{CGI.escapeHTML(@link_titles[:last])}
> +        #{CGI.escapeHTML(@link_titles[:next])}
> hunk ./nitro/lib/nitro/helper/pager.rb 274
> -
> +    pager_link_titles = Pager.link_titles.merge(options.delete(:pager_link_titles) || Pager.link_titles)
> hunk ./nitro/lib/nitro/helper/pager.rb 278
> -        pager = Pager.new(request, per_page, items.size, pager_key)
> +        pager = Pager.new(request, per_page, items.size, pager_key, pager_link_titles)
> hunk ./nitro/lib/nitro/helper/pager.rb 284
> -        pager = Pager.new(request, per_page, collection.count, pager_key)
> +        pager = Pager.new(request, per_page, collection.count, pager_key, pager_link_titles)
> }
>
> Context:
>
> [dispatcher-add_rule-patch
> bryan.a.soto at gmail.com**20060630000135
>  Duplicate of George's add_route to add_rule patch as his repo was unavailable.
> ]
> [og-tc-sti-find
> bryan.a.soto at gmail.com**20060629233146
>  Adds a test case for manveru's sti fix.
> ]
> [og-entity-from-string
> bryan.a.soto at gmail.com**20060629232926
>  Modifies the definition of Og::Entity::entity_from_string slightly. It assumed only one manager would ever exist. Not a valid assumption in the test suite.
> ]
> [og-thread-safe-options
> bryan.a.soto at gmail.com**20060629221018
>  Adds the ability to specify :thread_safe as an option to Og.setup with the default begin true. Without this patch, thread_safe was true and it couldn't be modified without modifying the Og code.
> ]
> [og-aggregate-fix
> bryan.a.soto at gmail.com**20060629220858
>  Accomodates the fact that the Postgres adapter doesn't return an Array like the other stores by modifying the aggregation code.
> ]
> [fix for schema_inheritance queries
> manveru at weez-int.com**20060627085447]
> [localization-support for table-helper
> m.fellinger at gmail.com**20060420100859]
> [update-facets-1.4.2
> bryan.a.soto at gmail.com**20060622000622
>  Updates to Facets-1.4.2. Uses facets/core/module/class_extension instead of classinherit. No other changes needed to pass tests.
> ]
> [fix problem with :plural_name in admin controller
> neokolor at gmx.de**20060618220838]
> [hack-enh-fcgi-rewind
> bryan.a.soto at gmail.com**20060609220620
>  A small hack to ensure that you can access the raw_body of a request, by converting the incoming stream to a StringIO if it doesn't respond to :rewind.
> ]
> [nitro-sendfile-support
> bryan.a.soto at gmail.com**20060608201940
>  Adds sendfile supported. Submitted by Jan A. Lambert . Tested on IE 6.0 and FireFox 1.5.
> ]
> [nitro-dispatcher-router-strip_path
> bryan.a.soto at gmail.com**20060608020204
>  Makes the dispatcher honour the Router.strip_path setting. This, basically, makes Nitro ignore a portion of the url that could represent a directory name.
>
>  Basically, without the setting:
>  /my_nitro_apps_directory/nitro-app => looks for a controller mapped to / or /my_nitro_apps_directory.
>
>  With Router.strip_path = '/my_nitro_apps_directory'
>  /my_nitro_apps_directory/nitro-app => looks for a controller mapped to / or /nitro-app.
>
>  Many thanks to Vagabond for the patch.
>
> ]
> [bugfix-mongrel-adapter-request-uri
> bryan.a.soto at gmail.com**20060605211152
>  Removes REQUEST_URI header manipulation and accomodates an upcoming change in Mongrel 0.3.13. An ugly class check... Lovely.
> ]
> [fixup-guilles-patch
> bryan.a.soto at gmail.com**20060518004717
>  Brings Guille's patch up to date by resolving a conflict.
> ]
> [og/lib/glue/taggable.rb refactor + support for passing :condition parameter
> Guillaume Pierronnet **20060512094822
>  SqlStore#prepare_statement moved to public scope
>  test case on taggable runs fine
> ]
> [test-fix-og-tc_ez
> bryan.a.soto at gmail.com**20060518001436
>  Fixes up the ez testcase Manveru submitted to make it run and pass.
> ]
> [ez-fix for trueclass and testcase for it
> m.fellinger at gmail.com**20060513111040]
> [error-page-fix for source-preview
> m.fellinger at gmail.com**20060513110945]
> [to_rexml and to_xml
> fabian at fabian-buch.de**20060517200153
>
>  to_rexml returns a REXML::Element object
>  to_xml returns an XML-String
>
> ]
> [introduction of OgObject.to_xml method
> fabian at fabian-buch.de**20060517143323
>
>  to_xml returns a REXML object of an Og object
>
>  example usage:
>    User[1].to_xml
>
>  returns a REXML object, so use to_s to get it as String containing XML
>
> ]
> [test-fix-nitro-tc_render
> bryan.a.soto at gmail.com**20060517223259
>  Makes the TestController subclass Nitro::Controller so it get's the #encode_url method.
> ]
> [adds-some-blank-lines-to-og-tc_setup
> bryan.a.soto at gmail.com**20060517222452]
> [gabrielle-renzis-linefeeds
> bryan.a.soto at gmail.com**20060517221725
>  Switches line endings from Dos to Unix.
> ]
> [crash-early Og.setup
> surrender_it at yahoo.it**20060505123051
>  This patch removed the Exception handling routine from Og.setup.
>  With the current code if an exception is raised during the execution of
>  the method there is no crash, but the crash will happen every time someone references Og.manager, cause that variable won't be initialized .
>  This patch makes Og.setup crash as soon as possible, thus allowing faster diagnostics.
>
>  The patch also adds a simple test case for Og.setup, which seem not tested anywhere, but the test case may need some love  for proper setup (it relies on glycerin as of now, cause loading CONFIG.rb seem to raise a strange exception related to mixed use of facets 1.3 and 1.0 on my box)
>
> ]
> [better test logging
> surrender_it at yahoo.it**20060505010206
>  As of now tests for Og swallow every error ingormation while not in $DBG mode,
>  this one-line patch allows visualization of logged informations at WARN level and more, which has little impact on the visualization of the tests and still allows quick diagnostics while staying out of $DBG
> ]
> [george-glue-tag-fixup
> bryan.a.soto at gmail.com**20060516203704]
> [Minor.
> George Moschovitis **20060516150705]
> [Moved examples, spark, flare to a separate repository.
> George Moschovitis **20060516073246]
> [Removed wee helper and example. Not really useful in Nitro, better use wee standalone + Og if you like Wee components (or wait for a Nitro solution).
> George Moschovitis **20060516072823]
> [Removed controller_name from Render, user controller.mount_path instead.
> George Moschovitis **20060516072350]
> [Fixed: dont overwrite already defined methods in the new scaffold code.
> George Moschovitis **20060516071615]
> [Improved R (encode_url) handles strings and entities. Redirect implicitly uses R. Form builder implicitly uses R for form actions.
> George Moschovitis **20060516071508]
> [Fixed name_to_jsfile (absolute url).
> George Moschovitis **20060515065439]
> [Reenabled format in Logger.
> George Moschovitis **20060513042251]
> [Added some :nodoc: all to cleanup generated RDocs.
> George Moschovitis **20060513034929]
> [Bumbed version number in many files.
> George Moschovitis **20060512154849]
> [Wrote a detailed parts howto in RDoc form in parts.rb as requested by Jonas. To be improved after community feedback.
> George Moschovitis **20060512153754]
> [Initial code for the new scaffolding system, *very* under construction, ignore for the moment.
> George Moschovitis **20060511101456]
> [Added a model macro to controllers that links related classes. At the moment it is used in the sweeper do delete cached pages from the linked controller base directory.
> George Moschovitis **20060510102836]
> [add-ostruct-require
> bryan.a.soto at gmail.com**20060512060511
>  Adds missing ostruct require to testcase.rb.
> ]
> [more-mongrel-adapter-fixes
> bryan.a.soto at gmail.com**20060510222051
>  Fixes up static file serving.
> ]
> [nitro-proto-page
> bryan.a.soto at gmail.com**20060510060546
>  Updates the intro page with a link to the rubyforge page for examples to be downloaded as suggested by James Britt.
> ]
> [mongrel-fixes
> bryan.a.soto at gmail.com**20060510060439
>  Gives the Mongrel adapter some need updating.
> ]
> [test-suite-fixup
> bryan.a.soto at gmail.com**20060509063643
>  Fixes the full test suite so that .\script\test.rb completes.
> ]
> [Fixed Session.current and Controller.current
> George Moschovitis **20060508075645]
> [Store current context in a thread local variable, accesible through Context.current
> George Moschovitis **20060507160715]
> [Removed glue/attribute, use Facet's cattr instead. Updated nitro/og source to make compatible.
> George Moschovitis **20060507085703]
> [Added script adapter file.
> George Moschovitis **20060507062034]
> [Removed flexob and references to it, use Facet's OpenObject instead.
> George Moschovitis **20060507061555]
> [Renamed to ScriptAdapter, keep ConsoleAdapter alias.
> George Moschovitis **20060506103921]
> [Yeah, the ConsoleAdapter kinda works now ;-)
> George Moschovitis **20060506102010]
> [Introduced new nitro command/runner that will replace the current spaghetti code. Based on facetes Console::Command. Also introduced the ConsoleAdapter.
> George Moschovitis **20060506095503]
> [Removed accumulate (in facets).
> George Moschovitis **20060506080052]
> [Some small updates.
> George Moschovitis **20060506075526]
> [TAG 0.30.0
> George Moschovitis **20060506075053]
> Patch bundle hash:
> 06ef741847a82d5f512c5d3c3d28c08fa9067f4f
>
>
>
> New patches:
>
> [helper/pager.rb: fix problems with nav output if we have no results
> neokolor at gmx.de**20060707230829] {
> hunk ./nitro/lib/nitro/helper/pager.rb 58
> +  def has_pages?
> +    return @page_count > 1
> +  end
> +
> hunk ./nitro/lib/nitro/helper/pager.rb 195
> +    return nav unless has_pages?
> }
>
> Context:
>
> [dispatcher-add_rule-patch
> bryan.a.soto at gmail.com**20060630000135
>  Duplicate of George's add_route to add_rule patch as his repo was unavailable.
> ]
> [og-tc-sti-find
> bryan.a.soto at gmail.com**20060629233146
>  Adds a test case for manveru's sti fix.
> ]
> [og-entity-from-string
> bryan.a.soto at gmail.com**20060629232926
>  Modifies the definition of Og::Entity::entity_from_string slightly. It assumed only one manager would ever exist. Not a valid assumption in the test suite.
> ]
> [og-thread-safe-options
> bryan.a.soto at gmail.com**20060629221018
>  Adds the ability to specify :thread_safe as an option to Og.setup with the default begin true. Without this patch, thread_safe was true and it couldn't be modified without modifying the Og code.
> ]
> [og-aggregate-fix
> bryan.a.soto at gmail.com**20060629220858
>  Accomodates the fact that the Postgres adapter doesn't return an Array like the other stores by modifying the aggregation code.
> ]
> [fix for schema_inheritance queries
> manveru at weez-int.com**20060627085447]
> [localization-support for table-helper
> m.fellinger at gmail.com**20060420100859]
> [update-facets-1.4.2
> bryan.a.soto at gmail.com**20060622000622
>  Updates to Facets-1.4.2. Uses facets/core/module/class_extension instead of classinherit. No other changes needed to pass tests.
> ]
> [fix problem with :plural_name in admin controller
> neokolor at gmx.de**20060618220838]
> [hack-enh-fcgi-rewind
> bryan.a.soto at gmail.com**20060609220620
>  A small hack to ensure that you can access the raw_body of a request, by converting the incoming stream to a StringIO if it doesn't respond to :rewind.
> ]
> [nitro-sendfile-support
> bryan.a.soto at gmail.com**20060608201940
>  Adds sendfile supported. Submitted by Jan A. Lambert . Tested on IE 6.0 and FireFox 1.5.
> ]
> [nitro-dispatcher-router-strip_path
> bryan.a.soto at gmail.com**20060608020204
>  Makes the dispatcher honour the Router.strip_path setting. This, basically, makes Nitro ignore a portion of the url that could represent a directory name.
>
>  Basically, without the setting:
>  /my_nitro_apps_directory/nitro-app => looks for a controller mapped to / or /my_nitro_apps_directory.
>
>  With Router.strip_path = '/my_nitro_apps_directory'
>  /my_nitro_apps_directory/nitro-app => looks for a controller mapped to / or /nitro-app.
>
>  Many thanks to Vagabond for the patch.
>
> ]
> [bugfix-mongrel-adapter-request-uri
> bryan.a.soto at gmail.com**20060605211152
>  Removes REQUEST_URI header manipulation and accomodates an upcoming change in Mongrel 0.3.13. An ugly class check... Lovely.
> ]
> [fixup-guilles-patch
> bryan.a.soto at gmail.com**20060518004717
>  Brings Guille's patch up to date by resolving a conflict.
> ]
> [og/lib/glue/taggable.rb refactor + support for passing :condition parameter
> Guillaume Pierronnet **20060512094822
>  SqlStore#prepare_statement moved to public scope
>  test case on taggable runs fine
> ]
> [test-fix-og-tc_ez
> bryan.a.soto at gmail.com**20060518001436
>  Fixes up the ez testcase Manveru submitted to make it run and pass.
> ]
> [ez-fix for trueclass and testcase for it
> m.fellinger at gmail.com**20060513111040]
> [error-page-fix for source-preview
> m.fellinger at gmail.com**20060513110945]
> [to_rexml and to_xml
> fabian at fabian-buch.de**20060517200153
>
>  to_rexml returns a REXML::Element object
>  to_xml returns an XML-String
>
> ]
> [introduction of OgObject.to_xml method
> fabian at fabian-buch.de**20060517143323
>
>  to_xml returns a REXML object of an Og object
>
>  example usage:
>    User[1].to_xml
>
>  returns a REXML object, so use to_s to get it as String containing XML
>
> ]
> [test-fix-nitro-tc_render
> bryan.a.soto at gmail.com**20060517223259
>  Makes the TestController subclass Nitro::Controller so it get's the #encode_url method.
> ]
> [adds-some-blank-lines-to-og-tc_setup
> bryan.a.soto at gmail.com**20060517222452]
> [gabrielle-renzis-linefeeds
> bryan.a.soto at gmail.com**20060517221725
>  Switches line endings from Dos to Unix.
> ]
> [crash-early Og.setup
> surrender_it at yahoo.it**20060505123051
>  This patch removed the Exception handling routine from Og.setup.
>  With the current code if an exception is raised during the execution of
>  the method there is no crash, but the crash will happen every time someone references Og.manager, cause that variable won't be initialized .
>  This patch makes Og.setup crash as soon as possible, thus allowing faster diagnostics.
>
>  The patch also adds a simple test case for Og.setup, which seem not tested anywhere, but the test case may need some love  for proper setup (it relies on glycerin as of now, cause loading CONFIG.rb seem to raise a strange exception related to mixed use of facets 1.3 and 1.0 on my box)
>
> ]
> [better test logging
> surrender_it at yahoo.it**20060505010206
>  As of now tests for Og swallow every error ingormation while not in $DBG mode,
>  this one-line patch allows visualization of logged informations at WARN level and more, which has little impact on the visualization of the tests and still allows quick diagnostics while staying out of $DBG
> ]
> [george-glue-tag-fixup
> bryan.a.soto at gmail.com**20060516203704]
> [Minor.
> George Moschovitis **20060516150705]
> [Moved examples, spark, flare to a separate repository.
> George Moschovitis **20060516073246]
> [Removed wee helper and example. Not really useful in Nitro, better use wee standalone + Og if you like Wee components (or wait for a Nitro solution).
> George Moschovitis **20060516072823]
> [Removed controller_name from Render, user controller.mount_path instead.
> George Moschovitis **20060516072350]
> [Fixed: dont overwrite already defined methods in the new scaffold code.
> George Moschovitis **20060516071615]
> [Improved R (encode_url) handles strings and entities. Redirect implicitly uses R. Form builder implicitly uses R for form actions.
> George Moschovitis **20060516071508]
> [Fixed name_to_jsfile (absolute url).
> George Moschovitis **20060515065439]
> [Reenabled format in Logger.
> George Moschovitis **20060513042251]
> [Added some :nodoc: all to cleanup generated RDocs.
> George Moschovitis **20060513034929]
> [Bumbed version number in many files.
> George Moschovitis **20060512154849]
> [Wrote a detailed parts howto in RDoc form in parts.rb as requested by Jonas. To be improved after community feedback.
> George Moschovitis **20060512153754]
> [Initial code for the new scaffolding system, *very* under construction, ignore for the moment.
> George Moschovitis **20060511101456]
> [Added a model macro to controllers that links related classes. At the moment it is used in the sweeper do delete cached pages from the linked controller base directory.
> George Moschovitis **20060510102836]
> [add-ostruct-require
> bryan.a.soto at gmail.com**20060512060511
>  Adds missing ostruct require to testcase.rb.
> ]
> [more-mongrel-adapter-fixes
> bryan.a.soto at gmail.com**20060510222051
>  Fixes up static file serving.
> ]
> [nitro-proto-page
> bryan.a.soto at gmail.com**20060510060546
>  Updates the intro page with a link to the rubyforge page for examples to be downloaded as suggested by James Britt.
> ]
> [mongrel-fixes
> bryan.a.soto at gmail.com**20060510060439
>  Gives the Mongrel adapter some need updating.
> ]
> [test-suite-fixup
> bryan.a.soto at gmail.com**20060509063643
>  Fixes the full test suite so that .\script\test.rb completes.
> ]
> [Fixed Session.current and Controller.current
> George Moschovitis **20060508075645]
> [Store current context in a thread local variable, accesible through Context.current
> George Moschovitis **20060507160715]
> [Removed glue/attribute, use Facet's cattr instead. Updated nitro/og source to make compatible.
> George Moschovitis **20060507085703]
> [Added script adapter file.
> George Moschovitis **20060507062034]
> [Removed flexob and references to it, use Facet's OpenObject instead.
> George Moschovitis **20060507061555]
> [Renamed to ScriptAdapter, keep ConsoleAdapter alias.
> George Moschovitis **20060506103921]
> [Yeah, the ConsoleAdapter kinda works now ;-)
> George Moschovitis **20060506102010]
> [Introduced new nitro command/runner that will replace the current spaghetti code. Based on facetes Console::Command. Also introduced the ConsoleAdapter.
> George Moschovitis **20060506095503]
> [Removed accumulate (in facets).
> George Moschovitis **20060506080052]
> [Some small updates.
> George Moschovitis **20060506075526]
> [TAG 0.30.0
> George Moschovitis **20060506075053]
> Patch bundle hash:
> 505fc7e46b1501429ce1830b5080abb1dcce58a6
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sat Jul  8 03:01:17 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 8 Jul 2006 10:01:17 +0300
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: <1152323590.6667.17.camel@localhost.localdomain>
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
	<55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>
	<3ff63f9b0607070120p2ceaacecw58257b8a2974adb2@mail.gmail.com>
	
	<716700c90607071100g29beeedbh64f9afe0d009fe17@mail.gmail.com>
	<1152323590.6667.17.camel@localhost.localdomain>
Message-ID: 

Hello Zed,

thanks for this great info. I will revisit the mongrel adapter (and
especially the upload/file upload features) when I finish my current
projects and hopefully release 0.31.0

thanks for   your great work,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sat Jul  8 03:26:14 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 8 Jul 2006 10:26:14 +0300
Subject: [Nitro] Publishable#path to allow easy access to
	Publishable.mount_path ?
In-Reply-To: 
References: 
Message-ID: 

Hah, I was thinking about adding something like that ;-)
thanks, even though I am not sure about the name path.

-g.

On 7/8/06, gabriele renzi  wrote:
> Hi everyone,
>
> I always thought that instances of Nitro::Controller and other
> Publishables had an instance variable @mount_path, possibly because the
> documentation says so.. but the documentation was obviously referring to
> a _class_ instance variable, and thus to access that information in a
> view it is necessary to write
>
>   self.class.mount_path
> or
>   ThisController.mount_path #losing modularity
>
> Would it be possible to apply this micro patch that adds a "path" method
> to Publishable objects? Simple test attached (both for
> Controller.mount_path and Controller#path, a two-liner :)
>
>
>
>
> --
> blog en: http://www.riffraff.info
> blog it: http://riffraff.blogsome.com
> jabber : rff.rff at gmail dot com
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sat Jul  8 03:39:34 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 8 Jul 2006 10:39:34 +0300
Subject: [Nitro] Publishable#path to allow easy access to
	Publishable.mount_path ?
In-Reply-To: 
References: 
	
Message-ID: 

manually added as publishable.mount_path to my repo, thanks.

-g.

On 7/8/06, George Moschovitis  wrote:
> Hah, I was thinking about adding something like that ;-)
> thanks, even though I am not sure about the name path.
>
> -g.
>
> On 7/8/06, gabriele renzi  wrote:
> > Hi everyone,
> >
> > I always thought that instances of Nitro::Controller and other
> > Publishables had an instance variable @mount_path, possibly because the
> > documentation says so.. but the documentation was obviously referring to
> > a _class_ instance variable, and thus to access that information in a
> > view it is necessary to write
> >
> >   self.class.mount_path
> > or
> >   ThisController.mount_path #losing modularity
> >
> > Would it be possible to apply this micro patch that adds a "path" method
> > to Publishable objects? Simple test attached (both for
> > Controller.mount_path and Controller#path, a two-liner :)
> >
> >
> >
> >
> > --
> > blog en: http://www.riffraff.info
> > blog it: http://riffraff.blogsome.com
> > jabber : rff.rff at gmail dot com
> >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
> >
> >
>
>
> --
> http://www.gmosx.com
> http://www.nitroproject.org
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From zimba.tm at gmail.com  Sat Jul  8 05:32:41 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Sat, 8 Jul 2006 11:32:41 +0200
Subject: [Nitro] mongrel adapter vs. large request
In-Reply-To: 
References: <716700c90607061331l3784f4c5lb6006212324450c4@mail.gmail.com>
	<3ff63f9b0607070024g53d16578nca074b2e384ab728@mail.gmail.com>
	<55c107bf0607070054h941f52cv853b9eb49eab7f5@mail.gmail.com>
	<3ff63f9b0607070120p2ceaacecw58257b8a2974adb2@mail.gmail.com>
	
	<716700c90607071100g29beeedbh64f9afe0d009fe17@mail.gmail.com>
	<1152323590.6667.17.camel@localhost.localdomain>
	
Message-ID: <3ff63f9b0607080232y34960b49ub006b461fa2bf737@mail.gmail.com>

Thanks Zed,

the mongrel adapter is basically a rip from the webrick adapter. This
is why some inconcistencies like that appear :-p

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From zimba.tm at gmail.com  Sat Jul  8 07:52:58 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Sat, 8 Jul 2006 13:52:58 +0200
Subject: [Nitro] devlab.ch ?
Message-ID: <3ff63f9b0607080452o24381ff8i4ba8a0905c8d6fd@mail.gmail.com>

I'm thinking of buying devlab.ch. Do you think it's worth the 35CHF ?

If you have any other available domain names ideas, I'm listening.

What I want to do, is to build something similar to rubyforge. I also
want to provide rubyforge mirroring. All this on a new separate
vserver.

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From william.full.moon at gmail.com  Sat Jul  8 07:56:36 2006
From: william.full.moon at gmail.com (* William)
Date: Sat, 8 Jul 2006 21:56:36 +1000
Subject: [Nitro] textile / nitro wiki question
Message-ID: <007801c6a285$952f6860$85af07ca@ghostgum>

Hi gang!
 
I've got two questions; one about textile and one about the textile used on
the Nitro wiki.
 
  1/.  Fortunately I came to answer my first question.  So I can pass
       on the answer which (was) ...

       *  To escape textile meta-characters use the 

                pair
 
       To get an asterisk(*); say ...

              This is *SQL*:  select * from User; 

       Works a treat, all on one line the way I like it.


  2/.  The redcloth or whatever textile used on the Nitro wiki 
       seems to do some things "different".  Is there a reason
       for this?  Does 'official nitro' also suffer this 'problem'?
 
In the mean time, when thing are not working as expected, I check here:

See textile tester ... http://www.textism.com/tools/textile/index.php

I think it would be nice to match the reference copy myself.

Aloha,
        Will.
 
 
____________________________________________________________________________
____
  (2006) Information proprietary and confidential intended for direct
recipient(s) and mutually agreed co-respondents.
  http://adroit-process.blogspot.com/

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/383 - Release Date: 07-Jul-2006
 


From lasso at lassoweb.se  Sat Jul  8 08:08:57 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Sat, 08 Jul 2006 14:08:57 +0200
Subject: [Nitro] Conditionals in templating
Message-ID: <44AFA059.3070105@lassoweb.se>

Hi list!

I'm a bit confused again. This time I've run into some trouble when
using conditionals in my template. Consider the following:

controller.rb:

class Foo
     def bar
         @bgcolor = 'cccccc'
     end
end


bar.xhtml:


     
         
         
     
         
     
     
-tags and just one -tag. The only way I get it to
work is by replacing the  with the the very ugly-looking:


     

     


There must be a better way, right? Shouldn't conditional logic be
applied BEFORE REXML gets hold of the data?


/Lasso


________________________________________
Lars Olsson
lasso at lassoweb.se
http://www.lassoweb.se/


From surrender_it at yahoo.it  Sat Jul  8 08:33:57 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sat, 08 Jul 2006 14:33:57 +0200
Subject: [Nitro] Publishable#path to allow easy access to
 Publishable.mount_path ?
In-Reply-To: 
References: 	
	
Message-ID: 

George Moschovitis ha scritto:
> manually added as publishable.mount_path to my repo, thanks.

I thought #path was ok because it goes hand on hand with request.path, 
but mount_path is good enough, thank you :)
Anyway, could you please add the small tests like those found in my patch ?


From m.fellinger at gmail.com  Sat Jul  8 09:56:40 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Sat, 8 Jul 2006 22:56:40 +0900
Subject: [Nitro] Conditionals in templating
In-Reply-To: <44AFA059.3070105@lassoweb.se>
References: <44AFA059.3070105@lassoweb.se>
Message-ID: <9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>

these two are just messed up from gmail?

         wrote:
> Hi list!
>
> I'm a bit confused again. This time I've run into some trouble when
> using conditionals in my template. Consider the following:
>
> controller.rb:
>
> class Foo
>      def bar
>          @bgcolor = 'cccccc'
>      end
> end
>
>
> bar.xhtml:
>
> 
>      
>                      
>          
>      
>          
>      
>      
> 
> Now, this triggers an error when REXML reads the template (since there
> are TWO -tags and just one -tag. The only way I get it to
> work is by replacing the  with the the very ugly-looking:
>
> 
>      
> 
>      
> 
>
> There must be a better way, right? Shouldn't conditional logic be
> applied BEFORE REXML gets hold of the data?
>
>
> /Lasso
>
>
> ________________________________________
> Lars Olsson
> lasso at lassoweb.se
> http://www.lassoweb.se/
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>

From transfire at gmail.com  Sat Jul  8 11:56:51 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 8 Jul 2006 11:56:51 -0400
Subject: [Nitro] devlab.ch ?
In-Reply-To: <3ff63f9b0607080452o24381ff8i4ba8a0905c8d6fd@mail.gmail.com>
References: <3ff63f9b0607080452o24381ff8i4ba8a0905c8d6fd@mail.gmail.com>
Message-ID: <4b6f054f0607080856u2c384316t84e816337fdfcc72@mail.gmail.com>

On 7/8/06, Jonas Pfenniger  wrote:
> I'm thinking of buying devlab.ch. Do you think it's worth the 35CHF ?
>
> If you have any other available domain names ideas, I'm listening.
>
> What I want to do, is to build something similar to rubyforge. I also
> want to provide rubyforge mirroring. All this on a new separate
> vserver.

Cool!

If you use 1&1 you can get upto 5 free .info domains.

T.

From lasso at lassoweb.se  Sat Jul  8 12:02:55 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Sat, 08 Jul 2006 18:02:55 +0200
Subject: [Nitro] Conditionals in templating
In-Reply-To: <9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>
References: <44AFA059.3070105@lassoweb.se>
	<9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>
Message-ID: <44AFD72F.4010507@lassoweb.se>

No not Gmail, just bad typing on my part. Let's try cut 'n' paste 
instead ;) This doesn't work:


     
         
     

     
     
     
     
     
     


But if I replace  with

     
     
     
     
     

REXML doesn't complain. But the fix looks *very wrong*


/lasso

________________________________________
Lars Olsson
lasso at lassoweb.se
http://www.lassoweb.se/



Michael Fellinger skrev:
> these two are just messed up from gmail?
> 
>              
References: <44AEF012.8040707@gmx.de>
	
Message-ID: <44B0D2C2.9090204@gmx.de>

here again.

zipped.

Felix

George Moschovitis wrote:
> Please,  resend the patches in tgz or zip format because Gmail messes
> up with the plain text files. Thanks!
>
> -g.
>
> On 7/8/06, jfwittmann  wrote:
>   
>> Hello,
>>
>> here are three patches.
>>
>> - the first two output-fix patches fix problems with the table and pager
>> helper if they get no results.
>> - the third patch add a new option to the pager helper to set the
>> navigation titles.
>>
>> additional, I add a simple test app to show the changes.
>>
>> cu,
>>
>> Felix
>>
>>     

-------------- next part --------------
A non-text attachment was scrubbed...
Name: patches_nitro-dev.zip
Type: application/x-zip-compressed
Size: 11662 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060709/5f629166/attachment-0001.bin 

From george.moschovitis at gmail.com  Sun Jul  9 12:09:14 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 9 Jul 2006 19:09:14 +0300
Subject: [Nitro] Publishable#path to allow easy access to
	Publishable.mount_path ?
In-Reply-To: 
References: 
	
	
	
Message-ID: 

i think the request.path is different to controller.mount_path

> Anyway, could you please add the small tests like those found in my patch ?

Could you please help me here and send me a patch with these tests against
repo.nitroproject.org? That would be very helpful. I am quite busy at
the moment working on the next update of np.org

thanks in advance,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sun Jul  9 12:16:42 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 9 Jul 2006 19:16:42 +0300
Subject: [Nitro] Conditionals in templating
In-Reply-To: <44AFD72F.4010507@lassoweb.se>
References: <44AFA059.3070105@lassoweb.se>
	<9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>
	<44AFD72F.4010507@lassoweb.se>
Message-ID: 

On 7/8/06, Lars Olsson  wrote:
> No not Gmail, just bad typing on my part. Let's try cut 'n' paste
> instead ;) This doesn't work:
>
> 
>      
>          
>      
>
>      
>      
>      
>      
>      
>      
> 

I use something like this:



...


Anyone does something better? If not please add this to oxywtf...

regards,
George.


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sun Jul  9 13:06:44 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 9 Jul 2006 20:06:44 +0300
Subject: [Nitro] teaser
Message-ID: 

http://nitroproject.org/blog/articles/4

regards,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From lasso at lassoweb.se  Sun Jul  9 16:28:03 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Sun, 09 Jul 2006 22:28:03 +0200
Subject: [Nitro] Conditionals in templating
In-Reply-To: 
References: <44AFA059.3070105@lassoweb.se>	<9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>	<44AFD72F.4010507@lassoweb.se>
	
Message-ID: <44B166D3.10408@lassoweb.se>

Hi!

Thanks for the tip, George. Unfortunatly REXML complains about your code
too. (Ijust pasted it into an empty template.)

Error raised by /lib/ruby/1.8/rexml/parsers/baseparser.rb, line 228:
---
Path: /test
malformed XML: missing tag start Line: Position: Last 80 unconsumed
characters:  ... 
---

Btw, I forgot to mention that I'm using the beta gems provided by Bryan.


Any clues to what might be wrong?


Kindly

/lasso

________________________________________
Lars Olsson
lasso at lassoweb.nu
http://www.lassoweb.nu/



George Moschovitis skrev:
> On 7/8/06, Lars Olsson  wrote:
>> No not Gmail, just bad typing on my part. Let's try cut 'n' paste
>> instead ;) This doesn't work:
>>
>> 
>>      
>>          
>>      
>>
>>      
>>      
>>      
>>      
>>      
>>      
>> 
> 
> I use something like this:
> 
> 
> 
> ...
> 
> 
> Anyone does something better? If not please add this to oxywtf...
> 
> regards,
> George.
> 
> 


From lasso at lassoweb.se  Sun Jul  9 16:32:39 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Sun, 09 Jul 2006 22:32:39 +0200
Subject: [Nitro] Conditionals in templating
In-Reply-To: <44B166D3.10408@lassoweb.se>
References: <44AFA059.3070105@lassoweb.se>	<9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>	<44AFD72F.4010507@lassoweb.se>	
	<44B166D3.10408@lassoweb.se>
Message-ID: <44B167E7.6000808@lassoweb.se>

oops...The error gets raised from from line 338, not 228. Sorry.

/lasso



Lars Olsson skrev:
> Hi!
> 
> Thanks for the tip, George. Unfortunatly REXML complains about your code
> too. (Ijust pasted it into an empty template.)
> 
> Error raised by /lib/ruby/1.8/rexml/parsers/baseparser.rb, line 228:
> ---
> Path: /test
> malformed XML: missing tag start Line: Position: Last 80 unconsumed
> characters:  ... 
> ---
> 
> Btw, I forgot to mention that I'm using the beta gems provided by Bryan.
> 
> 
> Any clues to what might be wrong?
> 
> 
> Kindly
> 
> /lasso
> 
> ________________________________________
> Lars Olsson
> lasso at lassoweb.nu
> http://www.lassoweb.nu/

From george.moschovitis at gmail.com  Sun Jul  9 16:52:12 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 9 Jul 2006 23:52:12 +0300
Subject: [Nitro] Conditionals in templating
In-Reply-To: <44B167E7.6000808@lassoweb.se>
References: <44AFA059.3070105@lassoweb.se>
	<9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>
	<44AFD72F.4010507@lassoweb.se>
	
	<44B166D3.10408@lassoweb.se> <44B167E7.6000808@lassoweb.se>
Message-ID: 

so does this work for u?

-g.

On 7/9/06, Lars Olsson  wrote:
> oops...The error gets raised from from line 338, not 228. Sorry.
>
> /lasso
>
>
>
> Lars Olsson skrev:
> > Hi!
> >
> > Thanks for the tip, George. Unfortunatly REXML complains about your code
> > too. (Ijust pasted it into an empty template.)
> >
> > Error raised by /lib/ruby/1.8/rexml/parsers/baseparser.rb, line 228:
> > ---
> > Path: /test
> > malformed XML: missing tag start Line: Position: Last 80 unconsumed
> > characters:  ... 
> > ---
> >
> > Btw, I forgot to mention that I'm using the beta gems provided by Bryan.
> >
> >
> > Any clues to what might be wrong?
> >
> >
> > Kindly
> >
> > /lasso
> >
> > ________________________________________
> > Lars Olsson
> > lasso at lassoweb.nu
> > http://www.lassoweb.nu/
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From william.full.moon at gmail.com  Mon Jul 10 04:25:51 2006
From: william.full.moon at gmail.com (* William)
Date: Mon, 10 Jul 2006 18:25:51 +1000
Subject: [Nitro] what Nitro sites do we know?
Message-ID: <012101c6a3fa$788c4d10$85af07ca@ghostgum>

Hello list people!
 
I have a question, and in true form I thought I'd offer an inducement.
Hopefully females on this list have a sense of humour -- or even an urge to
get into the entertainment business.
 
   HYPERLINK
"http://www.smh.com.au/news/technology/geek-goddesses-or-calendar-girls/2006
/07/09/1152383607217.html"http://www.smh.com.au/news/technology/geek-goddess
es-or-calendar-girls/2006/07/09/1152383607217.html
 
The real query is of course, what sites are using Nitro out there on the
wide web?  I thought I'd do a little census to put on the wiki.
 
People planning to start a site -- I am available for business coaching and
development.  
 
What else?  Oh I also thought to list any "Nitro Parts" available or even
under-construction since I'm asking. 
 
Thanks in advance ...
 
   /Will
 
____________________________________________________________________________
____
  (2006) Information proprietary and confidential intended for direct
recipient(s) and mutually agreed co-respondents.
  HYPERLINK
"http://adroit-process.blogspot.com/"http://adroit-process.blogspot.com/
 


--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/383 - Release Date: 07-Jul-2006



-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/383 - Release Date: 07-Jul-2006
 


From lasso at lassoweb.se  Mon Jul 10 05:23:49 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Mon, 10 Jul 2006 11:23:49 +0200
Subject: [Nitro] Conditionals in templating
In-Reply-To: 
References: <44AFA059.3070105@lassoweb.se>	<9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>	<44AFD72F.4010507@lassoweb.se>		<44B166D3.10408@lassoweb.se>
	<44B167E7.6000808@lassoweb.se>
	
Message-ID: <44B21CA5.7020509@lassoweb.se>

No. It doesn't work for me. REXML complains that my XML is malformed. I 
did some tests with a plain REXML script (no nitro) and the behaviour is 
consistent.

Actually, I don't quite understand how your code is supposed to work 
unless the  part are parsed before the rest of the document.

Example of "non-working" template:

1.	
2.		
3.			title
4.		
5.		
6.		
7.		
8.	

(The template works fine when replacing  with )

Does line 5 get "processed" before REXML gets the file? Does the style 
variable exist when line 6 is reached?


Kindly

/lasso

________________________________________
Lars Olsson
lasso at lassoweb.se
http://www.lassoweb.se/

George Moschovitis skrev:
> so does this work for u?
> 
> -g.

From william.full.moon at gmail.com  Mon Jul 10 07:21:27 2006
From: william.full.moon at gmail.com (* William)
Date: Mon, 10 Jul 2006 21:21:27 +1000
Subject: [Nitro] repository?
In-Reply-To: 
Message-ID: <013e01c6a413$00f57ce0$85af07ca@ghostgum>

Hi ho

>> repo.nitroproject.org

Is there an FTP server on there?

Also, for some reason Firefox is not recognising newlines on the text files.
So everything wraps 'blatt'.  

Further investigation shows that the directory for 

   http://repo.nitroproject.org/

Is not coming from the web server (or the web server is showing me the wrong
stuff).  I can't get a HTTP file listing.  For example (spark folder):

   Connecting to:  repo.nitroproject.org [207.245.118.196]
   >>>> GET /spark HTTP/1.1
   Host: repo.nitroproject.org
   User-Agent: GetRightPro/6.0beta7
   Accept: */*
   Accept-Encoding:
   HTTP/1.0 301 Moved Permanently
   Date: Mon, 10 Jul 2006 11:08:01 GMT
   Location: http://repo.nitroproject.org/spark/
   Content-Length: 0
   Server: lighttpd/1.4.6
   Connection: close

A 301 error comes over the wire.  

Anyway it is all very interesting.  I think an FTP link is the simple
solution.  Or a nice ZIP file for a one-file-download.

Aloha,
       Will.
    

-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of George Moschovitis
Sent: Monday, 10 July 2006 02:09
To: General discussion about Nitro
Subject: Re: [Nitro] Publishable#path to allow easy access
toPublishable.mount_path ?
Importance: Low

i think the request.path is different to controller.mount_path

> Anyway, could you please add the small tests like those found in my patch
?

Could you please help me here and send me a patch with these tests against
repo.nitroproject.org? That would be very helpful. I am quite busy at the
moment working on the next update of np.org

thanks in advance,
George.

--
http://www.gmosx.com
http://www.nitroproject.org
_______________________________________________
Nitro-general mailing list
Nitro-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/383 - Release Date: 07-Jul-2006
 


From zimba.tm at gmail.com  Mon Jul 10 07:32:32 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Mon, 10 Jul 2006 13:32:32 +0200
Subject: [Nitro] repository?
In-Reply-To: <013e01c6a413$00f57ce0$85af07ca@ghostgum>
References: 
	<013e01c6a413$00f57ce0$85af07ca@ghostgum>
Message-ID: <3ff63f9b0607100432i702dcabgb3f3e525d518eb1@mail.gmail.com>

On 10/07/06, * William  wrote:
> Hi ho
>
> >> repo.nitroproject.org

darcs get http://repo.nitroproject.org

is the simplest way

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From george.moschovitis at gmail.com  Mon Jul 10 08:02:01 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 10 Jul 2006 15:02:01 +0300
Subject: [Nitro] Conditionals in templating
In-Reply-To: <44B21CA5.7020509@lassoweb.se>
References: <44AFA059.3070105@lassoweb.se>
	<9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>
	<44AFD72F.4010507@lassoweb.se>
	
	<44B166D3.10408@lassoweb.se> <44B167E7.6000808@lassoweb.se>
	
	<44B21CA5.7020509@lassoweb.se>
Message-ID: 

that code  was out of my mind perhaps it is wrong, I will have to look
it up and get back to you. Something similar to this *does* work.

-g.

On 7/10/06, Lars Olsson  wrote:
> No. It doesn't work for me. REXML complains that my XML is malformed. I
> did some tests with a plain REXML script (no nitro) and the behaviour is
> consistent.
>
> Actually, I don't quite understand how your code is supposed to work
> unless the  part are parsed before the rest of the document.
>
> Example of "non-working" template:
>
> 1.      
> 2.              
> 3.                      title
> 4.              
> 5.              
> 6.              
> 7.              
> 8.      
>
> (The template works fine when replacing  with )
>
> Does line 5 get "processed" before REXML gets the file? Does the style
> variable exist when line 6 is reached?
>
>
> Kindly
>
> /lasso
>
> ________________________________________
> Lars Olsson
> lasso at lassoweb.se
> http://www.lassoweb.se/
>
> George Moschovitis skrev:
> > so does this work for u?
> >
> > -g.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Mon Jul 10 11:42:23 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 10 Jul 2006 18:42:23 +0300
Subject: [Nitro] Ruby warning question
Message-ID: 

Dear devs,

i have the following code:

Scaffold.model Ticket::Comment

and get  this error:

/home/gmosx/Code/plasma/part/tickets.rb:12: warning: toplevel constant
Comment referenced by Ticket::Comment

Anyone has any idea what may be causing this warning?

regards,
George.


-- 
http://www.gmosx.com
http://www.nitroproject.org

From john at oxyliquit.de  Mon Jul 10 12:15:52 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 10 Jul 2006 18:15:52 +0200
Subject: [Nitro] [Oxy] pending questions 2006-07-03
In-Reply-To: <44AD1D53.5090906@gmail.com>
References: <005901c69f1d$732413d0$52af07ca@ghostgum>
	<44AC89C7.4090400@gmail.com> 
	<44AD1D53.5090906@gmail.com>
Message-ID: 

Hi,

>> It can and will be fixed, but not before my exam ends.

Done (or should be). Please have a look, I tested a bit and
it seemed to work. But due to some heavy changes I can't be
totally sure everything works just as before.

Everyone else, also tell me if you find stuff which you don't
like, or it'll never get changed ;)

To everyone:

Aidan gave me some very nice feature requests, like better
stats (you like em? I can make them less... informal again
if you like :P) and the idea to utilize a 'Trackback'
to let Oxyliquit know of Sites posting stuff about Nitro.

This will probably be the next big feature, if you don't
tell me something even more intriguing ;D

> No, seriously, best of luck.  Let us know how you do.

Well .... ^^; unfortunately it didn't go all too well... but
thats life for you :P

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From john at oxyliquit.de  Mon Jul 10 12:16:06 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 10 Jul 2006 18:16:06 +0200
Subject: [Nitro] search Nitro-General
Message-ID: 

Hi,

as I couldn't find a way to nicely search the Nitro-General
Mailing List, I just made my own.
It's in alpha state, don't abuse, but use.

About search queries:

> Parses a query, which should be single words separated by
> the boolean operators "&" and, "|" or, and "!" not, which
> can be grouped using parenthesis.

Words seperated by spaces are revamped to ORs.
The querying is very fragile, report any weird behaviour.

link:
http://www.oxyliquit.de/mail/search

I just used the archive of the ML. To update this I'll
probably have to make a cron job to run over it every ...,
what is an acceptable update time?

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From william.full.moon at gmail.com  Mon Jul 10 12:50:25 2006
From: william.full.moon at gmail.com (* William)
Date: Tue, 11 Jul 2006 02:50:25 +1000
Subject: [Nitro] search Nitro-General
In-Reply-To: 
Message-ID: <001601c6a440$f5b48af0$c3af07ca@ghostgum>

Interesting ...

Anyway I guess you only need to once-a-month get the full-text archive to
your machine.  A Ruby HTTP client could do that.  You know the archive name
as soon as you know the month.

I have one suggestion.   I did a query on "part"-s.  The results look very
interesting.  My request is that the query is displayed and included in a
URI (the way google does this).

That way I can paste the URI in here:

   http://www.oxyliquit.de/mail/search

* You don't know what keywords or search expression I used.  It is better if
the URI (browser address bar) shows the query; such as:

   http://www.oxyliquit.de/mail/search?part%20AND%20anything

Just so I can paste it into an email and say, Look at -ALL- -THESE- :-)

Cool idea Jo.

-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of Jonathan Buch
Sent: Tuesday, 11 July 2006 02:16
To: nitro-general at rubyforge.org
Subject: [Nitro] search Nitro-General
Importance: Low

Hi,

as I couldn't find a way to nicely search the Nitro-General Mailing List, I
just made my own.
It's in alpha state, don't abuse, but use.

About search queries:

> Parses a query, which should be single words separated by the boolean 
> operators "&" and, "|" or, and "!" not, which can be grouped using 
> parenthesis.

Words seperated by spaces are revamped to ORs.
The querying is very fragile, report any weird behaviour.

link:
http://www.oxyliquit.de/mail/search

I just used the archive of the ML. To update this I'll probably have to make
a cron job to run over it every ..., what is an acceptable update time?

Jo

--
Feel the love
http://pinkjuice.com/pics/ruby.png
_______________________________________________
Nitro-general mailing list
Nitro-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/383 - Release Date: 07-Jul-2006
 


From george.moschovitis at gmail.com  Mon Jul 10 12:50:55 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 10 Jul 2006 19:50:55 +0300
Subject: [Nitro] search Nitro-General
In-Reply-To: 
References: 
Message-ID: 

Very nice,

can you give us some details, how this works?

best regards,
George.


On 7/10/06, Jonathan Buch  wrote:
> Hi,
>
> as I couldn't find a way to nicely search the Nitro-General
> Mailing List, I just made my own.
> It's in alpha state, don't abuse, but use.
>
> About search queries:
>
> > Parses a query, which should be single words separated by
> > the boolean operators "&" and, "|" or, and "!" not, which
> > can be grouped using parenthesis.
>
> Words seperated by spaces are revamped to ORs.
> The querying is very fragile, report any weird behaviour.
>
> link:
> http://www.oxyliquit.de/mail/search
>
> I just used the archive of the ML. To update this I'll
> probably have to make a cron job to run over it every ...,
> what is an acceptable update time?
>
> Jo
>
> --
> Feel the love
> http://pinkjuice.com/pics/ruby.png
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From john at oxyliquit.de  Mon Jul 10 13:44:43 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 10 Jul 2006 19:44:43 +0200
Subject: [Nitro] search Nitro-General
In-Reply-To: <001601c6a440$f5b48af0$c3af07ca@ghostgum>
References: <001601c6a440$f5b48af0$c3af07ca@ghostgum>
Message-ID: 

Hi,

> Anyway I guess you only need to once-a-month get the full-text archive to
> your machine.  A Ruby HTTP client could do that.  You know the archive  
> name as soon as you know the month.

Then I would have to parse the last month on every first of the month...
Well, as it takes like 5 seconds to parse a month worth of posts, I guess
I can just do it every week..

>    http://www.oxyliquit.de/mail/search?part%20AND%20anything
> Just so I can paste it into an email and say, Look at -ALL- -THESE- :-)

Well, almost. This is an undocumented feature, but you can use GET.

http://www.oxyliquit.de/mail/search?q=part%26William%26database

Not really sure why that's post there anyway.. gotta ask my bro after
the exams...

And, don't use ' AND ', use ' & '. This is search engine specific.

Note to self: make more than 5 search results, use pager.

> Cool idea Jo.

Thanks!

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From surrender_it at yahoo.it  Mon Jul 10 13:55:51 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Mon, 10 Jul 2006 19:55:51 +0200
Subject: [Nitro] Publishable#path to allow easy access to
 Publishable.mount_path ?
In-Reply-To: 
References: 			
	
Message-ID: 

George Moschovitis ha scritto:
> i think the request.path is different to controller.mount_path
> 
> 
>>Anyway, could you please add the small tests like those found in my patch ?
> 
> 
> Could you please help me here and send me a patch with these tests against
> repo.nitroproject.org? That would be very helpful. I am quite busy at
> the moment working on the next update of np.org

sure, I will ASAP, just give me some time to recover from world-cup party :)

-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From john at oxyliquit.de  Mon Jul 10 14:02:48 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 10 Jul 2006 20:02:48 +0200
Subject: [Nitro] search Nitro-General
In-Reply-To: 
References: 
	
Message-ID: 

Hi,

> Very nice,

Thanks! ^_^

> can you give us some details, how this works?

No, I'm greedy ;D

The mail stuff is actually very simple, it boils down to:

* parsing the mbox format:
     @mbox = open(url).read.split(/\n*^From\s.*\d\d\d\d$/)[1..-1]

* getting from, id, time, subject, header, body:
     header, body = raw.split(/\n\n/, 2)
     header[/^From:\s*(.*)$/]; from = $1
     header[/^Message-ID:\s*<(.*)>\s*$/]; id = $1
     header[/^Date:\s*(.*)\s*$/]; date = DateTime.parse($1)
     header[/^Subject:\s*(.*)\s*$/]; subject = $1

* saving each of those as a Og Model into the database
       class IndexedMail
         include ModEMailParser
         property :id, String, :unique => true
         property :email_from, String
         property :subject, String
         property :date, Time
         property :header, String
         property :body, String

         def initialize(raw)
           @email_from, @id, @date, @subject, @header, @body =  
parse_mail(raw)
         end
       end

* providing access to be able to search for information
       IndexedMail.find(:sql => "SELECT * FROM
           findogindexedmail('#{Og.escape(@query)}')
           ORDER BY rank LIMIT 5", :select => '*').each do |m|
         @results << [m, m.find_rank]
       end
   findogindexedmail('text') is just stored procedure hiding some
   unessessary search details. The select => '*' here is a workaround,
   to get the rank field provided by the search.

The other stuff like the nitro layer (templates, internal stuff) I had
already avalable from Oxy so it took like an hour or two to implement this.

George: I LOVE NITRO! Thank you, you can take the hands from your ears
again ;D

Well, this were some details, too much? ;P

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From george.moschovitis at gmail.com  Mon Jul 10 14:10:17 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 10 Jul 2006 21:10:17 +0300
Subject: [Nitro] Publishable#path to allow easy access to
	Publishable.mount_path ?
In-Reply-To: 
References: 
	
	
	
	
	
Message-ID: 

> sure, I will ASAP, just give me some time to recover from world-cup party :)


ooh, congrats (even though I would like Germany to win ;-)))

regards,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From lists at davidlangston.com  Mon Jul 10 14:45:11 2006
From: lists at davidlangston.com (Dave Langston)
Date: Mon, 10 Jul 2006 11:45:11 -0700
Subject: [Nitro] search Nitro-General
In-Reply-To: 
References: 
Message-ID: <44B2A037.1010007@davidlangston.com>

This will be convenient to use when at oxy... great job.  Also, as an 
alternative you can use gmane.org's archive of the c.l.r.n.g newsgroup 
with its search function at:

    http://news.gmane.org/gmane.comp.lang.ruby.nitro.general
           (search box at bottom of page)

Once you find an interesting message, you can see the entire thread it 
belongs to by going to the actual message from the search results and 
then clicking on the message subject line near the top...

rgds,
Dave Langston


Jonathan Buch wrote:
> Hi,
> 
> as I couldn't find a way to nicely search the Nitro-General
> Mailing List, I just made my own.
> It's in alpha state, don't abuse, but use.
> 
> About search queries:
> 
>> Parses a query, which should be single words separated by
>> the boolean operators "&" and, "|" or, and "!" not, which
>> can be grouped using parenthesis.
> 
> Words seperated by spaces are revamped to ORs.
> The querying is very fragile, report any weird behaviour.
> 
> link:
> http://www.oxyliquit.de/mail/search
> 
> I just used the archive of the ML. To update this I'll
> probably have to make a cron job to run over it every ...,
> what is an acceptable update time?
> 
> Jo
> 


From george.moschovitis at gmail.com  Mon Jul 10 14:56:52 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 10 Jul 2006 21:56:52 +0300
Subject: [Nitro] to_xml/to_json in Facets
Message-ID: 

Tom,

I would like to see he following methods added somehow in Facets:

obj.to_xml
obj.to_json
enumerable.to_xml
enumerable.to_json
hash.from_xml
hash.from_json

Could we perhaps integrate this kind of functionality in Facets? This
is really imporant in order for Nitro to move down the RESTful/Ajax
path. And they would be extermely useful additions for any kind of
project that leverages Facets.


regards,
George.


PS: I think someone posted a patch with to_xml in this list. But an
well-thought, integrated solution through facets would be better.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From john at oxyliquit.de  Mon Jul 10 15:38:00 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 10 Jul 2006 21:38:00 +0200
Subject: [Nitro] search Nitro-General
In-Reply-To: <44B2A037.1010007@davidlangston.com>
References:  <44B2A037.1010007@davidlangston.com>
Message-ID: 

Hi,

> This will be convenient to use when at oxy... great job.  Also, as an
> alternative you can use gmane.org's archive of the c.l.r.n.g newsgroup
> with its search function at:
>
>     http://news.gmane.org/gmane.comp.lang.ruby.nitro.general
>            (search box at bottom of page)
>
> Once you find an interesting message, you can see the entire thread it
> belongs to by going to the actual message from the search results and
> then clicking on the message subject line near the top...

Huh, well... so there was a search available :)

Well, I have currently no ambitions to write a full threading thingy
like this which involves more parsing of email than I'm able to atm ;)

Anyway, gmane seems to be ugly but useful to me, I'll try to remember.

Anyone care to write a full mail browser in Nitro as part which I can
plug 'n play into Oxy? ;D
This is partly a joke, yes :P

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From james.britt at gmail.com  Mon Jul 10 16:51:28 2006
From: james.britt at gmail.com (James Britt)
Date: Mon, 10 Jul 2006 13:51:28 -0700
Subject: [Nitro] to_xml/to_json in Facets
In-Reply-To: 
References: 
Message-ID: <44B2BDD0.8090806@gmail.com>

George Moschovitis wrote:
> Tom,
> 
> I would like to see he following methods added somehow in Facets:

There exists a Ruby json gem that adds to_json to core objects.

Is there a reason to have this behavior part of Facets, or can Nitro 
just use the existing json gem?


What is the purpose of to_xml?  json is a reasonably well-specified set 
of object serialization rules.  Not so with XML, which defines syntax 
but not the semantics.    In picking an XML format, a lot depends on a 
shared understanding among clients.  It seems like something better left 
to to application designer.



-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://yourelevatorpitch.com  - Finding Business Focus

From transfire at gmail.com  Mon Jul 10 17:17:07 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 10 Jul 2006 17:17:07 -0400
Subject: [Nitro] to_xml/to_json in Facets
In-Reply-To: <44B2BDD0.8090806@gmail.com>
References: 
	<44B2BDD0.8090806@gmail.com>
Message-ID: <4b6f054f0607101417g172585e4hfdb5a8a0d2858c50@mail.gmail.com>

require 'facet/json'
require 'facet/xoxo'

From george.moschovitis at gmail.com  Mon Jul 10 17:25:35 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 11 Jul 2006 00:25:35 +0300
Subject: [Nitro] to_xml/to_json in Facets
In-Reply-To: <4b6f054f0607101417g172585e4hfdb5a8a0d2858c50@mail.gmail.com>
References: 
	<44B2BDD0.8090806@gmail.com>
	<4b6f054f0607101417g172585e4hfdb5a8a0d2858c50@mail.gmail.com>
Message-ID: 

Do they handle the serialization of enumerations ? array/hash?
Can they also convert from xml/json to hash ?


-g.

On 7/11/06, TRANS  wrote:
> require 'facet/json'
> require 'facet/xoxo'
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Mon Jul 10 17:26:43 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 11 Jul 2006 00:26:43 +0300
Subject: [Nitro] search Nitro-General
In-Reply-To: 
References:  <44B2A037.1010007@davidlangston.com>
	
Message-ID: 

How do you do the indexing/crawling/searching?

out of curiosity,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From bryan.a.soto at gmail.com  Mon Jul 10 17:46:45 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 10 Jul 2006 14:46:45 -0700
Subject: [Nitro] Ruby warning question
In-Reply-To: 
References: 
Message-ID: 

On 7/10/06, George Moschovitis  wrote:
> Anyone has any idea what may be causing this warning?
>

=====
$ irb
irb(main):001:0> class Comment; end
=> nil
irb(main):002:0> class Ticket; end
=> nil
irb(main):003:0> Ticket::Comment
(irb):3: warning: toplevel constant Comment referenced by Ticket::Comment
=> Comment
irb(main):004:0> class Ticket; class Comment; end; end
=> nil
irb(main):005:0> Ticket::Comment
=> Ticket::Comment
irb(main):006:0>
=====

You might not be referring to what you think you're referring to.

bryan

From james.britt at gmail.com  Mon Jul 10 20:21:35 2006
From: james.britt at gmail.com (James Britt)
Date: Mon, 10 Jul 2006 17:21:35 -0700
Subject: [Nitro] [BUG] Og still ignores :port setting for MySQL connections ;
 fix included here
In-Reply-To: <44793842.5060508@neurogami.com>
References: <44793842.5060508@neurogami.com>
Message-ID: <44B2EF0F.4090804@gmail.com>

James Britt wrote:
This problem still exists in Og 0.30.1

> 
> It appears that Og 0.30 includes store/mysql.rb, where MysqlStore 
> defines 'create' and 'destroy' class methods.  These do not grab the 
> :port option, hence the problem.
> 
> Here's my modified version (sorry, no diff/ptch tool handy) that creates 
> a :port option if none exists, and passes on:
> 
> class MysqlStore < SqlStore
>    extend MysqlUtils
>    include MysqlUtils
> 
>    def self.create(options)
>      options[:port] ||= 3306 # NEW
>      # gmosx: system is used to avoid shell expansion.
>      system 'mysqladmin', '-f', "--user=#{options[:user]}",
>          "--password=#{options[:password]}",
>          "--host=#{options[:address]}",
>          "--port=#{options[:port]}", # NEW
>          'create', options[:name]
>      super
>    end
> 
>    def self.destroy(options)
>      options[:port] ||= 3306  # NEW
>      system 'mysqladmin', '-f', "--user=#{options[:user]}",
>          "--password=#{options[:password]}", 'drop',
>          "--host=#{options[:address]}",
>          "--port=#{options[:port]}" ,  # NEW
>          options[:name]
>      super
>    end
> 
> 
> 
> 


-- 
James Britt

"Programs must be written for people to read, and only incidentally
  for machines to execute."
   - H. Abelson and G. Sussman
   (in "The Structure and Interpretation of Computer Programs)

From bryan.a.soto at gmail.com  Mon Jul 10 20:47:37 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 10 Jul 2006 17:47:37 -0700
Subject: [Nitro] [BUG] Og still ignores :port setting for MySQL
	connections ; fix included here
In-Reply-To: <44B2EF0F.4090804@gmail.com>
References: <44793842.5060508@neurogami.com> <44B2EF0F.4090804@gmail.com>
Message-ID: 

On 7/10/06, James Britt  wrote:
> James Britt wrote:
> This problem still exists in Og 0.30.1
>

Thanks for reminding me.

From william.full.moon at gmail.com  Mon Jul 10 23:54:29 2006
From: william.full.moon at gmail.com (* William)
Date: Tue, 11 Jul 2006 13:54:29 +1000
Subject: [Nitro] to_xml/to_json in Facets
In-Reply-To: <44B2BDD0.8090806@gmail.com>
Message-ID: <005901c6a49d$ba32a5e0$c3af07ca@ghostgum>

In fact it isn't legal XML without a XSD schema or a DTD.

What you really want for something like this is OWL or RDF.

:-) 

-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of James Britt
Sent: Tuesday, 11 July 2006 06:51
To: General discussion about Nitro
Subject: Re: [Nitro] to_xml/to_json in Facets
Importance: Low


What is the purpose of to_xml?  json is a reasonably well-specified set 
of object serialization rules.  Not so with XML, which defines syntax 
but not the semantics.    In picking an XML format, a lot depends on a 
shared understanding among clients.  It seems like something better left 
to to application designer.

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/384 - Release Date: 10-Jul-2006
 


From william.full.moon at gmail.com  Mon Jul 10 23:56:59 2006
From: william.full.moon at gmail.com (* William)
Date: Tue, 11 Jul 2006 13:56:59 +1000
Subject: [Nitro] Publishable#path to allow easy access
	toPublishable.mount_path ?
In-Reply-To: 
Message-ID: <005a01c6a49e$13f09600$c3af07ca@ghostgum>

Yes also congratulations to Italia

It is some consolation that Australia were knocked out by the winners.  

:-) 

-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of George Moschovitis
Sent: Tuesday, 11 July 2006 04:10
To: General discussion about Nitro
Subject: Re: [Nitro] Publishable#path to allow easy access
toPublishable.mount_path ?
Importance: Low

> sure, I will ASAP, just give me some time to recover from world-cup 
> party :)


ooh, congrats (even though I would like Germany to win ;-)))

regards,
George.

--
http://www.gmosx.com
http://www.nitroproject.org
_______________________________________________
Nitro-general mailing list
Nitro-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/384 - Release Date: 10-Jul-2006
 


From jlsysinc at alltel.net  Tue Jul 11 00:51:25 2006
From: jlsysinc at alltel.net (Jon A. Lambert)
Date: Tue, 11 Jul 2006 00:51:25 -0400
Subject: [Nitro] Ruby warning question
References: 
Message-ID: <012601c6a4a5$aaa83fb0$0200000a@agamemnon>

George Moschovitis wrote:
> /home/gmosx/Code/plasma/part/tickets.rb:12: warning: toplevel constant
> Comment referenced by Ticket::Comment
> 
> Anyone has any idea what may be causing this warning?

Specifically no.  Generally yes.

Foo = 10
class Bar
end
puts Bar::Foo 

--
J. Lambert

From manveru at weez-int.com  Tue Jul 11 01:17:49 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Tue, 11 Jul 2006 14:17:49 +0900
Subject: [Nitro] search Nitro-General
In-Reply-To: 
References:  <44B2A037.1010007@davidlangston.com>
	
Message-ID: <200607111417.49124.manveru@weez-int.com>

On Tuesday 11 July 2006 04:38, Jonathan Buch wrote:
> Hi,
>
> > This will be convenient to use when at oxy... great job.  Also, as an
> > alternative you can use gmane.org's archive of the c.l.r.n.g newsgroup
> > with its search function at:
> >
> >     http://news.gmane.org/gmane.comp.lang.ruby.nitro.general
> >            (search box at bottom of page)
> >
> > Once you find an interesting message, you can see the entire thread it
> > belongs to by going to the actual message from the search results and
> > then clicking on the message subject line near the top...
>
> Huh, well... so there was a search available :)
>
> Well, I have currently no ambitions to write a full threading thingy
> like this which involves more parsing of email than I'm able to atm ;)
>
> Anyway, gmane seems to be ugly but useful to me, I'll try to remember.
>
> Anyone care to write a full mail browser in Nitro as part which I can
> plug 'n play into Oxy? ;D
> This is partly a joke, yes :P
>
> Jo

site:rubyforge.org/pipermail/nitro-general/ words you search for

just as a small pointer - this is how i used to search the ML :)
and yeah, there are APIs for google-search avaialabe, so this wouldn't have 
taken any effort...
however, since you had your fun building it i appreciate it nonetheless ;)

btw, i thought about an IMAP-client based on nitro... anybody interested?

From manveru at weez-int.com  Tue Jul 11 01:20:24 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Tue, 11 Jul 2006 14:20:24 +0900
Subject: [Nitro] repository?
In-Reply-To: <3ff63f9b0607100432i702dcabgb3f3e525d518eb1@mail.gmail.com>
References: 
	<013e01c6a413$00f57ce0$85af07ca@ghostgum>
	<3ff63f9b0607100432i702dcabgb3f3e525d518eb1@mail.gmail.com>
Message-ID: <200607111420.25007.manveru@weez-int.com>

On Monday 10 July 2006 20:32, Jonas Pfenniger wrote:
> On 10/07/06, * William  wrote:
> > Hi ho
> >
> > >> repo.nitroproject.org
>
> darcs get http://repo.nitroproject.org
>
> is the simplest way

This is only for people who have darcs... i know of some who don't because 
it's taking too long to compile ;)
no, i won't name anyone... not even vag^H^H^H

I really think a JIT-compression/download would be very nice...

From bryan.a.soto at gmail.com  Tue Jul 11 02:14:18 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 10 Jul 2006 23:14:18 -0700
Subject: [Nitro] [PATCH] table-helper and pager-helper
In-Reply-To: <44B0D2C2.9090204@gmx.de>
References: <44AEF012.8040707@gmx.de>
	
	<44B0D2C2.9090204@gmx.de>
Message-ID: 

On 7/9/06, jfwittmann  wrote:
> here again.
>
> zipped.
>

Thanks for that. Darcs is fussy about checksums. I like the idea of
the patch, but it seems to be missing the definition for
Pager.link_titles?

undefined method `link_titles' for Nitro::Pager:Class

In file 'K:/devlab/nitrohq/nitro/lib/nitro/helper/pager.rb' :

274: # #{pager.links}
275:
276: def paginate(items, options = {})
277: per_page = options.delete(:per_page) || Pager.per_page
278: pager_key = options.delete(:pager_key) || Pager.key
279: pager_link_titles =
Pager.link_titles.merge(options.delete(:pager_link_titles) ||
Pager.link_titles)
280: case items
281: when Array
282: items = items.dup
283: pager = Pager.new(request, per_page, items.size, pager_key,
pager_link_titles)
284: items = items.slice(pager.offset, pager.per_page) || []

From aglarond at gmail.com  Tue Jul 11 02:36:09 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Tue, 11 Jul 2006 08:36:09 +0200
Subject: [Nitro] search Nitro-General
In-Reply-To: 
References: 
	
	
Message-ID: <55c107bf0607102336s52c0bda7i35fd60cfa358b319@mail.gmail.com>

Hey Kashia,

On 7/10/06, Jonathan Buch  wrote:
>
>
> The mail stuff is actually very simple, it boils down to:
>
> * parsing the mbox format:
>      @mbox = open(url).read.split(/\n*^From\s.*\d\d\d\d$/)[1..-1]
>
> * getting from, id, time, subject, header, body:
>      header, body = raw.split(/\n\n/, 2)
>      header[/^From:\s*(.*)$/]; from = $1
>      header[/^Message-ID:\s*<(.*)>\s*$/]; id = $1
>      header[/^Date:\s*(.*)\s*$/]; date = DateTime.parse($1)
>      header[/^Subject:\s*(.*)\s*$/]; subject = $1
>


Did you not know about TMail ?  It
does all the parsing for you, and might even catch some cases you haven't
thought of.  I've got something similar to the following that I use to parse
mbox-format mailboxes:

 mbox = TMail::UNIXMbox.new("#{datapath}/#{u}/#{u}.mbox")
 mbox.each do |m|
    mess = TMail::Mail.new(m)
        begin
          recv = mess["Received"]
          recv_date = recv.first.to_s.split(/;/).last
          rdate = Time.parse(recv_date)
          diff = now - rdate
          mess.port.remove and next if diff > 2592000 or diff < 0
          next if ! rdate.nil? && rdate < yest
          date = mess.strftime("%a %b %d %H:%M")
          from = mess.from_addrs
          subj = mess.subject
          msgi = mess.message_id
        rescue TypeError, RangeError, ArgumentError, SyntaxError
          mess.port.remove and next
        end
  end


I use it to parse a spam-quarantine box, hence all the precautions and
rescues.

Too late for you now, but maybe it could help somebody else...

- Dimitri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060711/f983ae00/attachment.html 

From lasso at lassoweb.se  Tue Jul 11 03:02:46 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Tue, 11 Jul 2006 09:02:46 +0200
Subject: [Nitro] Conditionals in templating
In-Reply-To: 
References: <44AFA059.3070105@lassoweb.se>	<9c00d3e00607080656w373d2be5k8dfef2c8051e7641@mail.gmail.com>	<44AFD72F.4010507@lassoweb.se>		<44B166D3.10408@lassoweb.se>
	<44B167E7.6000808@lassoweb.se>		<44B21CA5.7020509@lassoweb.se>
	
Message-ID: <44B34D16.2020500@lassoweb.se>

Thanks for helping out, George. I did find a solution that looks a 
little bit better, but it still feels a *wrong*:


	
		title
	

	
	
	" ?>

	' ?>



Improvements are very welcome :)


Kindly

/lasso

________________________________________
Lars Olsson
lasso at lassoweb.se
http://www.lassoweb.se/



George Moschovitis skrev:
> that code  was out of my mind perhaps it is wrong, I will have to look
> it up and get back to you. Something similar to this *does* work.
> 
> -g.

From transfire at gmail.com  Tue Jul 11 03:14:13 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 11 Jul 2006 03:14:13 -0400
Subject: [Nitro] to_xml/to_json in Facets
In-Reply-To: <44B2BDD0.8090806@gmail.com>
References: 
	<44B2BDD0.8090806@gmail.com>
Message-ID: <4b6f054f0607110014x29aa07bq9db4c73e5e888337@mail.gmail.com>

On 7/10/06, James Britt  wrote:
> George Moschovitis wrote:
> > Tom,
> >
> > I would like to see he following methods added somehow in Facets:
>
> There exists a Ruby json gem that adds to_json to core objects.
>
> Is there a reason to have this behavior part of Facets, or can Nitro
> just use the existing json gem?

its an interesting question. one that induces me to take a step back
and wonder if perhaps i have gone to far with facets. perhaps i have
included too much? yet when i tried the opposite approach with
calibre, having all the separate libs as independent gems, that
didin;t seem to go over well at all. so where do you draw the line?

t.

From manveru at weez-int.com  Tue Jul 11 03:16:36 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Tue, 11 Jul 2006 16:16:36 +0900
Subject: [Nitro] Conditionals in templating
In-Reply-To: <44B34D16.2020500@lassoweb.se>
References: <44AFA059.3070105@lassoweb.se>
	
	<44B34D16.2020500@lassoweb.se>
Message-ID: <200607111616.36395.manveru@weez-int.com>

On Tuesday 11 July 2006 16:02, Lars Olsson wrote:
> Thanks for helping out, George. I did find a solution that looks a
> little bit better, but it still feels a *wrong*:
>
> 
> 	
> 		title
> 	
>
> 	
>
> 	" ?>
>
> 	' ?>
> 
>
>
> Improvements are very welcome :)



maybe this?
I am a bit unsure about using #{} inside #{} so i left that out and use + 
instead

>
>
> Kindly
>
> /lasso
>
> ________________________________________
> Lars Olsson
> lasso at lassoweb.se
> http://www.lassoweb.se/
>
> George Moschovitis skrev:
> > that code  was out of my mind perhaps it is wrong, I will have to look
> > it up and get back to you. Something similar to this *does* work.
> >
> > -g.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

From zimba.tm at gmail.com  Tue Jul 11 03:51:43 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Tue, 11 Jul 2006 09:51:43 +0200
Subject: [Nitro] repository?
In-Reply-To: <200607111420.25007.manveru@weez-int.com>
References: 
	<013e01c6a413$00f57ce0$85af07ca@ghostgum>
	<3ff63f9b0607100432i702dcabgb3f3e525d518eb1@mail.gmail.com>
	<200607111420.25007.manveru@weez-int.com>
Message-ID: <3ff63f9b0607110051t18c69641ic44aabc57e82bb2e@mail.gmail.com>

On 11/07/06, Michael Fellinger  wrote:
> This is only for people who have darcs... i know of some who don't because
> it's taking too long to compile ;)
> no, i won't name anyone... not even vag^H^H^H
>
> I really think a JIT-compression/download would be very nice...

If someone want to build a Rake or Reap task that I can run
periodically, I can put it on the server. Or just a cgi script for JIT
compression.

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From george.moschovitis at gmail.com  Tue Jul 11 08:03:23 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 11 Jul 2006 15:03:23 +0300
Subject: [Nitro] Ruby warning question
In-Reply-To: 
References: 
	
Message-ID: 

I resolved this, in the process I made a big backwards uncompatible
change to Og. As detailed in a previous mail to this list, the
polymorphic relation feature of Og now returns classes in the form
'ArticleComment' instead of 'Article::Comment'.

This allows for far greater flexibility in rearanging your code, and
just works more of the time.

-g.

On 7/11/06, Bryan Soto  wrote:
> On 7/10/06, George Moschovitis  wrote:
> > Anyone has any idea what may be causing this warning?
> >
>
> =====
> $ irb
> irb(main):001:0> class Comment; end
> => nil
> irb(main):002:0> class Ticket; end
> => nil
> irb(main):003:0> Ticket::Comment
> (irb):3: warning: toplevel constant Comment referenced by Ticket::Comment
> => Comment
> irb(main):004:0> class Ticket; class Comment; end; end
> => nil
> irb(main):005:0> Ticket::Comment
> => Ticket::Comment
> irb(main):006:0>
> =====
>
> You might not be referring to what you think you're referring to.
>
> bryan
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Tue Jul 11 08:15:50 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 11 Jul 2006 15:15:50 +0300
Subject: [Nitro] to_xml/to_json in Facets
In-Reply-To: <4b6f054f0607110014x29aa07bq9db4c73e5e888337@mail.gmail.com>
References: 
	<44B2BDD0.8090806@gmail.com>
	<4b6f054f0607110014x29aa07bq9db4c73e5e888337@mail.gmail.com>
Message-ID: 

facets is ok as it is. You have not gone to far.

-g.

On 7/11/06, TRANS  wrote:
> On 7/10/06, James Britt  wrote:
> > George Moschovitis wrote:
> > > Tom,
> > >
> > > I would like to see he following methods added somehow in Facets:
> >
> > There exists a Ruby json gem that adds to_json to core objects.
> >
> > Is there a reason to have this behavior part of Facets, or can Nitro
> > just use the existing json gem?
>
> its an interesting question. one that induces me to take a step back
> and wonder if perhaps i have gone to far with facets. perhaps i have
> included too much? yet when i tried the opposite approach with
> calibre, having all the separate libs as independent gems, that
> didin;t seem to go over well at all. so where do you draw the line?
>
> t.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From john at oxyliquit.de  Tue Jul 11 10:50:18 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 11 Jul 2006 16:50:18 +0200
Subject: [Nitro] search Nitro-General
In-Reply-To: <55c107bf0607102336s52c0bda7i35fd60cfa358b319@mail.gmail.com>
References: 
	
	
	<55c107bf0607102336s52c0bda7i35fd60cfa358b319@mail.gmail.com>
Message-ID: 

Hi,

> Did you not know about TMail  
> ?  It
> does all the parsing for you, and might even catch some cases you haven't
> thought of.

I'm quite sure that's the case ;D

Problem is though, If I hunt for libs to do the job, they might work a
little better, handling corner cases and such, but when I spend 2 hours
of evaluating and searching for libraries when I can do almost the same
job in like 5 minutes using dirty regex, I always go for the 5 minutes.

As a side project, I tried to write a full mbox and mail parser using
RFC2822, RFC4155, RFC2234... (yuck, I still remember the numbers).  It
failed but it was a nice try :)
Nothing beats own experience :P

> Too late for you now, but maybe it could help somebody else...

It's never too late! I won't use it in the current project, but I'm pretty
sure the need to parse an mbox will arise again :)

Thank you for that lib, I'll try to remember it.

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From john at oxyliquit.de  Tue Jul 11 10:50:18 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 11 Jul 2006 16:50:18 +0200
Subject: [Nitro] to_xml/to_json in Facets
In-Reply-To: <4b6f054f0607110014x29aa07bq9db4c73e5e888337@mail.gmail.com>
References: 
	<44B2BDD0.8090806@gmail.com>
	<4b6f054f0607110014x29aa07bq9db4c73e5e888337@mail.gmail.com>
Message-ID: 

Hi,

> its an interesting question. one that induces me to take a step back
> and wonder if perhaps i have gone to far with facets. perhaps i have
> included too much?

Well, not from a Nitro point of view.  Having those functions prevents
Nitro from having too many unnessessary dependencies to other libs.  I
think to_xml/to_json doesn't add too much baggage and they are useable
in many different occasions.

Just my point of view :)

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From john at oxyliquit.de  Tue Jul 11 10:50:15 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 11 Jul 2006 16:50:15 +0200
Subject: [Nitro] Ruby warning question
In-Reply-To: 
References: 
	
	
Message-ID: 

Hi,

> I resolved this, in the process I made a big backwards uncompatible
> change to Og. As detailed in a previous mail to this list, the
> polymorphic relation feature of Og now returns classes in the form
> 'ArticleComment' instead of 'Article::Comment'.
>
> This allows for far greater flexibility in rearanging your code, and
> just works more of the time.

Yeah, it's an acceptable change. But (to put this on to your todo), if
you come around to integrate PostgreSQL again, please also test this
whole polymorphic stuff with PostgreSQL. At the moment this fails, due
to the psql adapter and how it creates constraints.

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From john at oxyliquit.de  Tue Jul 11 10:50:19 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 11 Jul 2006 16:50:19 +0200
Subject: [Nitro] search Nitro-General
In-Reply-To: <200607111417.49124.manveru@weez-int.com>
References:  <44B2A037.1010007@davidlangston.com>
	
	<200607111417.49124.manveru@weez-int.com>
Message-ID: 

Hi,

> site:rubyforge.org/pipermail/nitro-general/ words you search for
>
> just as a small pointer - this is how i used to search the ML :)

'used to', how do you do that now?

> and yeah, there are APIs for google-search avaialabe, so this wouldn't  
> have taken any effort...

That'd be another interesting project, yes :P
But it would've taken even more effort to learn those APIs...

> however, since you had your fun building it i appreciate it nonetheless  
> ;)

^_^

> btw, i thought about an IMAP-client based on nitro... anybody interested?

Nice as well.... *think*

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From transfire at gmail.com  Tue Jul 11 13:03:38 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 11 Jul 2006 13:03:38 -0400
Subject: [Nitro] Facets 1.5 and Reap 7
Message-ID: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>

Hi,

I'm about to release the next versions of Facets and Reap. A few points:

1. This version of Facets deprecates the use of BasicObject. It's is
still there for backward compatability sake, but none of the other
Facets use it any longer. This is a fairly substantial change,  so
it's important to test you apps well against this new version so we
can squash any gotchyas this may cause. Also, given the magnituted of
the change, after further development of this 1.5 series I *might* go
ahead and bump Facets to 2.0.

2. Reap 7 is now transitioned to use of the meta/ dir and
meta/projectinfo instead of just ProjectInfo. It's still supports the
later so there's no need to take immediate action, but the future is
with meta/. You can also move your Reapfile there (btw reapfile,
Reapfile and REAPFILE all work, same with projectinfo, ProjectInfo,
and PROJECTINFO).

3. Reap 7 is now dependent on Sow for generating packages. Sow is
still a bit rough around the edges, but being based on Reap's old
project task, it has a good jump. (I'll release the next versionof
this at the same time, I just have one thing left to do to it first).

4. Currently per-file comment testing can be done via the command
'rubytest'. I plan to add some imporvements to that soon. But right
now I'm wondering if that's really the best name for this command?

5. The scaffold and template tasks have been separated into their own
command called 'seed'. I went back forth on this, and finally decided
it's just a whole lot easier that way.

6. I would like to see Reap become fully Windows compatible. It likely
would not be too difficult, mostly pathname issues. Is there any one
who would like to work on that?

T.

From james.britt at gmail.com  Tue Jul 11 13:26:55 2006
From: james.britt at gmail.com (James Britt)
Date: Tue, 11 Jul 2006 10:26:55 -0700
Subject: [Nitro] Elements STILL killing 
In-Reply-To: 
References: <125e06d10a652057e009301a6afd0ec6@oggu.de>
	
Message-ID: <44B3DF5F.9070908@gmail.com>

>> is wrong and isn't working at least in Firefox.
> 
> 
> I am curious why this is wrong, any idea?
> 
> 
>>George's solution (I guess) in Flare is for example like this:
>>...
>>the individual webapps.
> 
> 
> you are right, I 'll somehow fix this.
>  
> 
>>For everyone having the same problem: commenting out 165 and 166 in
>>elements.rb has no real negative side-effects.
> 
> 
> there *are* side-effects. A better fix will be available for the next version.
> 


My pages are broken because Nitro 0.30.1 is collapsing my textarea elements.

Is there an option someplace to turn this off?

-- 
James Britt

"Judge a man by his questions, rather than his answers."
  - Voltaire

From neokolor at gmx.de  Tue Jul 11 14:52:13 2006
From: neokolor at gmx.de (jfwittmann)
Date: Tue, 11 Jul 2006 20:52:13 +0200
Subject: [Nitro] [PATCH] table-helper and pager-helper
In-Reply-To: <bd93db0d0607102314n2b0400a4o61d2073059b360a3@mail.gmail.com>
References: <44AEF012.8040707@gmx.de>	<afd80b540607072356u4b273a70k7f9212d48ec20047@mail.gmail.com>	<44B0D2C2.9090204@gmx.de>
	<bd93db0d0607102314n2b0400a4o61d2073059b360a3@mail.gmail.com>
Message-ID: <44B3F35D.9000503@gmx.de>

Hi Bryan,

I think I do a mistake. I used the tool beyond compare to diff my dev 
version with the glycerin version.   
Probably  I forget to move some code lines.
here I send you the patch again. I hope it will now work.

Felix


Bryan Soto wrote:
> On 7/9/06, jfwittmann <neokolor at gmx.de> wrote:
>   
>> here again.
>>
>> zipped.
>>
>>     
>
> Thanks for that. Darcs is fussy about checksums. I like the idea of
> the patch, but it seems to be missing the definition for
> Pager.link_titles?
>
> undefined method `link_titles' for Nitro::Pager:Class
>   

-------------- next part --------------
A non-text attachment was scrubbed...
Name: pager_helper_add_option.zip
Type: application/x-zip-compressed
Size: 4425 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060711/9ea8224b/attachment.bin 

From john at oxyliquit.de  Tue Jul 11 14:53:09 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 11 Jul 2006 20:53:09 +0200
Subject: [Nitro] Elements STILL killing <textarea>
In-Reply-To: <44B3DF5F.9070908@gmail.com>
References: <125e06d10a652057e009301a6afd0ec6@oggu.de>
	
	<44B3DF5F.9070908@gmail.com>
Message-ID: 

Hi,

I think you have to modify the compiler pipeline, throw out the
Cleanup thingy.

> My pages are broken because Nitro 0.30.1 is collapsing my textarea  
> elements.
>
> Is there an option someplace to turn this off?


Look at glue/lib/glue/html.rb

The cleanup rules are specified there....

Hope that helps.

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From james.britt at gmail.com  Tue Jul 11 15:10:46 2006
From: james.britt at gmail.com (James Britt)
Date: Tue, 11 Jul 2006 12:10:46 -0700
Subject: [Nitro] Elements STILL killing 
In-Reply-To: 
References: <125e06d10a652057e009301a6afd0ec6@oggu.de>		<44B3DF5F.9070908@gmail.com>
	
Message-ID: <44B3F7B6.7060406@gmail.com>

Jonathan Buch wrote:
> Hi,
> 
> I think you have to modify the compiler pipeline, throw out the
> Cleanup thingy.
> 
> 
>>My pages are broken because Nitro 0.30.1 is collapsing my textarea  
>>elements.
>>
>>Is there an option someplace to turn this off?
> 
> 
> 
> Look at glue/lib/glue/html.rb
> 
> The cleanup rules are specified there....

Ah.

  out.gsub! /]*)><\/textarea>/, '<textarea \1>#{}'

Interesting.



> 
> Hope that helps.

Yes.  It appears this is my own dopey fault. :(

The element munging is happening in my post-pipeline transformation plugin.


Thanks!

-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists


From bryan.a.soto at gmail.com  Tue Jul 11 15:51:58 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 11 Jul 2006 12:51:58 -0700
Subject: [Nitro] [PATCH] table-helper and pager-helper
In-Reply-To: <44B3F35D.9000503@gmx.de>
References: <44AEF012.8040707@gmx.de>
	
	<44B0D2C2.9090204@gmx.de>
	
	<44B3F35D.9000503@gmx.de>
Message-ID: 

On 7/11/06, jfwittmann  wrote:
> Hi Bryan,
>
> I think I do a mistake. I used the tool beyond compare to diff my dev
> version with the glycerin version.
> Probably  I forget to move some code lines.
> here I send you the patch again. I hope it will now work.
>

Ah yes, that seemed to do it. Thanks. Barring problems, I'll probably
apply this today.

bryan

From transfire at gmail.com  Tue Jul 11 17:14:17 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 11 Jul 2006 17:14:17 -0400
Subject: [Nitro] Facets 1.5 and Reap 7
In-Reply-To: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>
References: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>
Message-ID: <4b6f054f0607111414g4cf972edobca9a44699007b1b@mail.gmail.com>

Released to the wild! Sow 0.6, Reap 7.0 and Facets 1.5 --only as
tar.gz source package though.

Some of the Webpages have been updated too.

  http://sow.rubyforge.org
  http://reap.rubyforge.org
  http://facets.rubyforge.org

T.

From george.moschovitis at gmail.com  Tue Jul 11 17:41:31 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 12 Jul 2006 00:41:31 +0300
Subject: [Nitro] Facets 1.5 and Reap 7
In-Reply-To: <4b6f054f0607111414g4cf972edobca9a44699007b1b@mail.gmail.com>
References: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>
	<4b6f054f0607111414g4cf972edobca9a44699007b1b@mail.gmail.com>
Message-ID: 

Havent checked them yet, but congrats anyway ;-)
Ooh, and the sites look  nice. But hopefully I will be able to beat
them with the new np.org update later this week.

regards,
George.

On 7/12/06, TRANS  wrote:
> Released to the wild! Sow 0.6, Reap 7.0 and Facets 1.5 --only as
> tar.gz source package though.
>
> Some of the Webpages have been updated too.
>
>   http://sow.rubyforge.org
>   http://reap.rubyforge.org
>   http://facets.rubyforge.org
>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Tue Jul 11 17:41:50 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 12 Jul 2006 00:41:50 +0300
Subject: [Nitro] Facets 1.5 and Reap 7
In-Reply-To: 
References: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>
	<4b6f054f0607111414g4cf972edobca9a44699007b1b@mail.gmail.com>
	
Message-ID: 

On 7/12/06, George Moschovitis  wrote:
> Havent checked them yet, but congrats anyway ;-)
> Ooh, and the sites look  nice. But hopefully I will be able to beat
> them with the new np.org update later this week.

;-) ;-) ;-)

-- 
http://www.gmosx.com
http://www.nitroproject.org

From john at oxyliquit.de  Tue Jul 11 19:48:38 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Wed, 12 Jul 2006 01:48:38 +0200
Subject: [Nitro] search Nitro-General
In-Reply-To: 
References:  <44B2A037.1010007@davidlangston.com>
	
	
Message-ID: 

Hi,

> How do you do the indexing/crawling/searching?
>
> out of curiosity,
> George.

You asked for it, more info attached. ;D Thought I'd just make
it to a kind of tutorial.

Skip right to Part 3.

I have someone reading over it, (thanks olle!) after that I'll
put it up on Oxy as well.

(It is already marked up for that, shouldn't be too  hard to
read though)

Hope you enjoy!

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mail_howto.md
Type: application/octet-stream
Size: 29096 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060712/f62097ed/attachment-0001.obj 

From manveru at weez-int.com  Tue Jul 11 21:28:05 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Wed, 12 Jul 2006 10:28:05 +0900
Subject: [Nitro] Facets 1.5 and Reap 7
In-Reply-To: <4b6f054f0607111414g4cf972edobca9a44699007b1b@mail.gmail.com>
References: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>
	<4b6f054f0607111414g4cf972edobca9a44699007b1b@mail.gmail.com>
Message-ID: <200607121028.05421.manveru@weez-int.com>

On Wednesday 12 July 2006 06:14, TRANS wrote:
> Released to the wild! Sow 0.6, Reap 7.0 and Facets 1.5 --only as
> tar.gz source package though.
>
> Some of the Webpages have been updated too.
>
>   http://sow.rubyforge.org
>   http://reap.rubyforge.org
>   http://facets.rubyforge.org

Will hopefully be able to try them later today, congrats anyway :)
one thing i really appreciate on your projects is the excellent 
documentation... no idea how you manage to write good code _and_ document all 
of it in one go - but i'm really impressed...

>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

From matt at kiatoa.com  Wed Jul 12 00:28:54 2006
From: matt at kiatoa.com (Matthew Welland)
Date: Tue, 11 Jul 2006 21:28:54 -0700
Subject: [Nitro] Best combo of nitro, og etc. versions for learning nitro?
Message-ID: <200607112128.54886.matt@kiatoa.com>

I'm trying to learn nitro and I'm having trouble finding a combination of 
versions that work with the downloadable examples. Can anyone suggest a combo 
of versions that will work well with the examples on Kubuntu? I want to redo 
my web site in nitro (beta.kiatoa.com).

I'm most interested in how to do login session and file uploads right now. The 
latest error I ran into was with the gallery example:

Error
Path: /save
unable to open image `pload/label.jpg': No such file or directory:
Reload this page. Go to the referer or the home page.
In file '/usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/glue/magick.rb' :

19:
20: def self.generate_thumbnail(src, tname, geostring)
21: ext = File.extname(src)
22: dst = "#{File.join(File.dirname(src), File.basename(src, 
ext))}_#{tname}#{ext}"
23:
24: thumb = Magick::Image.read(File.join(Nitro::Server.public_root, 
src)).first
25: thumb.change_geometry!(geostring) do |cols, rows, thumb|
26: thumb.resize!(cols, rows)
27: end
28: thumb.write(File.join(Nitro::Server.public_root, dst))
29:

-- 

From bryan.a.soto at gmail.com  Wed Jul 12 02:00:37 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 11 Jul 2006 23:00:37 -0700
Subject: [Nitro] Facets 1.5 and Reap 7
In-Reply-To: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>
References: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>
Message-ID: 

On 7/11/06, TRANS  wrote:
> 6. I would like to see Reap become fully Windows compatible. It likely
> would not be too difficult, mostly pathname issues. Is there any one
> who would like to work on that?
>

I can help out a bit with that. First though, as an aside, on Windows,
files like bin/reap aren't executable. Gems handles that by creating a
file, continuing with the reap example, named reap.cmd in the ruby/bin
directory containing

# begin file
@ruby "c:\ruby\bin\reap" %*
# end file

Basically, it allows a Windows user to type reap at the command line
and run the program. Not sure if that's something you care about in
terms of packaging. Just a heads up for other people trying things
out.

Anyway, first bug report.

1) Uninstalled gem version of reap.

C:\Documents and Settings\Bryan\Desktop\sow-0.6>gem list --local reap

*** LOCAL GEMS ***

2) Run reap

C:\Documents and Settings\Bryan\Desktop\sow-0.6>reap --version
c:/ruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
`gem_original_require': no such file to load -- reap/class/release
(LoadError)
        from c:/ruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
`require'
        from c:/ruby/lib/ruby/site_ruby/1.8/reap/tasks.rb:577:in `release'
        from c:/ruby/lib/ruby/site_ruby/1.8/reap/reap.rb:77
        from c:/ruby/lib/ruby/site_ruby/1.8/reap/projectinfo.rb:203:in `each'
        from c:/ruby/lib/ruby/site_ruby/1.8/reap/reap.rb:74
        from c:/ruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
`require'
        from c:/ruby/lib/ruby/site_ruby/1.8/reap/bin/reap.rb:11
        from c:/ruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
`require'
        from c:/ruby/bin/reap:3

A quick check confirms there is no reap/class/release.rb in the
reap-7.0.0 archive.

bryan

P.S. It might be helpful for Windows users if you create a zip archive
as well. Zip is standard on Windows now.

From manveru at weez-int.com  Wed Jul 12 03:47:40 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Wed, 12 Jul 2006 16:47:40 +0900
Subject: [Nitro] search Nitro-General
In-Reply-To: 
References: 
	<200607111417.49124.manveru@weez-int.com>
	
Message-ID: <200607121647.40828.manveru@weez-int.com>

On Tuesday 11 July 2006 23:50, Jonathan Buch wrote:
> Hi,
>
> > site:rubyforge.org/pipermail/nitro-general/ words you search for
> >
> > just as a small pointer - this is how i used to search the ML :)
>
> 'used to', how do you do that now?

Well... now i use oxywtf ;)

>
> > and yeah, there are APIs for google-search avaialabe, so this wouldn't
> > have taken any effort...
>
> That'd be another interesting project, yes :P
> But it would've taken even more effort to learn those APIs...
>
> > however, since you had your fun building it i appreciate it nonetheless
> > ;)
>
> ^_^
>
> > btw, i thought about an IMAP-client based on nitro... anybody interested?
>
> Nice as well.... *think*
>
> Jo

From surrender_it at yahoo.it  Wed Jul 12 07:36:37 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Wed, 12 Jul 2006 13:36:37 +0200
Subject: [Nitro] [Patch]: mount_path test
Message-ID: 

this was the micro test I added in my own patch, redone against 
repo.nitroproject. This is very very basic, but much better than 
nothing. It is put in tc_controller, but possibly we could have a 
tc_publishable some day.

George, please note that there are 3 failing tests (not related to 
this), and that tc_rss.rb should be removed/changed since helper/rss is 
no more available (I recall helper/feed has its own testsuite).



-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From george.moschovitis at gmail.com  Wed Jul 12 08:11:10 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 12 Jul 2006 15:11:10 +0300
Subject: [Nitro] [Patch]: mount_path test
In-Reply-To: 
References: 
Message-ID: 

there is no attachment ;-)

-g.

On 7/12/06, gabriele renzi  wrote:
> this was the micro test I added in my own patch, redone against
> repo.nitroproject. This is very very basic, but much better than
> nothing. It is put in tc_controller, but possibly we could have a
> tc_publishable some day.
>
> George, please note that there are 3 failing tests (not related to
> this), and that tc_rss.rb should be removed/changed since helper/rss is
> no more available (I recall helper/feed has its own testsuite).
>
>
>
> --
> blog en: http://www.riffraff.info
> blog it: http://riffraff.blogsome.com
> jabber : rff.rff at gmail dot com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From john at oxyliquit.de  Wed Jul 12 08:24:34 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Wed, 12 Jul 2006 14:24:34 +0200
Subject: [Nitro] search Nitro-General
In-Reply-To: <200607121647.40828.manveru@weez-int.com>
References: 
	<200607111417.49124.manveru@weez-int.com>
	
	<200607121647.40828.manveru@weez-int.com>
Message-ID: 

Haha! Result ;D

> Well... now i use oxywtf ;)

Now, for that nice answer, a treat, look at the *Standard SQL*
parts in the attached reviewed tutorial, I know you love it ;D

Jo


-- 
Feel the love
http://pinkjuice.com/pics/ruby.png
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mail_howto.md
Type: application/octet-stream
Size: 31594 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060712/99888bb5/attachment-0001.obj 

From transfire at gmail.com  Wed Jul 12 09:39:31 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 12 Jul 2006 09:39:31 -0400
Subject: [Nitro] Facets 1.5 and Reap 7
In-Reply-To: 
References: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>
	
Message-ID: <4b6f054f0607120639g3e8eab37r1bbed9fc72a1c5fc@mail.gmail.com>

On 7/12/06, Bryan Soto  wrote:
> On 7/11/06, TRANS  wrote:
> > 6. I would like to see Reap become fully Windows compatible. It likely
> > would not be too difficult, mostly pathname issues. Is there any one
> > who would like to work on that?
> >
>
> I can help out a bit with that. First though, as an aside, on Windows,
> files like bin/reap aren't executable. Gems handles that by creating a
> file, continuing with the reap example, named reap.cmd in the ruby/bin
> directory containing
>
> # begin file
> @ruby "c:\ruby\bin\reap" %*
> # end file
>
> Basically, it allows a Windows user to type reap at the command line
> and run the program. Not sure if that's something you care about in
> terms of packaging. Just a heads up for other people trying things
> out.
>
> Anyway, first bug report.
>
> 1) Uninstalled gem version of reap.
>
> C:\Documents and Settings\Bryan\Desktop\sow-0.6>gem list --local reap
>
> *** LOCAL GEMS ***
>
> 2) Run reap
>
> C:\Documents and Settings\Bryan\Desktop\sow-0.6>reap --version
> c:/ruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
> `gem_original_require': no such file to load -- reap/class/release
> (LoadError)
>         from c:/ruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
> `require'
>         from c:/ruby/lib/ruby/site_ruby/1.8/reap/tasks.rb:577:in `release'
>         from c:/ruby/lib/ruby/site_ruby/1.8/reap/reap.rb:77
>         from c:/ruby/lib/ruby/site_ruby/1.8/reap/projectinfo.rb:203:in `each'
>         from c:/ruby/lib/ruby/site_ruby/1.8/reap/reap.rb:74
>         from c:/ruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
> `require'
>         from c:/ruby/lib/ruby/site_ruby/1.8/reap/bin/reap.rb:11
>         from c:/ruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
> `require'
>         from c:/ruby/bin/reap:3
>
> A quick check confirms there is no reap/class/release.rb in the
> reap-7.0.0 archive.
>
> bryan
>
> P.S. It might be helpful for Windows users if you create a zip archive
> as well. Zip is standard on Windows now.

Big, big thank you!

I got the .bat files in place for Sow and Reap. I'll have a .zip file
for the next release. I still need to convert Sow's use of an external
tar call to an internal lib (only option at the moment is Austin's
minitar). So that's next on my todo list. Maybe Windows support will
be ready sooner than I thought!

I've fixed the above bug. Actually, I'm still not exactly sure why is
was excluding release.rb, I know it was because 'release' ended up the
exclude setting to the FileList, but that should have only effected a
top level project file/dir --not a file in a subdir and esspecially
not one ending in .rb. So I'll have to keep my eye out for that.
Anyway I removed 'release' and it seems to be okay now.

I'll relase update later today.
T.

From transfire at gmail.com  Wed Jul 12 09:43:36 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 12 Jul 2006 09:43:36 -0400
Subject: [Nitro] Facets 1.5 and Reap 7
In-Reply-To: <200607121028.05421.manveru@weez-int.com>
References: <4b6f054f0607111003w425b634cp11b20f03def6ec79@mail.gmail.com>
	<4b6f054f0607111414g4cf972edobca9a44699007b1b@mail.gmail.com>
	<200607121028.05421.manveru@weez-int.com>
Message-ID: <4b6f054f0607120643r49cc09f5gead53c9320e97217@mail.gmail.com>

On 7/11/06, Michael Fellinger  wrote:
> On Wednesday 12 July 2006 06:14, TRANS wrote:
> > Released to the wild! Sow 0.6, Reap 7.0 and Facets 1.5 --only as
> > tar.gz source package though.
> >
> > Some of the Webpages have been updated too.
> >
> >   http://sow.rubyforge.org
> >   http://reap.rubyforge.org
> >   http://facets.rubyforge.org
>
> Will hopefully be able to try them later today, congrats anyway :)
> one thing i really appreciate on your projects is the excellent
> documentation... no idea how you manage to write good code _and_ document all
> of it in one go - but i'm really impressed...

Thanks! Actaully I beat myself pretty hard for not having _better_
documentation. So I espeically appreciate your saying that :)

Congrats may be a bit premature though, the latest releases arn't
official releases quit yet. But with any luck (and bug reports) they
will be soon.

Thanks again,
T.

From surrender_it at yahoo.it  Wed Jul 12 11:59:55 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Wed, 12 Jul 2006 17:59:55 +0200
Subject: [Nitro] [Patch]: mount_path test
In-Reply-To: 
References: 
	
Message-ID: 

George Moschovitis ha scritto:
> there is no attachment ;-)

oh, damn, I hate when I do this :)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.patch.gz
Type: application/octet-stream
Size: 3716 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060712/da53e473/attachment.obj 

From george.moschovitis at gmail.com  Wed Jul 12 12:15:43 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 12 Jul 2006 19:15:43 +0300
Subject: [Nitro] Fecets Dictionary and POLS
Message-ID: 

Hello Tom,

I would expect this to work:

TYPE_VALUES = Dictionary.new(
    0 => :defect,
    1 => :enhancement,
    2 => :task
)

but I have to use the:

Dictionary[0, :defect, 1, :enhancement,....]

notation instead.

any chance this can be rectified?

thanks,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From james.britt at gmail.com  Wed Jul 12 15:08:08 2006
From: james.britt at gmail.com (James Britt)
Date: Wed, 12 Jul 2006 12:08:08 -0700
Subject: [Nitro] Question about parts
Message-ID: <44B54898.9090203@gmail.com>

I'm writing some docs about about Nitro parts, mostly going by George's 
comments in the source for part.rb

The example has part files all going into subfolders of part/, and the 
main run.rb file explicitly loading the part code by specifying the path.

Can't this happen automatically?  If the part/ folder is assumed by 
Nitro to hold parts code, and assumes that the ruby files in part/ are 
there to handle loading of the actual source, is there a reason Nitro 
shouldn't just automatically load whatever it finds in part/ ?

As is is, there seems to be nothing special about that folder name, 
other than being descriptive of what it contains. (There is 
Part.require, but it doesn't seem terribly useful, and seems out of step 
with the code comments, which say that run.rb is optional and intended 
for running a part to demo its behavior.)

I would think that if you are putting code into the part/ folder then 
you intend to load it, and Nitro should just automagically scan the 
top-level of that folder and auto-require what it finds.  Less work for 
the developer

And if you have parts that you do not want automagically included, put 
them someplace else and load them explicitly.



-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://www.jamesbritt.com     - Playing with Better Toys
http://www.30secondrule.com   - Building Better Tools

From george.moschovitis at gmail.com  Wed Jul 12 15:53:21 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 12 Jul 2006 22:53:21 +0300
Subject: [Nitro] Question about parts
In-Reply-To: <44B54898.9090203@gmail.com>
References: <44B54898.9090203@gmail.com>
Message-ID: 

James,

typically I put the parts directory *outside* of my Webapp. Moreover,
this part directory contains many many parts that are not all used by
the current web application. The dir structure goes like this:

part/
part/content.rb
part/content/model/article.rb
part/content/model/cateory.rb
part/content/controller.rb
part/content/api/blogger.rb
part/wiki.rb
part/wiki/model/page.rb
...

just

require 'parts/content'

to include CMS functionality in your Web App. Please note that you can
have multiple part
(or lib for that matter) directories in your LOAD_PATH.

But, in a future version of Nitro (when I am sure that proposed
directory names are absolutely finalized) Nitro will lookup for a
part, lib, or app directory inside your current application and
automatically load all files it finds as you suggest.

Be patient (or provide a patch)

George.




On 7/12/06, James Britt  wrote:
> I'm writing some docs about about Nitro parts, mostly going by George's
> comments in the source for part.rb
>
> The example has part files all going into subfolders of part/, and the
> main run.rb file explicitly loading the part code by specifying the path.
>
> Can't this happen automatically?  If the part/ folder is assumed by
> Nitro to hold parts code, and assumes that the ruby files in part/ are
> there to handle loading of the actual source, is there a reason Nitro
> shouldn't just automatically load whatever it finds in part/ ?
>
> As is is, there seems to be nothing special about that folder name,
> other than being descriptive of what it contains. (There is
> Part.require, but it doesn't seem terribly useful, and seems out of step
> with the code comments, which say that run.rb is optional and intended
> for running a part to demo its behavior.)
>
> I would think that if you are putting code into the part/ folder then
> you intend to load it, and Nitro should just automagically scan the
> top-level of that folder and auto-require what it finds.  Less work for
> the developer
>
> And if you have parts that you do not want automagically included, put
> them someplace else and load them explicitly.
>
>
>
> --
> James Britt
>
> http://www.ruby-doc.org       - Ruby Help & Documentation
> http://www.artima.com/rubycs/ - The Journal By & For Rubyists
> http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
> http://www.jamesbritt.com     - Playing with Better Toys
> http://www.30secondrule.com   - Building Better Tools
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From transfire at gmail.com  Wed Jul 12 16:17:14 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 12 Jul 2006 16:17:14 -0400
Subject: [Nitro] Fecets Dictionary and POLS
In-Reply-To: 
References: 
Message-ID: <4b6f054f0607121317r3f7db898ra144f92afc897fa0@mail.gmail.com>

On 7/12/06, George Moschovitis  wrote:
> Hello Tom,
>
> I would expect this to work:
>
> TYPE_VALUES = Dictionary.new(
>     0 => :defect,
>     1 => :enhancement,
>     2 => :task
> )
>
> but I have to use the:
>
> Dictionary[0, :defect, 1, :enhancement,....]
>
> notation instead.
>
> any chance this can be rectified?

I tend to agree with you for sure, but the reason for this is because
that's the way Matz defines the Hash.new -- Dictionary just passes the
args right thru to it's super class. Luckily Dictionary offers a
number of other initializers to make life easier (see the docs for a
list of all of them). One of them is [] though, as you point out, and
it can do what you want:

Dictionary[0 => :defect, 1 => :enhancement,....]

That work?
T.

From george.moschovitis at gmail.com  Wed Jul 12 16:29:27 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 12 Jul 2006 23:29:27 +0300
Subject: [Nitro] Fecets Dictionary and POLS
In-Reply-To: <4b6f054f0607121317r3f7db898ra144f92afc897fa0@mail.gmail.com>
References: 
	<4b6f054f0607121317r3f7db898ra144f92afc897fa0@mail.gmail.com>
Message-ID: 

> Dictionary[0 => :defect, 1 => :enhancement,....]

You mean,

Dictionary[0, :defect, 1, :enhancement,....]

thats ok...

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From transfire at gmail.com  Wed Jul 12 17:29:32 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 12 Jul 2006 17:29:32 -0400
Subject: [Nitro] Fecets Dictionary and POLS
In-Reply-To: 
References: 
	<4b6f054f0607121317r3f7db898ra144f92afc897fa0@mail.gmail.com>
	
Message-ID: <4b6f054f0607121429x12d0b191ja24e941e19eafdf4@mail.gmail.com>

On 7/12/06, George Moschovitis  wrote:
> > Dictionary[0 => :defect, 1 => :enhancement,....]
>
> You mean,
>
> Dictionary[0, :defect, 1, :enhancement,....]

No. I mean

  Dictionary[0 => :defect, 1 => :enhancement,....]

T.

From bryan.a.soto at gmail.com  Wed Jul 12 18:44:13 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 12 Jul 2006 15:44:13 -0700
Subject: [Nitro] Best combo of nitro,
	og etc. versions for learning nitro?
In-Reply-To: <200607112128.54886.matt@kiatoa.com>
References: <200607112128.54886.matt@kiatoa.com>
Message-ID: 

On 7/11/06, Matthew Welland  wrote:
> I'm trying to learn nitro and I'm having trouble finding a combination of
> versions that work with the downloadable examples. Can anyone suggest a combo
> of versions that will work well with the examples on Kubuntu? I want to redo

I'm not sure about Kubuntu. I remember going through all the examples
for the 0.29 release and everything worked. It was examples-0.29
available on the rubyforge site. If that's the version you're using,
could you detail what you're doing and perhaps I could help.

From john at oxyliquit.de  Wed Jul 12 18:53:02 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Thu, 13 Jul 2006 00:53:02 +0200
Subject: [Nitro] Community Documentation
Message-ID: 

Hi,

interesting link:

http://www.onlamp.com/pub/a/onlamp/2006/07/06/rethinking-community-documentation.html

I must confess that I read it but didn't get all of it.

When you get to the third page though:

> Community documentation can do a lot better. There is potential for:
> * Online training* Funding and rating systems* Search facilities

I think that Oxyliquit already kind of tries to solve the
second and third point ;D

How about the first?  Well, it's #nitro, the IRC channel.

There is always room for improvement, if you have ideas,
comments, rants or even suggestions on the color scheme,
it's the community that makes up Oxyliquit :)

(Uhm, yes, I feel kind of fan-boy'ish now ;D)

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From james.britt at gmail.com  Wed Jul 12 19:04:36 2006
From: james.britt at gmail.com (James Britt)
Date: Wed, 12 Jul 2006 16:04:36 -0700
Subject: [Nitro] Question about parts
In-Reply-To: 
References: <44B54898.9090203@gmail.com>
	
Message-ID: <44B58004.4020606@gmail.com>

George Moschovitis wrote:
> James,
> 
> typically I put the parts directory *outside* of my Webapp. Moreover,
> this part directory contains many many parts that are not all used by
> the current web application. 

Interesting.  I'm having a hard time picturing the larger directory 
layout, because not all of my apps get deployed to the same places.  I 
would probably use Subversion's 'external' hook.

The dir structure goes like this:
> 
> part/
> part/content.rb
> part/content/model/article.rb
> part/content/model/cateory.rb
> part/content/controller.rb
> part/content/api/blogger.rb
> part/wiki.rb
> part/wiki/model/page.rb
> ...
> 
> just
> 
> require 'parts/content'
> 
> to include CMS functionality in your Web App. Please note that you can
> have multiple part
> (or lib for that matter) directories in your LOAD_PATH.


Right.  But, in my typical case, I would not have 'part/wiki' there 
unless I was going to use it, so auto-loading would work well for me.

> 
> But, in a future version of Nitro (when I am sure that proposed
> directory names are absolutely finalized) Nitro will lookup for a
> part, lib, or app directory inside your current application and
> automatically load all files it finds as you suggest.

Nice.

> 
> Be patient (or provide a patch)

I may do that.  I wrote a small lib that looks for vendor/plugin, and 
auto-requires what it finds.  Doing the same for a part dir would be simple.

Thanks.

-- 
James Britt

"Programs must be written for people to read, and only incidentally
  for machines to execute."
   - H. Abelson and G. Sussman
   (in "The Structure and Interpretation of Computer Programs)

From bryan.a.soto at gmail.com  Wed Jul 12 19:47:29 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 12 Jul 2006 16:47:29 -0700
Subject: [Nitro] Beta Nitro gems available: Round 2
Message-ID: 

If you have the first round beta gems installed:

# gem update -y --source=http://devlab.oree.ch/~bryan/gems && gem cleanup

If not, instructions are below the changelog.

Feedback appreciated. If you have any outstanding patches or bugs,
please bring them up. Barring show-stoppers, I'd like to release soon.
I'm now going to concentrate on ensuring the examples work. I'm sure
that will be appreciated. :)

Thanks,

Bryan

=====

Tue Jul 11 13:10:06 PDT 2006  bryan DOT a DOT soto AT gmail DOT com
  * bugfix-nitro-flaky-template-root
  When running from a directory other than the root directory of your
Nitro app, Template.root was set incorrectly because class Template
would determine whether to use template, src/template and default to
public all before Nitro had chdir'd to the root directory of your
Nitro app. Moved the requires to after the chdir.

Tue Jul 11 13:13:47 PDT 2006  bryan DOT a DOT soto AT gmail DOT com
  * bugfix-nitro-extra-arguments
  Commented out the extra concat as I never could figure out what the
code intended.

Tue Jul 11 11:43:29 PDT 2006  neokolor AT gmx DOT de
  * add option to set nav link titles

Fri Jul  7 16:25:48 PDT 2006  neokolor AT gmx DOT de
  * helper/table.rb: fix problem with table output

Fri Jul  7 16:08:29 PDT 2006  neokolor AT gmx DOT de
  * helper/pager.rb: fix problems with nav output if we have no results

Fri Jul  7 16:18:08 PDT 2006  neokolor AT gmx DOT de
  * helper/pager.rb: add option to set nav link titles

Sun Jun 18 05:42:52 PDT 2006  neokolor AT gmx DOT de
  * fix_in_relation_symbol_to_class

Mon Jul 10 18:07:02 PDT 2006  bryan DOT a DOT soto AT gmail DOT com
  * bugfix-nitro-ticket-43
  Fixed a bug/type as suggested by rff_rff AT yahoo DOT it
  http://devlab.oree.ch/trac/glycerin/ticket/43

Mon Jul 10 18:03:50 PDT 2006  bryan DOT a DOT soto AT gmail DOT com
  * bugfix-nitro-ticket-34
  Fixed as per billk AT cts DOT com suggestion on Ticket #34.
  http://devlab.oree.ch/trac/glycerin/ticket/34

Mon Jul 10 17:56:56 PDT 2006  bryan DOT a DOT soto AT gmail DOT com
  * og-mysql-add-port-option
  From jbritt, makes Mysql respect port option when creating and
dropping databases.

Fri Jun 30 12:07:11 PDT 2006  George Moschovitis 
  * Applied james britt's mongrel patch.

---------- Forwarded message ----------
From: Bryan Soto 
Date: Jun 29, 2006 5:49 PM
Subject: Beta Nitro gems available
To: General discussion about Nitro 


If you have 0.30 installed:

# gem update -y --no-rdoc --no-ri --source=http://devlab.oree.ch/~bryan/gems

If you don't have them installed, there's some weirdness with -y and
--source, so:

# gem install nitro og glue facets gen RedCloth ruby-breakpoint
daemons --no-rdoc --no-ri --source=http://devlab.oree.ch/~bryan/gems

or

# gem install -y nitro

then follow the gem update command line at the beginning and finish it
off with a

# gem cleanup

The --no-rdoc and --no-ri flags are optional. It just goes faster when
specified that way.

Please let the list know of any bugs.

Thanks,

Bryan

From william.full.moon at gmail.com  Wed Jul 12 20:09:05 2006
From: william.full.moon at gmail.com (* William)
Date: Thu, 13 Jul 2006 10:09:05 +1000
Subject: [Nitro] Question about parts
In-Reply-To: <44B58004.4020606@gmail.com>
References: <44B54898.9090203@gmail.com>
	
	<44B58004.4020606@gmail.com>
Message-ID: <9e03c3c60607121709p5968d7eay218b5c7b492a3a7b@mail.gmail.com>

Hi James / George.

Can I comment that may be BOTH these approaches are on track.  You want
parts to be in a place so different applications can share them.  You might
want to load a configuration (say)

     Require 'config'

and then James auto-loader could be in the 'config.rb' as a way to have a
"part that loads parts".

.. ciao /w


On 13/07/06, James Britt  wrote:
>
> George Moschovitis wrote:
> > James,
> >
> > typically I put the parts directory *outside* of my Webapp. Moreover,
> > this part directory contains many many parts that are not all used by
> > the current web application.
>
> Interesting.  I'm having a hard time picturing the larger directory
> layout, because not all of my apps get deployed to the same places.  I
> would probably use Subversion's 'external' hook.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060713/dae1b53c/attachment.html 

From james.britt at gmail.com  Wed Jul 12 20:26:22 2006
From: james.britt at gmail.com (James Britt)
Date: Wed, 12 Jul 2006 17:26:22 -0700
Subject: [Nitro] Question about parts
In-Reply-To: <9e03c3c60607121709p5968d7eay218b5c7b492a3a7b@mail.gmail.com>
References: <44B54898.9090203@gmail.com>		<44B58004.4020606@gmail.com>
	<9e03c3c60607121709p5968d7eay218b5c7b492a3a7b@mail.gmail.com>
Message-ID: <44B5932E.7090907@gmail.com>

* William wrote:
> Hi James / George.
> 
> Can I comment that may be BOTH these approaches are on track.  You want
> parts to be in a place so different applications can share them.  You might
> want to load a configuration (say)
> 
>     Require 'config'
> 
> and then James auto-loader could be in the 'config.rb' as a way to have a
> "part that loads parts".

Well,  I have that now.

To use autoloading with share parts, the base part file could just go do 
the needed require:

[approot]/part/foo.rb

could perhaps have:

$:.unshift(/base/dir/for/system/wide/parts/foo)

require 'model/baz'
require 'controller/bar'



-- 
James Britt

"I often work by avoidance."
- Brian Eno

From james.britt at gmail.com  Wed Jul 12 20:49:31 2006
From: james.britt at gmail.com (James Britt)
Date: Wed, 12 Jul 2006 17:49:31 -0700
Subject: [Nitro] Beta Nitro gems available: Round 2
In-Reply-To: 
References: 
Message-ID: <44B5989B.6070100@gmail.com>

Bryan Soto wrote:
> If you have the first round beta gems installed:
> 
> # gem update -y --source=http://devlab.oree.ch/~bryan/gems && gem cleanup


Well, I suggest not doing that without first being aware of what 
happens.  I was stupid and not paying attention, then caught this:

Cleaning up installed gems...
Attempting uninstall on rubyful_soup-1.0.3
Successfully uninstalled rubyful_soup version 1.0.3
Attempting uninstall on reap-5.0.0
Successfully uninstalled reap version 5.0.0
Removing reap
Removing rubytest
Attempting uninstall on mongrel-0.3.4-mswin32
Successfully uninstalled mongrel version 0.3.4
Attempting uninstall on rails-0.13.1
Successfully uninstalled rails version 0.13.1
Attempting uninstall on actionwebservice-0.8.1
Successfully uninstalled actionwebservice version 0.8.1
Attempting uninstall on switchtower-0.9.0
...

And so on. Until I hit CTRL-C.

If you need or want earlier versions of gems,  know that "gem cleanup" 
will uninstall them.   If you have code that is tied to earlier versions 
of a gem, it will break.

I don't understand the reason for such suggesting such general commands, 
as they often have side-effects that, while OK for one person, are not 
so OK for others.

Does this command not install the newest beta gem?

  gem install Nitro --source=http://devlab.oree.ch/~bryan/gems

Does having earlier versions of Nitro around interfere with anything?


> 
> or
> 
> # gem install -y nitro
> 
> then follow the gem update command line at the beginning and finish it
> off with a
> 
> # gem cleanup

See above for same warnings.

-- 
James Britt

"A principle or axiom is of no value without the rules for applying it."
   - Len Bullard

From bryan.a.soto at gmail.com  Wed Jul 12 21:15:59 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 12 Jul 2006 18:15:59 -0700
Subject: [Nitro] Beta Nitro gems available: Round 2
In-Reply-To: <44B5989B.6070100@gmail.com>
References: 
	<44B5989B.6070100@gmail.com>
Message-ID: 

On 7/12/06, James Britt  wrote:
> I don't understand the reason for such suggesting such general commands,
> as they often have side-effects that, while OK for one person, are not
> so OK for others.
>

My apologies. It's what I typically do and I wrote it without even
thinking about.

From aglarond at gmail.com  Thu Jul 13 03:59:08 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Thu, 13 Jul 2006 09:59:08 +0200
Subject: [Nitro] Beta Nitro gems available: Round 2
In-Reply-To: 
References: 
Message-ID: <55c107bf0607130059s37efa11er19f7ec13b9597573@mail.gmail.com>

Hey Bryan,

On 7/13/06, Bryan Soto  wrote:
>
> If you have the first round beta gems installed:
>
> # gem update -y --source=http://devlab.oree.ch/~bryan/gems
>
> Feedback appreciated.


I don't know if this is a local issue or not, but I thought I'd report it in
case somebody else runs into it:

While generating documentation for gen-0.30.0.2
... MESSAGE:   Unhandled special: Special: type=17, text="This"
... RDOC args: --ri --op /usr/lib/ruby/gems/1.8/doc/gen-0.30.0.2/ri --quiet
lib
(continuing with the rest of the installation)


While generating documentation for glue-0.30.0.2
... MESSAGE:   Unhandled special: Special: type=17, text="Will"
... RDOC args: --ri --op /usr/lib/ruby/gems/1.8/doc/glue-0.30.0.2/ri --quiet
lib
(continuing with the rest of the installation)


While generating documentation for nitro-0.30.0.2
... MESSAGE:   Unhandled special: Special: type=17, text="Global"
... RDOC args: --ri --op /usr/lib/ruby/gems/1.8/doc/nitro-0.30.0.2/ri--quiet lib
(continuing with the rest of the installation)

While generating documentation for og-0.30.0.2
... MESSAGE:   Unhandled special: Special: type=17, text="The"
... RDOC args: --ri --op /usr/lib/ruby/gems/1.8/doc/og-0.30.0.2/ri --quiet
lib
(continuing with the rest of the installation)


Don't have time at the moment to track it down...

- Dimitri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060713/104befb1/attachment-0001.html 

From william.full.moon at gmail.com  Thu Jul 13 04:03:59 2006
From: william.full.moon at gmail.com (* William)
Date: Thu, 13 Jul 2006 18:03:59 +1000
Subject: [Nitro] Question about parts
In-Reply-To: <44B5932E.7090907@gmail.com>
Message-ID: <007501c6a652$e90c3be0$e1e7dbcb@ghostgum>

Excellent !!

James Britt >>>>>>>>>>>>
 
> You might want to load a configuration (say)
> 
>     Require 'config'
> 
> and then James auto-loader could be in the 'config.rb' as a way to 
> have a "part that loads parts".

Well,  I have that now.

To use autoloading with share parts, the base part file could just go do the
needed require:

[approot]/part/foo.rb

could perhaps have:

$:.unshift(/base/dir/for/system/wide/parts/foo)

require 'model/baz'
require 'controller/bar'


-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/385 - Release Date: 11-Jul-2006
 


From surrender_it at yahoo.it  Thu Jul 13 07:42:08 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Thu, 13 Jul 2006 13:42:08 +0200
Subject: [Nitro] Question about parts
In-Reply-To: <44B5932E.7090907@gmail.com>
References: <44B54898.9090203@gmail.com>		<44B58004.4020606@gmail.com>	<9e03c3c60607121709p5968d7eay218b5c7b492a3a7b@mail.gmail.com>
	<44B5932E.7090907@gmail.com>
Message-ID: 

James Britt ha scritto:
> * William wrote:
> 
>>Hi James / George.
>>
>>Can I comment that may be BOTH these approaches are on track.  You want
>>parts to be in a place so different applications can share them.  You might
>>want to load a configuration (say)
>>
>>    Require 'config'
>>
>>and then James auto-loader could be in the 'config.rb' as a way to have a
>>"part that loads parts".
> 
> 
> Well,  I have that now.
> 
> To use autoloading with share parts, the base part file could just go do 
> the needed require:
> 
> [approot]/part/foo.rb
> 
> could perhaps have:
> 
> $:.unshift(/base/dir/for/system/wide/parts/foo)
> 
> require 'model/baz'
> require 'controller/bar'

can I suggest not modifying $: ?
I have more than one part with

part/foo.rb
part/foo/controller.rb
part/baz.rb
part/baz/controller.rb

unshifting both foo/ and baz/ would make require 'controller' quite 
opaque...

-- 

blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From william.full.moon at gmail.com  Thu Jul 13 09:17:47 2006
From: william.full.moon at gmail.com (* William)
Date: Thu, 13 Jul 2006 23:17:47 +1000
Subject: [Nitro] Question about parts
In-Reply-To: 
Message-ID: <00a401c6a67e$bf0ce020$e1e7dbcb@ghostgum>

Oops, I didn't think about that
 
I agree no shift in that top-directory area.  I can't count the failed shell
and make that go psychotic with a shift the base directory.  

I really think a 'config' is smarter, if you want to share configs, etc,
that is possible too.

/w


-----Original Message-----
From: nitro-general-bounces at rubyforge.org [mailto:nitro-general-

James Britt ha scritto:
> * William wrote:
> 
>>Can I comment that may be BOTH these approaches are on track.  You 
>>want parts to be in a place so different applications can share them.  
>>You might want to load a configuration (say)
>>
>>    Require 'config'
>>
>>and then James auto-loader could be in the 'config.rb' as a way to 
>>have a "part that loads parts".

....
> could perhaps have:
> 
> $:.unshift(/base/dir/for/system/wide/parts/foo)
> 
> require 'model/baz'
> require 'controller/bar'

can I suggest not modifying $: ?
I have more than one part with

part/foo.rb
part/foo/controller.rb
part/baz.rb
part/baz/controller.rb

unshifting both foo/ and baz/ would make require 'controller' quite
opaque...

-- 

blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com

_______________________________________________
Nitro-general mailing list
Nitro-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.10/385 - Release Date: 11-Jul-2006
 


From george.moschovitis at gmail.com  Thu Jul 13 13:43:24 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 13 Jul 2006 20:43:24 +0300
Subject: [Nitro] Og stuff
Message-ID: 

Dear devs,

I am working on the Og refactoring again, and making great progress.
But more about this in a np.org blog post. For the moment I have a
question.

I would like to make the Og mysql adapter use innodb tables by default
and enable transactions.

Anyone sees any problems with that?

thanks,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From james.britt at gmail.com  Thu Jul 13 14:52:55 2006
From: james.britt at gmail.com (James Britt)
Date: Thu, 13 Jul 2006 11:52:55 -0700
Subject: [Nitro] Og stuff
In-Reply-To: 
References: 
Message-ID: <44B69687.7040308@gmail.com>

George Moschovitis wrote:
> ... 
> Anyone sees any problems with that?

Does this affect unit testing?  And is it easy to switch off for special 
cases?


-- 
James Britt

"I was born not knowing and have had only a little
  time to change that here and there."
  - Richard P. Feynman

From george.moschovitis at gmail.com  Thu Jul 13 15:02:23 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 13 Jul 2006 22:02:23 +0300
Subject: [Nitro] Og stuff
In-Reply-To: <44B69687.7040308@gmail.com>
References: 
	<44B69687.7040308@gmail.com>
Message-ID: 

> Does this affect unit testing?

I dont see how...

> And is it easy to switch off for special
> cases?

Not implemented yet.

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From aidan at infurious.com  Thu Jul 13 17:19:22 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Thu, 13 Jul 2006 22:19:22 +0100
Subject: [Nitro] [OT] cambrianhouse.com
Message-ID: 

Hi all,

I've registered a few ideas with cambrianhouse.com and was hoping to  
get support from the Nitro community for them - ideally if they kick  
off, I'd like to have them implemented using Nitro.

Take a look at them:

http://www.cambrianhouse.com/idea-warz/idea-promoter/ideas-id/H7vYj67/
http://www.cambrianhouse.com/idea-warz/idea-promoter/ideas-id/F30E3oX/

You can of course vote for other ideas too :-)

Thanks,

Aidan

From william.full.moon at gmail.com  Thu Jul 13 23:44:50 2006
From: william.full.moon at gmail.com (* William)
Date: Fri, 14 Jul 2006 13:44:50 +1000
Subject: [Nitro] nitroproject -- reusable bits
Message-ID: <003a01c6a6f7$df8b2950$20af07ca@ghostgum>

 
Greetings all,
 
Thanks to George the wiki now has reusable RedCloth macros.  Also wiki
anchors now go to the wiki pages not & no longer return to nitroproject
base.
 
For example I've added an alphabetic index macro:
 
    http://www.nitroproject.org/wiki/pages/menu(alpha)

    [ A | B | C | ...... | Y | Z ]
 
Which can be included in your wiki page with the syntax.

    {{i menu(alpha)}}

And then set bookmarks on the page with ...

     
     Albert ....
   
     
     
     Catherine .... is in the Australian outback.

Click on the C above and the page jumps to the line for "Catherine".
Similarly clicking "B" jumps to "Catherine" also because there are no
B-words in the list.

It is a simple macro, which can be very useful for readers.

The principle very useful for menus, re-useable panels and the like.  If
others take this up, we can list include (fragment) pages so others can know
of previously invented wheels.

Aloha,
        Will.
_______________________________________________________________________
  (2006) Information proprietary and confidential intended for 
         direct recipient(s) and mutually agreed co-respondents.
  http://adroit-process.blogspot.com/

--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.0/388 - Release Date: 13-Jul-2006
 

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.0/388 - Release Date: 13-Jul-2006
 


From zimba.tm at gmail.com  Fri Jul 14 05:09:56 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Fri, 14 Jul 2006 11:09:56 +0200
Subject: [Nitro] Og stuff
In-Reply-To: 
References: 
	<44B69687.7040308@gmail.com>
	
Message-ID: <3ff63f9b0607140209s3bae32cfu951a5428f1180c92@mail.gmail.com>

+1 all new MySQL instances support InnoDB. Just make sure that Og is
able to detect the table type and use it's own transaction if it's not
InnoDB

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From zimba.tm at gmail.com  Fri Jul 14 05:18:59 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Fri, 14 Jul 2006 11:18:59 +0200
Subject: [Nitro] [wish] np.org <-> nitro-general gateway ?
Message-ID: <3ff63f9b0607140218o7b31ec02r2405037cae61fbe2@mail.gmail.com>

It would be nice to have new blog-posts and tickets appear in the mailing list.

Inversely, it would be nice that tagged subjects could create new
tickets and that the following answer create new ticket answer.

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From zimba.tm at gmail.com  Fri Jul 14 05:22:01 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Fri, 14 Jul 2006 11:22:01 +0200
Subject: [Nitro] Og and DBI
Message-ID: <3ff63f9b0607140222y5e9fd775v1a981fddddc3db8c@mail.gmail.com>

Hi,

I'm wondering... Wouldn't it make sense to use DBI for all Og sql
adapters ? DBI already takes care of unifying the different databases,
so Og could directly benefit from all of them. Since Og is an
ObjectGraph, the backend is not very important. I also think that it
would solve a whole bunch of bugs.

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From aglarond at gmail.com  Fri Jul 14 05:23:46 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Fri, 14 Jul 2006 11:23:46 +0200
Subject: [Nitro] Og and DBI
In-Reply-To: <3ff63f9b0607140222y5e9fd775v1a981fddddc3db8c@mail.gmail.com>
References: <3ff63f9b0607140222y5e9fd775v1a981fddddc3db8c@mail.gmail.com>
Message-ID: <55c107bf0607140223w1935628ci53a801af7e252688@mail.gmail.com>

On 7/14/06, Jonas Pfenniger  wrote:
>
>
> I'm wondering... Wouldn't it make sense to use DBI for all Og sql
> adapters ? DBI already takes care of unifying the different databases,
> so Og could directly benefit from all of them. Since Og is an
> ObjectGraph, the backend is not very important. I also think that it
> would solve a whole bunch of bugs.
>

+1

Now would be the time to do it, too - with the Og refactoring and all...

- Dimitri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060714/6d75f2fd/attachment-0001.html 

From william.full.moon at gmail.com  Fri Jul 14 05:43:04 2006
From: william.full.moon at gmail.com (* William)
Date: Fri, 14 Jul 2006 19:43:04 +1000
Subject: [Nitro] Og and DBI
In-Reply-To: <55c107bf0607140223w1935628ci53a801af7e252688@mail.gmail.com>
Message-ID: <006e01c6a729$eb2499e0$20af07ca@ghostgum>

+1 yes makes sense
 
and could save future efforts with alternatives and options.
 
/will.
 
 On 7/14/06, Jonas Pfenniger  wrote:


I'm wondering... Wouldn't it make sense to use DBI for all Og sql
adapters ? DBI already takes care of unifying the different databases,
so Og could directly benefit from all of them. Since Og is an
ObjectGraph, the backend is not very important. I also think that it 
would solve a whole bunch of bugs.



+1

Now would be the time to do it, too - with the Og refactoring and all...

- Dimitri



-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.0/388 - Release Date: 13-Jul-2006
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060714/945cfee7/attachment.html 

From william.full.moon at gmail.com  Fri Jul 14 05:55:51 2006
From: william.full.moon at gmail.com (* William)
Date: Fri, 14 Jul 2006 19:55:51 +1000
Subject: [Nitro] Question about parts
In-Reply-To: 
Message-ID: <007401c6a72b$b3e1f480$20af07ca@ghostgum>

James & George

I've made an effort to put this structure on the wiki.  Additions are
welcome.

   http://www.nitroproject.org/wiki/pages/Nitro+Parts

I have set up an alpha-index for a list of Parts when you slot new ones in.

:-)

-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of George Moschovitis
Sent: Thursday, 13 July 2006 05:53
To: General discussion about Nitro
Subject: Re: [Nitro] Question about parts
Importance: Low

James,

typically I put the parts directory *outside* of my Webapp. Moreover, this
part directory contains many many parts that are not all used by the current
web application. The dir structure goes like this:

part/
part/content.rb
part/content/model/article.rb
part/content/model/cateory.rb
part/content/controller.rb
part/content/api/blogger.rb
part/wiki.rb
part/wiki/model/page.rb
...

just

require 'parts/content'

to include CMS functionality in your Web App. Please note that you can have
multiple part (or lib for that matter) directories in your LOAD_PATH.

But, in a future version of Nitro (when I am sure that proposed directory
names are absolutely finalized) Nitro will lookup for a part, lib, or app
directory inside your current application and automatically load all files
it finds as you suggest.

Be patient (or provide a patch)

George.




On 7/12/06, James Britt  wrote:
> I'm writing some docs about about Nitro parts, mostly going by 
> George's comments in the source for part.rb
>
> The example has part files all going into subfolders of part/, and the 
> main run.rb file explicitly loading the part code by specifying the path.
>
> Can't this happen automatically?  If the part/ folder is assumed by 
> Nitro to hold parts code, and assumes that the ruby files in part/ are 
> there to handle loading of the actual source, is there a reason Nitro 
> shouldn't just automatically load whatever it finds in part/ ?
>
> As is is, there seems to be nothing special about that folder name, 
> other than being descriptive of what it contains. (There is 
> Part.require, but it doesn't seem terribly useful, and seems out of 
> step with the code comments, which say that run.rb is optional and 
> intended for running a part to demo its behavior.)
>
> I would think that if you are putting code into the part/ folder then 
> you intend to load it, and Nitro should just automagically scan the 
> top-level of that folder and auto-require what it finds.  Less work 
> for the developer
>
> And if you have parts that you do not want automagically included, put 
> them someplace else and load them explicitly.
>
>
>
> --
> James Britt
>
> http://www.ruby-doc.org       - Ruby Help & Documentation
> http://www.artima.com/rubycs/ - The Journal By & For Rubyists
> http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
> http://www.jamesbritt.com     - Playing with Better Toys
> http://www.30secondrule.com   - Building Better Tools
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.nitroproject.org
_______________________________________________
Nitro-general mailing list
Nitro-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.0/388 - Release Date: 13-Jul-2006
 


From transfire at gmail.com  Fri Jul 14 06:39:38 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 14 Jul 2006 06:39:38 -0400
Subject: [Nitro] Og and DBI
In-Reply-To: <006e01c6a729$eb2499e0$20af07ca@ghostgum>
References: <55c107bf0607140223w1935628ci53a801af7e252688@mail.gmail.com>
	<006e01c6a729$eb2499e0$20af07ca@ghostgum>
Message-ID: <4b6f054f0607140339q1e8d1208l93802f16b8aaca5c@mail.gmail.com>

Though, one is then kind of stuck with how DBI does things. So it
might limit some flexibility for Og too.

BUT, daapter code is that's difficult really. The main thing is to a
have a small core of methods and use those for everything else --then
the adapters needs only define those methods. If you piece meal each
adapter to offer a large swath of Og functionality then it's a
maintanice nightmare.

T.

From zimba.tm at gmail.com  Fri Jul 14 06:39:54 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Fri, 14 Jul 2006 12:39:54 +0200
Subject: [Nitro] Og and DBI
In-Reply-To: <006e01c6a729$eb2499e0$20af07ca@ghostgum>
References: <55c107bf0607140223w1935628ci53a801af7e252688@mail.gmail.com>
	<006e01c6a729$eb2499e0$20af07ca@ghostgum>
Message-ID: <3ff63f9b0607140339g3030ab1du350d22c67953c73f@mail.gmail.com>

NOTES:

Interesting article : http://www.kitebird.com/articles/ruby-dbi.html

DBI already implements the abstract transactions

DBI already does SQL escaping.

DBI doesn't uniformize SQL. Especially table schemas must still be
produced custom.

I don't know about DBI's performances.

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From george.moschovitis at gmail.com  Fri Jul 14 09:32:49 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 14 Jul 2006 16:32:49 +0300
Subject: [Nitro] [wish] np.org <-> nitro-general gateway ?
In-Reply-To: <3ff63f9b0607140218o7b31ec02r2405037cae61fbe2@mail.gmail.com>
References: <3ff63f9b0607140218o7b31ec02r2405037cae61fbe2@mail.gmail.com>
Message-ID: 

> It would be nice to have new blog-posts and tickets appear in the mailing list.

this is already in my todo list ;-) stay tuned!

> Inversely, it would be nice that tagged subjects could create new
> tickets and that the following answer create new ticket answer.

hmm I am not sure about this...

-g.

>
> --
> Cheers,
>   zimbatm
>
> http://zimbatm.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Fri Jul 14 09:34:44 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 14 Jul 2006 16:34:44 +0300
Subject: [Nitro] Og and DBI
In-Reply-To: <3ff63f9b0607140339g3030ab1du350d22c67953c73f@mail.gmail.com>
References: <55c107bf0607140223w1935628ci53a801af7e252688@mail.gmail.com>
	<006e01c6a729$eb2499e0$20af07ca@ghostgum>
	<3ff63f9b0607140339g3030ab1du350d22c67953c73f@mail.gmail.com>
Message-ID: 

Og already does all that... And the new driver model is getting better
every day...
I dont want to spend soooo much time to recode all this with DBI.

-g.

On 7/14/06, Jonas Pfenniger  wrote:
> NOTES:
>
> Interesting article : http://www.kitebird.com/articles/ruby-dbi.html
>
> DBI already implements the abstract transactions
>
> DBI already does SQL escaping.
>
> DBI doesn't uniformize SQL. Especially table schemas must still be
> produced custom.
>
> I don't know about DBI's performances.
>
> --
> Cheers,
>   zimbatm
>
> http://zimbatm.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Fri Jul 14 09:36:09 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 14 Jul 2006 16:36:09 +0300
Subject: [Nitro] Question about parts
In-Reply-To: <007401c6a72b$b3e1f480$20af07ca@ghostgum>
References: 
	<007401c6a72b$b3e1f480$20af07ca@ghostgum>
Message-ID: 

wow, superb wiki pages ;-)

really makes me go and add more features to the wiki and np.org!!!

thanks,
George.

PS: stay tuned for a new np.org update ;-)


On 7/14/06, * William  wrote:
> James & George
>
> I've made an effort to put this structure on the wiki.  Additions are
> welcome.
>
>    http://www.nitroproject.org/wiki/pages/Nitro+Parts
>
> I have set up an alpha-index for a list of Parts when you slot new ones in.
>
> :-)
>
> -----Original Message-----
> From: nitro-general-bounces at rubyforge.org
> [mailto:nitro-general-bounces at rubyforge.org] On Behalf Of George Moschovitis
> Sent: Thursday, 13 July 2006 05:53
> To: General discussion about Nitro
> Subject: Re: [Nitro] Question about parts
> Importance: Low
>
> James,
>
> typically I put the parts directory *outside* of my Webapp. Moreover, this
> part directory contains many many parts that are not all used by the current
> web application. The dir structure goes like this:
>
> part/
> part/content.rb
> part/content/model/article.rb
> part/content/model/cateory.rb
> part/content/controller.rb
> part/content/api/blogger.rb
> part/wiki.rb
> part/wiki/model/page.rb
> ...
>
> just
>
> require 'parts/content'
>
> to include CMS functionality in your Web App. Please note that you can have
> multiple part (or lib for that matter) directories in your LOAD_PATH.
>
> But, in a future version of Nitro (when I am sure that proposed directory
> names are absolutely finalized) Nitro will lookup for a part, lib, or app
> directory inside your current application and automatically load all files
> it finds as you suggest.
>
> Be patient (or provide a patch)
>
> George.
>
>
>
>
> On 7/12/06, James Britt  wrote:
> > I'm writing some docs about about Nitro parts, mostly going by
> > George's comments in the source for part.rb
> >
> > The example has part files all going into subfolders of part/, and the
> > main run.rb file explicitly loading the part code by specifying the path.
> >
> > Can't this happen automatically?  If the part/ folder is assumed by
> > Nitro to hold parts code, and assumes that the ruby files in part/ are
> > there to handle loading of the actual source, is there a reason Nitro
> > shouldn't just automatically load whatever it finds in part/ ?
> >
> > As is is, there seems to be nothing special about that folder name,
> > other than being descriptive of what it contains. (There is
> > Part.require, but it doesn't seem terribly useful, and seems out of
> > step with the code comments, which say that run.rb is optional and
> > intended for running a part to demo its behavior.)
> >
> > I would think that if you are putting code into the part/ folder then
> > you intend to load it, and Nitro should just automagically scan the
> > top-level of that folder and auto-require what it finds.  Less work
> > for the developer
> >
> > And if you have parts that you do not want automagically included, put
> > them someplace else and load them explicitly.
> >
> >
> >
> > --
> > James Britt
> >
> > http://www.ruby-doc.org       - Ruby Help & Documentation
> > http://www.artima.com/rubycs/ - The Journal By & For Rubyists
> > http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
> > http://www.jamesbritt.com     - Playing with Better Toys
> > http://www.30secondrule.com   - Building Better Tools
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> --
> http://www.gmosx.com
> http://www.nitroproject.org
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
> --
> No virus found in this outgoing message.
> Checked by AVG Free Edition.
> Version: 7.1.394 / Virus Database: 268.10.0/388 - Release Date: 13-Jul-2006
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From surrender_it at yahoo.it  Fri Jul 14 09:35:45 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Fri, 14 Jul 2006 15:35:45 +0200
Subject: [Nitro] RedCloth dependency for Glue
Message-ID: 

Hi folks,
it seem that nitro (as found at repo.nitroproject.org now) depends on 
RedCloth (through Glue, but possibly the file should go straight into 
nitro) because it loads it and does not handle the case where it is 
missing.

I think it would be nice if this dependency could be added to the 
Nitro/Glue gem, or otherwise something like the attached patch could be 
applied.
Basically, the method gets defined conditionally as a message for the 
user, maybe it could just pass the string back, but I think not hiding 
problems is a better approach.


-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nocloth.patch.gz
Type: application/octet-stream
Size: 3896 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060714/6f65910a/attachment-0001.obj 

From transfire at gmail.com  Fri Jul 14 09:43:49 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 14 Jul 2006 09:43:49 -0400
Subject: [Nitro] Og and DBI
In-Reply-To: <4b6f054f0607140339q1e8d1208l93802f16b8aaca5c@mail.gmail.com>
References: <55c107bf0607140223w1935628ci53a801af7e252688@mail.gmail.com>
	<006e01c6a729$eb2499e0$20af07ca@ghostgum>
	<4b6f054f0607140339q1e8d1208l93802f16b8aaca5c@mail.gmail.com>
Message-ID: <4b6f054f0607140643y2d871cv5680fa6404402d9f@mail.gmail.com>

On 7/14/06, TRANS  wrote:

Eek.

s/daapter code is that's difficult really/adapter code ISN'T that
difficult really

T.

From aidan at infurious.com  Fri Jul 14 16:49:16 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Fri, 14 Jul 2006 21:49:16 +0100
Subject: [Nitro] Og and DBI
In-Reply-To: <4b6f054f0607140339q1e8d1208l93802f16b8aaca5c@mail.gmail.com>
References: <55c107bf0607140223w1935628ci53a801af7e252688@mail.gmail.com>
	<006e01c6a729$eb2499e0$20af07ca@ghostgum>
	<4b6f054f0607140339q1e8d1208l93802f16b8aaca5c@mail.gmail.com>
Message-ID: 

Dude, you have got to use a spell checker ;-)

Aidan

On 14/07/2006, at 11:39 AM, TRANS wrote:

> BUT, daapter code is that's difficult really. The main thing is to a
> have a small core of methods and use those for everything else --then
> the adapters needs only define those methods. If you piece meal each
> adapter to offer a large swath of Og functionality then it's a
> maintanice nightmare.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060714/dac4b135/attachment.html 

From aidan at infurious.com  Fri Jul 14 16:51:01 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Fri, 14 Jul 2006 21:51:01 +0100
Subject: [Nitro] Og and DBI
In-Reply-To: <4b6f054f0607140643y2d871cv5680fa6404402d9f@mail.gmail.com>
References: <55c107bf0607140223w1935628ci53a801af7e252688@mail.gmail.com>
	<006e01c6a729$eb2499e0$20af07ca@ghostgum>
	<4b6f054f0607140339q1e8d1208l93802f16b8aaca5c@mail.gmail.com>
	<4b6f054f0607140643y2d871cv5680fa6404402d9f@mail.gmail.com>
Message-ID: <8F52C739-4C40-44C9-B800-465E2A91630F@infurious.com>

And I really need to read all the mails before sending out pithy  
comments.

Aidan

On 14/07/2006, at 2:43 PM, TRANS wrote:

> On 7/14/06, TRANS  wrote:
>
> Eek.
>
> s/daapter code is that's difficult really/adapter code ISN'T that
> difficult really
>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


From george.moschovitis at gmail.com  Fri Jul 14 17:01:19 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 15 Jul 2006 00:01:19 +0300
Subject: [Nitro] Og and DBI
In-Reply-To: <4b6f054f0607140339q1e8d1208l93802f16b8aaca5c@mail.gmail.com>
References: <55c107bf0607140223w1935628ci53a801af7e252688@mail.gmail.com>
	<006e01c6a729$eb2499e0$20af07ca@ghostgum>
	<4b6f054f0607140339q1e8d1208l93802f16b8aaca5c@mail.gmail.com>
Message-ID: 

> BUT, daapter code is that's difficult really. The main thing is to a
> ...
> maintanice nightmare.

this is how the new adapters work. Have a look at the sqlite adapter i
just added to my repo.

-g.




>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Fri Jul 14 17:18:01 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 15 Jul 2006 00:18:01 +0300
Subject: [Nitro] nitroproject -- reusable bits
In-Reply-To: <003a01c6a6f7$df8b2950$20af07ca@ghostgum>
References: <003a01c6a6f7$df8b2950$20af07ca@ghostgum>
Message-ID: 

Very cool idea!
This validates the time it took me to remove the  tags
as you suggested.

keep up the great work.
-g.

On 7/14/06, * William  wrote:
>
> Greetings all,
>
> Thanks to George the wiki now has reusable RedCloth macros.  Also wiki
> anchors now go to the wiki pages not & no longer return to nitroproject
> base.
>
> For example I've added an alphabetic index macro:
>
>     http://www.nitroproject.org/wiki/pages/menu(alpha)
>
>     [ A | B | C | ...... | Y | Z ]
>
> Which can be included in your wiki page with the syntax.
>
>     {{i menu(alpha)}}
>
> And then set bookmarks on the page with ...
>
>      
>      Albert ....
>
>      
>      
>      Catherine .... is in the Australian outback.
>
> Click on the C above and the page jumps to the line for "Catherine".
> Similarly clicking "B" jumps to "Catherine" also because there are no
> B-words in the list.
>
> It is a simple macro, which can be very useful for readers.
>
> The principle very useful for menus, re-useable panels and the like.  If
> others take this up, we can list include (fragment) pages so others can know
> of previously invented wheels.
>
> Aloha,
>         Will.
> _______________________________________________________________________
>   (2006) Information proprietary and confidential intended for
>          direct recipient(s) and mutually agreed co-respondents.
>   http://adroit-process.blogspot.com/
>
> --
> No virus found in this outgoing message.
> Checked by AVG Free Edition.
> Version: 7.1.394 / Virus Database: 268.10.0/388 - Release Date: 13-Jul-2006
>
>
> --
> No virus found in this outgoing message.
> Checked by AVG Free Edition.
> Version: 7.1.394 / Virus Database: 268.10.0/388 - Release Date: 13-Jul-2006
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From john at oxyliquit.de  Fri Jul 14 18:02:05 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 15 Jul 2006 00:02:05 +0200
Subject: [Nitro] Og and DBI
In-Reply-To: <3ff63f9b0607140222y5e9fd775v1a981fddddc3db8c@mail.gmail.com>
References: <3ff63f9b0607140222y5e9fd775v1a981fddddc3db8c@mail.gmail.com>
Message-ID: 

Hi,

> I'm wondering... Wouldn't it make sense to use DBI for all Og sql
> adapters ? DBI already takes care of unifying the different databases,
> so Og could directly benefit from all of them. Since Og is an
> ObjectGraph, the backend is not very important. I also think that it
> would solve a whole bunch of bugs.

I'm sure it would take care of a few bugs, but... in german I'd say
"rein gefuehlsmaessig" (which means "just an unexplainable feeling")
... I somehow never liked to use DBI.  I tried it a few times, but
always kind of retreat from it.  I'm not sure there is a reason.

I know that Og can hide those details from me, but still.

Can there be written a DBI adapter?  If yes, someone could do it and
see how well it works out...

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From john at oxyliquit.de  Fri Jul 14 18:02:08 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 15 Jul 2006 00:02:08 +0200
Subject: [Nitro] nitroproject -- reusable bits
In-Reply-To: <003a01c6a6f7$df8b2950$20af07ca@ghostgum>
References: <003a01c6a6f7$df8b2950$20af07ca@ghostgum>
Message-ID: 

Hi,

> Thanks to George the wiki now has reusable RedCloth macros.  Also wiki
> anchors now go to the wiki pages not & no longer return to nitroproject
> base.

Very cool :D

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From john at oxyliquit.de  Fri Jul 14 18:02:09 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 15 Jul 2006 00:02:09 +0200
Subject: [Nitro] Question about parts
In-Reply-To: <007401c6a72b$b3e1f480$20af07ca@ghostgum>
References: <007401c6a72b$b3e1f480$20af07ca@ghostgum>
Message-ID: 

Hi,

> James & George
>
> I've made an effort to put this structure on the wiki.  Additions are
> welcome.
>
>    http://www.nitroproject.org/wiki/pages/Nitro+Parts
>
> I have set up an alpha-index for a list of Parts when you slot new ones  
> in.

Whoa, nice, definitly! :)

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From james.britt at gmail.com  Fri Jul 14 19:01:38 2006
From: james.britt at gmail.com (James Britt)
Date: Fri, 14 Jul 2006 16:01:38 -0700
Subject: [Nitro] Question about parts
In-Reply-To: 
References: <007401c6a72b$b3e1f480$20af07ca@ghostgum>
	
Message-ID: <44B82252.7020003@gmail.com>

Jonathan Buch wrote:
> Hi,
> 
> 
>>James & George
>>
>>I've made an effort to put this structure on the wiki.  Additions are
>>welcome.
>>
>>   http://www.nitroproject.org/wiki/pages/Nitro+Parts
>>
>>I have set up an alpha-index for a list of Parts when you slot new ones  
>>in.
> 
> 
> Whoa, nice, definitly! :)

Question:  If parts are spread out beyond the an application's root 
folder, how do you package and deploy that application?




-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://www.jamesbritt.com     - Playing with Better Toys
http://www.30secondrule.com   - Building Better Tools

From william.full.moon at gmail.com  Sat Jul 15 03:52:54 2006
From: william.full.moon at gmail.com (* William)
Date: Sat, 15 Jul 2006 17:52:54 +1000
Subject: [Nitro] components in Nitro
In-Reply-To: 
Message-ID: <006101c6a7e3$b14ebce0$bcaf07ca@ghostgum>


Hello all.

I recall this discussion (below) has some sound ideas for a GUI component
framework.  Is a web interface a WUI?

We could do worse than to parallel something like this effort for thick
client GUI framework (see web):

    http://freeride.rubyforge.org/wiki/wiki.pl?GUIFrameworkProject/GUICore

Ignoring why a Ruby wiki is in perl, ... This kind of structure would slot
into the Part structure for example and lends it self to reusable widgets,
controls and buttons (are you with me?).

That's only the core framework.  It makes sense to match on a core profile.
One step at a time.

Also there is some chance for a reusable logical pattern for a WUI and GUI
clients.  So if a GUI framework works with both Ruby apps and web apps, that
give us a lot of leverage.

A small side point before winding up.  It occurs to me that I want a local
Nitro Admin client, not via a web based application.  I'd rather use rsh, X,
or something.  Having said this, it still makes sense to reuse chunks of
code in the Admin Application and share with the User Nitro Web app.  


Several people these days have gui painters that will work AS both a thin or
thick client.  This could be a step towards a more general potentiality.  

Have fun,

   Will.


-----Original Message-----
From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of itsme213
Sent: Friday, 30 June 2006 15:43
To: nitro-general at rubyforge.org
Subject: [Nitro] A uniform structure for components in Nitro -- long post
Importance: Low

Here is something that may have been obvious to others all along. Any and
all comments or flames welcome ...

The context for this ... I was wondering what it would take to give Nitro a
full-fledged component model, where views (and controllers) can be composed
from generic parts either automatically for the default case, or with a DSL
that only specified the delta for the non-default 

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.1/389 - Release Date: 14-Jul-2006
 


From transfire at gmail.com  Sun Jul 16 10:20:30 2006
From: transfire at gmail.com (TRANS)
Date: Sun, 16 Jul 2006 10:20:30 -0400
Subject: [Nitro] REd dollare signs
Message-ID: <4b6f054f0607160720l6cd30621kf3ad80736a89cd1@mail.gmail.com>

What do red dollar signs ($) mean when recording with darcs?

T.

From john at oxyliquit.de  Sun Jul 16 13:24:09 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Sun, 16 Jul 2006 19:24:09 +0200
Subject: [Nitro] REd dollare signs
In-Reply-To: <4b6f054f0607160720l6cd30621kf3ad80736a89cd1@mail.gmail.com>
References: <4b6f054f0607160720l6cd30621kf3ad80736a89cd1@mail.gmail.com>
Message-ID: 

Hi,

> What do red dollar signs ($) mean when recording with darcs?

Those show where the end of line is.  For example you make a few
corrections and leave some whitespace after the line (bad) you
can see it while `darcs record`ing and correct it.

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From transfire at gmail.com  Sun Jul 16 15:04:59 2006
From: transfire at gmail.com (TRANS)
Date: Sun, 16 Jul 2006 15:04:59 -0400
Subject: [Nitro] REd dollare signs
In-Reply-To: 
References: <4b6f054f0607160720l6cd30621kf3ad80736a89cd1@mail.gmail.com>
	
Message-ID: <4b6f054f0607161204k27da14abwe881eaa8b9f364a9@mail.gmail.com>

On 7/16/06, Jonathan Buch  wrote:
> Hi,
>
> > What do red dollar signs ($) mean when recording with darcs?
>
> Those show where the end of line is.  For example you make a few
> corrections and leave some whitespace after the line (bad) you
> can see it while `darcs record`ing and correct it.

Ah cool thanks. Red's a bit ominous though --worried me.

T.

From transfire at gmail.com  Sun Jul 16 16:04:41 2006
From: transfire at gmail.com (TRANS)
Date: Sun, 16 Jul 2006 16:04:41 -0400
Subject: [Nitro] Annotation Notation
Message-ID: <4b6f054f0607161304w46d80355q86b6946361b4b559@mail.gmail.com>

Hi,

On Ruby-talk I posted concerning the difficulties of using
method_missing, esspecially in context Annotations. Ara posted a very
simple implementation that works like so:

   class X
     ann :x, :class => String, :default => 10, :foo => :bar
   end

   X.ann :x, :class      #=> String
   X.ann :x, :default    #=> 10
   X.ann :x, :foo        #=> :bar

To change one.

   X.ann :x, :default => 20

That's it. No method missing stuff at all. Now, no doubt the notation

  X.ann.x.default

has a certain elegance to it that's better, unfortuately there's *a
lot* more complication involved in it's implementation --and it can
NEVER work perfectly. For instance try,

   X.ann :msg, :send => "hello?"

and see where that gets you. And there are other methods that do the
same, plus the classes that are behind that have a potential to bomb
becuase many methods have been forced to be private to get out of the
way of method_missing, if some other tool tryies to use them as
public... well, sorry no can do. You'll have to change your code to
use #send. It really is an ugly world down there :-(

So I was thinking maybe we could transistion to this simpler notation?
At least it's staright-foward with the setting ans getter form being
so simliar:

   X.ann :msg, :send => "hello?"
   X.ann :msg, :send #=> "hello?"

Maybe we can even offer a short-short notation like:

   X^:msg/:send

or

   X-:msg/:send

or

  X<:msg>:send

or something.

T.

From zimba.tm at gmail.com  Sun Jul 16 16:37:38 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Sun, 16 Jul 2006 22:37:38 +0200
Subject: [Nitro] Annotation Notation
In-Reply-To: <4b6f054f0607161304w46d80355q86b6946361b4b559@mail.gmail.com>
References: <4b6f054f0607161304w46d80355q86b6946361b4b559@mail.gmail.com>
Message-ID: <3ff63f9b0607161337s67f7f41alc5d1f94bd8326b86@mail.gmail.com>

Why not leave method_missing and allow that notation for attributes
like send, class,. .. ?

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From george.moschovitis at gmail.com  Sun Jul 16 17:11:40 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 17 Jul 2006 00:11:40 +0300
Subject: [Nitro] Annotation Notation
In-Reply-To: <4b6f054f0607161304w46d80355q86b6946361b4b559@mail.gmail.com>
References: <4b6f054f0607161304w46d80355q86b6946361b4b559@mail.gmail.com>
Message-ID: 

I always thought that your implementation of annotations is a bit
overcomplicated, but I really love the notation:

X.ann.x.default

plus I am certain that there must be an elegant/efficient/robust way
to do this in ruby.

regards,
George.



On 7/16/06, TRANS  wrote:
> Hi,
>
> On Ruby-talk I posted concerning the difficulties of using
> method_missing, esspecially in context Annotations. Ara posted a very
> simple implementation that works like so:
>
>    class X
>      ann :x, :class => String, :default => 10, :foo => :bar
>    end
>
>    X.ann :x, :class      #=> String
>    X.ann :x, :default    #=> 10
>    X.ann :x, :foo        #=> :bar
>
> To change one.
>
>    X.ann :x, :default => 20
>
> That's it. No method missing stuff at all. Now, no doubt the notation
>
>   X.ann.x.default
>
> has a certain elegance to it that's better, unfortuately there's *a
> lot* more complication involved in it's implementation --and it can
> NEVER work perfectly. For instance try,
>
>    X.ann :msg, :send => "hello?"
>
> and see where that gets you. And there are other methods that do the
> same, plus the classes that are behind that have a potential to bomb
> becuase many methods have been forced to be private to get out of the
> way of method_missing, if some other tool tryies to use them as
> public... well, sorry no can do. You'll have to change your code to
> use #send. It really is an ugly world down there :-(
>
> So I was thinking maybe we could transistion to this simpler notation?
> At least it's staright-foward with the setting ans getter form being
> so simliar:
>
>    X.ann :msg, :send => "hello?"
>    X.ann :msg, :send #=> "hello?"
>
> Maybe we can even offer a short-short notation like:
>
>    X^:msg/:send
>
> or
>
>    X-:msg/:send
>
> or
>
>   X<:msg>:send
>
> or something.
>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From transfire at gmail.com  Sun Jul 16 17:49:44 2006
From: transfire at gmail.com (TRANS)
Date: Sun, 16 Jul 2006 17:49:44 -0400
Subject: [Nitro] Annotation Notation
In-Reply-To: <3ff63f9b0607161337s67f7f41alc5d1f94bd8326b86@mail.gmail.com>
References: <4b6f054f0607161304w46d80355q86b6946361b4b559@mail.gmail.com>
	<3ff63f9b0607161337s67f7f41alc5d1f94bd8326b86@mail.gmail.com>
Message-ID: <4b6f054f0607161449o7ede51fh164e5b6780adfc81@mail.gmail.com>

On 7/16/06, Jonas Pfenniger  wrote:
> Why not leave method_missing and allow that notation for attributes
> like send, class,. .. ?

You can't, the problems are multi-fold:

First, certain methods are absolutely fundamental, you can get away
from them, essentially: __id__, __send__, object_id, instance_eval,
dup, and a couple of others. Generally you want to keep anything
starting with 'instance_' too and most operators. If you don't keep
these, there's a good chance something will bomb eventaully.

So even if you let that go and exepct there are some exceptional
"keyword" methods you can use. You have the problem of dynamic method
additions to Kernel and Object, or any subclass for that matter (like
Hash for OpenObject --I just realized that! Aaagh!). If a public
method is added to one of these after you've defined your original
class then that method is going to show up and prevent method_missing
from being called. To get around this I had to create a monitor on
Kerenl and Object's ::method_added callback, such that everytime a
method is added to them, a list of "filtered" classes is looped
through and checked to see if they have special methiod visibility
requirements. That's a hell of lot of overhead for such a minor thing.
(And now with Hash I just mentioned it becomes even worse).

And 3rdly, if that weren't enough, if the subclass is at all
substantial i.e. has a lot of methods, say like OpenObject as a
subclass of Hash, all I can do is make most of them private and hope
for the best. A few of course just can't be private, #each for
instance is too important --but I had to privatize all other
enumerable methods, which suck in itself. And all #to_ methods also
are too important too. So there's lot of exceptions then, but still
lots previously public methods that are now private. What happens is
that any time an OpenObject is the explicit reciever of one of these
now private methods (even if via 'self.'), it will bomb unless you use
send(:meth) instead. -- For instance, I recently snagged the Hash#diff
method from ActiveSupport, it uses 'dup.delete_if' --so that will bomb
on an OpenObject.

So the bottom line is that there is NO robust way to do this --I've
done everything possible in attempting to achieve it. #method_missing
is fine for highly controlled usecases, but as a general open-ended
accessor it is riddled with complications.

I think some serious consideration needs to be given to a new noation.

T.

From transfire at gmail.com  Sun Jul 16 17:51:55 2006
From: transfire at gmail.com (TRANS)
Date: Sun, 16 Jul 2006 17:51:55 -0400
Subject: [Nitro] Annotation Notation
In-Reply-To: <4b6f054f0607161449o7ede51fh164e5b6780adfc81@mail.gmail.com>
References: <4b6f054f0607161304w46d80355q86b6946361b4b559@mail.gmail.com>
	<3ff63f9b0607161337s67f7f41alc5d1f94bd8326b86@mail.gmail.com>
	<4b6f054f0607161449o7ede51fh164e5b6780adfc81@mail.gmail.com>
Message-ID: <4b6f054f0607161451y77aab2e8ld78f126ead4d6a47@mail.gmail.com>

s/are some exceptional "keyword" methods you can use/
are some exceptional "keyword" methods you CAN'T use

From zimba.tm at gmail.com  Mon Jul 17 05:32:07 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Mon, 17 Jul 2006 11:32:07 +0200
Subject: [Nitro] Og entity managing strategy
Message-ID: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>

Hi list,

I would like to change how Og::Manager#manage_classes works. I'm going
to expose the actual and intended implementation. Tell me what you
think.

== Methods
* Resolve target : Find the real classes when you give a symbol for the relation
* Resolve polymorphic markers : If the target of a relation is an
object, the class is
polymorphic. Polymorphic is when you can have relation to various
objects (I think)
* Resolve polymorphic relation : Creates a new specialized class is
the class is polymorphic.
* Resolve names : resolve the key name of the relations. For example
if you don't specify a name, it will choose a plural for has_many and
singular for belongs_to relations.

== Actual implementation

* Look for all manageable classes
* Resolve targets
* Resolve polymorphic markers
* Resolve polymorphic relations

These operations will generate new classes.

So again:
* Look for all manageable classes
* Resolve targets
* Resolve names

The problem with this implementation is that you can't fire up two
managers and select which classes you want to manage with each of
them.

== Proposed implementation

The proposed implementation would be to change that method to make it
recursive. So that when you say, manage X and Y, it will also manage
their dependencies and generated classes. It would also require to
change some of the resolve methods to return the new classes.

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From transfire at gmail.com  Mon Jul 17 08:04:36 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 17 Jul 2006 08:04:36 -0400
Subject: [Nitro] New Development Versions of Facets, Reap and Sow
Message-ID: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>

That took longer then expected!

Here's the word:

  Facets: 1.6.9

I've done a major reanalysis of OpenObject and rewrote it three times.
I finally got it to a point of relative beauty. As I've mentioned
elsewhere, their are irreconcilable difficulties with a class like
this, but I think OpenObject is now just about a good as it can get.
It's also a hell of lot faster than it used to be, as it is now a full
on subclass of Hash.

I rewrote Annotations too. This still has the same interface as before
--that question is still up for consideration. The one noticeable
feature this new version currently lacks is singleton annotations and
the special :self key. I'll add those for the next version.

I entertained adding an explore/ directory for usable but expire
mental libs, but finally decided against it. Figure we may as well
just add them to core or more and if they don't work out, well then we
deprecate them --no need to make them so special.

Bunch of other little things. See ChangeLog:

  http://rubyforge.org/frs/shownotes.php?release_id=6033

Sorry for such ugly changelog entries, but I find it a pain to record
all these dinky changes --I go too fast for revision control? :-(  We
need better ChangeLog system --one that works more like RDoc and can
see *real* changes from source to source.

  Reap 7.2.0 and Sow 0.7.2

(The similar version numbers are a coincidence btw)

With this release all task entries now must be in a tasks: section in
your projectinfo file. Eg.

  ---
  name: foo
  title: Foo
  # etc. ...
  tasks:
    package: !!package
      dir: ../package

    rdoc: !!rdoc
      dir: doc/api

and so on. See http://reap.rubyforge.org for a new example.

Beyond that you can do a couple of cool new things. Firstly you can
move your ProjectInfo file into a meta/ directory. I go ahead anc
downcase it to 'projectinfo' at that point since it doesn't need to
stand out like that any longer. Next you can move all the tasks:
entries to a separate file called task or tasks (either should work).
That's a nice way to SOC. Also remove the version, date and status
entries from projectinfo and instead make a file in meta/ called
'version' and in it put: "number [date] status" Eg.

  1.6.9 [06-07-17 06:37:17] beta

Hmm...I will probably change [] to (), but anyhow...

also, the projectinfo file can now use a revision section:

  revision:
    tool: darcs
    exclude:
      - doc/api

With this you can currently define a !!changelog and a !!version task.
Which create those files respectively. Eg. in task file or in
projectinfo tasks: section.

  version: !!version
    dir: meta

  changelog: !!changelog
    dir: doc

Finally and VERY IMPORTANTLY, you need to add a package: section to
project info that's separate from the package task. Like so:

  package:
    dependencies: [ facets, sow ]
    executables : [ reap, rubytest, seed ]
    exclude:
      - doc/api

  tasks:
    package: !!package
      distribute: [ tar.gz, gem ]
      dir: ../package

The main part holds the "meta" info while the task just describes a
particular packaging action. Sow uses the main package section, and
like task above you can move that section to a separate file called
'package' if you prefer.

Okay, it's working for me, but I'm sure there will be bugs --always
backup your projects first! And let me know what you come across.

Thanks,
T.

Okay next. I got to get these repositories "out there"....

From george.moschovitis at gmail.com  Mon Jul 17 08:40:42 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 17 Jul 2006 15:40:42 +0300
Subject: [Nitro] Og entity managing strategy
In-Reply-To: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
References: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
Message-ID: 

Before you proceed with your changes have a look at the just updated
repo.nitroproject.org repo. I made some changes to the methods you
refer.

Moreover, I will try to release a new version of Nitro in the
following days (ie this week). I think it would be wise to wait 2-3
days until I release and then work on your idea afterwards.

In the meantime, make sure that you use and test the repo version to
help me catch bugs and/or problems.


regards,
George.


On 7/17/06, Jonas Pfenniger  wrote:
> Hi list,
>
> I would like to change how Og::Manager#manage_classes works. I'm going
> to expose the actual and intended implementation. Tell me what you
> think.
>
> == Methods
> * Resolve target : Find the real classes when you give a symbol for the relation
> * Resolve polymorphic markers : If the target of a relation is an
> object, the class is
> polymorphic. Polymorphic is when you can have relation to various
> objects (I think)
> * Resolve polymorphic relation : Creates a new specialized class is
> the class is polymorphic.
> * Resolve names : resolve the key name of the relations. For example
> if you don't specify a name, it will choose a plural for has_many and
> singular for belongs_to relations.
>
> == Actual implementation
>
> * Look for all manageable classes
> * Resolve targets
> * Resolve polymorphic markers
> * Resolve polymorphic relations
>
> These operations will generate new classes.
>
> So again:
> * Look for all manageable classes
> * Resolve targets
> * Resolve names
>
> The problem with this implementation is that you can't fire up two
> managers and select which classes you want to manage with each of
> them.
>
> == Proposed implementation
>
> The proposed implementation would be to change that method to make it
> recursive. So that when you say, manage X and Y, it will also manage
> their dependencies and generated classes. It would also require to
> change some of the resolve methods to return the new classes.
>
> --
> Cheers,
>   zimbatm
>
> http://zimbatm.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Mon Jul 17 17:26:57 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 18 Jul 2006 00:26:57 +0300
Subject: [Nitro] PostgreSQL request
Message-ID: 

Dear devs,

can someone that uses PostgreSQL post some info to the list (and the
wiki) about how he setups Postgres (autorization, socket/tcp, etc
etc)?

thanks in advance,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From ak at navel.gr  Tue Jul 18 02:37:09 2006
From: ak at navel.gr (Anastasios Koutoumanos)
Date: Tue, 18 Jul 2006 09:37:09 +0300
Subject: [Nitro] PostgreSQL request
In-Reply-To: 
References: 
Message-ID: <13906fa50607172337q48c3a62aq20ed2bd357709432@mail.gmail.com>

check on this: http://developer.apple.com/internet/opensource/postgres.html

T.

On 7/18/06, George Moschovitis  wrote:
>
> Dear devs,
>
> can someone that uses PostgreSQL post some info to the list (and the
> wiki) about how he setups Postgres (autorization, socket/tcp, etc
> etc)?
>
> thanks in advance,
> George.
>
> --
> http://www.gmosx.com
> http://www.nitroproject.org
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060718/ca880703/attachment.html 

From zimba.tm at gmail.com  Tue Jul 18 04:17:53 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Tue, 18 Jul 2006 10:17:53 +0200
Subject: [Nitro] New Development Versions of Facets, Reap and Sow
In-Reply-To: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>
References: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>
Message-ID: <3ff63f9b0607180117y32b87776gd138d2b1d75c8da9@mail.gmail.com>

All that looks really great :-)



-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From zimba.tm at gmail.com  Tue Jul 18 04:22:10 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Tue, 18 Jul 2006 10:22:10 +0200
Subject: [Nitro] Og entity managing strategy
In-Reply-To: 
References: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
	
Message-ID: <3ff63f9b0607180122o41ab7631kf81540b15c3d923c@mail.gmail.com>

n 17/07/06, George Moschovitis  wrote:
> Before you proceed with your changes have a look at the just updated
> repo.nitroproject.org repo. I made some changes to the methods you
> refer.

W00t :-)

> Moreover, I will try to release a new version of Nitro in the
> following days (ie this week). I think it would be wise to wait 2-3
> days until I release and then work on your idea afterwards.

Okay

> In the meantime, make sure that you use and test the repo version to
> help me catch bugs and/or problems.

Sure !

Do you see any problem that could arrise with my logic ? The main
concern I've found is with the Glue::Taggable code. The Tag Entity is
shared by all taggable entities. This would mean that all taggable
entities have to be managed by the same manager.

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From aglarond at gmail.com  Tue Jul 18 04:44:53 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Tue, 18 Jul 2006 10:44:53 +0200
Subject: [Nitro] Og entity managing strategy
In-Reply-To: <3ff63f9b0607180122o41ab7631kf81540b15c3d923c@mail.gmail.com>
References: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
	
	<3ff63f9b0607180122o41ab7631kf81540b15c3d923c@mail.gmail.com>
Message-ID: <55c107bf0607180144m1e6ca297m7499e13ff486b430@mail.gmail.com>

On 7/18/06, Jonas Pfenniger  wrote:
>
>
> Do you see any problem that could arrise with my logic ? The main
> concern I've found is with the Glue::Taggable code. The Tag Entity is
> shared by all taggable entities. This would mean that all taggable
> entities have to be managed by the same manager.
>

How would you get around this?  It seems logical to me that you would search
only within one manager (read here as 'store') for all tags,  or you'd have
to implement some kind of search_all_managers in order to retrieve all tags
for a given taggable entity.  Or would your search restrict itself to only
tags within the same store?  If so, what would happen when the same tag is
stored in two different stores?

I've found that the Taggable code is too complex for my uses.  I don't need
a join table to tell me which tags relate to which entities.  I just need a
simple YAML field to hold an array (or hash of arrays for multiple taggable
entities).  I like to have simple things like this stored in flat text files
(or managed by Kirby).

I re-implemented tags in the controller when I was hacking on Nixel.  I've
yet to abstract this out to Taggable itself.  If (when) I get around to
doing this, would you be interested in the code?  Do you see any drawbacks
to the approach outlined above (tag relation stored as YAML)?

- Dimitri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060718/3eca4899/attachment.html 

From manveru at weez-int.com  Tue Jul 18 05:26:24 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Tue, 18 Jul 2006 18:26:24 +0900
Subject: [Nitro] RedCloth dependency for Glue
In-Reply-To: 
References: 
Message-ID: <200607181826.24894.manveru@weez-int.com>

On Friday 14 July 2006 22:35, gabriele renzi wrote:
> Hi folks,
> it seem that nitro (as found at repo.nitroproject.org now) depends on
> RedCloth (through Glue, but possibly the file should go straight into
> nitro) because it loads it and does not handle the case where it is
> missing.

I'm still for dropping that dependency at all...

>
> I think it would be nice if this dependency could be added to the
> Nitro/Glue gem, or otherwise something like the attached patch could be
> applied.
> Basically, the method gets defined conditionally as a message for the
> user, maybe it could just pass the string back, but I think not hiding
> problems is a better approach.

From nusgnaf at gmail.com  Tue Jul 18 05:36:15 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Tue, 18 Jul 2006 05:36:15 -0400
Subject: [Nitro] Beta Nitro gems available: Round 2
In-Reply-To: <55c107bf0607130059s37efa11er19f7ec13b9597573@mail.gmail.com>
References: 
	<55c107bf0607130059s37efa11er19f7ec13b9597573@mail.gmail.com>
Message-ID: <716700c90607180236m64c5fcc8pe13b6172fd887174@mail.gmail.com>

while testing these beta gems under live mode I hit the following bug:
lib/nitro/caching/fragments.rb: "setting" doesn't generate class
variables implicitely, so @@cache should be fixed.
lib/nitro/caching/output.rb: cattr_accessor doesn't initilize default
value, known issue.
please review attached patch.
-------------- next part --------------
=== fragments.rb
==================================================================

--- fragments.rb	(revision 4)
+++ fragments.rb	(revision 5)
@@ -19,11 +19,11 @@
     setting :cache, :default => Glue::MemoryCache.new, :doc => 'The cache used to store the fragments'
     
     def self.get(name, options = {})
-      return @@cache.get(name, options)
+      return self.cache.get(name, options)
     end
     
     def self.put(name, content = nil, options = {})
-      @@cache.put(name, content, options)
+      self.cache.put(name, content, options)
       return content
     end
     
-------------- next part --------------
=== lib/nitro/caching/output.rb
==================================================================
--- lib/nitro/caching/output.rb	(revision 5)
+++ lib/nitro/caching/output.rb	(revision 6)
@@ -6,6 +6,7 @@
 
 module Caching
 
+  setting :output_cache_root, :default => 'public', :doc => 'The directory where cached pages are generated'
   # The Output caching subsystem stores whole pages in the 
   # filesystem to be served directly be the front web server
   # (Lighttpd, Apache, etc) for optimal performance.
@@ -22,9 +23,6 @@
 
     def self.included(base) # :nodoc: 
       base.extend(ClassMethods)
-      base.module_eval do
-        cattr_accessor :output_cache_root, 'public'
-      end
     end
 
     module ClassMethods
@@ -55,7 +53,7 @@
         filename.gsub!(/\/+$/, '')
         # unless req.path is like /go/to.html
         filename << '.html' unless (filename.split('/').last || filename).include? '.'
-        return output_cache_root + '/' + filename                     
+        return Caching.output_cache_root + '/' + filename                     
       end
       
     end

From zimba.tm at gmail.com  Tue Jul 18 05:47:57 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Tue, 18 Jul 2006 11:47:57 +0200
Subject: [Nitro] Og entity managing strategy
In-Reply-To: <55c107bf0607180144m1e6ca297m7499e13ff486b430@mail.gmail.com>
References: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
	
	<3ff63f9b0607180122o41ab7631kf81540b15c3d923c@mail.gmail.com>
	<55c107bf0607180144m1e6ca297m7499e13ff486b430@mail.gmail.com>
Message-ID: <3ff63f9b0607180247n7372a91eweee8ee9540d24ee@mail.gmail.com>

On 18/07/06, Dimitri Aivaliotis  wrote:
> How would you get around this?

If an Entity is already managed, the manager won't take over that
entity unless you force it.

Take two classes :

class Article
  attr_accessor :subject, String
  attr_accessor :body, String
  many_to_many Tag
end

class Tag
  attr_accessor :name, String
end

manager1 = Og::Manager.new
manager2 = Og::Manager.new

manager1.manage_class Tag

# Since Tag doesn't have any relation, it won't try to manage Article

manager2.manage_class Article

# Finds out that Tag has a N<->M relationship with Article but it's
already managed. The in-between generated class is not managed so it
will manage it.

# Thus you can't optimize the queries to get all tags with on article,
but the collection would still be available


-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From zimba.tm at gmail.com  Tue Jul 18 05:58:40 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Tue, 18 Jul 2006 11:58:40 +0200
Subject: [Nitro] Og entity managing strategy
In-Reply-To: <3ff63f9b0607180247n7372a91eweee8ee9540d24ee@mail.gmail.com>
References: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
	
	<3ff63f9b0607180122o41ab7631kf81540b15c3d923c@mail.gmail.com>
	<55c107bf0607180144m1e6ca297m7499e13ff486b430@mail.gmail.com>
	<3ff63f9b0607180247n7372a91eweee8ee9540d24ee@mail.gmail.com>
Message-ID: <3ff63f9b0607180258o4ee64119gdd7f0464891d2b9e@mail.gmail.com>

George, why do you use the command-line to create and destroy Msql databases ?


-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From transfire at gmail.com  Tue Jul 18 07:05:52 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 18 Jul 2006 07:05:52 -0400
Subject: [Nitro] New Development Versions of Facets, Reap and Sow
In-Reply-To: <3ff63f9b0607180117y32b87776gd138d2b1d75c8da9@mail.gmail.com>
References: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>
	<3ff63f9b0607180117y32b87776gd138d2b1d75c8da9@mail.gmail.com>
Message-ID: <4b6f054f0607180405p48cbb4dfuffb056e6a43df253@mail.gmail.com>

Changelog:

  http://rubyforge.org/frs/shownotes.php?release_id=6039

T.

From george.moschovitis at gmail.com  Tue Jul 18 08:41:11 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 18 Jul 2006 15:41:11 +0300
Subject: [Nitro] New Development Versions of Facets, Reap and Sow
In-Reply-To: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>
References: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>
Message-ID: 

> feature this new version currently lacks is singleton annotations and
> the special :self key. I'll add those for the next version.

Hmm, the special :self key is used throughout Nitro/Og. Any workaround?

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Tue Jul 18 08:42:38 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 18 Jul 2006 15:42:38 +0300
Subject: [Nitro] Og entity managing strategy
In-Reply-To: <3ff63f9b0607180258o4ee64119gdd7f0464891d2b9e@mail.gmail.com>
References: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
	
	<3ff63f9b0607180122o41ab7631kf81540b15c3d923c@mail.gmail.com>
	<55c107bf0607180144m1e6ca297m7499e13ff486b430@mail.gmail.com>
	<3ff63f9b0607180247n7372a91eweee8ee9540d24ee@mail.gmail.com>
	<3ff63f9b0607180258o4ee64119gdd7f0464891d2b9e@mail.gmail.com>
Message-ID: 

> George, why do you use the command-line to create and destroy Msql databases ?

Hmm, why not?

is there an alternative?

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From aglarond at gmail.com  Tue Jul 18 09:35:47 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Tue, 18 Jul 2006 15:35:47 +0200
Subject: [Nitro] Og entity managing strategy
In-Reply-To: 
References: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
	
	<3ff63f9b0607180122o41ab7631kf81540b15c3d923c@mail.gmail.com>
	<55c107bf0607180144m1e6ca297m7499e13ff486b430@mail.gmail.com>
	<3ff63f9b0607180247n7372a91eweee8ee9540d24ee@mail.gmail.com>
	<3ff63f9b0607180258o4ee64119gdd7f0464891d2b9e@mail.gmail.com>
	
Message-ID: <55c107bf0607180635yfc0ec72s9f247f7341c85bf6@mail.gmail.com>

On 7/18/06, George Moschovitis  wrote:
>
> > George, why do you use the command-line to create and destroy Msql
> databases ?
>
> Hmm, why not?
>
> is there an alternative?
>

Once you have the connection set up, you could just use SQL:

CREATE DATABASE ;
DROP DATABASE ;

same in PostgreSQL.

- Dimitri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060718/97caad1d/attachment.html 

From transfire at gmail.com  Tue Jul 18 12:05:33 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 18 Jul 2006 12:05:33 -0400
Subject: [Nitro] New Development Versions of Facets, Reap and Sow
In-Reply-To: 
References: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>
	
Message-ID: <4b6f054f0607180905k751542a2qe49b14556f502600@mail.gmail.com>

On 7/18/06, George Moschovitis  wrote:
> > feature this new version currently lacks is singleton annotations and
> > the special :self key. I'll add those for the next version.
>
> Hmm, the special :self key is used throughout Nitro/Og. Any workaround?

I'll fix this evening. No problemo.

We should still further discuss notation. I came across a use case
last night that highlights something Ara was saying. I've managed
Cut-based AOP system in pure Ruby. In so doing I added a more versitle
joining system whcih would work wonders with annotations. But
method_missing leads too:

  cut :C < X do
    pointcut { |meth|  ann.send(meth).somekey ...
  end

And if for some reason my key was a varaible too, it would be worse:

  cut :C < X do
    pointcut { |meth|  ann.send(meth).send(key) ...
  end

So that highlight's the need for a lookup method vs. using send, which
is not only less efficient, in Ruby 1.8 it won't even work! One has to
do:

  cut :C < X do
    pointcut { |meth|  ann(meth)[key]
  end

But as it is [] doesn't provide annotation "heritage", so that doesn't
work either --maybe that can be "fixed" but I'll have to look at it.

T.

From george.moschovitis at gmail.com  Tue Jul 18 13:29:35 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 18 Jul 2006 20:29:35 +0300
Subject: [Nitro] New Development Versions of Facets, Reap and Sow
In-Reply-To: <4b6f054f0607180905k751542a2qe49b14556f502600@mail.gmail.com>
References: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>
	
	<4b6f054f0607180905k751542a2qe49b14556f502600@mail.gmail.com>
Message-ID: 

Ergh ;-) This sounds gree.. erm chinese to me ;-)

-g.


On 7/18/06, TRANS  wrote:
> On 7/18/06, George Moschovitis  wrote:
> > > feature this new version currently lacks is singleton annotations and
> > > the special :self key. I'll add those for the next version.
> >
> > Hmm, the special :self key is used throughout Nitro/Og. Any workaround?
>
> I'll fix this evening. No problemo.
>
> We should still further discuss notation. I came across a use case
> last night that highlights something Ara was saying. I've managed
> Cut-based AOP system in pure Ruby. In so doing I added a more versitle
> joining system whcih would work wonders with annotations. But
> method_missing leads too:
>
>   cut :C < X do
>     pointcut { |meth|  ann.send(meth).somekey ...
>   end
>
> And if for some reason my key was a varaible too, it would be worse:
>
>   cut :C < X do
>     pointcut { |meth|  ann.send(meth).send(key) ...
>   end
>
> So that highlight's the need for a lookup method vs. using send, which
> is not only less efficient, in Ruby 1.8 it won't even work! One has to
> do:
>
>   cut :C < X do
>     pointcut { |meth|  ann(meth)[key]
>   end
>
> But as it is [] doesn't provide annotation "heritage", so that doesn't
> work either --maybe that can be "fixed" but I'll have to look at it.
>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From zimba.tm at gmail.com  Tue Jul 18 13:42:12 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Tue, 18 Jul 2006 19:42:12 +0200
Subject: [Nitro] Og entity managing strategy
In-Reply-To: 
References: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
	
	<3ff63f9b0607180122o41ab7631kf81540b15c3d923c@mail.gmail.com>
	<55c107bf0607180144m1e6ca297m7499e13ff486b430@mail.gmail.com>
	<3ff63f9b0607180247n7372a91eweee8ee9540d24ee@mail.gmail.com>
	<3ff63f9b0607180258o4ee64119gdd7f0464891d2b9e@mail.gmail.com>
	
Message-ID: <3ff63f9b0607181042j86ef99bx208b44f6ed6c87a4@mail.gmail.com>

On 18/07/06, George Moschovitis  wrote:
> > George, why do you use the command-line to create and destroy Msql databases ?
>
> Hmm, why not?

Because the mysql command-line tool is not necessarily available.
Since mysql-ruby is, it's most safe to only use that one.

> is there an alternative?

The CREATE DATABASE and DROP DATABASE sql commands

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From george.moschovitis at gmail.com  Tue Jul 18 13:48:38 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 18 Jul 2006 20:48:38 +0300
Subject: [Nitro] Og entity managing strategy
In-Reply-To: <3ff63f9b0607181042j86ef99bx208b44f6ed6c87a4@mail.gmail.com>
References: <3ff63f9b0607170232k1c9b2ab3h59b80bcf05121e83@mail.gmail.com>
	
	<3ff63f9b0607180122o41ab7631kf81540b15c3d923c@mail.gmail.com>
	<55c107bf0607180144m1e6ca297m7499e13ff486b430@mail.gmail.com>
	<3ff63f9b0607180247n7372a91eweee8ee9540d24ee@mail.gmail.com>
	<3ff63f9b0607180258o4ee64119gdd7f0464891d2b9e@mail.gmail.com>
	
	<3ff63f9b0607181042j86ef99bx208b44f6ed6c87a4@mail.gmail.com>
Message-ID: 

Ok, please add this to the Tickets section of np.org and I will fix it.

thanks,
George.

On 7/18/06, Jonas Pfenniger  wrote:
> On 18/07/06, George Moschovitis  wrote:
> > > George, why do you use the command-line to create and destroy Msql databases ?
> >
> > Hmm, why not?
>
> Because the mysql command-line tool is not necessarily available.
> Since mysql-ruby is, it's most safe to only use that one.
>
> > is there an alternative?
>
> The CREATE DATABASE and DROP DATABASE sql commands
>
> --
> Cheers,
>   zimbatm
>
> http://zimbatm.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Tue Jul 18 13:53:04 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 18 Jul 2006 20:53:04 +0300
Subject: [Nitro] Unified Repo
Message-ID: 

Dear devs,

I would like to go back to a unified repo in the following days. I
will probably have fully working mysql/sqlite/adapters ready in 1-2
days. So I would like to ask bryan help me merge missing patches to my
repo. And I would also like to ask more of you do download and
experiment with my repo over the following days. Please do send
patches against my version of the repo.

I will be able to look after repo.nitroproject.org on a day-to-day basis.

regards,
George.


-- 
http://www.gmosx.com
http://www.nitroproject.org

From john at oxyliquit.de  Wed Jul 19 06:03:58 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Wed, 19 Jul 2006 12:03:58 +0200
Subject: [Nitro] Devlab gems
Message-ID: 

Hi,

what happened to the devlab gems?  I think they should be posted
as the new stable on rubyforge before the new branch from George
gets stable to feed the masses (0.30.5 ?) :)

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From aidan at infurious.com  Wed Jul 19 06:17:16 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Wed, 19 Jul 2006 11:17:16 +0100
Subject: [Nitro] Devlab gems
In-Reply-To: 
References: 
Message-ID: 

+1

On 19/07/2006, at 11:03 AM, Jonathan Buch wrote:

> Hi,
>
> what happened to the devlab gems?  I think they should be posted
> as the new stable on rubyforge before the new branch from George
> gets stable to feed the masses (0.30.5 ?) :)

From aidan at infurious.com  Wed Jul 19 06:21:49 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Wed, 19 Jul 2006 11:21:49 +0100
Subject: [Nitro] Bug in mongrel adapter (plus fix)
Message-ID: <33DECE09-1909-4386-8C53-2C270DA5A966@infurious.com>

Hi all,

When you send a large post to Nitro via Mongrel, Mongrel saves the  
data in a Tempfile.  The mongrel adapter didn't take that into  
consideration.  Here's a fix:

--- old_mongrel.rb      Wed Jul 19 06:19:25 2006
+++ mongrel.rb  Wed Jul 19 06:13:33 2006
@@ -123,6 +123,8 @@
          context.in = if req.body.is_a? StringIO
                         req.body
+                     elsif req.body.kind_of?(Tempfile)
+                       req.body.to_io
                       else
                         StringIO.new(req.body.to_s)
                       end

hope this helps others :-)

Aidan

From nusgnaf at gmail.com  Wed Jul 19 07:54:28 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Wed, 19 Jul 2006 07:54:28 -0400
Subject: [Nitro] Bug in mongrel adapter (plus fix)
In-Reply-To: <33DECE09-1909-4386-8C53-2C270DA5A966@infurious.com>
References: <33DECE09-1909-4386-8C53-2C270DA5A966@infurious.com>
Message-ID: <716700c90607190454g6ea5ee4bofdf34576cb6ca5aa@mail.gmail.com>

>From previous discussion, mongrel will pass req.body as StringIO or
Tempfile, which we can directly pass to contex wrapper. I have using
the attatched patch without problem, please review it.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mongrel.patch
Type: application/octet-stream
Size: 696 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060719/c9230278/attachment-0001.obj 

From nusgnaf at gmail.com  Wed Jul 19 08:01:11 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Wed, 19 Jul 2006 08:01:11 -0400
Subject: [Nitro] [PATCH]caching fragment fix
Message-ID: <716700c90607190501w4db8f3aeq4bad9dcba43050f5@mail.gmail.com>

Kill the usage of @@cache in lib/nitro/caching/fragments.rb.Use
self.cache instead.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 20060719225854-3674d-d1ab1f14a659574235b68b33ff4812a95db8f35a
Type: application/octet-stream
Size: 400 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060719/b4c22ab8/attachment.obj 

From aidan at infurious.com  Wed Jul 19 08:09:40 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Wed, 19 Jul 2006 13:09:40 +0100
Subject: [Nitro] Bug in mongrel adapter (plus fix)
References: <695F3E93-2A9A-43E2-B4B6-99E506EDB086@yoyo.org>
Message-ID: <6D017366-0C68-49E3-8F20-FABAE591DD75@infurious.com>

Sorry, I must have missed that discussion thread.  I like your fix  
much better :-)

Aidan

On 19/07/2006, at 12:54 PM, Fang Sun wrote:

>> From previous discussion, mongrel will pass req.body as StringIO or
> Tempfile, which we can directly pass to contex wrapper. I have using
> the attatched patch without problem, please review it.
> 
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From george.moschovitis at gmail.com  Wed Jul 19 11:30:05 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 19 Jul 2006 18:30:05 +0300
Subject: [Nitro] Bug in mongrel adapter (plus fix)
In-Reply-To: <716700c90607190454g6ea5ee4bofdf34576cb6ca5aa@mail.gmail.com>
References: <33DECE09-1909-4386-8C53-2C270DA5A966@infurious.com>
	<716700c90607190454g6ea5ee4bofdf34576cb6ca5aa@mail.gmail.com>
Message-ID: 

Please, in the future send us patch bundles using dacrs against
repo.nitroproject.org

thanks in advance,
George.

On 7/19/06, Fang Sun  wrote:
> >From previous discussion, mongrel will pass req.body as StringIO or
> Tempfile, which we can directly pass to contex wrapper. I have using
> the attatched patch without problem, please review it.
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Wed Jul 19 11:36:49 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 19 Jul 2006 18:36:49 +0300
Subject: [Nitro] Devlab gems
In-Reply-To: 
References: 
	
Message-ID: 

Hmm, if Bryan provides me with an updated version of RELEASES I for
his changes I could release that as 0.31.0

My version will be released as 0.40.0 over the following days.

regards,
George.


On 7/19/06, Aidan Rogers  wrote:
> +1
>
> On 19/07/2006, at 11:03 AM, Jonathan Buch wrote:
>
> > Hi,
> >
> > what happened to the devlab gems?  I think they should be posted
> > as the new stable on rubyforge before the new branch from George
> > gets stable to feed the masses (0.30.5 ?) :)
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From transfire at gmail.com  Wed Jul 19 15:31:32 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 19 Jul 2006 15:31:32 -0400
Subject: [Nitro] New Development Versions of Facets, Reap and Sow
In-Reply-To: 
References: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>
	
	<4b6f054f0607180905k751542a2qe49b14556f502600@mail.gmail.com>
	
Message-ID: <4b6f054f0607191231t253ba96ewd5cc75abac1baf9f@mail.gmail.com>

Crap. Hadn't tried to build a gem lately and I've just releaized the
the Sow/Reap Gem builder is broke.

FYI.

I'm working on it...

T.

From george.moschovitis at gmail.com  Wed Jul 19 15:56:52 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 19 Jul 2006 22:56:52 +0300
Subject: [Nitro] New Development Versions of Facets, Reap and Sow
In-Reply-To: <4b6f054f0607191231t253ba96ewd5cc75abac1baf9f@mail.gmail.com>
References: <4b6f054f0607170504n6b788d30ua5be17fbba7ec0a7@mail.gmail.com>
	
	<4b6f054f0607180905k751542a2qe49b14556f502600@mail.gmail.com>
	
	<4b6f054f0607191231t253ba96ewd5cc75abac1baf9f@mail.gmail.com>
Message-ID: 

Good...

On 7/19/06, TRANS  wrote:
> Crap. Hadn't tried to build a gem lately and I've just releaized the
> the Sow/Reap Gem builder is broke.
>
> FYI.
>
> I'm working on it...
>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From bryan.a.soto at gmail.com  Wed Jul 19 16:12:30 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 19 Jul 2006 13:12:30 -0700
Subject: [Nitro] Devlab gems
In-Reply-To: 
References: 
Message-ID: 

On 7/19/06, Jonathan Buch  wrote:
> Hi,
>
> what happened to the devlab gems?  I think they should be posted
> as the new stable on rubyforge before the new branch from George
> gets stable to feed the masses (0.30.5 ?) :)
>

Still there. :)

I just need to get these latest cattr fixes in and update the blog
example (also cattr). Other than that, it should be ready to go.

From bryan.a.soto at gmail.com  Wed Jul 19 20:01:58 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 19 Jul 2006 17:01:58 -0700
Subject: [Nitro] Devlab gems
In-Reply-To: 
References: 
Message-ID: 

On 7/19/06, Jonathan Buch  wrote:
> Hi,
>
> what happened to the devlab gems?  I think they should be posted
> as the new stable on rubyforge before the new branch from George
> gets stable to feed the masses (0.30.5 ?) :)
>

Okay, I've added the recent patches moving from cattr to setting.
Other than the error messages when generating the rdoc/ri stuff, I'm
not aware of any other issues. I've also removed two examples (wee and
flash; wee as Nitro apparently doesn't support it anymore and flash as
ming-ruby has been untouched for two years and is out of date) and
verified the rest of them with mongrel and webrick.

Gems available as usual via

# gem update -y --source=http://devlab.oree.ch/~bryan/gems

Examples are also available at http://devlab.oree.ch/~bryan

Barring show-stoppers, this is what will be released, hopefully tomorrow.

I would like some advice as to the version number though. Should I go
with 30.5 as Kashia suggested? I'm not sure there's been enough done
to qualify as a 0.31.0 release. Any opinions?

Thanks,

bryan

PS - for the interested, here are the full changes since the 0.30 release.

Wed Jul 19 16:17:27 PDT 2006  bryan.a.soto at gmail.com
  * nitro-caching-fixes
  Updates Nitro caching classes from cattr to settings.

Mon Jul 17 13:50:50 PDT 2006  bryan.a.soto at gmail.com
  * bugfix-glue-validate_value
  Ticket #44: http://devlab.oree.ch/trac/glycerin/ticket/44
  Credit to matt DOT moriarity AT gmail DOT com.

Sun Jul 16 23:06:48 PDT 2006  bryan.a.soto at gmail.com
  * bugfix-nitro-mongrel-adapter
  Fixes multi-part form processing, which was broken by my previous
patch. Basically, image uploads work now for mongrel.

Tue Jul 11 13:10:06 PDT 2006  bryan.a.soto at gmail.com
  * bugfix-nitro-flaky-template-root
  When running from a directory other than the root directory of your
Nitro app, Template.root was set incorrectly because class Template
would determine whether to use template, src/template and default to
public all before Nitro had chdir'd to the root directory of your
Nitro app. Moved the requires to after the chdir.

Tue Jul 11 13:13:47 PDT 2006  bryan.a.soto at gmail.com
  * bugfix-nitro-extra-arguments
  Commented out the extra concat as I never could figure out what the
code intended.

Tue Jul 11 11:43:29 PDT 2006  neokolor at gmx.de
  * add option to set nav link titles

Fri Jul  7 16:25:48 PDT 2006  neokolor at gmx.de
  * helper/table.rb: fix problem with table output

Fri Jul  7 16:08:29 PDT 2006  neokolor at gmx.de
  * helper/pager.rb: fix problems with nav output if we have no results

Fri Jul  7 16:18:08 PDT 2006  neokolor at gmx.de
  * helper/pager.rb: add option to set nav link titles

Sun Jun 18 05:42:52 PDT 2006  neokolor at gmx.de
  * fix_in_relation_symbol_to_class

Mon Jul 10 18:07:02 PDT 2006  bryan.a.soto at gmail.com
  * bugfix-nitro-ticket-43
  Fixed a bug/type as suggested by rff_rff AT yahoo DOT it
  http://devlab.oree.ch/trac/glycerin/ticket/43

Mon Jul 10 18:03:50 PDT 2006  bryan.a.soto at gmail.com
  * bugfix-nitro-ticket-34
  Fixed as per billk AT cts DOT com suggestion on Ticket #34.
  http://devlab.oree.ch/trac/glycerin/ticket/34

Mon Jul 10 17:56:56 PDT 2006  bryan.a.soto at gmail.com
  * og-mysql-add-port-option
  From jbritt, makes Mysql respect port option when creating and
dropping databases.

Fri Jun 30 12:07:11 PDT 2006  George Moschovitis 
  * Applied james britt's mongrel patch.

Thu Jun 29 17:01:35 PDT 2006  bryan.a.soto at gmail.com
  * dispatcher-add_rule-patch
  Duplicate of George's add_route to add_rule patch as his repo was unavailable.

Thu Jun 29 16:31:46 PDT 2006  bryan.a.soto at gmail.com
  * og-tc-sti-find
  Adds a test case for manveru's sti fix.

Thu Jun 29 16:29:26 PDT 2006  bryan.a.soto at gmail.com
  * og-entity-from-string
  Modifies the definition of Og::Entity::entity_from_string slightly.
It assumed only one manager would ever exist. Not a valid assumption
in the test suite.

Thu Jun 29 15:10:18 PDT 2006  bryan.a.soto at gmail.com
  * og-thread-safe-options
  Adds the ability to specify :thread_safe as an option to Og.setup
with the default begin true. Without this patch, thread_safe was true
and it couldn't be modified without modifying the Og code.

Thu Jun 29 15:08:58 PDT 2006  bryan.a.soto at gmail.com
  * og-aggregate-fix
  Accomodates the fact that the Postgres adapter doesn't return an
Array like the other stores by modifying the aggregation code.

Tue Jun 27 01:54:47 PDT 2006  manveru at weez-int.com
  * fix for schema_inheritance queries

Thu Apr 20 03:08:59 PDT 2006  m.fellinger at gmail.com
  * localization-support for table-helper

Wed Jun 21 17:06:22 PDT 2006  bryan.a.soto at gmail.com
  * update-facets-1.4.2
  Updates to Facets-1.4.2. Uses facets/core/module/class_extension
instead of classinherit. No other changes needed to pass tests.

Sun Jun 18 15:08:38 PDT 2006  neokolor at gmx.de
  * fix problem with :plural_name in admin controller

Fri Jun  9 15:06:20 PDT 2006  bryan.a.soto at gmail.com
  * hack-enh-fcgi-rewind
  A small hack to ensure that you can access the raw_body of a
request, by converting the incoming stream to a StringIO if it doesn't
respond to :rewind.

Thu Jun  8 13:19:40 PDT 2006  bryan.a.soto at gmail.com
  * nitro-sendfile-support
  Adds sendfile supported. Submitted by Jan A. Lambert . Tested on IE 6.0 and FireFox 1.5.

Wed Jun  7 19:02:04 PDT 2006  bryan.a.soto at gmail.com
  * nitro-dispatcher-router-strip_path
  Makes the dispatcher honour the Router.strip_path setting. This,
basically, makes Nitro ignore a portion of the url that could
represent a directory name.

  Basically, without the setting:
  /my_nitro_apps_directory/nitro-app => looks for a controller mapped
to / or /my_nitro_apps_directory.

  With Router.strip_path = '/my_nitro_apps_directory'
  /my_nitro_apps_directory/nitro-app => looks for a controller mapped
to / or /nitro-app.

  Many thanks to Vagabond for the patch.


Mon Jun  5 14:11:52 PDT 2006  bryan.a.soto at gmail.com
  * bugfix-mongrel-adapter-request-uri
  Removes REQUEST_URI header manipulation and accomodates an upcoming
change in Mongrel 0.3.13. An ugly class check... Lovely.

Wed May 17 17:47:17 PDT 2006  bryan.a.soto at gmail.com
  * fixup-guilles-patch
  Brings Guille's patch up to date by resolving a conflict.

Fri May 12 02:48:22 PDT 2006  Guillaume Pierronnet

  * og/lib/glue/taggable.rb refactor + support for passing :condition parameter
  SqlStore#prepare_statement moved to public scope
  test case on taggable runs fine

Wed May 17 17:14:36 PDT 2006  bryan.a.soto at gmail.com
  * test-fix-og-tc_ez
  Fixes up the ez testcase Manveru submitted to make it run and pass.

Sat May 13 04:10:40 PDT 2006  m.fellinger at gmail.com
  * ez-fix for trueclass and testcase for it

Sat May 13 04:09:45 PDT 2006  m.fellinger at gmail.com
  * error-page-fix for source-preview

Wed May 17 13:01:53 PDT 2006  fabian at fabian-buch.de
  * to_rexml and to_xml

  to_rexml returns a REXML::Element object
  to_xml returns an XML-String


Wed May 17 07:33:23 PDT 2006  fabian at fabian-buch.de
  * introduction of OgObject.to_xml method

  to_xml returns a REXML object of an Og object

  example usage:
    User[1].to_xml

  returns a REXML object, so use to_s to get it as String containing XML


Wed May 17 15:32:59 PDT 2006  bryan.a.soto at gmail.com
  * test-fix-nitro-tc_render
  Makes the TestController subclass Nitro::Controller so it get's the
#encode_url method.

Wed May 17 15:24:52 PDT 2006  bryan.a.soto at gmail.com
  * adds-some-blank-lines-to-og-tc_setup

Wed May 17 15:17:25 PDT 2006  bryan.a.soto at gmail.com
  * gabrielle-renzis-linefeeds
  Switches line endings from Dos to Unix.

Fri May  5 05:30:51 PDT 2006  surrender_it at yahoo.it
  * crash-early Og.setup
  This patch removed the Exception handling routine from Og.setup.
  With the current code if an exception is raised during the execution of
  the method there is no crash, but the crash will happen every time
someone references Og.manager, cause that variable won't be
initialized .
  This patch makes Og.setup crash as soon as possible, thus allowing
faster diagnostics.

  The patch also adds a simple test case for Og.setup, which seem not
tested anywhere, but the test case may need some love  for proper
setup (it relies on glycerin as of now, cause loading CONFIG.rb seem
to raise a strange exception related to mixed use of facets 1.3 and
1.0 on my box)


Thu May  4 18:02:06 PDT 2006  surrender_it at yahoo.it
  * better test logging
  As of now tests for Og swallow every error ingormation while not in $DBG mode,
  this one-line patch allows visualization of logged informations at
WARN level and more, which has little impact on the visualization of
the tests and still allows quick diagnostics while staying out of $DBG

Tue May 16 13:37:04 PDT 2006  bryan.a.soto at gmail.com
  * george-glue-tag-fixup

Tue May 16 08:07:05 PDT 2006  George Moschovitis 
  * Minor.

Tue May 16 00:32:46 PDT 2006  George Moschovitis 
  * Moved examples, spark, flare to a separate repository.

Tue May 16 00:28:23 PDT 2006  George Moschovitis 
  * Removed wee helper and example. Not really useful in Nitro, better
use wee standalone + Og if you like Wee components (or wait for a
Nitro solution).

Tue May 16 00:23:50 PDT 2006  George Moschovitis 
  * Removed controller_name from Render, user controller.mount_path instead.

Tue May 16 00:16:15 PDT 2006  George Moschovitis 
  * Fixed: dont overwrite already defined methods in the new scaffold code.

Tue May 16 00:15:08 PDT 2006  George Moschovitis 
  * Improved R (encode_url) handles strings and entities. Redirect
implicitly uses R. Form builder implicitly uses R for form actions.

Sun May 14 23:54:39 PDT 2006  George Moschovitis 
  * Fixed name_to_jsfile (absolute url).

Fri May 12 21:22:51 PDT 2006  George Moschovitis 
  * Reenabled format in Logger.

Fri May 12 20:49:29 PDT 2006  George Moschovitis 
  * Added some :nodoc: all to cleanup generated RDocs.

Fri May 12 08:48:49 PDT 2006  George Moschovitis 
  * Bumbed version number in many files.

Fri May 12 08:37:54 PDT 2006  George Moschovitis 
  * Wrote a detailed parts howto in RDoc form in parts.rb as requested
by Jonas. To be improved after community feedback.

Thu May 11 03:14:56 PDT 2006  George Moschovitis 
  * Initial code for the new scaffolding system, *very* under
construction, ignore for the moment.

Wed May 10 03:28:36 PDT 2006  George Moschovitis 
  * Added a model macro to controllers that links related classes. At
the moment it is used in the sweeper do delete cached pages from the
linked controller base directory.

Thu May 11 23:05:11 PDT 2006  bryan.a.soto at gmail.com
  * add-ostruct-require
  Adds missing ostruct require to testcase.rb.

Wed May 10 15:20:51 PDT 2006  bryan.a.soto at gmail.com
  * more-mongrel-adapter-fixes
  Fixes up static file serving.

Tue May  9 23:05:46 PDT 2006  bryan.a.soto at gmail.com
  * nitro-proto-page
  Updates the intro page with a link to the rubyforge page for
examples to be downloaded as suggested by James Britt.

Tue May  9 23:04:39 PDT 2006  bryan.a.soto at gmail.com
  * mongrel-fixes
  Gives the Mongrel adapter some need updating.

Mon May  8 23:36:43 PDT 2006  bryan.a.soto at gmail.com
  * test-suite-fixup
  Fixes the full test suite so that .\script\test.rb completes.

Mon May  8 00:56:45 PDT 2006  George Moschovitis 
  * Fixed Session.current and Controller.current

Sun May  7 09:07:15 PDT 2006  George Moschovitis 
  * Store current context in a thread local variable, accesible
through Context.current

Sun May  7 01:57:03 PDT 2006  George Moschovitis 
  * Removed glue/attribute, use Facet's cattr instead. Updated
nitro/og source to make compatible.

Sat May  6 23:20:34 PDT 2006  George Moschovitis 
  * Added script adapter file.

Sat May  6 23:15:55 PDT 2006  George Moschovitis 
  * Removed flexob and references to it, use Facet's OpenObject instead.

Sat May  6 03:39:21 PDT 2006  George Moschovitis 
  * Renamed to ScriptAdapter, keep ConsoleAdapter alias.

Sat May  6 03:20:10 PDT 2006  George Moschovitis 
  * Yeah, the ConsoleAdapter kinda works now ;-)

Sat May  6 02:55:03 PDT 2006  George Moschovitis 
  * Introduced new nitro command/runner that will replace the current
spaghetti code. Based on facetes Console::Command. Also introduced the
ConsoleAdapter.

Sat May  6 01:00:52 PDT 2006  George Moschovitis 
  * Removed accumulate (in facets).

Sat May  6 00:55:26 PDT 2006  George Moschovitis 
  * Some small updates.

Sat May  6 00:50:53 PDT 2006  George Moschovitis 
  tagged 0.30.0

From transfire at gmail.com  Wed Jul 19 22:44:04 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 19 Jul 2006 22:44:04 -0400
Subject: [Nitro] Devlab gems
In-Reply-To: 
References: 
	
Message-ID: <4b6f054f0607191944jef93354vac1cfd966ade0f6f@mail.gmail.com>

On 7/19/06, Bryan Soto  wrote:
> On 7/19/06, Jonathan Buch  wrote:
> > Hi,
> >
> > what happened to the devlab gems?  I think they should be posted
> > as the new stable on rubyforge before the new branch from George
> > gets stable to feed the masses (0.30.5 ?) :)
> >
>
> Okay, I've added the recent patches moving from cattr to setting.
> Other than the error messages when generating the rdoc/ri stuff, I'm
> not aware of any other issues. I've also removed two examples (wee and
> flash; wee as Nitro apparently doesn't support it anymore and flash as
> ming-ruby has been untouched for two years and is out of date) and
> verified the rest of them with mongrel and webrick.
>
> Gems available as usual via
>
> # gem update -y --source=http://devlab.oree.ch/~bryan/gems
>
> Examples are also available at http://devlab.oree.ch/~bryan
>
> Barring show-stoppers, this is what will be released, hopefully tomorrow.
>
> I would like some advice as to the version number though. Should I go
> with 30.5 as Kashia suggested? I'm not sure there's been enough done
> to qualify as a 0.31.0 release. Any opinions?

Don't sweat the small stuff. It's a release so go to 0.31.

Good work btw.

T.

From manveru at weez-int.com  Thu Jul 20 02:36:31 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Thu, 20 Jul 2006 15:36:31 +0900
Subject: [Nitro] Devlab gems
In-Reply-To: <4b6f054f0607191944jef93354vac1cfd966ade0f6f@mail.gmail.com>
References: 
	
	<4b6f054f0607191944jef93354vac1cfd966ade0f6f@mail.gmail.com>
Message-ID: <200607201536.31237.manveru@weez-int.com>

On Thursday 20 July 2006 11:44, TRANS wrote:
> On 7/19/06, Bryan Soto  wrote:
> > On 7/19/06, Jonathan Buch  wrote:
> > > Hi,
> > >
> > > what happened to the devlab gems?  I think they should be posted
> > > as the new stable on rubyforge before the new branch from George
> > > gets stable to feed the masses (0.30.5 ?) :)
> >
> > Okay, I've added the recent patches moving from cattr to setting.
> > Other than the error messages when generating the rdoc/ri stuff, I'm
> > not aware of any other issues. I've also removed two examples (wee and
> > flash; wee as Nitro apparently doesn't support it anymore and flash as
> > ming-ruby has been untouched for two years and is out of date) and
> > verified the rest of them with mongrel and webrick.
> >
> > Gems available as usual via
> >
> > # gem update -y --source=http://devlab.oree.ch/~bryan/gems
> >
> > Examples are also available at http://devlab.oree.ch/~bryan
> >
> > Barring show-stoppers, this is what will be released, hopefully tomorrow.
> >
> > I would like some advice as to the version number though. Should I go
> > with 30.5 as Kashia suggested? I'm not sure there's been enough done
> > to qualify as a 0.31.0 release. Any opinions?
>
> Don't sweat the small stuff. It's a release so go to 0.31.

Exactly... we had releases with less new features already :)
though i really would like to see 0.32.0 (indicating it's a more or less 
stable release...)
there will never be any 0.3x.x again - so you can pick whatever you want... 
next stop - 0.40.0

>
> Good work btw.
>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

From nusgnaf at gmail.com  Thu Jul 20 04:57:10 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 20 Jul 2006 04:57:10 -0400
Subject: [Nitro] Bug in mongrel adapter (plus fix)
In-Reply-To: <6D017366-0C68-49E3-8F20-FABAE591DD75@infurious.com>
References: <695F3E93-2A9A-43E2-B4B6-99E506EDB086@yoyo.org>
	<6D017366-0C68-49E3-8F20-FABAE591DD75@infurious.com>
Message-ID: <716700c90607200157i1d44e79tc260312b5377d488@mail.gmail.com>

Hi, the discussion archieve is here
http://rubyforge.org/pipermail/nitro-general/2006-July/005294.html

On 7/19/06, Aidan Rogers  wrote:
> Sorry, I must have missed that discussion thread.  I like your fix
> much better :-)
>
> Aidan
>
> On 19/07/2006, at 12:54 PM, Fang Sun wrote:
>
> >> From previous discussion, mongrel will pass req.body as StringIO or
> > Tempfile, which we can directly pass to contex wrapper. I have using
> > the attatched patch without problem, please review it.
> > 
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>

From nusgnaf at gmail.com  Thu Jul 20 05:25:28 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 20 Jul 2006 05:25:28 -0400
Subject: [Nitro] Bug in mongrel adapter (plus fix)
In-Reply-To: 
References: <33DECE09-1909-4386-8C53-2C270DA5A966@infurious.com>
	<716700c90607190454g6ea5ee4bofdf34576cb6ca5aa@mail.gmail.com>
	
Message-ID: <716700c90607200225p4ab02131te24d404886c7a198@mail.gmail.com>

Hi, George, after reading the wonderful darcs usage tips on Oxyliquit
site, I am much clear now. Will use darcs to follow lastest
development.

On 7/19/06, George Moschovitis  wrote:
> Please, in the future send us patch bundles using dacrs against
> repo.nitroproject.org
>
> thanks in advance,
> George.

From nusgnaf at gmail.com  Thu Jul 20 05:27:44 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 20 Jul 2006 05:27:44 -0400
Subject: [Nitro] [PATCH]caching fragment fix
In-Reply-To: <716700c90607190501w4db8f3aeq4bad9dcba43050f5@mail.gmail.com>
References: <716700c90607190501w4db8f3aeq4bad9dcba43050f5@mail.gmail.com>
Message-ID: <716700c90607200227u33ab347dq33e180e11bd284f2@mail.gmail.com>

Hi, regenerate the patch use darcs against lastest repo, please review.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fragments.fix.tar.gz
Type: application/x-gzip
Size: 4307 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060720/c34eb102/attachment.gz 

From john at oxyliquit.de  Thu Jul 20 06:13:28 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Thu, 20 Jul 2006 12:13:28 +0200
Subject: [Nitro] Devlab gems
In-Reply-To: 
References: 
	
Message-ID: 

Hi,

> Barring show-stoppers, this is what will be released, hopefully tomorrow.

Very good, this will make some people happy :P

> I would like some advice as to the version number though. Should I go
> with 30.5 as Kashia suggested? I'm not sure there's been enough done
> to qualify as a 0.31.0 release. Any opinions?

Go for 0.31.0 (reason: read further :P)

> PS - for the interested, here are the full changes since the 0.30  
> release.

[Scroll down for hours]

Now I see that this is really a 0.31 or even 0.32 instead of 0.30.5 ;)

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From john at oxyliquit.de  Thu Jul 20 06:13:27 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Thu, 20 Jul 2006 12:13:27 +0200
Subject: [Nitro] Rolls (Rock?) status
Message-ID: 

Hi (TRANS?),

I'm just in the need of a package manager, and I thought of
Rolls.  Whats the status of it and or how useable is it?

Here's what I want to do basically:

* Setting up a folder on a webserver to download new versions
   of libraries.
* Install those packages in a certain folder.
* Maybe compile source packages to binaries
* Load a certain version of the package

Libraries would include:

* Ruby itself
* Nitro/Og
* Nitro Parts
* DB (sqlite)

Anyway, that's the plan, how can we accomplish that :)

Jo

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png

From george.moschovitis at gmail.com  Thu Jul 20 09:07:59 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 20 Jul 2006 16:07:59 +0300
Subject: [Nitro] Bug in mongrel adapter (plus fix)
In-Reply-To: <716700c90607200225p4ab02131te24d404886c7a198@mail.gmail.com>
References: <33DECE09-1909-4386-8C53-2C270DA5A966@infurious.com>
	<716700c90607190454g6ea5ee4bofdf34576cb6ca5aa@mail.gmail.com>
	
	<716700c90607200225p4ab02131te24d404886c7a198@mail.gmail.com>
Message-ID: 

> Hi, George, after reading the wonderful darcs usage tips on Oxyliquit
> site, I am much clear now. Will use darcs to follow lastest
> development.

Great ;-)

waiting for your contributions.

-g.

>
> On 7/19/06, George Moschovitis  wrote:
> > Please, in the future send us patch bundles using dacrs against
> > repo.nitroproject.org
> >
> > thanks in advance,
> > George.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Thu Jul 20 09:08:17 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 20 Jul 2006 16:08:17 +0300
Subject: [Nitro] [PATCH]caching fragment fix
In-Reply-To: <716700c90607200227u33ab347dq33e180e11bd284f2@mail.gmail.com>
References: <716700c90607190501w4db8f3aeq4bad9dcba43050f5@mail.gmail.com>
	<716700c90607200227u33ab347dq33e180e11bd284f2@mail.gmail.com>
Message-ID: 

Ok, will have a look at this later today!

-g.

On 7/20/06, Fang Sun  wrote:
> Hi, regenerate the patch use darcs against lastest repo, please review.
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Thu Jul 20 09:13:52 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 20 Jul 2006 16:13:52 +0300
Subject: [Nitro] Devlab gems
In-Reply-To: 
References: 
	
Message-ID: 

Great stuff, please release this as 0.31.0
Since I gave you admin rights on rubyforge.org/nitro I am wondering if
you could make the release yourself.

Hopefully you can help me integrate some of the changes to my own repo too!

-g.



On 7/20/06, Bryan Soto  wrote:
> On 7/19/06, Jonathan Buch  wrote:
> > Hi,
> >
> > what happened to the devlab gems?  I think they should be posted
> > as the new stable on rubyforge before the new branch from George
> > gets stable to feed the masses (0.30.5 ?) :)
> >
>
> Okay, I've added the recent patches moving from cattr to setting.
> Other than the error messages when generating the rdoc/ri stuff, I'm
> not aware of any other issues. I've also removed two examples (wee and
> flash; wee as Nitro apparently doesn't support it anymore and flash as
> ming-ruby has been untouched for two years and is out of date) and
> verified the rest of them with mongrel and webrick.
>
> Gems available as usual via
>
> # gem update -y --source=http://devlab.oree.ch/~bryan/gems
>
> Examples are also available at http://devlab.oree.ch/~bryan
>
> Barring show-stoppers, this is what will be released, hopefully tomorrow.
>
> I would like some advice as to the version number though. Should I go
> with 30.5 as Kashia suggested? I'm not sure there's been enough done
> to qualify as a 0.31.0 release. Any opinions?
>
> Thanks,
>
> bryan
>
> PS - for the interested, here are the full changes since the 0.30 release.
>
> Wed Jul 19 16:17:27 PDT 2006  bryan.a.soto at gmail.com
>   * nitro-caching-fixes
>   Updates Nitro caching classes from cattr to settings.
>
> Mon Jul 17 13:50:50 PDT 2006  bryan.a.soto at gmail.com
>   * bugfix-glue-validate_value
>   Ticket #44: http://devlab.oree.ch/trac/glycerin/ticket/44
>   Credit to matt DOT moriarity AT gmail DOT com.
>
> Sun Jul 16 23:06:48 PDT 2006  bryan.a.soto at gmail.com
>   * bugfix-nitro-mongrel-adapter
>   Fixes multi-part form processing, which was broken by my previous
> patch. Basically, image uploads work now for mongrel.
>
> Tue Jul 11 13:10:06 PDT 2006  bryan.a.soto at gmail.com
>   * bugfix-nitro-flaky-template-root
>   When running from a directory other than the root directory of your
> Nitro app, Template.root was set incorrectly because class Template
> would determine whether to use template, src/template and default to
> public all before Nitro had chdir'd to the root directory of your
> Nitro app. Moved the requires to after the chdir.
>
> Tue Jul 11 13:13:47 PDT 2006  bryan.a.soto at gmail.com
>   * bugfix-nitro-extra-arguments
>   Commented out the extra concat as I never could figure out what the
> code intended.
>
> Tue Jul 11 11:43:29 PDT 2006  neokolor at gmx.de
>   * add option to set nav link titles
>
> Fri Jul  7 16:25:48 PDT 2006  neokolor at gmx.de
>   * helper/table.rb: fix problem with table output
>
> Fri Jul  7 16:08:29 PDT 2006  neokolor at gmx.de
>   * helper/pager.rb: fix problems with nav output if we have no results
>
> Fri Jul  7 16:18:08 PDT 2006  neokolor at gmx.de
>   * helper/pager.rb: add option to set nav link titles
>
> Sun Jun 18 05:42:52 PDT 2006  neokolor at gmx.de
>   * fix_in_relation_symbol_to_class
>
> Mon Jul 10 18:07:02 PDT 2006  bryan.a.soto at gmail.com
>   * bugfix-nitro-ticket-43
>   Fixed a bug/type as suggested by rff_rff AT yahoo DOT it
>   http://devlab.oree.ch/trac/glycerin/ticket/43
>
> Mon Jul 10 18:03:50 PDT 2006  bryan.a.soto at gmail.com
>   * bugfix-nitro-ticket-34
>   Fixed as per billk AT cts DOT com suggestion on Ticket #34.
>   http://devlab.oree.ch/trac/glycerin/ticket/34
>
> Mon Jul 10 17:56:56 PDT 2006  bryan.a.soto at gmail.com
>   * og-mysql-add-port-option
>   From jbritt, makes Mysql respect port option when creating and
> dropping databases.
>
> Fri Jun 30 12:07:11 PDT 2006  George Moschovitis 
>   * Applied james britt's mongrel patch.
>
> Thu Jun 29 17:01:35 PDT 2006  bryan.a.soto at gmail.com
>   * dispatcher-add_rule-patch
>   Duplicate of George's add_route to add_rule patch as his repo was unavailable.
>
> Thu Jun 29 16:31:46 PDT 2006  bryan.a.soto at gmail.com
>   * og-tc-sti-find
>   Adds a test case for manveru's sti fix.
>
> Thu Jun 29 16:29:26 PDT 2006  bryan.a.soto at gmail.com
>   * og-entity-from-string
>   Modifies the definition of Og::Entity::entity_from_string slightly.
> It assumed only one manager would ever exist. Not a valid assumption
> in the test suite.
>
> Thu Jun 29 15:10:18 PDT 2006  bryan.a.soto at gmail.com
>   * og-thread-safe-options
>   Adds the ability to specify :thread_safe as an option to Og.setup
> with the default begin true. Without this patch, thread_safe was true
> and it couldn't be modified without modifying the Og code.
>
> Thu Jun 29 15:08:58 PDT 2006  bryan.a.soto at gmail.com
>   * og-aggregate-fix
>   Accomodates the fact that the Postgres adapter doesn't return an
> Array like the other stores by modifying the aggregation code.
>
> Tue Jun 27 01:54:47 PDT 2006  manveru at weez-int.com
>   * fix for schema_inheritance queries
>
> Thu Apr 20 03:08:59 PDT 2006  m.fellinger at gmail.com
>   * localization-support for table-helper
>
> Wed Jun 21 17:06:22 PDT 2006  bryan.a.soto at gmail.com
>   * update-facets-1.4.2
>   Updates to Facets-1.4.2. Uses facets/core/module/class_extension
> instead of classinherit. No other changes needed to pass tests.
>
> Sun Jun 18 15:08:38 PDT 2006  neokolor at gmx.de
>   * fix problem with :plural_name in admin controller
>
> Fri Jun  9 15:06:20 PDT 2006  bryan.a.soto at gmail.com
>   * hack-enh-fcgi-rewind
>   A small hack to ensure that you can access the raw_body of a
> request, by converting the incoming stream to a StringIO if it doesn't
> respond to :rewind.
>
> Thu Jun  8 13:19:40 PDT 2006  bryan.a.soto at gmail.com
>   * nitro-sendfile-support
>   Adds sendfile supported. Submitted by Jan A. Lambert  alltel DOT net>. Tested on IE 6.0 and FireFox 1.5.
>
> Wed Jun  7 19:02:04 PDT 2006  bryan.a.soto at gmail.com
>   * nitro-dispatcher-router-strip_path
>   Makes the dispatcher honour the Router.strip_path setting. This,
> basically, makes Nitro ignore a portion of the url that could
> represent a directory name.
>
>   Basically, without the setting:
>   /my_nitro_apps_directory/nitro-app => looks for a controller mapped
> to / or /my_nitro_apps_directory.
>
>   With Router.strip_path = '/my_nitro_apps_directory'
>   /my_nitro_apps_directory/nitro-app => looks for a controller mapped
> to / or /nitro-app.
>
>   Many thanks to Vagabond for the patch.
>
>
> Mon Jun  5 14:11:52 PDT 2006  bryan.a.soto at gmail.com
>   * bugfix-mongrel-adapter-request-uri
>   Removes REQUEST_URI header manipulation and accomodates an upcoming
> change in Mongrel 0.3.13. An ugly class check... Lovely.
>
> Wed May 17 17:47:17 PDT 2006  bryan.a.soto at gmail.com
>   * fixup-guilles-patch
>   Brings Guille's patch up to date by resolving a conflict.
>
> Fri May 12 02:48:22 PDT 2006  Guillaume Pierronnet
> 
>   * og/lib/glue/taggable.rb refactor + support for passing :condition parameter
>   SqlStore#prepare_statement moved to public scope
>   test case on taggable runs fine
>
> Wed May 17 17:14:36 PDT 2006  bryan.a.soto at gmail.com
>   * test-fix-og-tc_ez
>   Fixes up the ez testcase Manveru submitted to make it run and pass.
>
> Sat May 13 04:10:40 PDT 2006  m.fellinger at gmail.com
>   * ez-fix for trueclass and testcase for it
>
> Sat May 13 04:09:45 PDT 2006  m.fellinger at gmail.com
>   * error-page-fix for source-preview
>
> Wed May 17 13:01:53 PDT 2006  fabian at fabian-buch.de
>   * to_rexml and to_xml
>
>   to_rexml returns a REXML::Element object
>   to_xml returns an XML-String
>
>
> Wed May 17 07:33:23 PDT 2006  fabian at fabian-buch.de
>   * introduction of OgObject.to_xml method
>
>   to_xml returns a REXML object of an Og object
>
>   example usage:
>     User[1].to_xml
>
>   returns a REXML object, so use to_s to get it as String containing XML
>
>
> Wed May 17 15:32:59 PDT 2006  bryan.a.soto at gmail.com
>   * test-fix-nitro-tc_render
>   Makes the TestController subclass Nitro::Controller so it get's the
> #encode_url method.
>
> Wed May 17 15:24:52 PDT 2006  bryan.a.soto at gmail.com
>   * adds-some-blank-lines-to-og-tc_setup
>
> Wed May 17 15:17:25 PDT 2006  bryan.a.soto at gmail.com
>   * gabrielle-renzis-linefeeds
>   Switches line endings from Dos to Unix.
>
> Fri May  5 05:30:51 PDT 2006  surrender_it at yahoo.it
>   * crash-early Og.setup
>   This patch removed the Exception handling routine from Og.setup.
>   With the current code if an exception is raised during the execution of
>   the method there is no crash, but the crash will happen every time
> someone references Og.manager, cause that variable won't be
> initialized .
>   This patch makes Og.setup crash as soon as possible, thus allowing
> faster diagnostics.
>
>   The patch also adds a simple test case for Og.setup, which seem not
> tested anywhere, but the test case may need some love  for proper
> setup (it relies on glycerin as of now, cause loading CONFIG.rb seem
> to raise a strange exception related to mixed use of facets 1.3 and
> 1.0 on my box)
>
>
> Thu May  4 18:02:06 PDT 2006  surrender_it at yahoo.it
>   * better test logging
>   As of now tests for Og swallow every error ingormation while not in $DBG mode,
>   this one-line patch allows visualization of logged informations at
> WARN level and more, which has little impact on the visualization of
> the tests and still allows quick diagnostics while staying out of $DBG
>
> Tue May 16 13:37:04 PDT 2006  bryan.a.soto at gmail.com
>   * george-glue-tag-fixup
>
> Tue May 16 08:07:05 PDT 2006  George Moschovitis 
>   * Minor.
>
> Tue May 16 00:32:46 PDT 2006  George Moschovitis 
>   * Moved examples, spark, flare to a separate repository.
>
> Tue May 16 00:28:23 PDT 2006  George Moschovitis 
>   * Removed wee helper and example. Not really useful in Nitro, better
> use wee standalone + Og if you like Wee components (or wait for a
> Nitro solution).
>
> Tue May 16 00:23:50 PDT 2006  George Moschovitis 
>   * Removed controller_name from Render, user controller.mount_path instead.
>
> Tue May 16 00:16:15 PDT 2006  George Moschovitis 
>   * Fixed: dont overwrite already defined methods in the new scaffold code.
>
> Tue May 16 00:15:08 PDT 2006  George Moschovitis 
>   * Improved R (encode_url) handles strings and entities. Redirect
> implicitly uses R. Form builder implicitly uses R for form actions.
>
> Sun May 14 23:54:39 PDT 2006  George Moschovitis 
>   * Fixed name_to_jsfile (absolute url).
>
> Fri May 12 21:22:51 PDT 2006  George Moschovitis 
>   * Reenabled format in Logger.
>
> Fri May 12 20:49:29 PDT 2006  George Moschovitis 
>   * Added some :nodoc: all to cleanup generated RDocs.
>
> Fri May 12 08:48:49 PDT 2006  George Moschovitis 
>   * Bumbed version number in many files.
>
> Fri May 12 08:37:54 PDT 2006  George Moschovitis 
>   * Wrote a detailed parts howto in RDoc form in parts.rb as requested
> by Jonas. To be improved after community feedback.
>
> Thu May 11 03:14:56 PDT 2006  George Moschovitis 
>   * Initial code for the new scaffolding system, *very* under
> construction, ignore for the moment.
>
> Wed May 10 03:28:36 PDT 2006  George Moschovitis 
>   * Added a model macro to controllers that links related classes. At
> the moment it is used in the sweeper do delete cached pages from the
> linked controller base directory.
>
> Thu May 11 23:05:11 PDT 2006  bryan.a.soto at gmail.com
>   * add-ostruct-require
>   Adds missing ostruct require to testcase.rb.
>
> Wed May 10 15:20:51 PDT 2006  bryan.a.soto at gmail.com
>   * more-mongrel-adapter-fixes
>   Fixes up static file serving.
>
> Tue May  9 23:05:46 PDT 2006  bryan.a.soto at gmail.com
>   * nitro-proto-page
>   Updates the intro page with a link to the rubyforge page for
> examples to be downloaded as suggested by James Britt.
>
> Tue May  9 23:04:39 PDT 2006  bryan.a.soto at gmail.com
>   * mongrel-fixes
>   Gives the Mongrel adapter some need updating.
>
> Mon May  8 23:36:43 PDT 2006  bryan.a.soto at gmail.com
>   * test-suite-fixup
>   Fixes the full test suite so that .\script\test.rb completes.
>
> Mon May  8 00:56:45 PDT 2006  George Moschovitis 
>   * Fixed Session.current and Controller.current
>
> Sun May  7 09:07:15 PDT 2006  George Moschovitis 
>   * Store current context in a thread local variable, accesible
> through Context.current
>
> Sun May  7 01:57:03 PDT 2006  George Moschovitis 
>   * Removed glue/attribute, use Facet's cattr instead. Updated
> nitro/og source to make compatible.
>
> Sat May  6 23:20:34 PDT 2006  George Moschovitis 
>   * Added script adapter file.
>
> Sat May  6 23:15:55 PDT 2006  George Moschovitis 
>   * Removed flexob and references to it, use Facet's OpenObject instead.
>
> Sat May  6 03:39:21 PDT 2006  George Moschovitis 
>   * Renamed to ScriptAdapter, keep ConsoleAdapter alias.
>
> Sat May  6 03:20:10 PDT 2006  George Moschovitis 
>   * Yeah, the ConsoleAdapter kinda works now ;-)
>
> Sat May  6 02:55:03 PDT 2006  George Moschovitis 
>   * Introduced new nitro command/runner that will replace the current
> spaghetti code. Based on facetes Console::Command. Also introduced the
> ConsoleAdapter.
>
> Sat May  6 01:00:52 PDT 2006  George Moschovitis 
>   * Removed accumulate (in facets).
>
> Sat May  6 00:55:26 PDT 2006  George Moschovitis 
>   * Some small updates.
>
> Sat May  6 00:50:53 PDT 2006  George Moschovitis 
>   tagged 0.30.0
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Thu Jul 20 09:14:22 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 20 Jul 2006 16:14:22 +0300
Subject: [Nitro] Devlab gems
In-Reply-To: 
References: 
	
	
Message-ID: 

> Great stuff, please release this as 0.31.0

As I probably already told you, my repo will be released as 0.40.0

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From nusgnaf at gmail.com  Thu Jul 20 09:58:30 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 20 Jul 2006 09:58:30 -0400
Subject: [Nitro] [PATCH]form helper doesn't work right if given block
Message-ID: <716700c90607200658j2ad77d03s226255606338a3bf@mail.gmail.com>

Rightnow if we give a block like
#{form :object =>@obj, ... do |e|
e.tag1
e.tag2
}
the rendered xhtml will miss all the attributes of @obj, I track down
to the following code in lib/nitro/helper/form.rb:line 210
++++++++++
if block_given?
yield b
else
b.all_attributes
end
------------------
so we lost control after the call of yield. The quick fix against
lastest repo is attached.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: form.fix.tar.gz
Type: application/x-gzip
Size: 4674 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060720/0f5ed647/attachment.gz 

From aidan at infurious.com  Thu Jul 20 13:03:39 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Thu, 20 Jul 2006 18:03:39 +0100
Subject: [Nitro] Bug in mongrel adapter (plus fix)
In-Reply-To: 
References: <33DECE09-1909-4386-8C53-2C270DA5A966@infurious.com>
	<716700c90607190454g6ea5ee4bofdf34576cb6ca5aa@mail.gmail.com>
	
	<716700c90607200225p4ab02131te24d404886c7a198@mail.gmail.com>
	
Message-ID: <204AA7B7-2206-4FBA-91F4-8D92DB728EB4@infurious.com>


> Hi, George, after reading the wonderful darcs usage tips on Oxyliquit
> site, I am much clear now. Will use darcs to follow lastest
> development.

I don't use darcs because I don't use the latest repository in my  
code, I haven't installed darcs on this machine and I'm up against it  
for time pressure - I just report 'em if I see 'em/fix 'em.

I'll get around to it again soon though, probably.

Aidan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060720/93692c5a/attachment.html 

From george.moschovitis at gmail.com  Thu Jul 20 13:17:46 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 20 Jul 2006 20:17:46 +0300
Subject: [Nitro] [PATCH]caching fragment fix
In-Reply-To: <716700c90607200227u33ab347dq33e180e11bd284f2@mail.gmail.com>
References: <716700c90607190501w4db8f3aeq4bad9dcba43050f5@mail.gmail.com>
	<716700c90607200227u33ab347dq33e180e11bd284f2@mail.gmail.com>
Message-ID: 

The patch seems corrupted, can you resend?

-g.

On 7/20/06, Fang Sun  wrote:
> Hi, regenerate the patch use darcs against lastest repo, please review.
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Thu Jul 20 13:18:09 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 20 Jul 2006 20:18:09 +0300
Subject: [Nitro] [PATCH]caching fragment fix
In-Reply-To: 
References: <716700c90607190501w4db8f3aeq4bad9dcba43050f5@mail.gmail.com>
	<716700c90607200227u33ab347dq33e180e11bd284f2@mail.gmail.com>
	
Message-ID: 

gzip: stdin: not in gzip format
tar: Child returned status 1
tar: Error exit delayed from previous errors

On 7/20/06, George Moschovitis  wrote:
> The patch seems corrupted, can you resend?
>
> -g.
>
> On 7/20/06, Fang Sun  wrote:
> > Hi, regenerate the patch use darcs against lastest repo, please review.
> >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
> >
> >
>
>
> --
> http://www.gmosx.com
> http://www.nitroproject.org
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Thu Jul 20 13:23:02 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 20 Jul 2006 20:23:02 +0300
Subject: [Nitro] [PATCH]form helper doesn't work right if given block
In-Reply-To: <716700c90607200658j2ad77d03s226255606338a3bf@mail.gmail.com>
References: <716700c90607200658j2ad77d03s226255606338a3bf@mail.gmail.com>
Message-ID: 

applied and added you to the CONTRIBUTORS list.

-g.

On 7/20/06, Fang Sun  wrote:
> Rightnow if we give a block like
> #{form :object =>@obj, ... do |e|
> e.tag1
> e.tag2
> }
> the rendered xhtml will miss all the attributes of @obj, I track down
> to the following code in lib/nitro/helper/form.rb:line 210
> ++++++++++
> if block_given?
> yield b
> else
> b.all_attributes
> end
> ------------------
> so we lost control after the call of yield. The quick fix against
> lastest repo is attached.
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From transfire at gmail.com  Thu Jul 20 14:42:39 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 20 Jul 2006 14:42:39 -0400
Subject: [Nitro] Rolls (Rock?) status
In-Reply-To: 
References: 
Message-ID: <4b6f054f0607201142r710e739fvf32920a57e47a7cb@mail.gmail.com>

On 7/20/06, Jonathan Buch  wrote:
> Hi (TRANS?),
>
> I'm just in the need of a package manager, and I thought of
> Rolls.  Whats the status of it and or how useable is it?
>
> Here's what I want to do basically:
>
> * Setting up a folder on a webserver to download new versions
>    of libraries.
> * Install those packages in a certain folder.
> * Maybe compile source packages to binaries
> * Load a certain version of the package
>
> Libraries would include:
>
> * Ruby itself
> * Nitro/Og
> * Nitro Parts
> * DB (sqlite)
>
> Anyway, that's the plan, how can we accomplish that :)

Hi,

Well, I'm not far off. But there still work to do. Today I've put
together a unviersal extensible project metadata lib (you were right
about the btw), which Reap, Sow and Roll can all use. Right now I'm
debating whether it should be it's own project or to included it with
one of the above three. I thought about adding it to Roll just to give
people incentive to try Roll but I'm not sure if that's a good idea.
Anyhow I'm not sure what to call it either. I've been using
"Rock::Metadata", but that seems a bit odd. Basically this is just a
glorified and more elegant version of the ProjectInfo class from Reap.

Speaking of Rock, Yesterday I developed the new package format, which
basically displaces what was Rock. The orioginal idea is still a neat
idea though, basically recursive tar archives, but I don't think it
needs it's own project as such. This new package format is much cooler
(more on that soon). I wrote to the gnu tar developers to see if
they'd add recursive archiving  to tar itself as an option. Haven't
heard anything yet though.

So what needs to happen ---

  1. Finish organizng/locating/naming  the "Rock" Spec code
  2. Update Sow to use it (mostly done actually)
  3. Make sure Sow's working including Gem generation.
  4. Clean-up package format and integrate with Sow.

All that actually won't take very long. I spend more time just
figuring out how to organize things then I do coding it seems!

The final big question then becomes: how to distribute these packages
and how to deal with automatic dependecies. For this I'm thinking Rain
again (maybe Sunshine?) I don't know yet. (You know it would be great
if Gems went for Roll's versioning format, then we wouldn;t have to
worry about a Ruby dedicated packaging at all --but I seriuosly doubt
that would ever happen).

BTW, the package format uses setup.rb so installation is to standarad
locations. We'll need to add an _option_ for installing to a central
location too.

Also, keep in mind that with Sow, packaing doesn't matter as much
anymore. The only reason I'm really pursuing a new package format is
to help facilitate Rolls and to have a Ruby equivalent of Jar. I mean
Sow still needs work, but if you're using a debian system for instance
you can download a source tarball (with meta/ info in it) and do:

  sow install -t deb foo.tar.gz

And youre done.

T.

From aidan at infurious.com  Thu Jul 20 20:35:14 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Fri, 21 Jul 2006 01:35:14 +0100
Subject: [Nitro] SyncBridge 1.0b released
Message-ID: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>

Hi all,

We finally released the 1.0b of SyncBridge today (https:// 
syncbridge.com).  This is a service + app that allows Mac users to  
share their calendars.  I'm posting here because it uses Nitro+Og for  
the server side components (both web pages and web service).

Thanks for the community's help in getting this app out there - as  
far as I know, it's the first commercial app using Nitro, so put a  
big tick in that check box :-)  (Please correct me if I'm wrong on  
that count!).

Aidan

From bryan.a.soto at gmail.com  Thu Jul 20 21:07:56 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 20 Jul 2006 18:07:56 -0700
Subject: [Nitro] SyncBridge 1.0b released
In-Reply-To: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
References: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
Message-ID: 

On 7/20/06, Aidan Rogers  wrote:
> We finally released the 1.0b of SyncBridge today (https://
> syncbridge.com).  This is a service + app that allows Mac users to
> share their calendars.  I'm posting here because it uses Nitro+Og for
> the server side components (both web pages and web service).
>

Congratulations Aidan! I wish you and your company all the best. :)

Bryan

From james.britt at gmail.com  Thu Jul 20 21:44:14 2006
From: james.britt at gmail.com (James Britt)
Date: Thu, 20 Jul 2006 18:44:14 -0700
Subject: [Nitro] SyncBridge 1.0b released
In-Reply-To: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
References: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
Message-ID: <44C0316E.1000909@gmail.com>

Aidan Rogers wrote:
> Hi all,
> 
> We finally released the 1.0b of SyncBridge today (https:// 
> syncbridge.com).  This is a service + app that allows Mac users to  
> share their calendars.  I'm posting here because it uses Nitro+Og for  
> the server side components (both web pages and web service).
> 
> Thanks for the community's help in getting this app out there - as  
> far as I know, it's the first commercial app using Nitro, so put a  
> big tick in that check box :-)  (Please correct me if I'm wrong on  
> that count!).

Wow,  congratulations!



-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://www.jamesbritt.com     - Playing with Better Toys
http://www.30secondrule.com   - Building Better Tools

From nusgnaf at gmail.com  Thu Jul 20 23:34:40 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 20 Jul 2006 23:34:40 -0400
Subject: [Nitro] [PATCH]caching fragment fix
In-Reply-To: 
References: <716700c90607190501w4db8f3aeq4bad9dcba43050f5@mail.gmail.com>
	<716700c90607200227u33ab347dq33e180e11bd284f2@mail.gmail.com>
	
	
Message-ID: <716700c90607202034i2fa536b5x2fda05cb40b8a6ff@mail.gmail.com>

Ah, I simply use bzip2 too much.
Regenerate against latest repo.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fragments.fix.tar.gz
Type: application/x-gzip
Size: 4809 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060720/07e7bbff/attachment.gz 

From transfire at gmail.com  Fri Jul 21 01:47:29 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 21 Jul 2006 01:47:29 -0400
Subject: [Nitro] Super Annotations
Message-ID: <4b6f054f0607202247v66e5eea7h35372873ac5d3869@mail.gmail.com>

Madness or a stroke of Genius?


  class Annotation
    def initialize( name, *args, &func )
      @name = name
      @args = args
      @func = func
    end
  end

  module Annotate
    def annotations
      @annotations ||= []
    end

    def define_annotation( sym, *args, &yld )
      if args.empty? and not block_given?
        if sym == :default
          @annotation_modes = nil
        else
          @annotation_modes = *sym.to_s.split('_and_')
        end
      else
        annotations << Annotation.new( sym, *args, &yld )
        if @annotation_modes
          @annotation_modes.each do |mode|
            annotations << Annotation.new( mode, *args, &yld )
          end
        end
      end
    end

    def attr( *args, &yld )
      define_annotation( :attr, *args, &yld )
    end

    def type( *args, &yld )
      define_annotation( :type, *args, &yld )
    end

    def method_missing( sym, *args, &yld )
      define_annotation( sym, *args, &yld )
    end
  end


Here's an example:


  class Metadata
    extend Annotate

    attr :date
    attr :title
    attr :contact
    attr :license
    attr :status
    attr :arch
    attr :author
    attr :email
    attr :homepage
    attr :created
    attr :scope
    attr :rules  # should be in sow package?
    attr :description

    type :rubyforge, :Rubyforge
    type :revision , :Revision

    refer :authors      , :author
    refer :architecture , :arch
    refer :platform     , :arch

    default :license , 'Ruby/GPL'
    default :status  , 'alpha'
    default :arch    , 'any'
    default :data    do Time.now.strftime("%Y-%m-%d") end
    default :title   do name.capitalize end
    default :contact do author end

  needs

    attr :name
    attr :version
    attr :license
    attr :summary

  end

  p Metadata.annotations


T.

From zimba.tm at gmail.com  Fri Jul 21 04:23:33 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Fri, 21 Jul 2006 10:23:33 +0200
Subject: [Nitro] Super Annotations
In-Reply-To: <4b6f054f0607202247v66e5eea7h35372873ac5d3869@mail.gmail.com>
References: <4b6f054f0607202247v66e5eea7h35372873ac5d3869@mail.gmail.com>
Message-ID: <3ff63f9b0607210123t2c357bcdp132899c2aa8b7a50@mail.gmail.com>

I like the "extend Annotation". This would allow .kind_of?(Annotation)
and avoid pollution the main objects. But I would miss the ability to
annotate using one line only.


-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From lasso at lassoweb.se  Fri Jul 21 05:44:29 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Fri, 21 Jul 2006 11:44:29 +0200
Subject: [Nitro] What does this error mean?
Message-ID: <44C0A1FD.6050705@lassoweb.se>

Hi list!

After revamping my web site using Nitro at home (WinXP Pro+Apache 
2.2+CGI) I uploaded it to my ISPs server (Gentoo+Apache2.0+FastCGI). 
Most things worked as aspected, but any page that tries to use Og 
doesn't work. I get the following error on all pages that use Og:

--- BEGIN DEBUG INFO ---
Error
Path: /guestbook
undefined local variable or method `ogmanager' for Post:Class
Reload this page. Go to the referer or the home page.
In file '/usr/lib/ruby/gems/1.8/gems/og-0.30.0/lib/og/entity.rb' :

514: yield(obj) if block_given?
515: end
516:
517: return obj
518: else
519: super	<-- Error here
520: end
521: end
522:
523: private
524:
--- END DEBUG INFO ---

And the stack trace...

--- BEGIN STACK TRACE ---
/usr/lib/ruby/gems/1.8/gems/og-0.30.0/lib/og/entity.rb:519:in 
`method_missing'
/usr/lib/ruby/gems/1.8/gems/og-0.30.0/lib/og/entity.rb:191:in `find'
./src/controllers/guestbookcontroller.rb:7:in `index'
(eval):6:in `index_action'
/usr/lib/ruby/gems/1.8/gems/nitro-0.30.0/lib/nitro/render.rb:129:in `render'
/usr/lib/ruby/gems/1.8/gems/nitro-0.30.0/lib/nitro/cgi.rb:42:in `process'
/usr/lib/ruby/gems/1.8/gems/nitro-0.30.0/lib/nitro/adapter/fastcgi.rb:35:in 
`start'
/usr/lib/ruby/gems/1.8/gems/nitro-0.30.0/lib/nitro/adapter/fastcgi.rb:33:in 
`start'
/usr/lib/ruby/gems/1.8/gems/nitro-0.30.0/lib/nitro/server/runner.rb:287:in 
`invoke'
/usr/lib/ruby/gems/1.8/gems/nitro-0.30.0/lib/nitro/server.rb:128:in `run'
/usr/lib/ruby/gems/1.8/gems/nitro-0.30.0/lib/nitro.rb:78:in `run'
/home/lassoweb/nitro_sites/lassoweb/run.rb:4
/usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:21:in `require'
/home/lassoweb/public_html/dispatch.fcgi:7
--- END STACK TRACE ---

I've tried using plain CGI on my ISPs server, but the problem persists. 
I've also checked that the relevant db gem (ruby-sqlite3) is installed). 
Now I'm stuck. Can anyone please try to explain to me what the generated 
error message means?


Kindly


/lasso

-- 
________________________________________
Lars Olsson
lasso at lassoweb.se
http://www.lassoweb.se/

From george.moschovitis at gmail.com  Fri Jul 21 08:50:16 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 21 Jul 2006 15:50:16 +0300
Subject: [Nitro] SyncBridge 1.0b released
In-Reply-To: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
References: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
Message-ID: 

> Thanks for the community's help in getting this app out there - as
> far as I know, it's the first commercial app using Nitro, so put a
> big tick in that check box :-)  (Please correct me if I'm wrong on
> that count!).

You are wrong, I have succesfully deployed several commercial
applications based on Nitro.

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Fri Jul 21 08:52:49 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 21 Jul 2006 15:52:49 +0300
Subject: [Nitro] Super Annotations
In-Reply-To: <4b6f054f0607202247v66e5eea7h35372873ac5d3869@mail.gmail.com>
References: <4b6f054f0607202247v66e5eea7h35372873ac5d3869@mail.gmail.com>
Message-ID: 

I am not sure if I like this...

-g.

On 7/21/06, TRANS  wrote:
> Madness or a stroke of Genius?
>
>
>   class Annotation
>     def initialize( name, *args, &func )
>       @name = name
>       @args = args
>       @func = func
>     end
>   end
>
>   module Annotate
>     def annotations
>       @annotations ||= []
>     end
>
>     def define_annotation( sym, *args, &yld )
>       if args.empty? and not block_given?
>         if sym == :default
>           @annotation_modes = nil
>         else
>           @annotation_modes = *sym.to_s.split('_and_')
>         end
>       else
>         annotations << Annotation.new( sym, *args, &yld )
>         if @annotation_modes
>           @annotation_modes.each do |mode|
>             annotations << Annotation.new( mode, *args, &yld )
>           end
>         end
>       end
>     end
>
>     def attr( *args, &yld )
>       define_annotation( :attr, *args, &yld )
>     end
>
>     def type( *args, &yld )
>       define_annotation( :type, *args, &yld )
>     end
>
>     def method_missing( sym, *args, &yld )
>       define_annotation( sym, *args, &yld )
>     end
>   end
>
>
> Here's an example:
>
>
>   class Metadata
>     extend Annotate
>
>     attr :date
>     attr :title
>     attr :contact
>     attr :license
>     attr :status
>     attr :arch
>     attr :author
>     attr :email
>     attr :homepage
>     attr :created
>     attr :scope
>     attr :rules  # should be in sow package?
>     attr :description
>
>     type :rubyforge, :Rubyforge
>     type :revision , :Revision
>
>     refer :authors      , :author
>     refer :architecture , :arch
>     refer :platform     , :arch
>
>     default :license , 'Ruby/GPL'
>     default :status  , 'alpha'
>     default :arch    , 'any'
>     default :data    do Time.now.strftime("%Y-%m-%d") end
>     default :title   do name.capitalize end
>     default :contact do author end
>
>   needs
>
>     attr :name
>     attr :version
>     attr :license
>     attr :summary
>
>   end
>
>   p Metadata.annotations
>
>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From transfire at gmail.com  Fri Jul 21 11:59:57 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 21 Jul 2006 11:59:57 -0400
Subject: [Nitro] SyncBridge 1.0b released
In-Reply-To: 
References: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
	
Message-ID: <4b6f054f0607210859y1bb2cb13ob41a49bc3ba8df93@mail.gmail.com>

On 7/21/06, George Moschovitis  wrote:
> > Thanks for the community's help in getting this app out there - as
> > far as I know, it's the first commercial app using Nitro, so put a
> > big tick in that check box :-)  (Please correct me if I'm wrong on
> > that count!).
>
> You are wrong, I have succesfully deployed several commercial
> applications based on Nitro.

First commercial app in Nitro that Nitro's creator didn't wirte then? ;-)

T.

From aidan at infurious.com  Fri Jul 21 12:19:04 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Fri, 21 Jul 2006 17:19:04 +0100
Subject: [Nitro] SyncBridge 1.0b released
In-Reply-To: 
References: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
	
Message-ID: 

Do you count?  You wrote the framework! ;-)

When I did a quick poll on the IRC channel, everyone thought that  
your sites were powered by advertising rather than being a paid for  
service, so I'm using that interpretation of commercial :-)

Anyway, I'll post a comment on my blog to say we're second (or third?  
anyone?).

Aidan

On 21/07/2006, at 1:50 PM, George Moschovitis wrote:

>
> You are wrong, I have succesfully deployed several commercial
> applications based on Nitro.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060721/02b3a708/attachment-0001.html 

From m.fellinger at gmail.com  Fri Jul 21 12:38:01 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Sat, 22 Jul 2006 01:38:01 +0900
Subject: [Nitro] SyncBridge 1.0b released
In-Reply-To: 
References: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
	
	
Message-ID: <9c00d3e00607210938l7e6c3b2ftc911f2cda52da866@mail.gmail.com>

Not sure in which category our application would fall... but we
discussed that on irc already :)
i, for one, would like to congratulate you and wish you all the best...
may you take over the mac-world ;)

On 7/22/06, Aidan Rogers  wrote:
>
> Do you count?  You wrote the framework! ;-)
>
> When I did a quick poll on the IRC channel, everyone thought that your sites
> were powered by advertising rather than being a paid for service, so I'm
> using that interpretation of commercial :-)
>
> Anyway, I'll post a comment on my blog to say we're second (or third?
> anyone?).
>
> Aidan
>
>
> On 21/07/2006, at 1:50 PM, George Moschovitis wrote:
>
>
>
>
>
> You are wrong, I have succesfully deployed several commercial
>
> applications based on Nitro.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>

From bryan.a.soto at gmail.com  Fri Jul 21 13:56:44 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 21 Jul 2006 10:56:44 -0700
Subject: [Nitro] What does this error mean?
In-Reply-To: <44C0A1FD.6050705@lassoweb.se>
References: <44C0A1FD.6050705@lassoweb.se>
Message-ID: 

On 7/21/06, Lars Olsson  wrote:
> --- BEGIN DEBUG INFO ---
> Error
> Path: /guestbook
> undefined local variable or method `ogmanager' for Post:Class
> Reload this page. Go to the referer or the home page.
> In file '/usr/lib/ruby/gems/1.8/gems/og-0.30.0/lib/og/entity.rb' :
>

That's odd. Does the database file exist? Does the fastcgi process
have write privs to it?

Bryan

From george.moschovitis at gmail.com  Fri Jul 21 14:22:07 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 21 Jul 2006 21:22:07 +0300
Subject: [Nitro] SyncBridge 1.0b released
In-Reply-To: 
References: <2AEEA4E1-6E2B-4A19-863A-339CDFFAD097@infurious.com>
	
	
Message-ID: 

> When I did a quick poll on the IRC channel, everyone thought that your sites
> were powered by advertising rather than being a paid for service, so I'm
> using that interpretation of commercial :-)

No, I use Nitro on custom intranet applications, they are not public
applications you can see.
Examples:

a  cashflow prediction application,
a virtual business simulation for educational purposes,
a gis visualization applidation,
a document management system (for scanned papers)

and stuff like that,

-g.


-- 
http://www.gmosx.com
http://www.nitroproject.org

From bryan.a.soto at gmail.com  Fri Jul 21 15:00:46 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 21 Jul 2006 12:00:46 -0700
Subject: [Nitro] [ANN] Nitro/Og 0.31.0
Message-ID: 

** Changes

= Og ==

* Increased size of ogtype field in STI tables to 50 from 30.

* Made Og (Mysql) respect the port option when creating and dropping databases.

* Added thread_safe to Og.setup options.

* Added #to_xml and #to_rexml method to Og managed objects.

== Nitro ==

* Mongrel adapter updated to work with latest versions of Mongrel.

* Fixed a long-standing bug where the template root wasn't determined
correctly (would result in blank pages).

* Added sendfile support.

* helper/pager.rb: add option to set nav link titles

* Fixed strip_path support. When this setting is set to a path, it
will be stripped from urls. Given,

  Router.strip_path = '/nitro-apps'

when the dispatcher gets a url like /nitro-apps/blog it will strip out
/nitro-apps and resolve to the controller for /blog.

* Wee helper is removed.

* Updated Nitro start page with links to the examples.

== Glue ==

Bugfixes. Refactored taggable.rb and removed attribute in favor of
Facets's cattr which was then replaced with settings.

** Contributors
Andrew Thompson
Bill Kelly
Fabian Buch
Fang Sun
Felix Wittmann
Gabrielle Renzi
Guillaume Pierronnet
James Britt
Jan A. Lambert
Matt Moriarity
Michael Fellinger

** Get it by running

# gem update -y nitro

or

# gem install -y nitro

Archives for manual installations and example apps (examples, spark
and flare) are also available for download from

http://rubyforge.org/frs/?group_id=418

From lasso at lassoweb.se  Fri Jul 21 16:29:37 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Fri, 21 Jul 2006 22:29:37 +0200
Subject: [Nitro] What does this error mean?
In-Reply-To: 
References: <44C0A1FD.6050705@lassoweb.se>
	
Message-ID: <44C13931.4070809@lassoweb.se>

Hi,

the db file doesn't exist, but using a prepopulated db doesn't seem to 
help. The directory in which the db file is supposed to live have 777 
permissions.

I'll try using mysql to check whether this might be related to directory 
permissions or not.

/lasso

________________________________________
Lars Olsson
lasso at lassoweb.se
http://www.lassoweb.se/



Bryan Soto skrev:
> On 7/21/06, Lars Olsson  wrote:
>> --- BEGIN DEBUG INFO ---
>> Error
>> Path: /guestbook
>> undefined local variable or method `ogmanager' for Post:Class
>> Reload this page. Go to the referer or the home page.
>> In file '/usr/lib/ruby/gems/1.8/gems/og-0.30.0/lib/og/entity.rb' :
>>
> 
> That's odd. Does the database file exist? Does the fastcgi process
> have write privs to it?
> 
> Bryan
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>

From bryan.a.soto at gmail.com  Fri Jul 21 17:54:02 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 21 Jul 2006 14:54:02 -0700
Subject: [Nitro] What does this error mean?
In-Reply-To: <44C13931.4070809@lassoweb.se>
References: <44C0A1FD.6050705@lassoweb.se>
	
	<44C13931.4070809@lassoweb.se>
Message-ID: 

On 7/21/06, Lars Olsson  wrote:
> Hi,
>
> the db file doesn't exist, but using a prepopulated db doesn't seem to
> help. The directory in which the db file is supposed to live have 777
> permissions.
>
> I'll try using mysql to check whether this might be related to directory
> permissions or not.
>

Well, it seems that the class isn't being enchanted which is the cause
of that missing ogmanager message. You might also try setting the
global variable $DBG to true before you're call to Og.setup. It will
generate a bit more logger messages and hopefully show what's
happening. Assuming you have access to the error logs...

Bryan

From george.moschovitis at gmail.com  Sat Jul 22 01:32:09 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 22 Jul 2006 08:32:09 +0300
Subject: [Nitro] [ANN] Nitro/Og 0.31.0
In-Reply-To: 
References: 
Message-ID: 

Thanks Bryan ;-)
One thing, could you please announce this to c.l.r too?
If you do, please make clear that this is a maintenance release.


thanks + great job,
George.

On 7/21/06, Bryan Soto  wrote:
> ** Changes
>
> = Og ==
>
> * Increased size of ogtype field in STI tables to 50 from 30.
>
> * Made Og (Mysql) respect the port option when creating and dropping databases.
>
> * Added thread_safe to Og.setup options.
>
> * Added #to_xml and #to_rexml method to Og managed objects.
>
> == Nitro ==
>
> * Mongrel adapter updated to work with latest versions of Mongrel.
>
> * Fixed a long-standing bug where the template root wasn't determined
> correctly (would result in blank pages).
>
> * Added sendfile support.
>
> * helper/pager.rb: add option to set nav link titles
>
> * Fixed strip_path support. When this setting is set to a path, it
> will be stripped from urls. Given,
>
>   Router.strip_path = '/nitro-apps'
>
> when the dispatcher gets a url like /nitro-apps/blog it will strip out
> /nitro-apps and resolve to the controller for /blog.
>
> * Wee helper is removed.
>
> * Updated Nitro start page with links to the examples.
>
> == Glue ==
>
> Bugfixes. Refactored taggable.rb and removed attribute in favor of
> Facets's cattr which was then replaced with settings.
>
> ** Contributors
> Andrew Thompson
> Bill Kelly
> Fabian Buch
> Fang Sun
> Felix Wittmann
> Gabrielle Renzi
> Guillaume Pierronnet
> James Britt
> Jan A. Lambert
> Matt Moriarity
> Michael Fellinger
>
> ** Get it by running
>
> # gem update -y nitro
>
> or
>
> # gem install -y nitro
>
> Archives for manual installations and example apps (examples, spark
> and flare) are also available for download from
>
> http://rubyforge.org/frs/?group_id=418
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From lasso at lassoweb.se  Sat Jul 22 04:36:45 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Sat, 22 Jul 2006 10:36:45 +0200
Subject: [Nitro] What does this error mean?
In-Reply-To: 
References: <44C0A1FD.6050705@lassoweb.se>		<44C13931.4070809@lassoweb.se>
	
Message-ID: <44C1E39D.5030301@lassoweb.se>

Hi,

mysql worked. It seems it wasn't enough to let Apache have write 
priviliges to the file, it actually had to be the __owner__ of the file 
in order for sqlite to work properly. After letting Apache create the db 
everything works as exspected.


Thanks for helping

/lasso

________________________________________
Lars Olsson
lasso at lassoweb.nu
http://www.lassoweb.nu/

Bryan Soto skrev:
> On 7/21/06, Lars Olsson  wrote:
>> Hi,
>>
>> the db file doesn't exist, but using a prepopulated db doesn't seem to
>> help. The directory in which the db file is supposed to live have 777
>> permissions.
>>
>> I'll try using mysql to check whether this might be related to directory
>> permissions or not.
>>
> 
> Well, it seems that the class isn't being enchanted which is the cause
> of that missing ogmanager message. You might also try setting the
> global variable $DBG to true before you're call to Og.setup. It will
> generate a bit more logger messages and hopefully show what's
> happening. Assuming you have access to the error logs...
> 
> Bryan
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
> 

-- 
________________________________________
Lars Olsson
lasso at lassoweb.se
http://www.lassoweb.se/

From fabian at oggu.de  Sat Jul 22 04:47:01 2006
From: fabian at oggu.de (Fabian Buch)
Date: Sat, 22 Jul 2006 10:47:01 +0200
Subject: [Nitro] nitroproject.org wiki not editable
Message-ID: 

the nitroproject.org wiki throws an Error if I want to edit a page  
(http://nitroproject.org/wiki/edit/Public+sites+powered+by+Nitro/ for  
example)

Ticket: http://nitroproject.org/tickets/read/24

Fabian

From george.moschovitis at gmail.com  Sat Jul 22 07:56:09 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 22 Jul 2006 14:56:09 +0300
Subject: [Nitro] nitroproject.org wiki not editable
In-Reply-To: 
References: 
Message-ID: 

fixed!

-g.

PS: please add more sites

On 7/22/06, Fabian Buch  wrote:
> the nitroproject.org wiki throws an Error if I want to edit a page
> (http://nitroproject.org/wiki/edit/Public+sites+powered+by+Nitro/ for
> example)
>
> Ticket: http://nitroproject.org/tickets/read/24
>
> Fabian
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sat Jul 22 14:19:30 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 22 Jul 2006 21:19:30 +0300
Subject: [Nitro] Facets inflect bug
Message-ID: 

Tom,

there is a problem with pluralize:

'forum'.pluralize yields forums instead of fora.

can you fix this? While you are at it, could you incorporate the
latest Rails inflector fixes in the inflector included in Facets?

thanks,
George.


-- 
http://www.gmosx.com
http://www.nitroproject.org

From transfire at gmail.com  Sat Jul 22 15:58:58 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 22 Jul 2006 15:58:58 -0400
Subject: [Nitro] Facets inflect bug
In-Reply-To: 
References: 
Message-ID: <4b6f054f0607221258t417b2664tdb0dce3fd53d9c85@mail.gmail.com>

On 7/22/06, George Moschovitis  wrote:
> Tom,
>
> there is a problem with pluralize:
>
> 'forum'.pluralize yields forums instead of fora.
>
> can you fix this? While you are at it, could you incorporate the
> latest Rails inflector fixes in the inflector included in Facets?

Okay.

Query: Where are you using pluralize?

T.

From transfire at gmail.com  Sat Jul 22 20:07:29 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 22 Jul 2006 20:07:29 -0400
Subject: [Nitro] Facets inflect bug
In-Reply-To: <4b6f054f0607221258t417b2664tdb0dce3fd53d9c85@mail.gmail.com>
References: 
	<4b6f054f0607221258t417b2664tdb0dce3fd53d9c85@mail.gmail.com>
Message-ID: <4b6f054f0607221707t2ff54819u23ef65269badd459@mail.gmail.com>

On 7/22/06, TRANS  wrote:
> On 7/22/06, George Moschovitis  wrote:
> > Tom,
> >
> > there is a problem with pluralize:
> >
> > 'forum'.pluralize yields forums instead of fora.
> >
> > can you fix this? While you are at it, could you incorporate the
> > latest Rails inflector fixes in the inflector included in Facets?

FYI, just rewrote the whole thing.

Thanks,
T.

From gwazy at zigg.us  Sat Jul 22 20:25:08 2006
From: gwazy at zigg.us (Will)
Date: Sat, 22 Jul 2006 18:25:08 -0600
Subject: [Nitro] Hello All!
Message-ID: <1153614309.12828.3.camel@localhost>

Hello guys I would like to introduce myself to you all, my name is will,
and I love programming and computers. :-) Manveru showed me nitro and I
have been in love with it when I first type the very basic but global
"Hello World!". I hope to contribute to the community and help nitro
grow with every one else!

Thanks and It will be fun talking and programming with you guys!
-will a.k.a. Gwazy


From transfire at gmail.com  Sun Jul 23 01:07:58 2006
From: transfire at gmail.com (TRANS)
Date: Sun, 23 Jul 2006 01:07:58 -0400
Subject: [Nitro] [ANN] Facets 1.6.20 beta
Message-ID: <4b6f054f0607222207x7d56fc4drbde5dbaf174d6ce5@mail.gmail.com>

Give that a try and let me know how it bombs with Nitro ;)

Added singleton annotations, and I'm wodering hwo :self as a key holds up.

Also completely redid the singular plural algorithm. Want to see?


class String

  @inflection_rules = []

  def self.inflection_rule( *args ) @inflection_rules << args end
  def self.inflection_rules() @inflection_rules end

  #

  def self.inflection_rules_by_singular
    return @inflection_rules_by_singular if @inflection_rules_by_singular

    sorted = inflection_rules.sort_by{ |b,s,pl| "#{b}#{s}".size }.reverse
    @inflection_rules_by_singular = sorted.collect do |b, s, pl|
      [ /(#{b})#{s}$/, '\1'+"#{pl}" ]
    end
  end

  #

  def self.inflection_rules_by_plural
    return @inflection_rules_by_plural if @inflection_rules_by_plural

    sorted = inflection_rules.sort_by{ |b,s,pl| "#{b}#{pl}".size }.reverse
    @inflection_rules_by_plural = sorted.collect do |b, s, pl|
      [ /(#{b})#{pl}$/, '\1'+"#{s}" ]
    end
  end

  # One argument means singular and plural are the same.

  inflection_rule 'equipment'
  inflection_rule 'information'
  inflection_rule 'money'
  inflection_rule 'species'
  inflection_rule 'series'
  inflection_rule 'fish'
  inflection_rule 'sheep'
  inflection_rule 'moose'

  # USAGE: +inflection_rule+ base, singular, plural
  #
  # IMPORTANT Each base+singular and base+plural should only appear once!

  inflection_rule 'quiz'         , ''     , 'zes'       # quiz
  inflection_rule 'matr'         , 'ix'   , 'ices'      # matrix
  inflection_rule 'vert|ind'     , 'ex'   , 'ices'      # vertex, index
  inflection_rule 'ox'           , ''     , 'en'        # ox
  inflection_rule 'm'            , 'ouse' , 'ice'       # mouse, louse
  inflection_rule 'l'            , 'ouse' , 'ice'       # mouse, louse
  inflection_rule 'x'            , ''     , 'es'        # fix, box
  inflection_rule 'ch'           , ''     , 'es'        # search, switch
  inflection_rule 'ss'           , ''     , 'es'        # process, address
  inflection_rule 'sh'           , ''     , 'es'        # dish, wish
  inflection_rule 'qu'           , 'y'    , 'ies'       # query, ability, agency
  inflection_rule 'hive'         , ''     , 's'         # archive, hive
  inflection_rule 'sis'          , ''     , 'ses'       # basis, diagnosis
  inflection_rule 't'            , 'um'   , 'a'         # datum, medium, forum
  inflection_rule 'i'            , 'um'   , 'a'         # datum, medium, forum
  inflection_rule 'r'            , 'um'   , 'a'         # datum, medium, forum
  inflection_rule 'potat'        , 'o'    , 'oes'       # potato
  inflection_rule 'buffal'       , 'o'    , 'oes'       # buffalo, tomato
  inflection_rule 'tomat'        , 'o'    , 'oes'       # buffalo, tomato
  inflection_rule 'p'            , 'erson', 'eople'     # person, salesperson
  inflection_rule 'm'            , 'an'   , 'en'        # man, woman, spokesman
  inflection_rule 'c'            , 'hild' , 'hildren'   # child
  inflection_rule 'bu'           , 's'    , 'ses'       # bus
  inflection_rule 'alias'        , ''     , 'es'        # alias
  inflection_rule 'status'       , ''     , 'es'        # status
  inflection_rule 'vir'          , 'us'   , 'i'         # virus
  inflection_rule 'octop'        , 'us'   , 'i'         # octopus
  inflection_rule 'ax'           , 'is'   , 'es'        # axis
  inflection_rule 'cris'         , 'is'   , 'es'        # crisis
  inflection_rule 'test'         , 'is'   , 'es'        # testis
  inflection_rule 'l'            , 'f'    , 'ves'       # half halves
  inflection_rule ''             , 'fe'   , 'ves'       # safe, wife
  inflection_rule 'a'            , 'y'    , 's'         #
  inflection_rule 'e'            , 'y'    , 's'         #
  inflection_rule 'i'            , 'y'    , 's'         #
  inflection_rule 'o'            , 'y'    , 's'         #
  inflection_rule 'u'            , 'y'    , 's'         #
  inflection_rule ''             , 'y'    , 'ies'       # query, ability, agency
  inflection_rule ''             , ''     , 's'         # --catch all--

  # Convert an English word from plurel to singular.
  #
  #   "boys".singular      #=> boy
  #   "tomatoes".singular  #=> tomato
  #

  def singular
    result = self.dup
    self.class.inflection_rules_by_plural.each do |(match, replacement)|
      break if result.gsub!(match, replacement)
    end
    return result
  end

  # Convert an English word from singular to plurel.
  #
  #   "boy".plural     #=> boys
  #   "tomato".plural  #=> tomatoes
  #

  def plural
    return self.dup if /s$/ =~ self
    result = self.dup
    self.class.inflection_rules_by_singular.each do |(match, replacement)|
      break if result.gsub!(match, replacement)
    end
    return result
  end

  # Railism alias for #singular.
  alias_method( :singularize, :singular )

  # Railism alias for #plural.
  alias_method( :pluralize, :plural )

end


T.

From zimba.tm at gmail.com  Sun Jul 23 03:28:16 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Sun, 23 Jul 2006 09:28:16 +0200
Subject: [Nitro] Hello All!
In-Reply-To: <1153614309.12828.3.camel@localhost>
References: <1153614309.12828.3.camel@localhost>
Message-ID: <3ff63f9b0607230028x18e811dag6cb81c776c3a9a6f@mail.gmail.com>

Hi Will, Welcome :-)

On 23/07/06, Will  wrote:
> Hello guys I would like to introduce myself to you all, my name is will,
> and I love programming and computers. :-) Manveru showed me nitro and I
> have been in love with it when I first type the very basic but global
> "Hello World!". I hope to contribute to the community and help nitro
> grow with every one else!
>
> Thanks and It will be fun talking and programming with you guys!
> -will a.k.a. Gwazy
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From george.moschovitis at gmail.com  Sun Jul 23 04:03:19 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 23 Jul 2006 11:03:19 +0300
Subject: [Nitro] Hello All!
In-Reply-To: <3ff63f9b0607230028x18e811dag6cb81c776c3a9a6f@mail.gmail.com>
References: <1153614309.12828.3.camel@localhost>
	<3ff63f9b0607230028x18e811dag6cb81c776c3a9a6f@mail.gmail.com>
Message-ID: 

> > Thanks and It will be fun talking and programming with you guys!
> > -will a.k.a. Gwazy

Welcome ;-)

It is nice to have you on-board!

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sun Jul 23 04:06:05 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 23 Jul 2006 11:06:05 +0300
Subject: [Nitro] [ANN] Facets 1.6.20 beta
In-Reply-To: <4b6f054f0607222207x7d56fc4drbde5dbaf174d6ce5@mail.gmail.com>
References: <4b6f054f0607222207x7d56fc4drbde5dbaf174d6ce5@mail.gmail.com>
Message-ID: 

> Give that a try and let me know how it bombs with Nitro ;)

hmm I am working on a project so I can afford bombing at the moment ;-)

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From fabian at oggu.de  Sun Jul 23 08:31:28 2006
From: fabian at oggu.de (Fabian Buch)
Date: Sun, 23 Jul 2006 14:31:28 +0200
Subject: [Nitro] nitroproject.org wiki not editable
In-Reply-To: 
References: 
	
Message-ID: <0B1C470D-3608-4C7B-BC8B-D18FD0F1BD7D@oggu.de>

Still (or again) throws "Error"

Am 22.07.2006 um 13:56 schrieb George Moschovitis:
> fixed!
>
> -g.
>
> PS: please add more sites
>
> On 7/22/06, Fabian Buch  wrote:
>> the nitroproject.org wiki throws an Error if I want to edit a page
>> (http://nitroproject.org/wiki/edit/Public+sites+powered+by+Nitro/ for
>> example)
>>
>> Ticket: http://nitroproject.org/tickets/read/24
>>
>> Fabian



From surrender_it at yahoo.it  Sun Jul 23 10:58:12 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sun, 23 Jul 2006 16:58:12 +0200
Subject: [Nitro] nitroproject - qualified title for feed
Message-ID: 

Hie people, I just noticed that nitroproject.org's feeds have a tiele 
like "Article Feed", "Ticket Feed" and so on.
I think it would be nice to qualify them as related to the website, i.e.

"Nitro Project's Article Feed"

What do other people and george think?


-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From george.moschovitis at gmail.com  Sun Jul 23 16:00:08 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 23 Jul 2006 22:00:08 +0200
Subject: [Nitro] nitroproject - qualified title for feed
In-Reply-To: 
References: 
Message-ID: 

Ok will improve the titles in the next update (I am working on this at
the moment ;-))

-g.

On 7/23/06, gabriele renzi  wrote:
> Hie people, I just noticed that nitroproject.org's feeds have a tiele
> like "Article Feed", "Ticket Feed" and so on.
> I think it would be nice to qualify them as related to the website, i.e.
>
> "Nitro Project's Article Feed"
>
> What do other people and george think?
>
>
> --
> blog en: http://www.riffraff.info
> blog it: http://riffraff.blogsome.com
> jabber : rff.rff at gmail dot com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sun Jul 23 17:10:47 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 24 Jul 2006 00:10:47 +0300
Subject: [Nitro] nitroproject.org wiki not editable
In-Reply-To: <0B1C470D-3608-4C7B-BC8B-D18FD0F1BD7D@oggu.de>
References: 
	
	<0B1C470D-3608-4C7B-BC8B-D18FD0F1BD7D@oggu.de>
Message-ID: 

Have you logged in?

-g.

On 7/23/06, Fabian Buch  wrote:
> Still (or again) throws "Error"
>
> Am 22.07.2006 um 13:56 schrieb George Moschovitis:
> > fixed!
> >
> > -g.
> >
> > PS: please add more sites
> >
> > On 7/22/06, Fabian Buch  wrote:
> >> the nitroproject.org wiki throws an Error if I want to edit a page
> >> (http://nitroproject.org/wiki/edit/Public+sites+powered+by+Nitro/ for
> >> example)
> >>
> >> Ticket: http://nitroproject.org/tickets/read/24
> >>
> >> Fabian
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sun Jul 23 17:17:39 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 24 Jul 2006 00:17:39 +0300
Subject: [Nitro] nitroproject.org wiki not editable
In-Reply-To: 
References: 
	
	<0B1C470D-3608-4C7B-BC8B-D18FD0F1BD7D@oggu.de>
	
Message-ID: 

Ok, now it works for sure :)

-g.

On 7/24/06, George Moschovitis  wrote:
> Have you logged in?
>
> -g.
>
> On 7/23/06, Fabian Buch  wrote:
> > Still (or again) throws "Error"
> >
> > Am 22.07.2006 um 13:56 schrieb George Moschovitis:
> > > fixed!
> > >
> > > -g.
> > >
> > > PS: please add more sites
> > >
> > > On 7/22/06, Fabian Buch  wrote:
> > >> the nitroproject.org wiki throws an Error if I want to edit a page
> > >> (http://nitroproject.org/wiki/edit/Public+sites+powered+by+Nitro/ for
> > >> example)
> > >>
> > >> Ticket: http://nitroproject.org/tickets/read/24
> > >>
> > >> Fabian
> >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> --
> http://www.gmosx.com
> http://www.nitroproject.org
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From m.fellinger at gmail.com  Sun Jul 23 19:29:14 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Mon, 24 Jul 2006 08:29:14 +0900
Subject: [Nitro] [PATCH] Console::Command in nitro/bin/nitro
Message-ID: <9c00d3e00607231629occ82edeh5216371a14ab63b8@mail.gmail.com>

just a quick patch for the wrong require-path... this is also wrong in
the beta-gems :|
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle
Type: application/octet-stream
Size: 318 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060724/cc08e90c/attachment.obj 

From bryan.a.soto at gmail.com  Mon Jul 24 01:00:19 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 23 Jul 2006 22:00:19 -0700
Subject: [Nitro] [PATCH] Console::Command in nitro/bin/nitro
In-Reply-To: <9c00d3e00607231629occ82edeh5216371a14ab63b8@mail.gmail.com>
References: <9c00d3e00607231629occ82edeh5216371a14ab63b8@mail.gmail.com>
Message-ID: 

On 7/23/06, Michael Fellinger  wrote:
> just a quick patch for the wrong require-path... this is also wrong in
> the beta-gems :|
>

And the release... I've applied, thanks.

From bryan.a.soto at gmail.com  Mon Jul 24 01:01:23 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 23 Jul 2006 22:01:23 -0700
Subject: [Nitro] [ANN] Nitro/Og 0.31.0
In-Reply-To: 
References: 
	
Message-ID: 

On 7/21/06, George Moschovitis  wrote:
> One thing, could you please announce this to c.l.r too?
> If you do, please make clear that this is a maintenance release.
>

I actually don't have access at the moment. If someone else could
forward, I'd appreciate it.

From surrender_it at yahoo.it  Mon Jul 24 04:31:36 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Mon, 24 Jul 2006 10:31:36 +0200
Subject: [Nitro] [ANN] Nitro/Og 0.31.0
In-Reply-To: 
References: 	
	
Message-ID: 

Bryan Soto ha scritto:
> On 7/21/06, George Moschovitis  wrote:
> 
>>One thing, could you please announce this to c.l.r too?
>>If you do, please make clear that this is a maintenance release.
>>
> 
> 
> I actually don't have access at the moment. If someone else could
> forward, I'd appreciate it.


doing it now via c.l.r.

-- 

blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From william.full.moon at gmail.com  Mon Jul 24 08:03:55 2006
From: william.full.moon at gmail.com (* William)
Date: Mon, 24 Jul 2006 22:03:55 +1000
Subject: [Nitro] filter on Tickets
Message-ID: <000601c6af19$a66ecbc0$b3af07ca@ghostgum>

Hi all ...
 
I really like the Tickets system.  I submitted a ticket today to filter
tickets -- Unfortunately Tickets are broken.  I had an ERROR page when I
clicked on [CREATE].  I don't love that so much, still I also like the
special 'cool aspect' of Nitro sites that allows evolution of the systems
like this!!!
 
I'd like to begin a discussion about "Deployment & Release Management" if
any one is interested.  In this context my idea is to release new,
evolutionary increments in a way that sustains function.  For example if I
get "error" -- I could roll-back to the last "good release".  When there is
enough interest we can document options and choices on the wiki.  Yea!!!
 
Things are moving, and making progress.  I can tell by the 'energy' and
communication in the group and wiki.  You guys are great.
 
Aloha,
           Will.
 
___________________
Hey !!
 
The tickets list is getting bigger.  It would great imho if this tickets
display could be filtered by category, and even a keywords.  
 
Apart form any obvious uses, a popular product like *Nitro* will generate a
lot of input and as Ghandi said once, "_You have to BE the change you want
to be._"
 
Love the work.  *Nitro* is looking really good.
 
Aloha /Will.
 
____________________________________________________________________________
____
  (2006) Information proprietary and confidential intended for direct
recipient(s) and mutually agreed co-respondents.
  HYPERLINK
"http://adroit-process.blogspot.com/"http://adroit-process.blogspot.com/
 

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.3/395 - Release Date: 21-Jul-2006
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060724/ff744cbe/attachment.html 

From george.moschovitis at gmail.com  Mon Jul 24 08:12:16 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 24 Jul 2006 15:12:16 +0300
Subject: [Nitro] [PATCH] Console::Command in nitro/bin/nitro
In-Reply-To: 
References: <9c00d3e00607231629occ82edeh5216371a14ab63b8@mail.gmail.com>
	
Message-ID: 

Please,

I would really lke to deprecate the devlab repo and keep jus tone repo
(repo.nitroproject.org)
This way we will not duplicate our work. Brian, is it possible that
you could provide a patch against my repo with some (even better all)
of the missing fixes?

thanks in advance,
George.

On 7/24/06, Bryan Soto  wrote:
> On 7/23/06, Michael Fellinger  wrote:
> > just a quick patch for the wrong require-path... this is also wrong in
> > the beta-gems :|
> >
>
> And the release... I've applied, thanks.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From lasso at lassoweb.se  Mon Jul 24 12:00:20 2006
From: lasso at lassoweb.se (Lars Olsson)
Date: Mon, 24 Jul 2006 18:00:20 +0200
Subject: [Nitro] Missing  declaration
Message-ID: <44C4EE94.4080204@lassoweb.se>

Hi list,

When validating my nitro pages against W3C validation service it 
complains about missing  declarations. My source .xhtml files 
contains this information and the  tag is still present after 
REXML has gone through the file (in lib/nitro/compiler/morphing.rb). 
After that it get "lost" somewhere and doesn't show up in the final 
page. Is this by design or a bug?

Kindly

/lasso

-- 
________________________________________
Lars Olsson
lasso at lassoweb.se
http://www.lassoweb.se/

From billk at cts.com  Mon Jul 24 12:08:34 2006
From: billk at cts.com (Bill Kelly)
Date: Mon, 24 Jul 2006 09:08:34 -0700
Subject: [Nitro] PostgreSQL request
References: 
Message-ID: <010501c6af3b$6d0233f0$6442a8c0@musicbox>

From: "George Moschovitis" 
>
> can someone that uses PostgreSQL post some info to the list (and the
> wiki) about how he setups Postgres (autorization, socket/tcp, etc
> etc)?

Hi,

For some reason I prefer password authorization instead of ident, so
I set up my /etc/postgresql/pg_hba.conf like this:

# TYPE  DATABASE    USER        IP-ADDRESS        IP-MASK           METHOD
# Database administrative login by UNIX sockets
local   all         postgres                                        ident sameuser
#
# All other connections by UNIX sockets
local   all         all                                             md5
#
# All IPv4 connections from localhost
host    all         all         127.0.0.1         255.255.255.255   md5
#
# All IPv6 localhost connections
host    all         all         ::1               ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff        md5
host    all         all         ::ffff:127.0.0.1/128                md5
#
# reject all other connection attempts
host    all         all         0.0.0.0           0.0.0.0           reject



Note that for whatever reason, the :password field in the Og.setup 
doesn't seem to work with postgres.  postgresql would prompt for the
password each time.

Someone on #nitro IRC referred me to the ~/.pgpass file.
http://pgsqld.active-venture.com/libpq-files.html

So, I created a ~/.pgpass file like:
*:*:*:dbuser:passwd


Since then Og + postgresql has worked fine for me.


Hope this helps,

Regards,

Bill



From george.moschovitis at gmail.com  Mon Jul 24 13:43:17 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 24 Jul 2006 20:43:17 +0300
Subject: [Nitro] Missing  declaration
In-Reply-To: <44C4EE94.4080204@lassoweb.se>
References: <44C4EE94.4080204@lassoweb.se>
Message-ID: 

There is a workaround, add this to your skin (ie main page element).

-g.

On 7/24/06, Lars Olsson  wrote:
> Hi list,
>
> When validating my nitro pages against W3C validation service it
> complains about missing  declarations. My source .xhtml files
> contains this information and the  tag is still present after
> REXML has gone through the file (in lib/nitro/compiler/morphing.rb).
> After that it get "lost" somewhere and doesn't show up in the final
> page. Is this by design or a bug?
>
> Kindly
>
> /lasso
>
> --
> ________________________________________
> Lars Olsson
> lasso at lassoweb.se
> http://www.lassoweb.se/
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From transfire at gmail.com  Tue Jul 25 09:03:57 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 25 Jul 2006 09:03:57 -0400
Subject: [Nitro] [ANN] Facets 1.6.20 beta
In-Reply-To: 
References: <4b6f054f0607222207x7d56fc4drbde5dbaf174d6ce5@mail.gmail.com>
	
Message-ID: <4b6f054f0607250603q2f6c7ae7n7628dec7f09f43a3@mail.gmail.com>

Made a few fixes/touch ups with 1.6.26.

T.

From transfire at gmail.com  Tue Jul 25 11:02:07 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 25 Jul 2006 11:02:07 -0400
Subject: [Nitro] Package format
Message-ID: <4b6f054f0607250802n77f263ecj87b0211031ff77cd@mail.gmail.com>

Well, I'm closing in on a package format. I've gone with the name Rock
and an extension .rbr. The neat thing about this package format is
that the package itself is an excutable ruby script. Just run it and
it installs the package. The system is almost totally decentralized.
The only central piece of info needed is a dependency list, so they
can be automatically installed too. All in all, it's starting to look
pretty good. With Rolls you'll be able to require these packages
directly too.

One problem I have though is how to decide if something is already
installed? The system uses setup.rb, so files are installed in
standard locations. It would be nice if I could accomodate that. And
also do uninstalls too. But I'm not sure if I can, and I may have to
install to a special location and tie it into the $LOAD_PATH like
gems. But I would prefer not. Anyone have any ideas?

Thanks,
T.

From aidan at infurious.com  Tue Jul 25 11:17:49 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Tue, 25 Jul 2006 16:17:49 +0100
Subject: [Nitro] Package format
In-Reply-To: <4b6f054f0607250802n77f263ecj87b0211031ff77cd@mail.gmail.com>
References: <4b6f054f0607250802n77f263ecj87b0211031ff77cd@mail.gmail.com>
Message-ID: <10349FE6-451E-4E5B-BB2F-390A2784246C@infurious.com>

Why not have a registry of some sort?  a mapping of packages to  
locations (and possibly versions)?  The installer could then first  
check the registry.  It could live by default in e.g. /usr/local/etc/ 
rock.reg and an alternative could be specified on the command line.

Aidan

On 25/07/2006, at 4:02 PM, TRANS wrote:

> One problem I have though is how to decide if something is already
> installed? The system uses setup.rb, so files are installed in
> standard locations. It would be nice if I could accomodate that. And
> also do uninstalls too. But I'm not sure if I can, and I may have to
> install to a special location and tie it into the $LOAD_PATH like
> gems. But I would prefer not. Anyone have any ideas?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060725/df24947e/attachment.html 

From transfire at gmail.com  Tue Jul 25 11:32:38 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 25 Jul 2006 11:32:38 -0400
Subject: [Nitro] Package format
In-Reply-To: <10349FE6-451E-4E5B-BB2F-390A2784246C@infurious.com>
References: <4b6f054f0607250802n77f263ecj87b0211031ff77cd@mail.gmail.com>
	<10349FE6-451E-4E5B-BB2F-390A2784246C@infurious.com>
Message-ID: <4b6f054f0607250832t2e461fb1odf4ba8a6a7fc2ec6@mail.gmail.com>

On 7/25/06, Aidan Rogers  wrote:
>
> Why not have a registry of some sort?  a mapping of packages to locations
> (and possibly versions)?  The installer could then first check the registry.
>  It could live by default in e.g. /usr/local/etc/rock.reg and an alternative
> could be specified on the command line.

That a good idea. I wan't giving that course credit, but you are
right. That's probably the best way to do it.

When setup.rb runs it creates a manifest of the files it installed in
"InstalledFiles" I could copy that to a rock registry orgranized by

  packagename/
    version/
      InstalledFiles

That way I'd know that package is installed and exactly what it
installed. Of course there may be issues with two packges installing
the same file, but maybe that's not something Rock should have to
worry about --eg. it's the programmer fault?

Thanks!
T.

From aidan at infurious.com  Tue Jul 25 11:34:44 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Tue, 25 Jul 2006 16:34:44 +0100
Subject: [Nitro] Package format
In-Reply-To: <4b6f054f0607250832t2e461fb1odf4ba8a6a7fc2ec6@mail.gmail.com>
References: <4b6f054f0607250802n77f263ecj87b0211031ff77cd@mail.gmail.com>
	<10349FE6-451E-4E5B-BB2F-390A2784246C@infurious.com>
	<4b6f054f0607250832t2e461fb1odf4ba8a6a7fc2ec6@mail.gmail.com>
Message-ID: <2A2B2C96-78A6-4ED9-A9A2-F725B7B844AD@infurious.com>


> That way I'd know that package is installed and exactly what it
> installed. Of course there may be issues with two packges installing
> the same file, but maybe that's not something Rock should have to
> worry about --eg. it's the programmer fault?

Realistically, that won't happen often.  If it does, it probably is  
the developer's issue (like namespace clashes).

Aidan

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060725/807e44e1/attachment.html 

From zimba.tm at gmail.com  Tue Jul 25 11:46:22 2006
From: zimba.tm at gmail.com (Jonas Pfenniger)
Date: Tue, 25 Jul 2006 17:46:22 +0200
Subject: [Nitro] Package format
In-Reply-To: <2A2B2C96-78A6-4ED9-A9A2-F725B7B844AD@infurious.com>
References: <4b6f054f0607250802n77f263ecj87b0211031ff77cd@mail.gmail.com>
	<10349FE6-451E-4E5B-BB2F-390A2784246C@infurious.com>
	<4b6f054f0607250832t2e461fb1odf4ba8a6a7fc2ec6@mail.gmail.com>
	<2A2B2C96-78A6-4ED9-A9A2-F725B7B844AD@infurious.com>
Message-ID: <3ff63f9b0607250846i4f8238bcg8aeca7d54d3e32b4@mail.gmail.com>

Why not look in every folder or ruby's load path for the registry ?
That way you can have multiple registries.

-- 
Cheers,
  zimbatm

http://zimbatm.oree.ch

From surrender_it at yahoo.it  Tue Jul 25 13:35:07 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Tue, 25 Jul 2006 19:35:07 +0200
Subject: [Nitro] Package format
In-Reply-To: <4b6f054f0607250832t2e461fb1odf4ba8a6a7fc2ec6@mail.gmail.com>
References: <4b6f054f0607250802n77f263ecj87b0211031ff77cd@mail.gmail.com>	<10349FE6-451E-4E5B-BB2F-390A2784246C@infurious.com>
	<4b6f054f0607250832t2e461fb1odf4ba8a6a7fc2ec6@mail.gmail.com>
Message-ID: 

TRANS ha scritto:
> On 7/25/06, Aidan Rogers  wrote:
> 
>>Why not have a registry of some sort?  a mapping of packages to locations
>>(and possibly versions)?  The installer could then first check the registry.
>> It could live by default in e.g. /usr/local/etc/rock.reg and an alternative
>>could be specified on the command line.
> 
> 
> That a good idea. I wan't giving that course credit, but you are
> right. That's probably the best way to do it.
> 
> When setup.rb runs it creates a manifest of the files it installed in
> "InstalledFiles" I could copy that to a rock registry orgranized by
> 
>   packagename/
>     version/
>       InstalledFiles

Ever loooked at python? They have this nice feature were you can put a 
foo.pth file in the python/site-packages/ directory, where the content 
of the lis basically just a line pointing to the real install directory 
for foo. Seems quite similar to what you're doing, bar versioning.

-- 

blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From transfire at gmail.com  Tue Jul 25 14:29:52 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 25 Jul 2006 14:29:52 -0400
Subject: [Nitro] Package format
In-Reply-To: <3ff63f9b0607250846i4f8238bcg8aeca7d54d3e32b4@mail.gmail.com>
References: <4b6f054f0607250802n77f263ecj87b0211031ff77cd@mail.gmail.com>
	<10349FE6-451E-4E5B-BB2F-390A2784246C@infurious.com>
	<4b6f054f0607250832t2e461fb1odf4ba8a6a7fc2ec6@mail.gmail.com>
	<2A2B2C96-78A6-4ED9-A9A2-F725B7B844AD@infurious.com>
	<3ff63f9b0607250846i4f8238bcg8aeca7d54d3e32b4@mail.gmail.com>
Message-ID: <4b6f054f0607251129o3a719987t94cd85a230c363f@mail.gmail.com>

On 7/25/06, Jonas Pfenniger  wrote:
> Why not look in every folder or ruby's load path for the registry ?
> That way you can have multiple registries.

Not sure how multiple registries would be helpful, bu ton reason is
because that would slow down startup time.

T.

From transfire at gmail.com  Tue Jul 25 14:33:06 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 25 Jul 2006 14:33:06 -0400
Subject: [Nitro] Package format
In-Reply-To: 
References: <4b6f054f0607250802n77f263ecj87b0211031ff77cd@mail.gmail.com>
	<10349FE6-451E-4E5B-BB2F-390A2784246C@infurious.com>
	<4b6f054f0607250832t2e461fb1odf4ba8a6a7fc2ec6@mail.gmail.com>
	
Message-ID: <4b6f054f0607251133i2b46424bv55df8e927288f4d9@mail.gmail.com>

On 7/25/06, gabriele renzi  wrote:
> TRANS ha scritto:
> > On 7/25/06, Aidan Rogers  wrote:
> >
> >>Why not have a registry of some sort?  a mapping of packages to locations
> >>(and possibly versions)?  The installer could then first check the registry.
> >> It could live by default in e.g. /usr/local/etc/rock.reg and an alternative
> >>could be specified on the command line.
> >
> >
> > That a good idea. I wan't giving that course credit, but you are
> > right. That's probably the best way to do it.
> >
> > When setup.rb runs it creates a manifest of the files it installed in
> > "InstalledFiles" I could copy that to a rock registry orgranized by
> >
> >   packagename/
> >     version/
> >       InstalledFiles
>
> Ever loooked at python? They have this nice feature were you can put a
> foo.pth file in the python/site-packages/ directory, where the content
> of the lis basically just a line pointing to the real install directory
> for foo. Seems quite similar to what you're doing, bar versioning.

Hmm... That seems very interesting, but I'm not quite sure what the
benefits of this are.

T.

From transfire at gmail.com  Tue Jul 25 15:07:40 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 25 Jul 2006 15:07:40 -0400
Subject: [Nitro] Be gone for a few days.
Message-ID: <4b6f054f0607251207o630d9d95keea84bd6003e2945@mail.gmail.com>

FYI, I'll be out of twon for a few days. (My great uncle needs me to
drive a car from Maryland down to Florida). Let you knwo wehn I get
back.

BTW I release Facets 1.7.0 today, incuding the .gem. I'm sure there's
still a few loose ends to tie up, but it's close enough that I went
ahead and released it.

Also, the new versions of Reap and Sow are just about ready. So I'll
release those shortly after I return.

T.

From george.moschovitis at gmail.com  Tue Jul 25 17:31:48 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 26 Jul 2006 00:31:48 +0300
Subject: [Nitro] Be gone for a few days.
In-Reply-To: <4b6f054f0607251207o630d9d95keea84bd6003e2945@mail.gmail.com>
References: <4b6f054f0607251207o630d9d95keea84bd6003e2945@mail.gmail.com>
Message-ID: 

Have a nice trip Tom ;-)

-g.



On 7/25/06, TRANS  wrote:
> FYI, I'll be out of twon for a few days. (My great uncle needs me to
> drive a car from Maryland down to Florida). Let you knwo wehn I get
> back.
>
> BTW I release Facets 1.7.0 today, incuding the .gem. I'm sure there's
> still a few loose ends to tie up, but it's close enough that I went
> ahead and released it.
>
> Also, the new versions of Reap and Sow are just about ready. So I'll
> release those shortly after I return.
>
> T.
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Tue Jul 25 17:36:03 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 26 Jul 2006 00:36:03 +0300
Subject: [Nitro] Little forum on NP.org
Message-ID: 

Title says it all. A little experiment at the moment, will be updated
early next week. For the moment add bug reports/suggestions tickets.
And of course add some nice forum discussions.

regards,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From john at oxyliquit.de  Wed Jul 26 16:56:13 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Wed, 26 Jul 2006 22:56:13 +0200
Subject: [Nitro] Hello All!
In-Reply-To: <1153614309.12828.3.camel@localhost>
References: <1153614309.12828.3.camel@localhost>
Message-ID: 

Heya!

> Hello guys I would like to introduce myself to you all, my name is will,
> and I love programming and computers. :-) Manveru showed me nitro and I
> have been in love with it when I first type the very basic but global
> "Hello World!". I hope to contribute to the community and help nitro
> grow with every one else!

Thx manv! :P

> Thanks and It will be fun talking and programming with you guys!

Welcome on board and have fun and I wish you a smooth bumpy ride :)

Jo
PS to everyone: I'm on vacation, almost without internet access, so I'm  
kind of lacking behind.

From lionel.orry at gmail.com  Thu Jul 27 03:16:33 2006
From: lionel.orry at gmail.com (Lionel Orry)
Date: Thu, 27 Jul 2006 09:16:33 +0200
Subject: [Nitro] Hello All!
In-Reply-To: 
References: <1153614309.12828.3.camel@localhost>
	
Message-ID: <44C86851.5080809@gmail.com>



Jonathan Buch wrote:
> [...]
> PS to everyone: I'm on vacation, almost without internet access, so I'm  
> kind of lacking behind.

I had to say, being on vacation with a computer or any other communication support is not being on vacation.
Lucky Jo, you will be able to let fresh oxygen enter your brain again ;-)

Have a nice time.

Lionel


From john at oxyliquit.de  Thu Jul 27 09:46:05 2006
From: john at oxyliquit.de (Jonathan Buch)
Date: Thu, 27 Jul 2006 15:46:05 +0200
Subject: [Nitro] Hello All!
In-Reply-To: <44C86851.5080809@gmail.com>
References: <1153614309.12828.3.camel@localhost>
	 <44C86851.5080809@gmail.com>
Message-ID: 

Hi,

> I had to say, being on vacation with a computer or any other  
> communication support is not being on vacation.
> Lucky Jo, you will be able to let fresh oxygen enter your brain again

Very much true, although I already got a sunburn.  I was exactly 30
minutes in the sun..... *sigh* I should do this more often ;)

> Have a nice time.

Thank you very much.

Jo

From george.moschovitis at gmail.com  Thu Jul 27 16:22:55 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 27 Jul 2006 23:22:55 +0300
Subject: [Nitro] np.org bug fixes
Message-ID: 

Dear devs,

I have fixed the icon and ticket bugs on np.org. So go ahead, add you
icons and test the forum. Report any bugs with a ticket ;-)

regards,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From surrender_it at yahoo.it  Sat Jul 29 07:31:14 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sat, 29 Jul 2006 13:31:14 +0200
Subject: [Nitro] Is repo in a working state?
Message-ID: 

Hi people,

I was trying to merge some stuff from devlab into a local copy of 
repo.nitroproject.org [0].
All the expected test were passing[1] but I was not able to run any of 
the demo apps and it seem that not even a summy "gen app foo" is able to 
work. Then I noticed that it doesn't work either with a clean copy of 
the main repository.

But I'm trying this on win32 cause my laptop died this night (insert 
random bad word here), so could someone please confirm this on another OS?



[0] you can get a bundle of the 35 original patches (out of 58) plus the 
additional conflict-fix ones at:
http://www.riffraff.info/devlab-merge.patch.gz

[1] I mean:
clean repo, nitro tests:
  40 tests, 134 assertions, 3 failures, 1 errors
my repo, nitro tests:
  47 tests, 154 assertions, 3 failures, 1 errors
yeah, the errors and failures are the same, and 1 failure is expected 
due to a bug in ruby < HEAD.


-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From surrender_it at yahoo.it  Sat Jul 29 08:50:56 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sat, 29 Jul 2006 14:50:56 +0200
Subject: [Nitro] Is repo in a working state?
In-Reply-To: 
References: 
Message-ID: 

gabriele renzi ha scritto:



auto follow up: now everything seem to work fine, sorry for making noise.

George, could you consider applying the patch or some patches included 
in it?


From surrender_it at yahoo.it  Sat Jul 29 10:33:44 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sat, 29 Jul 2006 16:33:44 +0200
Subject: [Nitro] [PATCH] form_errors helper
Message-ID: 

Hi people,

The new "edit" page for scaffoldings is something like:

  
     Show all?
     show form
   
     show errors
     show form{
      "Attributes for Foo"
      f.attributes
      "Relations for foo"
      f.relations
     }
   

I've made an helper out of the "show errors" part, so the view code is 
cleaner, and this is a functionality that could be useful to others.

   * This patch refactors the code used in the scaffolded view to be 
usable as a simple helper method (part of FormHelper).
   It basically creates a div containing a  list of  tags, each 
containing a string "name: message" for the errors in flash[:ERRORS]
Tests attached.


Also, why is the form view different on the two sides of the else?
We could possibly use the same code in both and push the new 
functionality, i.e. highlight of what the field mean, into  form_for.
Possibly with an option ( :field_headers=>true or something).
What do others think?

-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: form_errors.patch.gz
Type: application/octet-stream
Size: 5054 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060729/9e30ddd5/attachment.obj 

From george.moschovitis at gmail.com  Sat Jul 29 14:38:26 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 29 Jul 2006 21:38:26 +0300
Subject: [Nitro] Is repo in a working state?
In-Reply-To: 
References:  
Message-ID: 

> George, could you consider applying the patch or some patches included
> in it?

I tried to apply a similar patch from bsoto some days ago but got some
problems. I will try your patch tommorow when I get back to my laptop
;-)

btw, use prefer the forum on np.org ;-)

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sat Jul 29 14:40:52 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 29 Jul 2006 21:40:52 +0300
Subject: [Nitro] [PATCH] form_errors helper
In-Reply-To: 
References: 
Message-ID: 

If you want to work on the admin screens to improve them, please do so.
I will work on other things till wednesday, so you can work
exclusively on this if you want. After wednesday I will try to work on
wrapping up nitro version 0.40.0 for release.

regards,
George.


On 7/29/06, gabriele renzi  wrote:
> Hi people,
>
> The new "edit" page for scaffoldings is something like:
>
>   
>      Show all?
>      show form
>    
>      show errors
>      show form{
>       "Attributes for Foo"
>       f.attributes
>       "Relations for foo"
>       f.relations
>      }
>    
>
> I've made an helper out of the "show errors" part, so the view code is
> cleaner, and this is a functionality that could be useful to others.
>
>    * This patch refactors the code used in the scaffolded view to be
> usable as a simple helper method (part of FormHelper).
>    It basically creates a div containing a  list of  tags, each
> containing a string "name: message" for the errors in flash[:ERRORS]
> Tests attached.
>
>
> Also, why is the form view different on the two sides of the else?
> We could possibly use the same code in both and push the new
> functionality, i.e. highlight of what the field mean, into  form_for.
> Possibly with an option ( :field_headers=>true or something).
> What do others think?
>
> --
> blog en: http://www.riffraff.info
> blog it: http://riffraff.blogsome.com
> jabber : rff.rff at gmail dot com
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From surrender_it at yahoo.it  Sat Jul 29 15:54:38 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sat, 29 Jul 2006 21:54:38 +0200
Subject: [Nitro] [PATCH] form_errors helper
In-Reply-To: 
References: 
	
Message-ID: 

George Moschovitis ha scritto:
> If you want to work on the admin screens to improve them, please do so.
> I will work on other things till wednesday, so you can work
> exclusively on this if you want. After wednesday I will try to work on
> wrapping up nitro version 0.40.0 for release.

Ok, but I have a small problem, due to #render behaving differently from 
  the old version.
I tried looking into the code.. imho it should work  :) but there are 
two failing tests for it, which were working with 0.31, and they are 
quite important to me (i.e redirect('foo') redirects to /foo instead of 
self/foo).
It would be nice if you could take a look at this.

-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From surrender_it at yahoo.it  Sat Jul 29 16:46:04 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sat, 29 Jul 2006 22:46:04 +0200
Subject: [Nitro] [PATCH] form_errors helper
In-Reply-To: 
References: 	
	
Message-ID: 

gabriele renzi ha scritto:
> George Moschovitis ha scritto:

> Ok, but I have a small problem, due to #render behaving differently from 
>   the old version.

forgot to say: replacing lnitro/render.rb with the one from devlab makes 
the errors go away.
It would reintroduce the @base instance variable but imho that could 
still be useful, even if mounth_path is better in most cases.

I think there are still soime uses to it, i.e. thinking of mounting a 
controller over another, for example,  it could be possible to 
autogenerate breadcrumbs.


From george.moschovitis at gmail.com  Sun Jul 30 03:07:34 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 30 Jul 2006 10:07:34 +0300
Subject: [Nitro] [PATCH] form_errors helper
In-Reply-To: 
References: 
	
	 
Message-ID: 

Can you explain a bit better what you mean? I removed base along other
things while simplifying the code. And I think render works more
corectly now, especially handling relative/absolute urls. Can you
explain the problem in more detail please?

thanks in advance!
George.

On 7/29/06, gabriele renzi  wrote:
> gabriele renzi ha scritto:
> > George Moschovitis ha scritto:
>
> > Ok, but I have a small problem, due to #render behaving differently from
> >   the old version.
>
> forgot to say: replacing lnitro/render.rb with the one from devlab makes
> the errors go away.
> It would reintroduce the @base instance variable but imho that could
> still be useful, even if mounth_path is better in most cases.
>
> I think there are still soime uses to it, i.e. thinking of mounting a
> controller over another, for example,  it could be possible to
> autogenerate breadcrumbs.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Sun Jul 30 03:59:59 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 30 Jul 2006 10:59:59 +0300
Subject: [Nitro] Is repo in a working state?
In-Reply-To: 
References: 
Message-ID: 

The patchfile is corrupted, can you please zip and attach this?

thanks,
George.


On 7/29/06, gabriele renzi  wrote:
> Hi people,
>
> I was trying to merge some stuff from devlab into a local copy of
> repo.nitroproject.org [0].
> All the expected test were passing[1] but I was not able to run any of
> the demo apps and it seem that not even a summy "gen app foo" is able to
> work. Then I noticed that it doesn't work either with a clean copy of
> the main repository.
>
> But I'm trying this on win32 cause my laptop died this night (insert
> random bad word here), so could someone please confirm this on another OS?
>
>
>
> [0] you can get a bundle of the 35 original patches (out of 58) plus the
> additional conflict-fix ones at:
> http://www.riffraff.info/devlab-merge.patch.gz
>
> [1] I mean:
> clean repo, nitro tests:
>   40 tests, 134 assertions, 3 failures, 1 errors
> my repo, nitro tests:
>   47 tests, 154 assertions, 3 failures, 1 errors
> yeah, the errors and failures are the same, and 1 failure is expected
> due to a bug in ruby < HEAD.
>
>
> --
> blog en: http://www.riffraff.info
> blog it: http://riffraff.blogsome.com
> jabber : rff.rff at gmail dot com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From surrender_it at yahoo.it  Sun Jul 30 04:33:50 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sun, 30 Jul 2006 10:33:50 +0200
Subject: [Nitro] [PATCH] form_errors helper
In-Reply-To: 
References: 		
	
	
Message-ID: 

George Moschovitis ha scritto:
> Can you explain a bit better what you mean? I removed base along other
> things while simplifying the code. And I think render works more
> corectly now, especially handling relative/absolute urls. Can you
> explain the problem in more detail please?

well, if you run the tests you should see two errors :)
My problem is this: I have a controller Bar mounted at

  /foo/bar/

and I am executing
  /foo/bar/act1

in the action Bar#act1 I do a
  redirect "act2"
this should send me to
  /foo/bar/act2
but I get an error because it redirects to /act2




-- 

blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From surrender_it at yahoo.it  Sun Jul 30 04:41:12 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Sun, 30 Jul 2006 10:41:12 +0200
Subject: [Nitro] Is repo in a working state?
In-Reply-To: 
References: 
	
Message-ID: 

George Moschovitis ha scritto:
> The patchfile is corrupted, can you please zip and attach this?

sure here it is.

The test for nitro are:
54 tests, 164 assertions, 3 failures, 1 errors



I hope you can merge something even if not all the patches.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: devlab_sync.zip
Type: application/octet-stream
Size: 34610 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060730/fea3a7f8/attachment-0001.obj 

From george.moschovitis at gmail.com  Sun Jul 30 09:24:07 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 30 Jul 2006 16:24:07 +0300
Subject: [Nitro] Is repo in a working state?
In-Reply-To: 
References: 
	
	
Message-ID: 

> I hope you can merge something even if not all the patches.

I merged quite a lot of patches from the bundle. I havet merged your
gen stuff because I got some conflicts with my code. Please continue
working on this, and I 'll merge them later this week.

thanks a lot to you, bsoto and all the commiters for the bug fixes.

-g.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From manveru at weez-int.com  Sun Jul 30 20:23:38 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Mon, 31 Jul 2006 09:23:38 +0900
Subject: [Nitro] Is repo in a working state?
In-Reply-To: 
References:  
	
Message-ID: <200607310923.38550.manveru@weez-int.com>

On Sunday 30 July 2006 03:38, George Moschovitis wrote:
> > George, could you consider applying the patch or some patches included
> > in it?
>
> I tried to apply a similar patch from bsoto some days ago but got some
> problems. I will try your patch tommorow when I get back to my laptop
> ;-)
>
> btw, use prefer the forum on np.org ;-)

Uhm, since i'm not a big forum-fan...
is there some way to mirror that stuff to the mailinglist?

>
> -g.

From james.britt at gmail.com  Sun Jul 30 23:08:24 2006
From: james.britt at gmail.com (James Britt)
Date: Sun, 30 Jul 2006 20:08:24 -0700
Subject: [Nitro] Is repo in a working state?
In-Reply-To: <200607310923.38550.manveru@weez-int.com>
References: 
		
	<200607310923.38550.manveru@weez-int.com>
Message-ID: <44CD7428.4070107@gmail.com>

Michael Fellinger wrote:
> On Sunday 30 July 2006 03:38, George Moschovitis wrote:
> 
>>>George, could you consider applying the patch or some patches included
>>>in it?
>>
>>I tried to apply a similar patch from bsoto some days ago but got some
>>problems. I will try your patch tommorow when I get back to my laptop
>>;-)
>>
>>btw, use prefer the forum on np.org ;-)
> 
> 
> Uhm, since i'm not a big forum-fan...
> is there some way to mirror that stuff to the mailinglist?

Or something.  There is no way I will follow a Web forum; too much work 
to have to go look for stuff instead of having it sent right to me.



-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://yourelevatorpitch.com  - Finding Business Focus

From manveru at weez-int.com  Mon Jul 31 00:58:25 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Mon, 31 Jul 2006 13:58:25 +0900
Subject: [Nitro] [PATCH] JS autocomplete and require_files
Message-ID: <200607311358.25589.manveru@weez-int.com>

There was a bug in the javascript autocomplete...
so when you have an action that takes an argument:

#### controller.rb ####

class MyController < Nitro::Controller
  helper :javascript

  def mail_new address = 'bar'
    do_something_with_bar
  end

  def address_auto_complete
    build {
      ul {
        li 'manveru'
        li 'michael'
        }
      }
  end
end

#### controller.rb ####

now you get two problems:
1)
auto-complete tries to call /mail_new/address_auto_complete - 
not /address_auto_complete - this is _not_ amusing...
i added a mandatory / not sure if that's a good solution, but it doesn't break 
anything as far as i can see.

2)
every time one reloads, Javascript.require_files is extended and so the list 
of the javascript-files that are included is growing and growing... over time 
that is a significant problem - i haven't found out where exactly that 
happens, and my patch is just a quick workaround, calling uniq! on 
Javascript.require_files - but it's still being populated :|
this is a odd thing also because it doesn't happen everytime... i think it's 
triggered by reloading sourcefiles or something similar...


thanks for your time :)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.tar.gz
Type: application/x-tgz
Size: 4673 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060731/670c65b9/attachment.bin 

From manveru at weez-int.com  Mon Jul 31 01:06:15 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Mon, 31 Jul 2006 14:06:15 +0900
Subject: [Nitro] [PATCH] JS autocomplete and require_files
In-Reply-To: <200607311358.25589.manveru@weez-int.com>
References: <200607311358.25589.manveru@weez-int.com>
Message-ID: <200607311406.15149.manveru@weez-int.com>

FSKN... just found that i broke something... will resend the fixed one later

On Monday 31 July 2006 13:58, Michael Fellinger wrote:
> There was a bug in the javascript autocomplete...
> so when you have an action that takes an argument:
>
> #### controller.rb ####
>
> class MyController < Nitro::Controller
>   helper :javascript
>
>   def mail_new address = 'bar'
>     do_something_with_bar
>   end
>
>   def address_auto_complete
>     build {
>       ul {
>         li 'manveru'
>         li 'michael'
>         }
>       }
>   end
> end
>
> #### controller.rb ####
>
> now you get two problems:
> 1)
> auto-complete tries to call /mail_new/address_auto_complete -
> not /address_auto_complete - this is _not_ amusing...
> i added a mandatory / not sure if that's a good solution, but it doesn't
> break anything as far as i can see.
>
> 2)
> every time one reloads, Javascript.require_files is extended and so the
> list of the javascript-files that are included is growing and growing...
> over time that is a significant problem - i haven't found out where exactly
> that happens, and my patch is just a quick workaround, calling uniq! on
> Javascript.require_files - but it's still being populated :|
> this is a odd thing also because it doesn't happen everytime... i think
> it's triggered by reloading sourcefiles or something similar...
>
>
> thanks for your time :)

From manveru at weez-int.com  Mon Jul 31 01:18:49 2006
From: manveru at weez-int.com (Michael Fellinger)
Date: Mon, 31 Jul 2006 14:18:49 +0900
Subject: [Nitro] [PATCH] JS autocomplete and require_files
In-Reply-To: <200607311406.15149.manveru@weez-int.com>
References: <200607311358.25589.manveru@weez-int.com>
	<200607311406.15149.manveru@weez-int.com>
Message-ID: <200607311418.49325.manveru@weez-int.com>

OK, sorry for the problems, this patch works now without problems...
i forgot that .uniq! returns nil :|
wonder why it worked first though.

I also can now say that the trigger for the addition of the same js-files in 
the Javascript.require_files is to update a file while the application is 
running - don't ask me why... :)
would be nice if someone could investigate this further - my feeling tells me 
that it has something to do with the aspects...

On Monday 31 July 2006 14:06, Michael Fellinger wrote:
> FSKN... just found that i broke something... will resend the fixed one
> later
>
> On Monday 31 July 2006 13:58, Michael Fellinger wrote:
> > There was a bug in the javascript autocomplete...
> > so when you have an action that takes an argument:
> >
> > #### controller.rb ####
> >
> > class MyController < Nitro::Controller
> >   helper :javascript
> >
> >   def mail_new address = 'bar'
> >     do_something_with_bar
> >   end
> >
> >   def address_auto_complete
> >     build {
> >       ul {
> >         li 'manveru'
> >         li 'michael'
> >         }
> >       }
> >   end
> > end
> >
> > #### controller.rb ####
> >
> > now you get two problems:
> > 1)
> > auto-complete tries to call /mail_new/address_auto_complete -
> > not /address_auto_complete - this is _not_ amusing...
> > i added a mandatory / not sure if that's a good solution, but it doesn't
> > break anything as far as i can see.
> >
> > 2)
> > every time one reloads, Javascript.require_files is extended and so the
> > list of the javascript-files that are included is growing and growing...
> > over time that is a significant problem - i haven't found out where
> > exactly that happens, and my patch is just a quick workaround, calling
> > uniq! on Javascript.require_files - but it's still being populated :|
> > this is a odd thing also because it doesn't happen everytime... i think
> > it's triggered by reloading sourcefiles or something similar...
> >
> >
> > thanks for your time :)
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.tar.gz
Type: application/x-tgz
Size: 4674 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060731/8766664a/attachment-0001.bin 

From surrender_it at yahoo.it  Mon Jul 31 11:28:49 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Mon, 31 Jul 2006 17:28:49 +0200
Subject: [Nitro] use facets' __DIR__ ?
Message-ID: 

Hi everyone,

It seem that the most common idiom in Nitro/Og is this:

something(File.dirname(__FILE__), other, stuff)

I have :

$ grep File.dirname(__FILE__) * -r | grep -v darcs| grep -v patch| wc -l
183

Could we possibly replace it with facets/core/kernel/__dir__.rb, which 
gives us the nice __DIR__ method?

It makes code a little more readable, IMVHO.

-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From george.moschovitis at gmail.com  Mon Jul 31 12:13:31 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 31 Jul 2006 19:13:31 +0300
Subject: [Nitro] [PATCH] JS autocomplete and require_files
In-Reply-To: <200607311418.49325.manveru@weez-int.com>
References: <200607311358.25589.manveru@weez-int.com>
	<200607311406.15149.manveru@weez-int.com>
	<200607311418.49325.manveru@weez-int.com>
Message-ID: 

> I also can now say that the trigger for the addition of the same js-files in
> the Javascript.require_files is to update a file while the application is
> running - don't ask me why... :)
> would be nice if someone could investigate this further - my feeling tells me
> that it has something to do with the aspects...

Ok, I will check this out,

thanks,
George.

-- 
http://www.gmosx.com
http://www.nitroproject.org

From george.moschovitis at gmail.com  Mon Jul 31 12:22:04 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 31 Jul 2006 19:22:04 +0300
Subject: [Nitro] use facets' __DIR__ ?
In-Reply-To: 
References: 
Message-ID: 

Ok, why not ;-) how about providing a patch with this single change? ;-)

-g.

On 7/31/06, gabriele renzi  wrote:
> Hi everyone,
>
> It seem that the most common idiom in Nitro/Og is this:
>
> something(File.dirname(__FILE__), other, stuff)
>
> I have :
>
> $ grep File.dirname(__FILE__) * -r | grep -v darcs| grep -v patch| wc -l
> 183
>
> Could we possibly replace it with facets/core/kernel/__dir__.rb, which
> gives us the nice __DIR__ method?
>
> It makes code a little more readable, IMVHO.
>
> --
> blog en: http://www.riffraff.info
> blog it: http://riffraff.blogsome.com
> jabber : rff.rff at gmail dot com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

From surrender_it at yahoo.it  Mon Jul 31 13:20:42 2006
From: surrender_it at yahoo.it (gabriele renzi)
Date: Mon, 31 Jul 2006 19:20:42 +0200
Subject: [Nitro] @template_root problem with subclasses and specific
	sequence of actions
Message-ID: 

Basically a code like this:

----8<------
class SuperController < Nitro::Controller
  def setup_template_root(path)
   @template_root=[somepath]
  end
end

class SubController < SuperController
  def setup_template_root(path)
   @template_root=[some other path]
  end
end

System.map['/super']=SuperController
System.map['/super/sub']=SubController
------->8--------


Notice that everything works fine if SubController does not inherit from 
SuperController.

Now, if I access
  /super/sub/index
and then
  /super/index/
everything works fine.

If, on the other hand, I access
  /super/index/
first, and just later
  /super/sub/index
everything goes mad, because the templates for the latter got searched 
in the template_root of the first.

I've been trying to fix this for a lot but I have no idea of what is 
happening. When the application starts the template_root is calculated 
in Publishable.mount_at and it seem to be set up correctly, but then 
things fail if I follow that specific sequence of access.
BTW, this seem somehow similar to a bug notifed from zimba.tm some 
months ago. If someone has any kind of clue, could you share it?


While we're at it: do I recall correctly that just doing
  @template_root =[something]
in a controller is deprecated ?
If it's not, then there is a bug in Publishable.mount_at where it gets 
reset to [] unconditionally.

Cheers

-- 
blog en: http://www.riffraff.info
blog it: http://riffraff.blogsome.com
jabber : rff.rff at gmail dot com


From george.moschovitis at gmail.com  Mon Jul 31 14:25:56 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 31 Jul 2006 21:25:56 +0300
Subject: [Nitro] @template_root problem with subclasses and specific
	sequence of actions
In-Reply-To: 
References: 
Message-ID: 

Hmm, I dont have access to my laptop at the moment, will have a look
at this tommorow,
thanks for reporting. If you someone can shed some more light into
this it would be nice.

regards,
George.



On 7/31/06, gabriele renzi  wrote:
> Basically a code like this:
>
> ----8<------
> class SuperController < Nitro::Controller
>   def setup_template_root(path)
>    @template_root=[somepath]
>   end
> end
>
> class SubController < SuperController
>   def setup_template_root(path)
>    @template_root=[some other path]
>   end
> end
>
> System.map['/super']=SuperController
> System.map['/super/sub']=SubController
> ------->8--------
>
>
> Notice that everything works fine if SubController does not inherit from
> SuperController.
>
> Now, if I access
>   /super/sub/index
> and then
>   /super/index/
> everything works fine.
>
> If, on the other hand, I access
>   /super/index/
> first, and just later
>   /super/sub/index
> everything goes mad, because the templates for the latter got searched
> in the template_root of the first.
>
> I've been trying to fix this for a lot but I have no idea of what is
> happening. When the application starts the template_root is calculated
> in Publishable.mount_at and it seem to be set up correctly, but then
> things fail if I follow that specific sequence of access.
> BTW, this seem somehow similar to a bug notifed from zimba.tm some
> months ago. If someone has any kind of clue, could you share it?
>
>
> While we're at it: do I recall correctly that just doing
>   @template_root =[something]
> in a controller is deprecated ?
> If it's not, then there is a bug in Publishable.mount_at where it gets
> reset to [] unconditionally.
>
> Cheers
>
> --
> blog en: http://www.riffraff.info
> blog it: http://riffraff.blogsome.com
> jabber : rff.rff at gmail dot com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


-- 
http://www.gmosx.com
http://www.nitroproject.org

