From george.moschovitis at gmail.com  Wed Aug  1 02:52:10 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Aug 2007 09:52:10 +0300
Subject: [Nitro] Facets SVN
In-Reply-To: <1185914350.423367.174310@o61g2000hsh.googlegroups.com>
References: <1185914350.423367.174310@o61g2000hsh.googlegroups.com>
Message-ID: <afd80b540707312352w59583ed8xccd2c53e6aef1558@mail.gmail.com>

Hello Tom,

I am on vacation these days (but I managed to find an internet connection)
I will checkout the code now.

-g.

On 7/31/07, Trans <transfire at gmail.com> wrote:
>
> Hi--
>
> I just imported Facets into the Rubyforge subversion repository. You
> should be able to get an anonymous checkout of the latest with:
>
>   svn checkout svn://rubyforge.org/var/svn/facets/trunk facets/
>
> Facets 2.0 is coming along nicely --still some things I'd like to get
> cleaned up, but a release (beta?) is Real Soon Now(tm).
>
> Btw, if you are interested in becoming a honest to goodness ruby spit-
> shining Facets developer, let me know.
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070801/07a52cb8/attachment.html 

From transfire at gmail.com  Wed Aug  1 20:27:16 2007
From: transfire at gmail.com (Trans)
Date: Thu, 02 Aug 2007 00:27:16 -0000
Subject: [Nitro] Facets SVN
In-Reply-To: <afd80b540707312352w59583ed8xccd2c53e6aef1558@mail.gmail.com>
References: <1185914350.423367.174310@o61g2000hsh.googlegroups.com>
	<afd80b540707312352w59583ed8xccd2c53e6aef1558@mail.gmail.com>
Message-ID: <1186014436.314436.242460@w3g2000hsg.googlegroups.com>


On Jul 31, 11:52 pm, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> Hello Tom,
>
> I am on vacation these days (but I managed to find an internet connection)
> I will checkout the code now.

good good. keep in mind that some of the old code is now in other
projects. you can check out blow already - i've upload it's source
too. i still have english to upload (of all the libs it probably needs
the most work) and also nemo.

by the way, for per-method requires, the new location is,
facets_by_method/. for example,

  require 'facets_by_method/string/lchomp'

i thought about keeping facets/core/ for backward compatibility, and
maybe I really ought too. but it's a bit of a misnomer now, b/c core
is actually what is in 'src/core/lib/facets/', which is defined by the
fact that if you do,

  require 'facets'

everything in there is automatically loaded.

but maybe backward compatibility is better than exacting semantics/
organization. hmmm...

t.


From transfire at gmail.com  Wed Aug  1 22:52:22 2007
From: transfire at gmail.com (Trans)
Date: Thu, 02 Aug 2007 02:52:22 -0000
Subject: [Nitro] Facets SVN
In-Reply-To: <afd80b540707312352w59583ed8xccd2c53e6aef1558@mail.gmail.com>
References: <1185914350.423367.174310@o61g2000hsh.googlegroups.com>
	<afd80b540707312352w59583ed8xccd2c53e6aef1558@mail.gmail.com>
Message-ID: <1186023142.090009.185480@19g2000hsx.googlegroups.com>



On Jul 31, 11:52 pm, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> Hello Tom,
>
> I am on vacation these days (but I managed to find an internet connection)
> I will checkout the code now.

english code is now in svn repo as well.

  svn checkout svn://rubyforge.org/var/svn/english

that just leaves nemo, and then we should have everything we need to
move forward with nitro 5-0.

t.


From george.moschovitis at gmail.com  Thu Aug  2 09:45:51 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 2 Aug 2007 16:45:51 +0300
Subject: [Nitro] Facets SVN
In-Reply-To: <1186023142.090009.185480@19g2000hsx.googlegroups.com>
References: <1185914350.423367.174310@o61g2000hsh.googlegroups.com>
	<afd80b540707312352w59583ed8xccd2c53e6aef1558@mail.gmail.com>
	<1186023142.090009.185480@19g2000hsx.googlegroups.com>
Message-ID: <afd80b540708020645n68a30afp4b875f43a9f8cfc2@mail.gmail.com>

>
> english code is now in svn repo as well.
>
>   svn checkout svn://rubyforge.org/var/svn/english
>
> that just leaves nemo, and then we should have everything we need to
> move forward with nitro 5-0.
>

nice, just came back from vacations, will have a couple of days before I
leave again ;-)

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070802/c3255ecd/attachment.html 

From george.moschovitis at gmail.com  Fri Aug  3 12:41:22 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Aug 2007 19:41:22 +0300
Subject: [Nitro] Patches?
Message-ID: <afd80b540708030941t4c5c5bf1t2886b1cc9ddf6216@mail.gmail.com>

Does anyone have any patches for inclusion in the main repo? Jonathan?

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070803/737ddddf/attachment-0001.html 

From john at oxyliquit.de  Sat Aug  4 17:57:09 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sun, 05 Aug 2007 00:57:09 +0300
Subject: [Nitro] Patches?
In-Reply-To: <afd80b540708030941t4c5c5bf1t2886b1cc9ddf6216@mail.gmail.com>
References: <afd80b540708030941t4c5c5bf1t2886b1cc9ddf6216@mail.gmail.com>
Message-ID: <op.twju19e3xegpx6@jo.local>

Hi,

> Does anyone have any patches for inclusion in the main repo? Jonathan?

no patches, I've been busy wrapping up my work here in Finland.  I hoped
to do some in the few days I had here (like 3 days off work) but it didn't
turn out to happen.

Jo
PS:  well, no patches wouldn't be true, I have tons of stuff I'd like to
      move to Og head, but that'll never happen before 0.50 as I want to
      see Og on solid testing ground before anything else.

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

From john at oxyliquit.de  Sat Aug  4 17:57:09 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sun, 05 Aug 2007 00:57:09 +0300
Subject: [Nitro] vacation time for me
Message-ID: <op.twj1wrzpxegpx6@jo.local>

Hi,

I'm leaving Finland tomorrow again, taking a break of probably a
week where I'm offline.  I can only recommend the finnish country,
nice place to be, friendly people, very nice environment and it
was a great time being an intern there.

Anyway.  Will probably be back in a week, if not, then know that
I'm still in denmark or northern germany.  At max I'll be off
until the 20th this month.

Take care!

Jo

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

From transfire at gmail.com  Sat Aug  4 20:58:45 2007
From: transfire at gmail.com (Trans)
Date: Sun, 05 Aug 2007 00:58:45 -0000
Subject: [Nitro] vacation time for me
In-Reply-To: <op.twj1wrzpxegpx6@jo.local>
References: <op.twj1wrzpxegpx6@jo.local>
Message-ID: <1186275525.047011.159040@o61g2000hsh.googlegroups.com>



On Aug 4, 2:57 pm, "Jonathan Buch" <j... at oxyliquit.de> wrote:
> Hi,
>
> I'm leaving Finland tomorrow again, taking a break of probably a
> week where I'm offline.  I can only recommend the finnish country,
> nice place to be, friendly people, very nice environment and it
> was a great time being an intern there.
>
> Anyway.  Will probably be back in a week, if not, then know that
> I'm still in denmark or northern germany.  At max I'll be off
> until the 20th this month.
>
> Take care!

Enjoy.

T.


From george.moschovitis at gmail.com  Thu Aug  9 08:58:23 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 9 Aug 2007 15:58:23 +0300
Subject: [Nitro] Preparing nitro 0.50
Message-ID: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>

Dear devs,

while waiting for Trans to make the finishing touches to facets 2 and during
my vacations I am working on preparing
Nitro for the upcoming 0.50 release. These days I am working a bit on tests.
I changed slightly the directory structure,
have a look in raw/test. To run the (existing) tests give:

$ rake test:raw

I am trying to fix some small bugs and annoyances along the way. If anyone
has something to report, let me know.

thanks,
-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070809/b95c76a7/attachment.html 

From ravage at fragmentized.net  Thu Aug  9 09:56:50 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Thu, 9 Aug 2007 14:56:50 +0100
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
Message-ID: <0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>

Hello!

First congrats on the great project!

I'm having an issue. When an error occurs outside the root controller  
it renders a blank page instead of error info.

One little question, is it possible to Model.find( :include =>  
[model1, model2, model3] ) to use many joins?
and is it possible to select just the fields one wants to see?
Model.find(:select => [:field1, :field2] )

Thank you,
Rui Miguel

On Aug 9, 2007, at Aug 9 - 13:58, George Moschovitis wrote:

> Dear devs,
>
> while waiting for Trans to make the finishing touches to facets 2  
> and during my vacations I am working on preparing
> Nitro for the upcoming 0.50 release. These days I am working a bit  
> on tests. I changed slightly the directory structure,
> have a look in raw/test. To run the (existing) tests give:
>
> $ rake test:raw
>
> I am trying to fix some small bugs and annoyances along the way. If  
> anyone has something to report, let me know.
>
> thanks,
> -g.
>
> -- 
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org
> _______________________________________________
> 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: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070809/9e6ced25/attachment.bin 

From george.moschovitis at gmail.com  Thu Aug  9 12:54:50 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 9 Aug 2007 19:54:50 +0300
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
Message-ID: <afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>

Welcome to the Nitro community!

Which version are you using? Please checkout the repository version:

darcs get http://repo.nitroproject.org

for more info:

http://www.nitroproject.org/repo

One little question, is it possible to Model.find( :include =>
> [model1, model2, model3] ) to use many joins?
> and is it possible to select just the fields one wants to see?
> Model.find(:select => [:field1, :field2] )


I haven't used this functionality lately. This should work as you describe.
A quick test revealed some bugs though.
 Gimme a day or so to fix them and post a patch.

thanks,
-g.





-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070809/a2ae674b/attachment.html 

From ravage at fragmentized.net  Thu Aug  9 14:29:20 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Thu, 9 Aug 2007 19:29:20 +0100
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
Message-ID: <757F1661-56A4-4154-95BF-D1D7A47E3348@fragmentized.net>

Thank you!

I'm using repo version.

Replicated the error using blog example, there you go:

DEBUG: SELECT * FROM ogcategory WHERE oid='aaKYGmrQur3lI4abFYZQ-0'
DEBUG: Rendering '/status_500'
ERROR: Error while handling '/categories/view/aaKYGmrQur3lI4abFYZQ-0'
ERROR: private method `render_action' called for #<Post::Controller: 
0x28cc36c>
   /Users/ravage/Desktop/Pendente/nitro-dev/script/lib/../../raw/lib/ 
raw/controller/publishable.rb:38:in `method_missing'
   /Users/ravage/Desktop/Pendente/nitro-dev/script/lib/../../raw/lib/ 
raw/controller/render.rb:82:in `render'
   /Users/ravage/Desktop/Pendente/nitro-dev/script/lib/../../raw/lib/ 
raw/controller/render.rb:237:in `handle_error'
   /Users/ravage/Desktop/Pendente/nitro-dev/script/lib/../../raw/lib/ 
raw/controller/render.rb:106:in `render_action'
   /Users/ravage/Desktop/Pendente/nitro-dev/script/lib/../../raw/lib/ 
raw/adapter.rb:69:in `send'
   /Users/ravage/Desktop/Pendente/nitro-dev/script/lib/../../raw/lib/ 
raw/adapter.rb:69:in `handle_context'
   /Users/ravage/Desktop/Pendente/nitro-dev/script/lib/../../raw/lib/ 
raw/adapter/webrick.rb:134:in `do_GET'

I get no error page, just a blank.

Hope it helps.

@ this time i'm more a user than a contributer, Ruby is still a new  
world to me... but i hope to be able to help with some bits.

Thanks,
Rui Miguel

On Aug 9, 2007, at Aug 9 - 17:54, George Moschovitis wrote:

> Welcome to the Nitro community!
>
> Which version are you using? Please checkout the repository version:
>
> darcs get http://repo.nitroproject.org
>
> for more info:
>
> http://www.nitroproject.org/repo
>
> One little question, is it possible to Model.find( :include =>
> [model1, model2, model3] ) to use many joins?
> and is it possible to select just the fields one wants to see?
> Model.find(:select => [:field1, :field2] )
>
> I haven't used this functionality lately. This should work as you  
> describe. A quick test revealed some bugs though.
>  Gimme a day or so to fix them and post a patch.
>
> thanks,
> -g.
>
>
>
>
>
> -- 
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org
> _______________________________________________
> 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: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070809/0edbd6d1/attachment.bin 

From george.moschovitis at gmail.com  Thu Aug  9 16:32:02 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 9 Aug 2007 23:32:02 +0300
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <757F1661-56A4-4154-95BF-D1D7A47E3348@fragmentized.net>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
	<757F1661-56A4-4154-95BF-D1D7A47E3348@fragmentized.net>
Message-ID: <afd80b540708091332o7fcd7d3dpb0c28ed8a3a74d43@mail.gmail.com>

Ok, I think I have fixed this. Pull from the repo, try again and let me know
if it works.

btw, you uncovered a nasty design issue in the render implementation. I will
have to provide a better fix.

thanks,
-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070809/c51c53d3/attachment.html 

From billk at cts.com  Thu Aug  9 17:45:19 2007
From: billk at cts.com (Bill Kelly)
Date: Thu, 9 Aug 2007 14:45:19 -0700
Subject: [Nitro] Preparing nitro 0.50
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com><0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
Message-ID: <061701c7dace$94b35f90$6442a8c0@musicbox>


From: George Moschovitis 
> 
> Welcome to the Nitro community!
> 
> Which version are you using? Please checkout the repository version:
> 
> darcs get http://repo.nitroproject.org

Apologies for the newbie question, but what is the recommended way to
try out the repo version, when I already have various nitro and og 
gems installed?

After the darcs get, I have:

/usr/src/repo.nitroproject.org/

  _darcs/  examples/  glue/  nitro/  og/  rakefile  rake_tasks/  raw/  script/


Do I need to install each of these projects, such as glue, og, raw,
and nitro individually into lib/ruby/site_ruby/1.8/... ?

Or is there an easier way to try out the repo version?  (Maybe 
install it in a ~/lib/ruby directory and add that directory to
RUBYOPT?)

Just wondering how the experts do it.


Thanks,

Bill



From ravage at fragmentized.net  Thu Aug  9 17:48:30 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Thu, 9 Aug 2007 22:48:30 +0100
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <afd80b540708091332o7fcd7d3dpb0c28ed8a3a74d43@mail.gmail.com>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
	<757F1661-56A4-4154-95BF-D1D7A47E3348@fragmentized.net>
	<afd80b540708091332o7fcd7d3dpb0c28ed8a3a74d43@mail.gmail.com>
Message-ID: <706D6A3D-2F60-4A29-8A98-72D191DE7113@fragmentized.net>

All is working now!

I'm using Nitro to develop a little application, so i'll be posting  
issues i find.

I have been looking into the code and i find it sometimes fragmented,  
mainly form_helper, xhtml_helpers and javascript..., the ones i  
looked more.
But i think these are issues to take care in the long run.

looking forward for 0.50 :)

Thanks for the fix,
Rui Miguel

On Aug 9, 2007, at Aug 9 - 21:32, George Moschovitis wrote:

> Ok, I think I have fixed this. Pull from the repo, try again and  
> let me know if it works.
>
> btw, you uncovered a nasty design issue in the render  
> implementation. I will have to provide a better fix.
>
> thanks,
> -g.
>
>
> -- 
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org
> _______________________________________________
> 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: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070809/1c57b042/attachment.bin 

From ravage at fragmentized.net  Thu Aug  9 17:58:02 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Thu, 9 Aug 2007 22:58:02 +0100
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <061701c7dace$94b35f90$6442a8c0@musicbox>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com><0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
	<061701c7dace$94b35f90$6442a8c0@musicbox>
Message-ID: <A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>

Check this http://www.arnebrasseur.net/2007/07/01/the-daily-nitro/en/

Hope it helps.

Greets,
Rui Miguel


On Aug 9, 2007, at Aug 9 - 22:45, Bill Kelly wrote:

> Apologies for the newbie question, but what is the recommended way to
> try out the repo version, when I already have various nitro and og
> gems installed?
>
> After the darcs get, I have:
>
> /usr/src/repo.nitroproject.org/
>
>   _darcs/  examples/  glue/  nitro/  og/  rakefile  rake_tasks/   
> raw/  script/
>
>
> Do I need to install each of these projects, such as glue, og, raw,
> and nitro individually into lib/ruby/site_ruby/1.8/... ?
>
> Or is there an easier way to try out the repo version?  (Maybe
> install it in a ~/lib/ruby directory and add that directory to
> RUBYOPT?)
>
> Just wondering how the experts do it.
>
>
> Thanks,
>
> Bill
>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070809/2e549e77/attachment.bin 

From billk at cts.com  Thu Aug  9 20:29:23 2007
From: billk at cts.com (Bill Kelly)
Date: Thu, 9 Aug 2007 17:29:23 -0700
Subject: [Nitro] Preparing nitro 0.50
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com><0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net><afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com><061701c7dace$94b35f90$6442a8c0@musicbox>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
Message-ID: <065601c7dae5$80a34c60$6442a8c0@musicbox>


From: "Rui Miguel" <ravage at fragmentized.net>
>
> Check this http://www.arnebrasseur.net/2007/07/01/the-daily-nitro/en/
> 
> Hope it helps.

Thanks.  Looks like the "-r glycerin" does the magic.


Wow, my app doesn't run at all anymore.  


Has "property" been completely eliminated in favor of
"attr_accessor" ?

Looks like the "setting" method is gone, too....


Time to look at some running example code I guess. :)


Regards,

Bill



From george.moschovitis at gmail.com  Fri Aug 10 03:13:20 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 10 Aug 2007 10:13:20 +0300
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <065601c7dae5$80a34c60$6442a8c0@musicbox>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
	<061701c7dace$94b35f90$6442a8c0@musicbox>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
Message-ID: <afd80b540708100013s879d5a6v5b4d6ff8a171786c@mail.gmail.com>

>
> Wow, my app doesn't run at all anymore.


There are many changes (for the better), have a look in the blog example.

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070810/c748ec95/attachment.html 

From arne at arnebrasseur.net  Fri Aug 10 03:25:00 2007
From: arne at arnebrasseur.net (arne at arnebrasseur.net)
Date: Fri, 10 Aug 2007 00:25:00 -0700 (PDT)
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <065601c7dae5$80a34c60$6442a8c0@musicbox>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com><0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net><afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com><061701c7dace$94b35f90$6442a8c0@musicbox>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
Message-ID: <3917.81.245.185.237.1186730700.squirrel@webmail.arnebrasseur.net>

> From: "Rui Miguel" <ravage at fragmentized.net>
>>
>> Check this http://www.arnebrasseur.net/2007/07/01/the-daily-nitro/en/
>>
>> Hope it helps.

Regarding that article, if you have any issues (or suggestions) with it
please report so I can improve it.

> Has "property" been completely eliminated in favor of
> "attr_accessor" ?

AFAIK yes. The new trend in Nitro is to stick to standard Ruby idioms, so
attr_accessor instead of property, include (or is) instead of helper.

(George, Jonathan, any others like this? interesting for the release
notes...)

> Looks like the "setting" method is gone, too....

Much code still uses settings from facets, I suppose that's what you're
looking for. Maybe it just isn't loaded where you are using it.

>
> Time to look at some running example code I guess. :)
>

There are quite a few breaking changes since 0.42. The blog example should
be a good starting point.

Good luck!
(ab)




From Reid.Thompson at ateb.com  Fri Aug 10 09:32:28 2007
From: Reid.Thompson at ateb.com (Reid Thompson)
Date: Fri, 10 Aug 2007 09:32:28 -0400
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <3917.81.245.185.237.1186730700.squirrel@webmail.arnebrasseur.net>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
	<061701c7dace$94b35f90$6442a8c0@musicbox>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
	<3917.81.245.185.237.1186730700.squirrel@webmail.arnebrasseur.net>
Message-ID: <1186752750.12191.5.camel@jhereg>


> >> Check this http://www.arnebrasseur.net/2007/07/01/the-daily-nitro/en/
> >>

> Regarding that article, if you have any issues (or suggestions) with it
> please report so I can improve it.
> 

rthompso at jhereg: ~/src/repo.nitroproject.org/examples/blog$ env|grep RUBY
RUBYOPT=-r/home/rthompso/src/repo.nitroproject.org//script/lib/glycerin -rubygems

rthompso at jhereg: ~/src/repo.nitroproject.org/examples/blog$ env|grep NITRO
NITRO_REPO=/home/rthompso/src/repo.nitroproject.org/


rthompso at jhereg: ~/src/repo.nitroproject.org/examples/blog$ ./app.rb
/home/rthompso/src/repo.nitroproject.org/script/lib/../../raw/lib/raw/view/table.rb:5: uninitialized constant Raw (NameError)
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in `gem_original_require'
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in `require'
        from /home/rthompso/src/repo.nitroproject.org/script/lib/../../nitro/lib/nitro/part/admin/controller.rb:4
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in `gem_original_require'
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in `require'
        from /home/rthompso/src/repo.nitroproject.org/script/lib/../../nitro/lib/nitro/part/admin.rb:3
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in `gem_original_require'
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in `require'
        from ./app.rb:8

From ravage at fragmentized.net  Fri Aug 10 10:37:42 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Fri, 10 Aug 2007 15:37:42 +0100
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <1186752750.12191.5.camel@jhereg>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
	<061701c7dace$94b35f90$6442a8c0@musicbox>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
	<3917.81.245.185.237.1186730700.squirrel@webmail.arnebrasseur.net>
	<1186752750.12191.5.camel@jhereg>
Message-ID: <E14516B1-BCAC-45CC-9D13-9FEDFC6112CA@fragmentized.net>

Care to share you app.rb?

All is working @ this side. Did you change anything in the default  
example?

Greets,
Rui Miguel

On Aug 10, 2007, at Aug 10 - 14:32, Reid Thompson wrote:

>
>>>> Check this http://www.arnebrasseur.net/2007/07/01/the-daily- 
>>>> nitro/en/
>>>>
>
>> Regarding that article, if you have any issues (or suggestions)  
>> with it
>> please report so I can improve it.
>>
>
> rthompso at jhereg: ~/src/repo.nitroproject.org/examples/blog$ env| 
> grep RUBY
> RUBYOPT=-r/home/rthompso/src/repo.nitroproject.org//script/lib/ 
> glycerin -rubygems
>
> rthompso at jhereg: ~/src/repo.nitroproject.org/examples/blog$ env| 
> grep NITRO
> NITRO_REPO=/home/rthompso/src/repo.nitroproject.org/
>
>
> rthompso at jhereg: ~/src/repo.nitroproject.org/examples/blog$ ./app.rb
> /home/rthompso/src/repo.nitroproject.org/script/lib/../../raw/lib/ 
> raw/view/table.rb:5: uninitialized constant Raw (NameError)
>         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> 27:in `gem_original_require'
>         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> 27:in `require'
>         from /home/rthompso/src/repo.nitroproject.org/script/ 
> lib/../../nitro/lib/nitro/part/admin/controller.rb:4
>         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> 27:in `gem_original_require'
>         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> 27:in `require'
>         from /home/rthompso/src/repo.nitroproject.org/script/ 
> lib/../../nitro/lib/nitro/part/admin.rb:3
>         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> 27:in `gem_original_require'
>         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> 27:in `require'
>         from ./app.rb:8
> _______________________________________________
> 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: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070810/7c6aa077/attachment.bin 

From Reid.Thompson at ateb.com  Fri Aug 10 10:47:33 2007
From: Reid.Thompson at ateb.com (Reid Thompson)
Date: Fri, 10 Aug 2007 10:47:33 -0400
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <E14516B1-BCAC-45CC-9D13-9FEDFC6112CA@fragmentized.net>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
	<061701c7dace$94b35f90$6442a8c0@musicbox>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
	<3917.81.245.185.237.1186730700.squirrel@webmail.arnebrasseur.net>
	<1186752750.12191.5.camel@jhereg>
	<E14516B1-BCAC-45CC-9D13-9FEDFC6112CA@fragmentized.net>
Message-ID: <1186757255.27382.1.camel@jhereg>


On Fri, 2007-08-10 at 15:37 +0100, Rui Miguel wrote:
> Care to share you app.rb?
> 
> All is working @ this side. Did you change anything in the default  
> example?
> 
> Greets,
> Rui Miguel
> 
> On Aug 10, 2007, at Aug 10 - 14:32, Reid Thompson wrote:
> 
> >
> >>>> Check this http://www.arnebrasseur.net/2007/07/01/the-daily- 
> >>>> nitro/en/
> >>>>
> >
> >> Regarding that article, if you have any issues (or suggestions)  
> >> with it
> >> please report so I can improve it.
> >>
> >
> > rthompso at jhereg: ~/src/repo.nitroproject.org/examples/blog$ env| 
> > grep RUBY
> > RUBYOPT=-r/home/rthompso/src/repo.nitroproject.org//script/lib/ 
> > glycerin -rubygems
> >
> > rthompso at jhereg: ~/src/repo.nitroproject.org/examples/blog$ env| 
> > grep NITRO
> > NITRO_REPO=/home/rthompso/src/repo.nitroproject.org/
> >
> >
> > rthompso at jhereg: ~/src/repo.nitroproject.org/examples/blog$ ./app.rb
> > /home/rthompso/src/repo.nitroproject.org/script/lib/../../raw/lib/ 
> > raw/view/table.rb:5: uninitialized constant Raw (NameError)
> >         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> > 27:in `gem_original_require'
> >         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> > 27:in `require'
> >         from /home/rthompso/src/repo.nitroproject.org/script/ 
> > lib/../../nitro/lib/nitro/part/admin/controller.rb:4
> >         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> > 27:in `gem_original_require'
> >         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> > 27:in `require'
> >         from /home/rthompso/src/repo.nitroproject.org/script/ 
> > lib/../../nitro/lib/nitro/part/admin.rb:3
> >         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> > 27:in `gem_original_require'
> >         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb: 
> > 27:in `require'
> >         from ./app.rb:8
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> 


rthompso at jhereg: ~/src/repo.nitroproject.org$ darcs whatsnew
{
hunk ./examples/blog/app.rb 5
-require "nitro_and_og"; include Nitro
+require '/home/rthompso/src/repo.nitroproject.org/script/lib/glycerin'
+#require "nitro_and_og"; include Nitro
hunk ./examples/blog/conf/debug.rb 2
-  Raw::Caching.enabled = false $
-  Raw::Template.strip_xml_comments = false $
+  Raw::Caching.enabled = false
+  Raw::Template.strip_xml_comments = false
hunk ./examples/blog/conf/debug.rb 10
-  Og.create_schema = true $
+  Og.create_schema = true
hunk ./examples/blog/conf/debug.rb 15
-    :adapter => :mysql,
-    :user => "root",
-    :password => ENV["DB_PASSWORD"], # SET your password here!
-    :evolve_schema => :full $
+    :adapter => :postgresql,
+    :user => "rthompso",
+    :password => "",
+    :evolve_schema => :full
hunk ./examples/blog/conf/debug.rb 39
-  $
+
hunk ./examples/blog/conf/debug.rb 44
-  $
+
}
rthompso at jhereg: ~/src/repo.nitroproject.org$ 

From Reid.Thompson at ateb.com  Fri Aug 10 10:50:15 2007
From: Reid.Thompson at ateb.com (Reid Thompson)
Date: Fri, 10 Aug 2007 10:50:15 -0400
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <1186757255.27382.1.camel@jhereg>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
	<afd80b540708090954h656ab055v2dea1c3f272c82c6@mail.gmail.com>
	<061701c7dace$94b35f90$6442a8c0@musicbox>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
	<3917.81.245.185.237.1186730700.squirrel@webmail.arnebrasseur.net>
	<1186752750.12191.5.camel@jhereg>
	<E14516B1-BCAC-45CC-9D13-9FEDFC6112CA@fragmentized.net>
	<1186757255.27382.1.camel@jhereg>
Message-ID: <1186757417.27382.2.camel@jhereg>



> 
> rthompso at jhereg: ~/src/repo.nitroproject.org$ darcs whatsnew
> {
> hunk ./examples/blog/app.rb 5
> -require "nitro_and_og"; include Nitro
> +require '/home/rthompso/src/repo.nitroproject.org/script/lib/glycerin'
> +#require "nitro_and_og"; include Nitro

reverted this -- running ok now -- sorry for the noise

From transfire at gmail.com  Fri Aug 10 11:47:09 2007
From: transfire at gmail.com (Trans)
Date: Fri, 10 Aug 2007 15:47:09 -0000
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <065601c7dae5$80a34c60$6442a8c0@musicbox>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
Message-ID: <1186760829.526328.119210@i38g2000prf.googlegroups.com>



On Aug 9, 5:29 pm, "Bill Kelly" <bi... at cts.com> wrote:
> From: "Rui Miguel" <rav... at fragmentized.net>
>
>
>
> > Check thishttp://www.arnebrasseur.net/2007/07/01/the-daily-nitro/en/
>
> > Hope it helps.
>
> Thanks.  Looks like the "-r glycerin" does the magic.
>
> Wow, my app doesn't run at all anymore.
>
> Has "property" been completely eliminated in favor of
> "attr_accessor" ?

I'm surprised you didn't leave property as an alias of attr_accessor.
I think that's a reasonable thing to do, since it's a simple thing,
and will help backward compatibility (and remains available should you
ever reconsider the attr_accessor decsision ;)

T.


From billk at cts.com  Fri Aug 10 15:25:06 2007
From: billk at cts.com (Bill Kelly)
Date: Fri, 10 Aug 2007 12:25:06 -0700
Subject: [Nitro] Preparing nitro 0.50
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com><A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net><065601c7dae5$80a34c60$6442a8c0@musicbox>
	<1186760829.526328.119210@i38g2000prf.googlegroups.com>
Message-ID: <07ab01c7db84$28d85260$6442a8c0@musicbox>


From: "Trans" <transfire at gmail.com>
> 
> On Aug 9, 5:29 pm, "Bill Kelly" <bi... at cts.com> wrote:
>> From: "Rui Miguel" <rav... at fragmentized.net>
>>
>> > Check thishttp://www.arnebrasseur.net/2007/07/01/the-daily-nitro/en/
>>
>> > Hope it helps.
>>
>> Thanks.  Looks like the "-r glycerin" does the magic.
>>
>> Wow, my app doesn't run at all anymore.
>>
>> Has "property" been completely eliminated in favor of
>> "attr_accessor" ?
> 
> I'm surprised you didn't leave property as an alias of attr_accessor.
> I think that's a reasonable thing to do, since it's a simple thing,
> and will help backward compatibility (and remains available should you
> ever reconsider the attr_accessor decsision ;)

Just to add my 2 cents...  I would have preferred to continue to use
"property", because overriding attr_accessor feels a little too 
magic to me; and I tend to regard my og-managed classes as distinct
from "normal" ruby classes.  

To me, I don't think my og-managed classes become more readable by
making their definitions harder to distinguish from normal ruby
classes.  I feel og-managed classes are really quite special, and
require special treatment, so I kind of liked that property showed
that they aren't really quite ordinary.

But... just my opinion.  :)


Regards,

Bill



From george.moschovitis at gmail.com  Fri Aug 10 17:52:53 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 11 Aug 2007 00:52:53 +0300
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <1186760829.526328.119210@i38g2000prf.googlegroups.com>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
	<1186760829.526328.119210@i38g2000prf.googlegroups.com>
Message-ID: <afd80b540708101452h5b689747ndc379f7ae13a835b@mail.gmail.com>

>
> I'm surprised you didn't leave property as an alias of attr_accessor.
> I think that's a reasonable thing to do, since it's a simple thing,
> and will help backward compatibility (and remains available should you
> ever reconsider the attr_accessor decsision ;)


well, I will add this again ;-)

-g.

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



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070811/0e871457/attachment.html 

From george.moschovitis at gmail.com  Fri Aug 10 20:05:02 2007
From: george.moschovitis at gmail.com (george.moschovitis at gmail.com)
Date: Sat, 11 Aug 2007 00:05:02 +0000 (UTC)
Subject: [Nitro] =?utf-8?q?=5BNP=2EORG=5D_Nitro/Og_Fora_daily_digest?=
Message-ID: <20070811000502.9E9668701@reizu.com>

>From gmosx.myopenid.com 
Date: Fri Aug 10 10:02:06 UTC 2007
Subject: Remove template stack functionality?
Message-ID: http://www.nitroproject.org/fora/posts/view/daetEUrYGr3jYMeJeZaaqU

Dear devs,

in the process of symplifying Nitro, I am considering removing the template stack functionality. I don't think anyone is using it (if you don't count the default error handler, which I would like to re-implement anyway).

any objections?

-g.

>From gmosx.myopenid.com 
Date: Fri Aug 10 09:53:40 UTC 2007
Subject: Remove template stack functionality?
Message-ID: http://www.nitroproject.org/fora/posts/view/csLqgYrYCr3ltneJeZaaqU

Dear devs,

in the process of symplifying Nitro, I am considering removing the template stack functionality. I don't think anyone is using it (if you don't count the default error handler, which I would like to reimplement anyway).

any objections?

-g.

>From gmosx.myopenid.com 
Date: Fri Aug 10 09:53:28 UTC 2007
Subject: Remove template stack functionality?
Message-ID: http://www.nitroproject.org/fora/posts/view/clGtKyrYCr3ktfeJeZaaqU

Dear devs,

in the process of symplifying Nitro, I am considering removing the template stack functionality. I don't think anyone is using it (if you don't count the default error handler, which I would like to reimplement anyway).

any objections?

-g.

>From gmosx.myopenid.com 
Date: Fri Aug 10 09:53:25 UTC 2007
Subject: Remove template stack functionality?
Message-ID: http://www.nitroproject.org/fora/posts/view/ckbHk2rYCr3jgDeJeZaaqU

Dear devs,

in the process of symplifying Nitro, I am considering removing the template stack functionality. I don't think anyone is using it (if you don't count the default error handler, which I would like to reimplement anyway).

any objections?

-g.




--
This mail is automatically generated from the http://nitroproject.org/fora
digest robot. It presents the discussions in the fora during the last 24 hours.
Do not reply to this email.

From james.britt at gmail.com  Fri Aug 10 20:47:40 2007
From: james.britt at gmail.com (James Britt)
Date: Fri, 10 Aug 2007 17:47:40 -0700
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <afd80b540708101452h5b689747ndc379f7ae13a835b@mail.gmail.com>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>	<065601c7dae5$80a34c60$6442a8c0@musicbox>	<1186760829.526328.119210@i38g2000prf.googlegroups.com>
	<afd80b540708101452h5b689747ndc379f7ae13a835b@mail.gmail.com>
Message-ID: <46BD072C.8020804@gmail.com>

George Moschovitis wrote:
>     I'm surprised you didn't leave property as an alias of attr_accessor.
>     I think that's a reasonable thing to do, since it's a simple thing,
>     and will help backward compatibility (and remains available should you
>     ever reconsider the attr_accessor decsision ;)
> 
> 
> well, I will add this again ;-)

But please add deprecation warnings for it.


What I like about Og is that you can code more or less plain Ruby and, 
with a few extra moves, get automagic persistence.



-- 
James Britt

"In physics the truth is rarely perfectly clear, and that is certainly
  universally the case in human affairs. Hence, what is not surrounded by
  uncertainty cannot be the truth."
  - R. Feynman

From james.britt at gmail.com  Fri Aug 10 20:49:30 2007
From: james.britt at gmail.com (James Britt)
Date: Fri, 10 Aug 2007 17:49:30 -0700
Subject: [Nitro] [NP.ORG] Nitro/Og Fora daily digest
In-Reply-To: <20070811000502.9E9668701@reizu.com>
References: <20070811000502.9E9668701@reizu.com>
Message-ID: <46BD079A.90405@gmail.com>

george.moschovitis at gmail.com wrote:
>>From gmosx.myopenid.com 
> Date: Fri Aug 10 10:02:06 UTC 2007
> Subject: Remove template stack functionality?
> Message-ID: http://www.nitroproject.org/fora/posts/view/daetEUrYGr3jYMeJeZaaqU
> 
> Dear devs,
> 
> in the process of symplifying Nitro, I am considering removing the template stack functionality. I don't think anyone is using it (if you don't count the default error handler, which I would like to re-implement anyway).
> 
> any objections?


Well, first explain what it is. ;)


I think I know, but I'm not sure, though I'm fairly certain that there 
are many sweet features in Og/Nitro/Raw that I've never made use of 
because of my ignorance.

James


From transfire at gmail.com  Fri Aug 10 21:16:35 2007
From: transfire at gmail.com (Trans)
Date: Sat, 11 Aug 2007 01:16:35 -0000
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <46BD072C.8020804@gmail.com>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
	<1186760829.526328.119210@i38g2000prf.googlegroups.com>
	<afd80b540708101452h5b689747ndc379f7ae13a835b@mail.gmail.com>
	<46BD072C.8020804@gmail.com>
Message-ID: <1186794995.359958.107900@w3g2000hsg.googlegroups.com>



On Aug 10, 5:47 pm, James Britt <james.br... at gmail.com> wrote:
> George Moschovitis wrote:
> >     I'm surprised you didn't leave property as an alias of attr_accessor.
> >     I think that's a reasonable thing to do, since it's a simple thing,
> >     and will help backward compatibility (and remains available should you
> >     ever reconsider the attr_accessor decsision ;)
>
> > well, I will add this again ;-)
>
> But please add deprecation warnings for it.

Well, not unless it in fact will be deprecated. I don;t necessarily
see anything wrong with supporting both. At the very least, it can
always be an optional require.

> What I like about Og is that you can code more or less plain Ruby and,
> with a few extra moves, get automagic persistence.

I agree. That is a benefit. But it's "not quite" b/c you still have to
specify the persistence class. Eg.

  attr_accessor :x, String

That "String" throws it off, so then is it really any better than
using #property?

I sometimes feel a little uncomfortable about the override of the attr
methods. I guess one of the reasons is that Ara has his own lib that
overrides them for defaults, like so:

  attr_accessor :x => "foo"

Which seems pretty useful too. An you can generally infer the class
from a default, but unfortunately not always. Obviously Ara's override
scheme is more limited than annotations. We can do the same with:

  attr_accessor :x, :default => "foo"

but, it's just an annotation and doesn't actually do what Ara's does
(ie. prime @x with "foo"). Rather it provides Og with some persistence
metadata. But what if we wanted both?

Stuff like this has made me concerned about annotation conflicts.
While it seems like annotations should have the potential to become a
powerful general purpose tool, the potential of such conflicts may be
a seriously limiting flaw. Can it be overcome? --some sort of
annotations namespaces perhaps? Or am I just worrying over a mole
hill?

T.


From transfire at gmail.com  Fri Aug 10 21:34:17 2007
From: transfire at gmail.com (Trans)
Date: Sat, 11 Aug 2007 01:34:17 -0000
Subject: [Nitro] CLI libs in Facets or not?
Message-ID: <1186796057.657304.115590@g4g2000hsf.googlegroups.com>

I'm am down to my final significant organizational decision for Facets
2.0 (Talk about your wild rides. I seriously doubt any code in the
universe has been poked and prodded and shoved around as much a
Facets' code has). The final question is whether to move the cli libs
out to their own project. This includes ansicode.rb, command.rb,
arguments.rb, progressbar.rb, and a few others. If I do keep them in
Facets, I'm considering moving them to a CLI:: namespace and put them
in a facets/cli/ subdir. On the other hand, and why this is a
question, the CLI libs are probably some of the most useful/reusable
code in the whole library. So maybe for that reason it merits a
separate project?

I'm not sure. I'm torn between the two choices. On one hand I like the
idea of these libs being readily available from Facets --don't need to
worry about installing something else, and so forth. On the other
hand, I can see how they can stand out as it a strong library in its
own right.

What is the Glycerider take?

T.


From george.moschovitis at gmail.com  Sat Aug 11 03:56:12 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 11 Aug 2007 10:56:12 +0300
Subject: [Nitro] [NP.ORG] Nitro/Og Fora daily digest
In-Reply-To: <46BD079A.90405@gmail.com>
References: <20070811000502.9E9668701@reizu.com> <46BD079A.90405@gmail.com>
Message-ID: <afd80b540708110056q1adb2054y3f4a43af30ae3954@mail.gmail.com>

>
> Well, first explain what it is. ;)


In Nitro we don't have a template_root (ie a directory where to look for
templates) but a template_root stack. Ie there are multiple directories.
If a template is not found in the top dir of the stack nitro walks the stack
to find another template_root dir. The bottom template_root dir is the
proto/template directory. The top dir may be app/template. Controllers in
parts (nitro reusable pieces of functionality) can insert a local template
in the stack. And generally you can manipulate this stack for special fx.

This mechanism allows you to 'override' (or 'overload') templates. Lets say
you want to change the default 50x error page template. Just put a template
called status_500.htmlx in app/template (that goes before proto/template).

But these days I am thinking that this complex mechanism is not really
necessary.

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070811/4d18c010/attachment-0001.html 

From george.moschovitis at gmail.com  Sat Aug 11 03:59:49 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 11 Aug 2007 10:59:49 +0300
Subject: [Nitro] CLI libs in Facets or not?
In-Reply-To: <1186796057.657304.115590@g4g2000hsf.googlegroups.com>
References: <1186796057.657304.115590@g4g2000hsf.googlegroups.com>
Message-ID: <afd80b540708110059i10846923t13533d447da55a2@mail.gmail.com>

As I have  previously said I preferred having just two projects:

core/more

but now that you have splitted facets in multiple, granular projects
(for example) English, I think it is appropriate for the cli utils to
have a separate project as well.

-g.


On 8/11/07, Trans <transfire at gmail.com> wrote:
>
> I'm am down to my final significant organizational decision for Facets
> 2.0 (Talk about your wild rides. I seriously doubt any code in the
> universe has been poked and prodded and shoved around as much a
>
> What is the Glycerider take?
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070811/e46d631e/attachment.html 

From george.moschovitis at gmail.com  Sat Aug 11 05:12:15 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 11 Aug 2007 12:12:15 +0300
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <1186794995.359958.107900@w3g2000hsg.googlegroups.com>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<A3DC1473-7089-4C59-8D35-B9E7580876B2@fragmentized.net>
	<065601c7dae5$80a34c60$6442a8c0@musicbox>
	<1186760829.526328.119210@i38g2000prf.googlegroups.com>
	<afd80b540708101452h5b689747ndc379f7ae13a835b@mail.gmail.com>
	<46BD072C.8020804@gmail.com>
	<1186794995.359958.107900@w3g2000hsg.googlegroups.com>
Message-ID: <afd80b540708110212w441cb0b8ia522e7e2476139e9@mail.gmail.com>

>
> Well, not unless it in fact will be deprecated. I don;t necessarily
> see anything wrong with supporting both. At the very least, it can
> always be an optional require.
>

I ve just seen that Nitro 0.50.0 allows property as an alias to
attr_accessor.

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070811/236a0674/attachment.html 

From george.moschovitis at gmail.com  Sat Aug 11 05:57:40 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 11 Aug 2007 12:57:40 +0300
Subject: [Nitro] Preparing nitro 0.50
In-Reply-To: <0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
References: <afd80b540708090558g2dc5d925od99d984ccda96395@mail.gmail.com>
	<0B49C90B-3CC0-4112-A13C-A674EB3C096A@fragmentized.net>
Message-ID: <afd80b540708110257t2a0f6390vb663b8c2ae3c1ffb@mail.gmail.com>

>
> and is it possible to select just the fields one wants to see?
> Model.find(:select => [:field1, :field2] )
>

I just fixed this in the repo so it IS possible.

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070811/53c0a4b5/attachment.html 

From transfire at gmail.com  Sat Aug 11 09:15:05 2007
From: transfire at gmail.com (Trans)
Date: Sat, 11 Aug 2007 13:15:05 -0000
Subject: [Nitro] CLI libs in Facets or not?
In-Reply-To: <afd80b540708110059i10846923t13533d447da55a2@mail.gmail.com>
References: <1186796057.657304.115590@g4g2000hsf.googlegroups.com>
	<afd80b540708110059i10846923t13533d447da55a2@mail.gmail.com>
Message-ID: <1186838105.863911.22600@b79g2000hse.googlegroups.com>



On Aug 11, 12:59 am, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> As I have  previously said I preferred having just two projects:
>
> core/more

Yes, to be sure I've come ''-this close to doing exactly that with
Facets. I seem endlessly torn between one project, two projects and
the 50+ little mini-projects approach. All have merits and downsides.

The mini-project approach, for instance, allows for selective usage of
the many more libraries (eg. if you just want annotations you just
install annotations), but it's a nightmare in version control if you
want to use a bunch of libraries at once, as is Nitro/Og's case, so I
ultimately decided against it. However I did see where some parts of
Facets stood out as beyond what one might expect from a supplement to
Ruby's core and standard libs. It's those libs I moved to their own
projects, namely Stick, English and Blow. Yes, in some respects it
would be nice to have everything under the Helios in one install (gem
install it_all ;), but in reality we have to draw some lines. I think
these three outcrops are a good compromise.

Of course, I have still been tempted to divide what remains between
core/more projects, Such a divide does have some organizational
benefits to us developers, and offers a little more choice to the end-
users. And I've considered a variety of naming schemes for this:
'facets_core', 'facets_more' with a 'facets' meta-package, or just
'core', 'more' and 'facets' as you suggest, or (what I was leaning
toward most recently) just two packages, 'facets' and 'ultrafacets',
since the later depends on the former. These later options has some
nice merits. But ultimately I've never been quite persuaded to return
to the nano/mega jaunt (as cool as those names were ;), and give up on
the single install and version number. Maybe ultimately this split
should happen (Facets 3.0? ;), but for now I think the off-shoot
projects are enough of a compromise.

Having said that though, I just want you to knwo, that I can totally
see why you favor core/more. 'core' and 'more' no doubt fit nicely
with rest of Nitro's names. And if Facets were only Nitro's "Glue",
I'd take that route. But Facets is also it's own project too (which is
actually a benefit to Nitro) and so I'm more inclined to keeping
'facets' in the package names.

Sorry to get a little long winded on this. I just wanted you to
understand my thought processes on the matter.

> but now that you have split facets in multiple, granular projects
> (for example) English, I think it is appropriate for the cli utils to
> have a separate project as well.

Okay, cool. That's what I was thinking too. "Clio" right? :)  My only
hesitation has been in considering: if I were making Ruby's standard
library (the Modus Operandi question of Facets), what would I make
sure to put in it? CLI libs are definitely on that list.

T.


From george.moschovitis at gmail.com  Sat Aug 11 09:35:55 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 11 Aug 2007 16:35:55 +0300
Subject: [Nitro] CLI libs in Facets or not?
In-Reply-To: <1186838105.863911.22600@b79g2000hse.googlegroups.com>
References: <1186796057.657304.115590@g4g2000hsf.googlegroups.com>
	<afd80b540708110059i10846923t13533d447da55a2@mail.gmail.com>
	<1186838105.863911.22600@b79g2000hse.googlegroups.com>
Message-ID: <afd80b540708110635s3691b175yf910adc5b7d8f0df@mail.gmail.com>

>
> Okay, cool. That's what I was thinking too. "Clio" right? :)  My only


Clio is ok ;-)

I am still waiting to see the rest of the projects in svn.

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070811/fca1be53/attachment.html 

From ravage at fragmentized.net  Sat Aug 11 23:05:44 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Sun, 12 Aug 2007 04:05:44 +0100
Subject: [Nitro] Regarding :include and :select and some SQL questions
Message-ID: <E2E209D8-D521-4900-9C97-869609B9EED6@fragmentized.net>

Hello!

Attached is included some Og tests which fail using these  
functionalities together. Hope i'm not doing anything wrong and  
appreciate correction if i am.

:select is very sensible and should be taken to attention if it gets  
included in Nitro because when used with :include doesn't work.

Implementation should be thought carefully. When using :select  
with :include it's necessary to know, to which model a field refers,  
which should be described when passing the :select argument to avoid  
ambiguity.

The current code breaks when passing two fields to :select.
I changed it a bit so it would work when passing two or more  
parameters, that's when i add the field ambiguity issue described above.

Regarding :include, would it be possible to change SQL generation  
into "LEFT JOIN table2 ON table1.field = table2.field?

Regarding Model.select, should this work,
Model.select("select nome from ogaluno where turma_oid not in (select  
oid from ogturma)"), returning only field "nome"? or how should it  
behave? Just curious. Or a more simple one like Model.select("SELECT  
* FROM oguser").


Regards,
Rui Miguel

-------------- next part --------------
A non-text attachment was scrubbed...
Name: og_tests.patch
Type: application/octet-stream
Size: 55723 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070812/54c7e8b5/attachment-0001.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070812/54c7e8b5/attachment-0001.bin 

From george.moschovitis at gmail.com  Sun Aug 12 04:33:33 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 12 Aug 2007 11:33:33 +0300
Subject: [Nitro] Regarding :include and :select and some SQL questions
In-Reply-To: <E2E209D8-D521-4900-9C97-869609B9EED6@fragmentized.net>
References: <E2E209D8-D521-4900-9C97-869609B9EED6@fragmentized.net>
Message-ID: <afd80b540708120133y648d9789hc123a0ee48eb191a@mail.gmail.com>

>
> The current code breaks when passing two fields to :select.


I fixed this yesterday, have you pulled the latest code? Anw, I will try
your code.

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070812/3a546510/attachment.html 

From george.moschovitis at gmail.com  Sun Aug 12 06:40:47 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 12 Aug 2007 13:40:47 +0300
Subject: [Nitro] Repository Gems
Message-ID: <afd80b540708120340g39f1586bjd4166f3a66cfe1c4@mail.gmail.com>

Dear devs,

I just added some gems generated for the repository code to the site:

http://www.nitroproject.org/repo/gems

Can you please download, test them and report any bugs?

thanks,
George

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070812/639df725/attachment.html 

From ravage at fragmentized.net  Sun Aug 12 07:58:16 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Sun, 12 Aug 2007 12:58:16 +0100
Subject: [Nitro] Repository Gems
In-Reply-To: <afd80b540708120340g39f1586bjd4166f3a66cfe1c4@mail.gmail.com>
References: <afd80b540708120340g39f1586bjd4166f3a66cfe1c4@mail.gmail.com>
Message-ID: <90258E1B-EA69-4944-8183-4D14B1A1A0A3@fragmentized.net>

Hello!

It complains about gen versions:

eflow:~/Desktop/Inbox ravage$ sudo gem install nitro-0.50.0.gem
ERROR:  While executing gem ... (RuntimeError)
     Error instaling nitro-0.50.0.gem:
         nitro requires gen = 0.50.0

Regards,
Rui Miguel

On Aug 12, 2007, at Aug 12 - 11:40, George Moschovitis wrote:

> Dear devs,
>
> I just added some gems generated for the repository code to the site:
>
> http://www.nitroproject.org/repo/gems
>
> Can you please download, test them and report any bugs?
>
> thanks,
> George
>
> -- 
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org
> _______________________________________________
> 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: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070812/408701dd/attachment.bin 

From george.moschovitis at gmail.com  Sun Aug 12 08:09:22 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 12 Aug 2007 15:09:22 +0300
Subject: [Nitro] Repository Gems
In-Reply-To: <90258E1B-EA69-4944-8183-4D14B1A1A0A3@fragmentized.net>
References: <afd80b540708120340g39f1586bjd4166f3a66cfe1c4@mail.gmail.com>
	<90258E1B-EA69-4944-8183-4D14B1A1A0A3@fragmentized.net>
Message-ID: <afd80b540708120509oe323e90y83ec0c8ed3264ef4@mail.gmail.com>

Fixed this, please try again.

-g.

On 8/12/07, Rui Miguel <ravage at fragmentized.net> wrote:
>
> Hello!
>
> It complains about gen versions:
>
> eflow:~/Desktop/Inbox ravage$ sudo gem install nitro-0.50.0.gem
> ERROR:  While executing gem ... (RuntimeError)
>      Error instaling nitro-0.50.0.gem:
>          nitro requires gen = 0.50.0
>
> Regards,
> Rui Miguel
>
> On Aug 12, 2007, at Aug 12 - 11:40, George Moschovitis wrote:
>
> > Dear devs,
> >
> > I just added some gems generated for the repository code to the site:
> >
> > http://www.nitroproject.org/repo/gems
> >
> > Can you please download, test them and report any bugs?
> >
> > thanks,
> > George
> >
> > --
> > http://www.me.gr
> > http://phidz.com
> > http://blog.gmosx.com
> > http://cull.gr
> > http://www.joy.gr
> > http://nitroproject.org
> > _______________________________________________
> > 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
>
>


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070812/27db3c67/attachment.html 

From ravage at fragmentized.net  Sun Aug 12 08:15:54 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Sun, 12 Aug 2007 13:15:54 +0100
Subject: [Nitro] Repository Gems
In-Reply-To: <afd80b540708120509oe323e90y83ec0c8ed3264ef4@mail.gmail.com>
References: <afd80b540708120340g39f1586bjd4166f3a66cfe1c4@mail.gmail.com>
	<90258E1B-EA69-4944-8183-4D14B1A1A0A3@fragmentized.net>
	<afd80b540708120509oe323e90y83ec0c8ed3264ef4@mail.gmail.com>
Message-ID: <813B4AFD-9D86-41AD-9EFC-BD57B04B9BBF@fragmentized.net>

Working as intended ;)

thanks,
Rui Miguel

On Aug 12, 2007, at Aug 12 - 13:09, George Moschovitis wrote:

> Fixed this, please try again.
>
> -g.
>
> On 8/12/07, Rui Miguel <ravage at fragmentized.net> wrote: Hello!
>
> It complains about gen versions:
>
> eflow:~/Desktop/Inbox ravage$ sudo gem install nitro-0.50.0.gem
> ERROR:  While executing gem ... (RuntimeError)
>      Error instaling nitro-0.50.0.gem:
>          nitro requires gen = 0.50.0
>
> Regards,
> Rui Miguel
>
> On Aug 12, 2007, at Aug 12 - 11:40, George Moschovitis wrote:
>
> > Dear devs,
> >
> > I just added some gems generated for the repository code to the  
> site:
> >
> > http://www.nitroproject.org/repo/gems
> >
> > Can you please download, test them and report any bugs?
> >
> > thanks,
> > George
> >
> > --
> > http://www.me.gr
> > http://phidz.com
> > http://blog.gmosx.com
> > http://cull.gr
> > http://www.joy.gr
> > http://nitroproject.org
> > _______________________________________________
> > 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
>
>
>
>
> -- 
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org
> _______________________________________________
> 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: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070812/aadaf7c8/attachment.bin 

From george.moschovitis at gmail.com  Sun Aug 12 08:20:41 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 12 Aug 2007 15:20:41 +0300
Subject: [Nitro] Repository Gems
In-Reply-To: <813B4AFD-9D86-41AD-9EFC-BD57B04B9BBF@fragmentized.net>
References: <afd80b540708120340g39f1586bjd4166f3a66cfe1c4@mail.gmail.com>
	<90258E1B-EA69-4944-8183-4D14B1A1A0A3@fragmentized.net>
	<afd80b540708120509oe323e90y83ec0c8ed3264ef4@mail.gmail.com>
	<813B4AFD-9D86-41AD-9EFC-BD57B04B9BBF@fragmentized.net>
Message-ID: <afd80b540708120520u48c9a8efl2937bcd977755801@mail.gmail.com>

>
> Working as intended ;)


Great... Is this easier than using darcs/glycerin.rb ?

-g.




-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070812/c4ae5c17/attachment.html 

From ravage at fragmentized.net  Sun Aug 12 08:28:32 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Sun, 12 Aug 2007 13:28:32 +0100
Subject: [Nitro] Repository Gems
In-Reply-To: <afd80b540708120520u48c9a8efl2937bcd977755801@mail.gmail.com>
References: <afd80b540708120340g39f1586bjd4166f3a66cfe1c4@mail.gmail.com>
	<90258E1B-EA69-4944-8183-4D14B1A1A0A3@fragmentized.net>
	<afd80b540708120509oe323e90y83ec0c8ed3264ef4@mail.gmail.com>
	<813B4AFD-9D86-41AD-9EFC-BD57B04B9BBF@fragmentized.net>
	<afd80b540708120520u48c9a8efl2937bcd977755801@mail.gmail.com>
Message-ID: <0285658E-0BCE-4724-814C-81BC095F7C8E@fragmentized.net>

I personally prefer a darcs pull and setting RUBYOPTS (arne's blog)  
also requiring darcs/glycerin ain't so bad, if one is following  
development branch.

IMHO i'd go for darcs pull and setting environment.

But this gems are awesome for someone who doesn't want to mess with  
darcs et al.

Regards,
Rui Miguel

On Aug 12, 2007, at Aug 12 - 13:20, George Moschovitis wrote:

> Working as intended ;)
>
> Great... Is this easier than using darcs/glycerin.rb ?
>
> -g.
>
>
>
>
> -- 
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org
> _______________________________________________
> 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: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070812/af4a7e21/attachment.bin 

From ravage at fragmentized.net  Sun Aug 12 09:41:48 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Sun, 12 Aug 2007 14:41:48 +0100
Subject: [Nitro] Regarding :include and :select and some SQL questions
In-Reply-To: <E2E209D8-D521-4900-9C97-869609B9EED6@fragmentized.net>
References: <E2E209D8-D521-4900-9C97-869609B9EED6@fragmentized.net>
Message-ID: <EF48317F-0875-4EC8-BE4C-F271E727865C@fragmentized.net>

Hello!

Somehow i missed your reply to this issue...

Did the Og tests sent passed on your side?

Thanks,
Rui Miguel

On Aug 12, 2007, at Aug 12 - 04:05, Rui Miguel wrote:

> Hello!
>
> Attached is included some Og tests which fail using these  
> functionalities together. Hope i'm not doing anything wrong and  
> appreciate correction if i am.
>
> :select is very sensible and should be taken to attention if it  
> gets included in Nitro because when used with :include doesn't work.
>
> Implementation should be thought carefully. When using :select  
> with :include it's necessary to know, to which model a field  
> refers, which should be described when passing the :select argument  
> to avoid ambiguity.
>
> The current code breaks when passing two fields to :select.
> I changed it a bit so it would work when passing two or more  
> parameters, that's when i add the field ambiguity issue described  
> above.
>
> Regarding :include, would it be possible to change SQL generation  
> into "LEFT JOIN table2 ON table1.field = table2.field?
>
> Regarding Model.select, should this work,
> Model.select("select nome from ogaluno where turma_oid not in  
> (select oid from ogturma)"), returning only field "nome"? or how  
> should it behave? Just curious. Or a more simple one like  
> Model.select("SELECT * FROM oguser").
>
>
> Regards,
> Rui Miguel
>
> <og_tests.patch>
> _______________________________________________
> 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: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070812/fa1bc043/attachment-0001.bin 

From george.moschovitis at gmail.com  Mon Aug 13 04:25:53 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 13 Aug 2007 11:25:53 +0300
Subject: [Nitro] Og table prefix
Message-ID: <afd80b540708130125l1a6b7856sb0dfe196f8199a7f@mail.gmail.com>

Dear devs,

I am thinking about droping the 'og' prefix from model tables by default
(the setting will stay). From what I see you can have safe table names by
using backtick (`) in mysql and double quotes (") in Postgresql. What do you
thinkg?

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070813/24962cc6/attachment.html 

From john at oxyliquit.de  Mon Aug 13 06:13:50 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 13 Aug 2007 12:13:50 +0200
Subject: [Nitro] Og table prefix
In-Reply-To: <afd80b540708130125l1a6b7856sb0dfe196f8199a7f@mail.gmail.com>
References: <afd80b540708130125l1a6b7856sb0dfe196f8199a7f@mail.gmail.com>
Message-ID: <op.twztcux7xegpx6@jo.local>

Hi,

> I am thinking about droping the 'og' prefix from model tables by default
> (the setting will stay). From what I see you can have safe table names by
> using backtick (`) in mysql and double quotes (") in Postgresql. What do you
> think?

I happen to like the prefix, as I often use psql and it's tab trigger to
autocomplete table names.  Having the 'og' prefix makes things easier for me.
As you (if you don't do more low level db stuff) never really specify the
table name at all, they could be hashes instead of good sounding names.  ;)

That said, I just see no real benefit other than 'looks slightly better from
the outside' and vote against.

Jo

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

From ravage at fragmentized.net  Mon Aug 13 07:54:46 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Mon, 13 Aug 2007 12:54:46 +0100
Subject: [Nitro] Og table prefix
In-Reply-To: <op.twztcux7xegpx6@jo.local>
References: <afd80b540708130125l1a6b7856sb0dfe196f8199a7f@mail.gmail.com>
	<op.twztcux7xegpx6@jo.local>
Message-ID: <7F82BE22-2427-4DF5-BB4D-E8238EAF1260@fragmentized.net>

Hello,

I like the default also, and think it's safer and more intuitive for  
databases with managed and unmanaged tables all together.

Rui Miguel

On Aug 13, 2007, at Aug 13 - 11:13, Jonathan Buch wrote:

> Hi,
>
>> I am thinking about droping the 'og' prefix from model tables by  
>> default
>> (the setting will stay). From what I see you can have safe table  
>> names by
>> using backtick (`) in mysql and double quotes (") in Postgresql.  
>> What do you
>> think?
>
> I happen to like the prefix, as I often use psql and it's tab  
> trigger to
> autocomplete table names.  Having the 'og' prefix makes things  
> easier for me.
> As you (if you don't do more low level db stuff) never really  
> specify the
> table name at all, they could be hashes instead of good sounding  
> names.  ;)
>
> That said, I just see no real benefit other than 'looks slightly  
> better from
> the outside' and vote against.
>
> 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

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070813/5c29da56/attachment.bin 

From Reid.Thompson at ateb.com  Mon Aug 13 08:09:03 2007
From: Reid.Thompson at ateb.com (Reid Thompson)
Date: Mon, 13 Aug 2007 08:09:03 -0400
Subject: [Nitro] Og table prefix
In-Reply-To: <7F82BE22-2427-4DF5-BB4D-E8238EAF1260@fragmentized.net>
References: <afd80b540708130125l1a6b7856sb0dfe196f8199a7f@mail.gmail.com>
	<op.twztcux7xegpx6@jo.local>
	<7F82BE22-2427-4DF5-BB4D-E8238EAF1260@fragmentized.net>
Message-ID: <1187006945.9453.4.camel@jhereg>


On Mon, 2007-08-13 at 12:54 +0100, Rui Miguel wrote:
> Hello,
> 
> I like the default also, and think it's safer and more intuitive for  
> databases with managed and unmanaged tables all together.
> 
> Rui Miguel
> 
> On Aug 13, 2007, at Aug 13 - 11:13, Jonathan Buch wrote:
> 
> > Hi,
> >
> >> I am thinking about droping the 'og' prefix from model tables by  
> >> default
> >> (the setting will stay). From what I see you can have safe table  
> >> names by
> >> using backtick (`) in mysql and double quotes (") in Postgresql.  
> >> What do you
> >> think?
> >
> > I happen to like the prefix, as I often use psql and it's tab  
> > trigger to
> > autocomplete table names.  Having the 'og' prefix makes things  
> > easier for me.
> > As you (if you don't do more low level db stuff) never really  
> > specify the
> > table name at all, they could be hashes instead of good sounding  
> > names.  ;)
> >
> > That said, I just see no real benefit other than 'looks slightly  
> > better from
> > the outside' and vote against.
> >
> > Jo

my .02 $, keep it as is, but make it easy for users to be aware that
Og.table_prefix = '' will generate the tables w/o the 'og' prefix.


From arne at arnebrasseur.net  Mon Aug 13 09:20:05 2007
From: arne at arnebrasseur.net (arne at arnebrasseur.net)
Date: Mon, 13 Aug 2007 06:20:05 -0700 (PDT)
Subject: [Nitro] Og table prefix
In-Reply-To: <1187006945.9453.4.camel@jhereg>
References: <afd80b540708130125l1a6b7856sb0dfe196f8199a7f@mail.gmail.com>
	<op.twztcux7xegpx6@jo.local>
	<7F82BE22-2427-4DF5-BB4D-E8238EAF1260@fragmentized.net>
	<1187006945.9453.4.camel@jhereg>
Message-ID: <3338.81.245.171.88.1187011205.squirrel@webmail.arnebrasseur.net>

> On Mon, 2007-08-13 at 12:54 +0100, Rui Miguel wrote:
>> Hello,
>>
>> I like the default also, and think it's safer and more intuitive for
>> databases with managed and unmanaged tables all together.
>>
>> > That said, I just see no real benefit other than 'looks slightly
>> > better from
>> > the outside' and vote against.
>> >
>> > Jo
>
> my .02 $, keep it as is, but make it easy for users to be aware that
> Og.table_prefix = '' will generate the tables w/o the 'og' prefix.

+1 for keeping the prefix. I see no compelling argument to change the
existing  behavior.

(ab)




From george.moschovitis at gmail.com  Mon Aug 13 12:16:34 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 13 Aug 2007 19:16:34 +0300
Subject: [Nitro] Og table prefix
In-Reply-To: <3338.81.245.171.88.1187011205.squirrel@webmail.arnebrasseur.net>
References: <afd80b540708130125l1a6b7856sb0dfe196f8199a7f@mail.gmail.com>
	<op.twztcux7xegpx6@jo.local>
	<7F82BE22-2427-4DF5-BB4D-E8238EAF1260@fragmentized.net>
	<1187006945.9453.4.camel@jhereg>
	<3338.81.245.171.88.1187011205.squirrel@webmail.arnebrasseur.net>
Message-ID: <afd80b540708130916m40d581c6qaff5cc064ef69835@mail.gmail.com>

Ok, I will leave the default prefix but I will still 'escape' the table name
(with backticks and/or quotes).
In my personal projects I plan to remove the prefix though.

thanks for the remarks,
-g.

On 8/13/07, arne at arnebrasseur.net <arne at arnebrasseur.net> wrote:
>
> > On Mon, 2007-08-13 at 12:54 +0100, Rui Miguel wrote:
> >> Hello,
> >>
> >> I like the default also, and think it's safer and more intuitive for
> >> databases with managed and unmanaged tables all together.
> >>
> >> > That said, I just see no real benefit other than 'looks slightly
> >> > better from
> >> > the outside' and vote against.
> >> >
> >> > Jo
> >
> > my .02 $, keep it as is, but make it easy for users to be aware that
> > Og.table_prefix = '' will generate the tables w/o the 'og' prefix.
>
> +1 for keeping the prefix. I see no compelling argument to change the
> existing  behavior.
>
> (ab)
>
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070813/04d32457/attachment.html 

From george.moschovitis at gmail.com  Mon Aug 13 12:18:00 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 13 Aug 2007 19:18:00 +0300
Subject: [Nitro] Vacations
Message-ID: <afd80b540708130918q6f71bd88we6b2b3bc0975635@mail.gmail.com>

Dear devs,

I will probably be offline for the next few days (but still, working on
Nitro ;-))

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070813/9dd8e7ad/attachment.html 

From george.moschovitis at gmail.com  Mon Aug 13 20:05:01 2007
From: george.moschovitis at gmail.com (george.moschovitis at gmail.com)
Date: Tue, 14 Aug 2007 00:05:01 +0000 (UTC)
Subject: [Nitro] =?utf-8?q?=5BNP=2EORG=5D_Nitro/Og_Fora_daily_digest?=
Message-ID: <20070814000501.9D5748700@reizu.com>

>From gmosx.myopenid.com 
Date: Mon Aug 13 08:24:48 UTC 2007
Subject: Drop the default og prefix from tables
Message-ID: http://www.nitroproject.org/fora/posts/view/cN2wV8sxyr3k-VeJeZaaqU

Dear devs,

I am thinking about droping the 'og' prefix from model tables by default (the setting will stay). From what I see you can have safe table names by using backtick (`) in mysql and double quotes (") in Postgresql. What do you thinkg?

-g.

>From gmosx.myopenid.com 
Date: Mon Aug 13 08:27:54 UTC 2007
Subject: Unicodechars in facets
Message-ID: http://www.nitroproject.org/fora/posts/view/aw9PtIsxCr3k-VeJeZaaqU

Tom, 

have you seen the unicode-chars gem ?

http://rubyforge.org/projects/unicodechars/

I think it would be a great inclusion in facets or something. I am not sure if this is up-to-date with the latest rails code.

-g.




--
This mail is automatically generated from the http://nitroproject.org/fora
digest robot. It presents the discussions in the fora during the last 24 hours.
Do not reply to this email.

From billk at cts.com  Wed Aug 15 20:34:24 2007
From: billk at cts.com (Bill Kelly)
Date: Wed, 15 Aug 2007 17:34:24 -0700
Subject: [Nitro] Glue::Template.root
Message-ID: <00c801c7df9d$3241a190$6442a8c0@musicbox>

Greetings and Salutations,

I have an older Nitro app with a line like:

  Glue::Template.root = 'template'

I tried Raw::Template.root, but no-go.

What's the correct way to set the template directory with the
repository version of Nitro?


Thanks!

Bill



From weather at speakeasy.net  Thu Aug 16 02:24:56 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Thu, 16 Aug 2007 02:24:56 -0400
Subject: [Nitro] Problem with Og post method
Message-ID: <200708160224.56992.weather@speakeasy.net>

Hello, I'm struggling a little with using the post method. I have two classes, 
Root and Account < Root, and I can't get post to work inside of the Root 
class. However, it does work in the Account class. When I try to utilize the 
method in Root, I get:

`method_missing': undefined method `post' for Root:Class (NoMethodError)

I'm using post with the following arguments:

post :to_world, :on => [:og_read], where to_world is a Root method

The only difference between the two classes, with regards to Og, is that 
Account is Timestamped and requires glue/timestamped (though I got the same 
errors in Root when I added the same data). Although post seems to work fine 
in Account, I did get the same NoMethodError for Account two or three times 
in a row, though I haven't been able to duplicate it. It happened right after 
I removed my original post method attempt from Root and returned it to 
Account.

I'd of course like to figure this issue out, if possible, but the purpose is 
trying to test whether or not I can place the post method in Root and have 
Account#og_read call to_world. This is probably easier to answer?

Thank you for any help,
Matt

From arne at arnebrasseur.net  Thu Aug 16 03:39:35 2007
From: arne at arnebrasseur.net (arne at arnebrasseur.net)
Date: Thu, 16 Aug 2007 00:39:35 -0700 (PDT)
Subject: [Nitro] Glue::Template.root
In-Reply-To: <00c801c7df9d$3241a190$6442a8c0@musicbox>
References: <00c801c7df9d$3241a190$6442a8c0@musicbox>
Message-ID: <1383.81.245.180.77.1187249975.squirrel@webmail.arnebrasseur.net>

> What's the correct way to set the template directory with the
> repository version of Nitro?

Raw::Template.root_dir = '...'

(ab)


From arne at arnebrasseur.net  Thu Aug 16 03:48:52 2007
From: arne at arnebrasseur.net (arne at arnebrasseur.net)
Date: Thu, 16 Aug 2007 00:48:52 -0700 (PDT)
Subject: [Nitro] Problem with Og post method
In-Reply-To: <200708160224.56992.weather@speakeasy.net>
References: <200708160224.56992.weather@speakeasy.net>
Message-ID: <1437.81.245.180.77.1187250532.squirrel@webmail.arnebrasseur.net>

Hi Matt,

Jonathan or George will probably have to answer this one, but just a guess
: where in the class definition did you make the call to post?

Og does some magic with the class, but only after it found out that the
class has to be managed by Og. Presumably the aspect methods are available
after that.

I can't test this where I'm now, and I'm no Og expert, but I'm thinking that

# this might not work
class Something
   post ....
   attr_accessor :name, String
end

# this might work
class Something
   attr_accessor :name, String
   post ....
end

(ab)

> Hello, I'm struggling a little with using the post method. I have two
> classes,
> Root and Account < Root, and I can't get post to work inside of the Root
> class. However, it does work in the Account class. When I try to utilize
> the
> method in Root, I get:
>
> `method_missing': undefined method `post' for Root:Class (NoMethodError)
>
> I'm using post with the following arguments:
>
> post :to_world, :on => [:og_read], where to_world is a Root method
>
> The only difference between the two classes, with regards to Og, is that
> Account is Timestamped and requires glue/timestamped (though I got the
> same
> errors in Root when I added the same data). Although post seems to work
> fine
> in Account, I did get the same NoMethodError for Account two or three
> times
> in a row, though I haven't been able to duplicate it. It happened right
> after
> I removed my original post method attempt from Root and returned it to
> Account.
>
> I'd of course like to figure this issue out, if possible, but the purpose
> is
> trying to test whether or not I can place the post method in Root and have
> Account#og_read call to_world. This is probably easier to answer?
>
> Thank you for any help,
> Matt
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From billk at cts.com  Thu Aug 16 04:05:52 2007
From: billk at cts.com (Bill Kelly)
Date: Thu, 16 Aug 2007 01:05:52 -0700
Subject: [Nitro] Glue::Template.root
References: <00c801c7df9d$3241a190$6442a8c0@musicbox>
	<1383.81.245.180.77.1187249975.squirrel@webmail.arnebrasseur.net>
Message-ID: <001601c7dfdc$43ea82b0$6442a8c0@musicbox>

From: <arne at arnebrasseur.net>
>
>> What's the correct way to set the template directory with the
>> repository version of Nitro?
> 
> Raw::Template.root_dir = '...'

Thanks!

Sorry about my newbie questions.


Regards,

Bill



From arne at arnebrasseur.net  Thu Aug 16 14:22:58 2007
From: arne at arnebrasseur.net (Arne Brasseur)
Date: Thu, 16 Aug 2007 20:22:58 +0200
Subject: [Nitro] Glue::Template.root
In-Reply-To: <001601c7dfdc$43ea82b0$6442a8c0@musicbox>
References: <00c801c7df9d$3241a190$6442a8c0@musicbox>	<1383.81.245.180.77.1187249975.squirrel@webmail.arnebrasseur.net>
	<001601c7dfdc$43ea82b0$6442a8c0@musicbox>
Message-ID: <46C49602.5020801@arnebrasseur.net>

Bill Kelly schreef:
> Thanks!
>
> Sorry about my newbie questions.
>   
Better to ask and look dumb for five minutes then not ask and stay dumb 
forever (or something)

:)

(ab)


From billk at cts.com  Thu Aug 16 15:08:31 2007
From: billk at cts.com (Bill Kelly)
Date: Thu, 16 Aug 2007 12:08:31 -0700
Subject: [Nitro] confused about app.dispatcher.root
Message-ID: <00fb01c7e038$d63f6de0$6442a8c0@musicbox>

Hi again,

I'm encountering an error trying to fetch the index page on
my site.  Essentially I have:

class StoreController < Nitro::Controller

  def index
    @time = Time.now
    redirect 'catalog'
  end

  # ...

end

app = Application.new
app.dispatcher.root = StoreController


...the error I'm getting is:


DEBUG: Rendering '/status_500'
ERROR: Error while handling StoreController#index(status_500)
ERROR: wrong number of arguments (1 for 0)


But when I looked at the blog and hello examples in the repository,
neither of their index methods accepted any arguments.

Then I noticed the blog actually sets app.dispatcher.root to
its Post *model* class, and the controller is defined at
Post::Controller.

So am I doing the wrong thing by setting app.dispatcher.root
directly to my controller?


Thanks,

Bill




From mvyver at gmail.com  Thu Aug 16 22:49:15 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Fri, 17 Aug 2007 12:49:15 +1000
Subject: [Nitro] OG vs Active Record
Message-ID: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>

Hi,
Thanks for all the effort that has gone into making Nitro/OG available.

I'm relatively new to Ruby and am trying to decide between diving into
ActiveRecord and OG.... I'd appreciate any advice.
The useage case/scenario is:  Single 'amateur' developer, large
numbers of databases, large databases (10'sGB/table, 10's tables/DB),
multiple ruby scripts accessing the dbserver directly (no web
interface - at this stage.)

I understand that this is a little like asking a parent if their child
is beautiful, but was hoping for some frank feedback on the following.

1) Am I right that OG largely overcomes the issues discussed in this
(long) AR thread (I summarize a key issue below):
http://rubyurl.com/xWb
and here:
http://aralbalkan.com/764

A summary of a key AR issue (for me) is:
<quote>
Whether its in migrations or in the DB schema, I have to specify all
the columns as well as their lengths and unique constraints to have
the database created.  Then I have to specify the validates_* rules
in the model to match.  Finally I have to keep both of them in sync
over time.
Why not do it all inside the model and have that set up all the
validates_* rules in a single step?
</quote>

2) Has anyone compared the performance of AR vs OG. The following
thread about AR's cpu useage, surprised me:
http://rubyurl.com/vFz

I see the same performance on my system (script attached).  Has anyone
run this test under OG?

3) How well is MySQL supported (bearing in mind that OG is at 0.5), or
is most development done with Postgresql?  On MySQL, how about the
MyISAM and Memory Engine/tables?

Thanks for any insights.
Regards
Mark
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ar_bench.rb
Type: application/x-ruby
Size: 2951 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070817/10a4e812/attachment.bin 

From mvyver at gmail.com  Fri Aug 17 03:50:55 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Fri, 17 Aug 2007 17:50:55 +1000
Subject: [Nitro] nitro 0.50.0 dev gem RunTimeError
Message-ID: <389c43e40708170050p1b1cee8dybf200bf846a43e83@mail.gmail.com>

Hi,
While I will probably spend the time to get to know Darcs it is great
to have the source packages as a gem.  Unfortunately I ran into the
following 'RunTimeError'.

laptop:/tmp # gem install nitro --install-dir /tmp --include-dependencies
ERROR:  While executing gem ... (RuntimeError)
    Error instaling nitro:
        nitro requires og = 0.50.0

laptop:/tmp # ls *.gem
glue-0.50.0.gem  nitro-0.50.0.gem  og-0.50.0.gem  raw-0.50.0.gem


I see the same error if I run:
gem install nitro --version 0.50.0 --install-dir /tmp --include-dependencies


Regards
Mark

From mvyver at gmail.com  Fri Aug 17 13:43:20 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Sat, 18 Aug 2007 03:43:20 +1000
Subject: [Nitro] Repository Gems
Message-ID: <389c43e40708171043w7124fe1euc4a64231e7aa34f@mail.gmail.com>

Hi,
This thread started before I joined, so apologies if this thread is
now disjoint.
Like Rui Miguel I ran into several 'RunTimeError's complaining about
required versions.  To install nitro I used the following sequence
(the nitro repository gems were installed in /tmp):

# gem install RedCloth --include-dependencies

# gem install ruby-breakpoint --include-dependencies

# gem install glue --install-dir /tmp

# gem install og --install-dir /tmp

# gem install raw --install-dir /tmp

# gem install nitro --install-dir /tmp

Thanks for the repository gems - useful to dip your toe into the water
without tackling Darcs.

Regards
Mark

From weather at speakeasy.net  Fri Aug 17 15:59:42 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Fri, 17 Aug 2007 15:59:42 -0400
Subject: [Nitro] The has_one/belongs_to relationship
Message-ID: <200708171559.43013.weather@speakeasy.net>

Hello, I've tried rather hard to figure this relationship out, but I just 
can't get it to work correctly on the has_one end. I have some classes that 
utilize the has_many/belongs_to relationship, and those work fine.

The two classes are:

class Account
	has_one :character, Character
end

class Character
	belongs_to :account, Account
end

When I initialize Character, I assign the passed account to its account 
variable, and the relationship is established (and saved properly to the 
database). When I try to assign the created character to Account#character, 
however, it treats the character variable as a local variable. For instance, 
in Account:

character = Character.create msg.capitalize, self
save!

Checking character outside of that method returns nil, and the character_oid 
database field is NULL. I tried using self.character and experienced the same 
results.

Both classes include the Og::EntityMixin module, and both relationships are 
defined after I define my properties.

I really appreciate any help with this -- let me know if I need to post more 
information.

Thanks,
Matt

From arne at arnebrasseur.net  Sat Aug 18 04:13:30 2007
From: arne at arnebrasseur.net (Arne Brasseur)
Date: Sat, 18 Aug 2007 10:13:30 +0200
Subject: [Nitro] confused about app.dispatcher.root
In-Reply-To: <00fb01c7e038$d63f6de0$6442a8c0@musicbox>
References: <00fb01c7e038$d63f6de0$6442a8c0@musicbox>
Message-ID: <46C6AA2A.8070609@arnebrasseur.net>

Bill Kelly schreef:
> Hi again,
>
> I'm encountering an error trying to fetch the index page on
> my site.  Essentially I have:
>
> class StoreController < Nitro::Controller
>
>   def index
>     @time = Time.now
>     redirect 'catalog'
>   end
>
>   # ...
>
> end
>
> app = Application.new
> app.dispatcher.root = StoreController
>
>
> ...the error I'm getting is:
>
>
> DEBUG: Rendering '/status_500'
> ERROR: Error while handling StoreController#index(status_500)
> ERROR: wrong number of arguments (1 for 0)
>   
Can you show us your catalog method? Does it take arguments? What's the 
URL you're loading?
> Then I noticed the blog actually sets app.dispatcher.root to
> its Post *model* class, and the controller is defined at
> Post::Controller.
>   
That's an alternative way of structuring your app. It's perfectly fine 
to just use the controller itself.
> So am I doing the wrong thing by setting app.dispatcher.root
> directly to my controller?
>   
Certainly not.

(ab)

-- 
Arne Brasseur
http://www.arnebrasseur.net
arne at arnebrasseur.net


From arne at arnebrasseur.net  Sat Aug 18 04:14:57 2007
From: arne at arnebrasseur.net (Arne Brasseur)
Date: Sat, 18 Aug 2007 10:14:57 +0200
Subject: [Nitro] nitro 0.50.0 dev gem RunTimeError
In-Reply-To: <389c43e40708170050p1b1cee8dybf200bf846a43e83@mail.gmail.com>
References: <389c43e40708170050p1b1cee8dybf200bf846a43e83@mail.gmail.com>
Message-ID: <46C6AA81.7040104@arnebrasseur.net>

Mark Van De Vyver schreef:
> Hi,
> While I will probably spend the time to get to know Darcs it is great
> to have the source packages as a gem.  Unfortunately I ran into the
> following 'RunTimeError'.
>
> laptop:/tmp # gem install nitro --install-dir /tmp --include-dependencies
> ERROR:  While executing gem ... (RuntimeError)
>     Error instaling nitro:
>         nitro requires og = 0.50.0
>   
I never installed gems from the local disk, looks like it has a hard 
time finding the other gems. Try installing og first.

(ab)

-- 
Arne Brasseur
http://www.arnebrasseur.net
arne at arnebrasseur.net


From arne at arnebrasseur.net  Sat Aug 18 04:27:05 2007
From: arne at arnebrasseur.net (Arne Brasseur)
Date: Sat, 18 Aug 2007 10:27:05 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708171559.43013.weather@speakeasy.net>
References: <200708171559.43013.weather@speakeasy.net>
Message-ID: <46C6AD59.1090304@arnebrasseur.net>

Matthew B Gardner schreef:
> Hello, I've tried rather hard to figure this relationship out, but I just 
> can't get it to work correctly on the has_one end. I have some classes that 
> utilize the has_many/belongs_to relationship, and those work fine.
>
> The two classes are:
>
> class Account
> 	has_one :character, Character
> end
>
> class Character
> 	belongs_to :account, Account
> end
>
> When I initialize Character, I assign the passed account to its account 
> variable, and the relationship is established (and saved properly to the 
> database). When I try to assign the created character to Account#character, 
> however, it treats the character variable as a local variable. For instance, 
> in Account:
>
> character = Character.create msg.capitalize, self
> save!
>
> Checking character outside of that method returns nil, and the character_oid 
> database field is NULL. I tried using self.character and experienced the same 
> results.
>   
I would love to help you but I don't know much about Og though. The lead 
developer George is currently on vacations, he should be back in a few 
days. If you don't get an answer by then please ask again.

The other Og guy Jonathan just finished his job in Finland, I imagine 
he's also taking a break.

I notice some new blood on the mailing list, welcome to Nitro! Please 
hang on if it takes a while to get your questions answered and issues 
solved. Things move slow around here sometimes. On the other hand I have 
spoken to George recently and he's planning to devote more time to 
Nitro, in fact he's working on it right now while on vacations.

My contract ends in a week, from September on I will be studying abroad. 
That should also give me more time for Nitro.

Please keep sending any stories of success or failure. If you have a 
small app that exhibits a problem, feel free to just send the whole app. 
We'll analyze it together, should be educational for all.

(ab)

-- 
Arne Brasseur
http://www.arnebrasseur.net
arne at arnebrasseur.net


From john at oxyliquit.de  Sat Aug 18 04:47:45 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 18 Aug 2007 10:47:45 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708171559.43013.weather@speakeasy.net>
References: <200708171559.43013.weather@speakeasy.net>
Message-ID: <op.tw8wdkc5xegpx6@jo.local>

Hi,

> When I try to assign the created character to Account#character,
> however, it treats the character variable as a local variable. For instance,
> in Account:
>character = Character.create msg.capitalize, self
> save!

I assume this is within a method in the Account# scope?

> Checking character outside of that method returns nil, and the character_oid
> database field is NULL. I tried using self.character and experienced the same
> results.

the Account#character= method is defined by Og and handles the rel saving.

self.character = Character.create()
p @character, @character_oid
# save!
update(:only => [:character_oid]) # slight optimization

I'm quite sure that this doesn't treat the character as the local variable.
If you tried that before without success, I'm not sure what else can be
wrong, could you try that once more?

Jo

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

From john at oxyliquit.de  Sat Aug 18 04:47:46 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 18 Aug 2007 10:47:46 +0200
Subject: [Nitro] Repository Gems
In-Reply-To: <389c43e40708171043w7124fe1euc4a64231e7aa34f@mail.gmail.com>
References: <389c43e40708171043w7124fe1euc4a64231e7aa34f@mail.gmail.com>
Message-ID: <op.tw8wrtaoxegpx6@jo.local>

Hi,

> This thread started before I joined, so apologies if this thread is
> now disjoint.
> Like Rui Miguel I ran into several 'RunTimeError's complaining about
> required versions.

thank you for reporting,  I'm sure George will figure out whats wrong.

Btw, George, didn't we say we wanted to relase an 0.49 first?  I'm still
for that, as to make some kinda 'release candidate' so we can make a
bugfix release out of 0.50 and not have to support a buggy release
until the next one.

Jo

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

From john at oxyliquit.de  Sat Aug 18 04:47:46 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 18 Aug 2007 10:47:46 +0200
Subject: [Nitro] confused about app.dispatcher.root
In-Reply-To: <00fb01c7e038$d63f6de0$6442a8c0@musicbox>
References: <00fb01c7e038$d63f6de0$6442a8c0@musicbox>
Message-ID: <op.tw8yavvmxegpx6@jo.local>

Hi,

[can't say much about the error, I think we talked on irc about that.]

My uneducated guess would be that the status_500 is either a standard
method on the controller or a template in the proto folder.
Maybe they have just been forgotten, George?

> Then I noticed the blog actually sets app.dispatcher.root to
> its Post *model* class, and the controller is defined at
> Post::Controller.
>
> So am I doing the wrong thing by setting app.dispatcher.root
> directly to my controller?

When using a model for the .dispatcher.root the controller has to be
`MyModel::Controller`.  The dispatcher will find it automatically
then.  (What it _could_ also do, is make the 'MyModel#link_to()'
work correctly when used...  Could be only my mind, I think though
someone on IRC recently metioned something like this.)

Other than that, passing a controller is perfectly ok here.  (should
be at least, everything else is a bug)

Jo

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

From arne at arnebrasseur.net  Sat Aug 18 05:20:04 2007
From: arne at arnebrasseur.net (Arne Brasseur)
Date: Sat, 18 Aug 2007 11:20:04 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <op.tw8wdkc5xegpx6@jo.local>
References: <200708171559.43013.weather@speakeasy.net>
	<op.tw8wdkc5xegpx6@jo.local>
Message-ID: <46C6B9C4.6060908@arnebrasseur.net>

Jonathan Buch schreef:
> Jo
>   
He's back! ^_^

-- 
Arne Brasseur
http://www.arnebrasseur.net
arne at arnebrasseur.net


From george.moschovitis at gmail.com  Sat Aug 18 05:46:04 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 18 Aug 2007 12:46:04 +0300
Subject: [Nitro] Back from vacations...
Message-ID: <afd80b540708180246m3a09e308yb917e7b6377719c6@mail.gmail.com>

Dear devs,

I am back from (short) vacations. I will try to go over the posts of the
last 5 days...

regards,
George.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070818/25a5e22c/attachment.html 

From george.moschovitis at gmail.com  Sat Aug 18 11:26:40 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 18 Aug 2007 18:26:40 +0300
Subject: [Nitro] Problem with Og post method
In-Reply-To: <200708160224.56992.weather@speakeasy.net>
References: <200708160224.56992.weather@speakeasy.net>
Message-ID: <afd80b540708180826se3ea13fjb2d3eb9725ee80f7@mail.gmail.com>

Please provide a simple example (small source file) that demonstrated the
problem.

-g.

On 8/16/07, Matthew B Gardner <weather at speakeasy.net> wrote:
>
> Hello, I'm struggling a little with using the post method. I have two
> classes,
> Root and Account < Root, and I can't get post to work inside of the Root
> class. However, it does work in the Account class. When I try to utilize
> the
> method in Root, I get:
>
> `method_missing': undefined method `post' for Root:Class (NoMethodError)
>
> I'm using post with the following arguments:
>
> post :to_world, :on => [:og_read], where to_world is a Root method
>
> The only difference between the two classes, with regards to Og, is that
> Account is Timestamped and requires glue/timestamped (though I got the
> same
> errors in Root when I added the same data). Although post seems to work
> fine
> in Account, I did get the same NoMethodError for Account two or three
> times
> in a row, though I haven't been able to duplicate it. It happened right
> after
> I removed my original post method attempt from Root and returned it to
> Account.
>
> I'd of course like to figure this issue out, if possible, but the purpose
> is
> trying to test whether or not I can place the post method in Root and have
> Account#og_read call to_world. This is probably easier to answer?
>
> Thank you for any help,
> Matt
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070818/88271cc3/attachment.html 

From george.moschovitis at gmail.com  Sat Aug 18 11:30:47 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 18 Aug 2007 18:30:47 +0300
Subject: [Nitro] nitro 0.50.0 dev gem RunTimeError
In-Reply-To: <389c43e40708170050p1b1cee8dybf200bf846a43e83@mail.gmail.com>
References: <389c43e40708170050p1b1cee8dybf200bf846a43e83@mail.gmail.com>
Message-ID: <afd80b540708180830y2fc1b9ffy491aca8307b0a904@mail.gmail.com>

You have to install them in order...

glue, og, raw, nitro

-g.

On 8/17/07, Mark Van De Vyver <mvyver at gmail.com> wrote:
>
> Hi,
> While I will probably spend the time to get to know Darcs it is great
> to have the source packages as a gem.  Unfortunately I ran into the
> following 'RunTimeError'.
>
> laptop:/tmp # gem install nitro --install-dir /tmp --include-dependencies
> ERROR:  While executing gem ... (RuntimeError)
>     Error instaling nitro:
>         nitro requires og = 0.50.0
>
> laptop:/tmp # ls *.gem
> glue-0.50.0.gem  nitro-0.50.0.gem  og-0.50.0.gem  raw-0.50.0.gem
>
>
> I see the same error if I run:
> gem install nitro --version 0.50.0 --install-dir /tmp
> --include-dependencies
>
>
> Regards
> Mark
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070818/b53f5415/attachment.html 

From george.moschovitis at gmail.com  Sat Aug 18 11:36:29 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 18 Aug 2007 18:36:29 +0300
Subject: [Nitro] confused about app.dispatcher.root
In-Reply-To: <00fb01c7e038$d63f6de0$6442a8c0@musicbox>
References: <00fb01c7e038$d63f6de0$6442a8c0@musicbox>
Message-ID: <afd80b540708180836w6f206e29w3b1f795faa2138e1@mail.gmail.com>

>
> class StoreController < Nitro::Controller


no need to extend Nitro::Controller (in fact there is no Nitro::Controller
class, it is Raw::Controller now)


  def index
>     @time = Time.now
>     redirect 'catalog'


where is catalog defined?

  end
>
>   # ...
>
> end
>
> app = Application.new
> app.dispatcher.root = StoreController


app.start ???




...the error I'm getting is:
>
>
> DEBUG: Rendering '/status_500'
> ERROR: Error while handling StoreController#index(status_500)
> ERROR: wrong number of arguments (1 for 0)



please show as the  error BEFORE 'Rendering /status_500'.

So am I doing the wrong thing by setting app.dispatcher.root
> directly to my controller?
>

you are doing this right.

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070818/a57c7eff/attachment-0001.html 

From george.moschovitis at gmail.com  Sat Aug 18 11:39:28 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 18 Aug 2007 18:39:28 +0300
Subject: [Nitro] Repository Gems
In-Reply-To: <389c43e40708171043w7124fe1euc4a64231e7aa34f@mail.gmail.com>
References: <389c43e40708171043w7124fe1euc4a64231e7aa34f@mail.gmail.com>
Message-ID: <afd80b540708180839k18acb42fse8c861d7c532e5d6@mail.gmail.com>

Thanks for this email. Btw, this reminded me to remove the ruby-breakpoint
dependency.

-g.

On 8/17/07, Mark Van De Vyver <mvyver at gmail.com> wrote:
>
> Hi,
> This thread started before I joined, so apologies if this thread is
> now disjoint.
> Like Rui Miguel I ran into several 'RunTimeError's complaining about
> required versions.  To install nitro I used the following sequence
> (the nitro repository gems were installed in /tmp):
>
> # gem install RedCloth --include-dependencies
>
> # gem install ruby-breakpoint --include-dependencies
>
> # gem install glue --install-dir /tmp
>
> # gem install og --install-dir /tmp
>
> # gem install raw --install-dir /tmp
>
> # gem install nitro --install-dir /tmp
>
> Thanks for the repository gems - useful to dip your toe into the water
> without tackling Darcs.
>
> Regards
> Mark
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070818/3b58386a/attachment.html 

From weather at speakeasy.net  Sat Aug 18 11:59:06 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sat, 18 Aug 2007 11:59:06 -0400
Subject: [Nitro] The has_one/belongs_to relationship
Message-ID: <200708181159.06393.weather@speakeasy.net>

Thank you for the response -- it's still not working, so I'll post the output 
from the additions you suggested.

> When I try to assign the created character to Account#character,
> however, it treats the character variable as a local variable. For instance,
>> in Account:
>>character = Character.create msg.capitalize, self
>> save!

>I assume this is within a method in the Account# scope?

Yes, this is in an Account method.

>> Checking character outside of that method returns nil, and the 
character_oid
>> database field is NULL. I tried using self.character and experienced the 
same
>> results.

>the Account#character= method is defined by Og and handles the rel saving.

>self.character = Character.create()
>p @character, @character_oid
># save!
>update(:only => [:character_oid]) # slight optimization

p @character #=> nil
p @character_oid #=> 5
update(:only => [:character_oid]) # slight optimization #=> 07-08-18 11:42:07 
[ERROR] (Account) Account#update unknown message - {:only=>[:character_oid]}

Despite the error message, it does setup its @character_oid in the database, 
but you can see that @character is still nil. Also, the corresponding 
Character isn't loaded and attached when the Account object is loaded from 
the database -- my has_many/belongs_to relationship does this (and rather 
beautifully, I might add) -- should I expect the same from this relationship?

Thank you for any assistance you can offer,
Matt

From weather at speakeasy.net  Sat Aug 18 12:05:32 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sat, 18 Aug 2007 12:05:32 -0400
Subject: [Nitro] Problem with Og post method
In-Reply-To: <afd80b540708180826se3ea13fjb2d3eb9725ee80f7@mail.gmail.com>
References: <200708160224.56992.weather@speakeasy.net>
	<afd80b540708180826se3ea13fjb2d3eb9725ee80f7@mail.gmail.com>
Message-ID: <200708181205.32571.weather@speakeasy.net>

$:.unshift "lib" if !$:.include? "lib"

# note that og and og/relation have already been required in another file

class Root
  include Og::EntityMixin

	property :name, VarChar(100)

	def initialize name
		@name   = name
		@dirty  = false
    		@remove = false
    
    		to_world
	end

  	def to_world
    		world.objects << self # world is a public method
  	end

	def dirty?
		@dirty
	end

	def dirty!
		@dirty = true
	end

	def clean!
		@dirty = false
	end
  
  	def remove!
    		@remove = true
  	end
  
  	def remove?
    		@remove
  	end

	post :to_world, :on => [:og_read]

end

Please let me know if more info is required. Thanks for any help,
Matt

On Saturday 18 August 2007 11:26, George Moschovitis wrote:
> Please provide a simple example (small source file) that demonstrated the
> problem.
>
> -g.
>
> On 8/16/07, Matthew B Gardner <weather at speakeasy.net> wrote:
> > Hello, I'm struggling a little with using the post method. I have two
> > classes,
> > Root and Account < Root, and I can't get post to work inside of the Root
> > class. However, it does work in the Account class. When I try to utilize
> > the
> > method in Root, I get:
> >
> > `method_missing': undefined method `post' for Root:Class (NoMethodError)
> >
> > I'm using post with the following arguments:
> >
> > post :to_world, :on => [:og_read], where to_world is a Root method
> >
> > The only difference between the two classes, with regards to Og, is that
> > Account is Timestamped and requires glue/timestamped (though I got the
> > same
> > errors in Root when I added the same data). Although post seems to work
> > fine
> > in Account, I did get the same NoMethodError for Account two or three
> > times
> > in a row, though I haven't been able to duplicate it. It happened right
> > after
> > I removed my original post method attempt from Root and returned it to
> > Account.
> >
> > I'd of course like to figure this issue out, if possible, but the purpose
> > is
> > trying to test whether or not I can place the post method in Root and
> > have Account#og_read call to_world. This is probably easier to answer?
> >
> > Thank you for any help,
> > Matt
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general

From john at oxyliquit.de  Sat Aug 18 16:19:07 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 18 Aug 2007 22:19:07 +0200
Subject: [Nitro] Back from vacations...
In-Reply-To: <afd80b540708180246m3a09e308yb917e7b6377719c6@mail.gmail.com>
References: <afd80b540708180246m3a09e308yb917e7b6377719c6@mail.gmail.com>
Message-ID: <op.tw9tp6ojxegpx6@jo.local>

Hi,

> Dear devs,

and dear users too, I love users.  ;)

> I am back from (short) vacations. I will try to go over the posts of the
> last 5 days...

ayy, me will be back (kind of) tomorrow, at least at home again.  Evil Uni
stuff starts a day after though *huff*

Jo

From john at oxyliquit.de  Sat Aug 18 16:19:14 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 18 Aug 2007 22:19:14 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <46C6B9C4.6060908@arnebrasseur.net>
References: <200708171559.43013.weather@speakeasy.net>
	<op.tw8wdkc5xegpx6@jo.local> <46C6B9C4.6060908@arnebrasseur.net>
Message-ID: <op.tw9trm17xegpx6@jo.local>

Hi,

> Jonathan Buch schreef:
>> Jo
>>
> He's back! ^_^

not fully, still wrong part of germany, getting slowly closer to home
though.  ;)

Jo


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

From john at oxyliquit.de  Sat Aug 18 16:19:14 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 18 Aug 2007 22:19:14 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708181159.06393.weather@speakeasy.net>
References: <200708181159.06393.weather@speakeasy.net>
Message-ID: <op.tw9unazvxegpx6@jo.local>

Hi,

>> self.character = Character.create()
>> p @character, @character_oid
>> # save!
>> update(:only => [:character_oid]) # slight optimization
>
> p @character #=> nil
> p @character_oid #=> 5
> update(:only => [:character_oid]) # slight optimization #=> 07-08-18 11:42:07
> [ERROR] (Account) Account#update unknown message - {:only=>[:character_oid]}

totally unexplainable to me, I always use update().  George, any recent work
 from you on this (don't have a clean repo from you readily accessable here).

> Despite the error message, it does setup its @character_oid in the database,
> but you can see that @character is still nil. Also, the corresponding
> Character isn't loaded and attached when the Account object is loaded from
> the database -- my has_many/belongs_to relationship does this (and rather
> beautifully, I might add) -- should I expect the same from this relationship?

Ahh... I see in the code that @character does not get set when applying,
only when you call .character then.  Though I think you're doing that.  (since
calling .character right after applying is kind of useless, as you already
had the character (when putting it in), but ... might be kind of misleading
performance wise that it pulls the object from db right after adding it.)

So yes, it does the 'correct thing' here (debatable).  Calling self.character
right after that instead of @character must pull it from the db and everything
works normally.  That the object is saved in @character is more of a
'implementation detail' anyway...

Pulling one object from the db (the account in this case) does not preload
the character.  When calling account.character it must get it though.
(The has_many doesn't preload either, that'd be crazy ^^;)

So, please tell if self.character does not return the character, that'd be a
major bug.

> Thank you for any assistance you can offer,

I allways hope to be of help.  :)

Jo

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

From john at oxyliquit.de  Sat Aug 18 16:24:42 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sat, 18 Aug 2007 22:24:42 +0200
Subject: [Nitro] Problem with Og post method
In-Reply-To: <200708181205.32571.weather@speakeasy.net>
References: <200708160224.56992.weather@speakeasy.net>
	<afd80b540708180826se3ea13fjb2d3eb9725ee80f7@mail.gmail.com>
	<200708181205.32571.weather@speakeasy.net>
Message-ID: <op.tw9uzgxxxegpx6@jo.local>

Hi,

> # note that og and og/relation have already been required in another file

Only Og has to be required, og/relation and aspects (post/pre stuff) gets
required within Og.  In fact Timestamped relies on aspects.
og/model/timestamped.rb  for an example for aspects.

Oh, but it's not pre/post it's before/after there.  Could you try using
these method names?

(I know next to nothing about aspects, so I'm not really able to help there)

Jo

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

From weather at speakeasy.net  Sat Aug 18 17:36:13 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sat, 18 Aug 2007 17:36:13 -0400
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <op.tw9unazvxegpx6@jo.local>
References: <200708181159.06393.weather@speakeasy.net>
	<op.tw9unazvxegpx6@jo.local>
Message-ID: <200708181736.13354.weather@speakeasy.net>

I think I'm working my way through it and it seems to be working correctly 
itself, for the most part. I do have some questions...


> > p @character #=> nil
> > p @character_oid #=> 5
> > update(:only => [:character_oid]) # slight optimization #=> 07-08-18
> > 11:42:07 [ERROR] (Account) Account#update unknown message -
> > {:only=>[:character_oid]}
>
> totally unexplainable to me, I always use update().  George, any recent
> work from you on this (don't have a clean repo from you readily accessable
> here).
>
> > Despite the error message, it does setup its @character_oid in the
> > database, but you can see that @character is still nil. Also, the
> > corresponding Character isn't loaded and attached when the Account object
> > is loaded from the database -- my has_many/belongs_to relationship does
> > this (and rather beautifully, I might add) -- should I expect the same
> > from this relationship?
>
> Ahh... I see in the code that @character does not get set when applying,
> only when you call .character then.  Though I think you're doing that. 
> (since calling .character right after applying is kind of useless, as you
> already had the character (when putting it in), but ... might be kind of
> misleading performance wise that it pulls the object from db right after
> adding it.)
>
> So yes, it does the 'correct thing' here (debatable).  Calling
> self.character right after that instead of @character must pull it from the
> db and everything works normally.  That the object is saved in @character
> is more of a 'implementation detail' anyway...
>

Is Character#account_oid not saved until I call Account#update? I can't say 
for sure, but I think the field was NULL in the db after Character#create, 
but was then set in the db after Account#update. Just trying to clarify how 
this works.

> Pulling one object from the db (the account in this case) does not preload
> the character.  When calling account.character it must get it though.
> (The has_many doesn't preload either, that'd be crazy ^^;)
>

I may have misspoken -- what I meant is that the has_many objects were loaded 
and attached when I loaded the object they belong to. The Character object is 
now also loading and attaching when I load the corresponding Account object. 
I'm assuming, from what you said, that this loading and attachment happens 
during Og#allocate?

> So, please tell if self.character does not return the character, that'd be
> a major bug.
>

As far as I can tell, it's working -- no bug squashing needed on this front. 
I'm still feeling this relationship out though -- I'm actually curious about 
why I can refer to Character#account by "account", but Account#character only 
by self.character and why it treats "character" as a local variable (but 
not "account" in the Character class).

Thanks again for the help -- I'll keep you posted (pardon the pun) on any 
further difficulties.

-Matt

From weather at speakeasy.net  Sat Aug 18 18:03:45 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sat, 18 Aug 2007 18:03:45 -0400
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <op.tw9unazvxegpx6@jo.local>
References: <200708181159.06393.weather@speakeasy.net>
	<op.tw9unazvxegpx6@jo.local>
Message-ID: <200708181803.45176.weather@speakeasy.net>

I may have spoken a little prematurely on this -- I am encountering a problem. 
When I call Character#create, or when a Character object is loaded (via it's 
corresponding Account being loaded), it's creating it's own account. To 
illustrate, this is what's happening:

a1 = Account.new
a1.character = Character.create self
a1.equal?(a1.character.account) #=> false

The same thing is happening when loaded from the database, so the account is 
pointing to the character, but the character is pointing to a different 
account (an account with the same attributes, but a different identity).

I'm assuming I'm doing this to myself, here is how I'm initializing Character:

class Account
	has_one :character, Character

	def foo
		self.character = Character.create msg.capitalize, self
	      	update(:only => [:character_oid]) # slight optimization
	end
end

class Character
	belongs_to :account, Account

	def initialize name, acc
		super(name)
		self.account = acc
	end
end

Let me know if that isn't enough information, or if it's unclear.

Thanks for any help (getting closer!),
Matt

From weather at speakeasy.net  Sat Aug 18 20:13:09 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sat, 18 Aug 2007 20:13:09 -0400
Subject: [Nitro] Problem with Og post method
In-Reply-To: <afd80b540708180826se3ea13fjb2d3eb9725ee80f7@mail.gmail.com>
References: <200708160224.56992.weather@speakeasy.net>
	<afd80b540708180826se3ea13fjb2d3eb9725ee80f7@mail.gmail.com>
Message-ID: <200708182013.09226.weather@speakeasy.net>

Jonathan made a reply that I didn't receive, so I'm replying to this message 
again to keep it in the same thread. He suggested that I use after() instead 
of post(). I had actually already tried that and again received a 
NoMethodError.

On Saturday 18 August 2007 11:26, George Moschovitis wrote:
> Please provide a simple example (small source file) that demonstrated the
> problem.
>
> -g.
>
> On 8/16/07, Matthew B Gardner <weather at speakeasy.net> wrote:
> > Hello, I'm struggling a little with using the post method. I have two
> > classes,
> > Root and Account < Root, and I can't get post to work inside of the Root
> > class. However, it does work in the Account class. When I try to utilize
> > the
> > method in Root, I get:
> >
> > `method_missing': undefined method `post' for Root:Class (NoMethodError)
> >
> > I'm using post with the following arguments:
> >
> > post :to_world, :on => [:og_read], where to_world is a Root method
> >
> > The only difference between the two classes, with regards to Og, is that
> > Account is Timestamped and requires glue/timestamped (though I got the
> > same
> > errors in Root when I added the same data). Although post seems to work
> > fine
> > in Account, I did get the same NoMethodError for Account two or three
> > times
> > in a row, though I haven't been able to duplicate it. It happened right
> > after
> > I removed my original post method attempt from Root and returned it to
> > Account.
> >
> > I'd of course like to figure this issue out, if possible, but the purpose
> > is
> > trying to test whether or not I can place the post method in Root and
> > have Account#og_read call to_world. This is probably easier to answer?
> >
> > Thank you for any help,
> > Matt
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general

From mvyver at gmail.com  Sun Aug 19 00:59:38 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Sun, 19 Aug 2007 14:59:38 +1000
Subject: [Nitro] Nitro>>Developers Center page: some errors
Message-ID: <389c43e40708182159w5b31fec4r9e7264febc6720ec@mail.gmail.com>

Hi,

After some reading around I've decided OG will be a worth a shot over AR.....
I encountered some problems working from the page:
http://www.nitroproject.org/dev

Details follow:

A) Under Get Started, the ink behind "The Core developers forum" to
http://www.nitroproject.org/fora/view/6
delivers and error page:

 Internal Server Error
 Path: /fora/view/6
 Reload this page. Go to the referer or the home page.
 Powered by Nitro version 0.50.0

B) The 'fora' link under 'Get Started' seems to be OK, however when I
click on "Post new topic" I get the openid sign-in page.  Providing my
openID yields the following page:

Server Error in '/' Application.
Object reference not set to an instance of an object.
Description: An unhandled exception occurred during the execution of
the current web request. Please review the stack trace for more
information about the error and where it originated in the code.

Exception Details: System.NullReferenceException: Object reference not
set to an instance of an object.

Source Error:

Line 67:             if
(State.Session.LastRequest.IsAnySimpleRegistrationFieldsRequestedOrRequired)
Line 68:             {
Line 69:                 int? profileid =
trust.RetrieveProfile(Context.User.Identity.Name,
State.Session.LastRequest.TrustRoot);
Line 70:
Line 71:                 if (profileid.HasValue)


Source File: e:\Inetpub\openid\login.aspx.cs    Line: 69

Stack Trace:

[NullReferenceException: Object reference not set to an instance of an object.]
   org.openid.User.DataUser.GetUserId(String username) in C:\Documents
and Settings\steven\My Documents\Visual Studio
2005\Projects\Org\orglib\User\DataUser.cs:175
   org.openid.Trust.ProfileTrust.RetrieveProfile(String username,
String requestURL) in C:\Documents and Settings\steven\My
Documents\Visual Studio
2005\Projects\Org\orglib\Trust\ProfileTrust.cs:94
   login.AutoShareData() in e:\Inetpub\openid\login.aspx.cs:69
   login.Page_Load(Object sender, EventArgs e) in
e:\Inetpub\openid\login.aspx.cs:29
   System.Web.Util.CalliHelper.EventArgFunctionCaller(IntPtr fp,
Object o, Object t, EventArgs e) +15
   System.Web.Util.CalliEventHandlerDelegateProxy.Callback(Object
sender, EventArgs e) +34
   System.Web.UI.Control.OnLoad(EventArgs e) +99
   System.Web.UI.Control.LoadRecursive() +47
   System.Web.UI.Page.ProcessRequestMain(Boolean
includeStagesBeforeAsyncPoint, Boolean includeStagesAfterAsyncPoint)
+1061


Version Information: Microsoft .NET Framework Version:2.0.50727.832;
ASP.NET Version:2.0.50727.832


Hope this helps.
Mark

From weather at speakeasy.net  Sun Aug 19 01:46:25 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sun, 19 Aug 2007 01:46:25 -0400
Subject: [Nitro] Problem with Og post method
In-Reply-To: <200708182013.09226.weather@speakeasy.net>
References: <200708160224.56992.weather@speakeasy.net>
	<afd80b540708180826se3ea13fjb2d3eb9725ee80f7@mail.gmail.com>
	<200708182013.09226.weather@speakeasy.net>
Message-ID: <200708190146.25964.weather@speakeasy.net>

I should also note that I have several classes that utilize the pre/post 
methods fine -- it's just the Root class that doesn't seem to like it.

On Saturday 18 August 2007 20:13, Matthew B Gardner wrote:
> Jonathan made a reply that I didn't receive, so I'm replying to this
> message again to keep it in the same thread. He suggested that I use
> after() instead of post(). I had actually already tried that and again
> received a
> NoMethodError.
>
> On Saturday 18 August 2007 11:26, George Moschovitis wrote:
> > Please provide a simple example (small source file) that demonstrated the
> > problem.
> >
> > -g.
> >
> > On 8/16/07, Matthew B Gardner <weather at speakeasy.net> wrote:
> > > Hello, I'm struggling a little with using the post method. I have two
> > > classes,
> > > Root and Account < Root, and I can't get post to work inside of the
> > > Root class. However, it does work in the Account class. When I try to
> > > utilize the
> > > method in Root, I get:
> > >
> > > `method_missing': undefined method `post' for Root:Class
> > > (NoMethodError)
> > >
> > > I'm using post with the following arguments:
> > >
> > > post :to_world, :on => [:og_read], where to_world is a Root method
> > >
> > > The only difference between the two classes, with regards to Og, is
> > > that Account is Timestamped and requires glue/timestamped (though I got
> > > the same
> > > errors in Root when I added the same data). Although post seems to work
> > > fine
> > > in Account, I did get the same NoMethodError for Account two or three
> > > times
> > > in a row, though I haven't been able to duplicate it. It happened right
> > > after
> > > I removed my original post method attempt from Root and returned it to
> > > Account.
> > >
> > > I'd of course like to figure this issue out, if possible, but the
> > > purpose is
> > > trying to test whether or not I can place the post method in Root and
> > > have Account#og_read call to_world. This is probably easier to answer?
> > >
> > > Thank you for any help,
> > > Matt
> > > _______________________________________________
> > > 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 george.moschovitis at gmail.com  Sun Aug 19 03:03:54 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 19 Aug 2007 10:03:54 +0300
Subject: [Nitro] OG vs Active Record
In-Reply-To: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
Message-ID: <afd80b540708190003h31648fbdif3a7c7e25ef26a22@mail.gmail.com>

>
> The useage case/scenario is:  Single 'amateur' developer, large
> numbers of databases, large databases (10'sGB/table, 10's tables/DB),
> multiple ruby scripts accessing the dbserver directly (no web
> interface - at this stage.)


I don't see any Og related issue here. This has to do with the actual RDBMS
used.

1) Am I right that OG largely overcomes the issues discussed in this
> (long) AR thread (I summarize a key issue below):
> http://rubyurl.com/xWb
> and here:
> http://aralbalkan.com/764


I would say yes.

I see the same performance on my system (script attached).  Has anyone
> run this test under OG?


Not me

3) How well is MySQL supported (bearing in mind that OG is at 0.5), or
> is most development done with Postgresql?  On MySQL, how about the
> MyISAM and Memory Engine/tables?


I am the main developer and I use MySQL.

regards,
George.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070819/d31e0617/attachment.html 

From george.moschovitis at gmail.com  Sun Aug 19 03:11:01 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 19 Aug 2007 10:11:01 +0300
Subject: [Nitro] Nitro>>Developers Center page: some errors
In-Reply-To: <389c43e40708182159w5b31fec4r9e7264febc6720ec@mail.gmail.com>
References: <389c43e40708182159w5b31fec4r9e7264febc6720ec@mail.gmail.com>
Message-ID: <afd80b540708190011o2d32204fxa8da3fc12bb8c88f@mail.gmail.com>

>
> A) Under Get Started, the ink behind "The Core developers forum" to
> http://www.nitroproject.org/fora/view/6
> delivers and error page:


will fix.

> B) The 'fora' link under 'Get Started' seems to be OK, however when I
> click on "Post new topic" I get the openid sign-in page.  Providing my
> openID yields the following page:
>
> Server Error in '/' Application.
> Object reference not set to an instance of an object.
> Description: An unhandled exception occurred during the execution of
> the current web request. Please review the stack trace for more
> information about the error and where it originated in the code.
>
> Exception Details: System.NullReferenceException: Object reference not
> set to an instance of an object.
>
> Source Error:
>
> Line 67:             if
> (
> State.Session.LastRequest.IsAnySimpleRegistrationFieldsRequestedOrRequired
> )
> Line 68:             {
> Line 69:                 int? profileid =
> trust.RetrieveProfile(Context.User.Identity.Name,
> State.Session.LastRequest.TrustRoot);
> Line 70:
> Line 71:                 if (profileid.HasValue)
>
>
> Source File: e:\Inetpub\openid\login.aspx.cs    Line: 69
>
> Stack Trace:
>
> [NullReferenceException: Object reference not set to an instance of an
> object.]
>    org.openid.User.DataUser.GetUserId(String username) in C:\Documents
> and Settings\steven\My Documents\Visual Studio
> 2005\Projects\Org\orglib\User\DataUser.cs:175
>    org.openid.Trust.ProfileTrust.RetrieveProfile(String username,
> String requestURL) in C:\Documents and Settings\steven\My
> Documents\Visual Studio
> 2005\Projects\Org\orglib\Trust\ProfileTrust.cs:94
>    login.AutoShareData() in e:\Inetpub\openid\login.aspx.cs:69
>    login.Page_Load(Object sender, EventArgs e) in
> e:\Inetpub\openid\login.aspx.cs:29
>    System.Web.Util.CalliHelper.EventArgFunctionCaller(IntPtr fp,
> Object o, Object t, EventArgs e) +15
>    System.Web.Util.CalliEventHandlerDelegateProxy.Callback(Object
> sender, EventArgs e) +34
>    System.Web.UI.Control.OnLoad(EventArgs e) +99
>    System.Web.UI.Control.LoadRecursive() +47
>    System.Web.UI.Page.ProcessRequestMain(Boolean
> includeStagesBeforeAsyncPoint, Boolean includeStagesAfterAsyncPoint)
> +1061
>
>
> Version Information: Microsoft .NET Framework Version:2.0.50727.832;
> ASP.NET Version:2.0.50727.832



Microsoft.NET ?? This looks like an error in your openid provider.

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070819/74c39cd3/attachment.html 

From george.moschovitis at gmail.com  Sun Aug 19 03:31:21 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 19 Aug 2007 10:31:21 +0300
Subject: [Nitro] Problem with Og post method
In-Reply-To: <200708181205.32571.weather@speakeasy.net>
References: <200708160224.56992.weather@speakeasy.net>
	<afd80b540708180826se3ea13fjb2d3eb9725ee80f7@mail.gmail.com>
	<200708181205.32571.weather@speakeasy.net>
Message-ID: <afd80b540708190031w7f117150k6c65b40201468388@mail.gmail.com>

The following works as expected in the Repository version of Og. If you have
further questions please let me
know.


require "og"

class Root
  attr_accessor :name, Og::VarChar(100)

  after :og_read, :call => :to_world

  # more verbose (and less efficient) alternative
  # after :og_read do
  #   to_world()
  # end

  def initialize(name = nil)
    @name   = name
    @dirty  = false
    @remove = false

    to_world
  end

  def to_world
    puts "----- POST ASPECT CALLED -----"
  end

  def dirty?
    @dirty
  end

  def dirty!
    @dirty = true
  end

  def clean!
    @dirty = false
  end

  def remove!
    @remove = true
  end

  def remove?
    @remove
  end
end

Og.start(
  :name => "test",
  :adapter => :mysql,
  :user => "root",
  :password => ",sql51e",
  :destroy => true,
  :evolve_schema => :full
)

# At the moment this is needed. Hopefuly will not be needed in
# the final release.

Aspects.setup


r = Root.create_with :name => "George", :dirty => true
r0 = Root.all.first
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070819/461fd832/attachment.html 

From george.moschovitis at gmail.com  Sun Aug 19 03:32:08 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 19 Aug 2007 10:32:08 +0300
Subject: [Nitro] Problem with Og post method
In-Reply-To: <afd80b540708190031w7f117150k6c65b40201468388@mail.gmail.com>
References: <200708160224.56992.weather@speakeasy.net>
	<afd80b540708180826se3ea13fjb2d3eb9725ee80f7@mail.gmail.com>
	<200708181205.32571.weather@speakeasy.net>
	<afd80b540708190031w7f117150k6c65b40201468388@mail.gmail.com>
Message-ID: <afd80b540708190032s63ab6f60yfab6b33ec7525a95@mail.gmail.com>

btw,

you can still use post instead of after, property instead of attr_accessor
etc etc...

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070819/e321be34/attachment.html 

From george.moschovitis at gmail.com  Sun Aug 19 03:34:53 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 19 Aug 2007 10:34:53 +0300
Subject: [Nitro] OG vs Active Record
In-Reply-To: <afd80b540708190003h31648fbdif3a7c7e25ef26a22@mail.gmail.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<afd80b540708190003h31648fbdif3a7c7e25ef26a22@mail.gmail.com>
Message-ID: <afd80b540708190034x7f2b7c53ja5df1d79b3851afd@mail.gmail.com>

>
> I see the same performance on my system (script attached).  Has anyone
> > run this test under OG?
>
>
> Not me
>


Og *may* be faster because it uses custom, precompiled
serializers/deserializers for each managed class.

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070819/ae4b3689/attachment-0001.html 

From weather at speakeasy.net  Sun Aug 19 03:53:04 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sun, 19 Aug 2007 03:53:04 -0400
Subject: [Nitro] Problem with Og post method
In-Reply-To: <afd80b540708190031w7f117150k6c65b40201468388@mail.gmail.com>
References: <200708160224.56992.weather@speakeasy.net>
	<200708181205.32571.weather@speakeasy.net>
	<afd80b540708190031w7f117150k6c65b40201468388@mail.gmail.com>
Message-ID: <200708190353.05044.weather@speakeasy.net>

Hello -

On Sunday 19 August 2007 03:31, George Moschovitis wrote:
> The following works as expected in the Repository version of Og. If you
> have further questions please let me
> know.
>
>
> require "og"
>
> class Root
>   attr_accessor :name, Og::VarChar(100)
>
>   after :og_read, :call => :to_world
>
>   # more verbose (and less efficient) alternative
>   # after :og_read do
>   #   to_world()
>   # end
>
>   def initialize(name = nil)
>     @name   = name
>     @dirty  = false
>     @remove = false
>
>     to_world
>   end
>
>   def to_world
>     puts "----- POST ASPECT CALLED -----"
>   end
>
>   def dirty?
>     @dirty
>   end
>
>   def dirty!
>     @dirty = true
>   end
>
>   def clean!
>     @dirty = false
>   end
>
>   def remove!
>     @remove = true
>   end
>
>   def remove?
>     @remove
>   end
> end
>
> Og.start(
>
>   :name => "test",
>   :adapter => :mysql,
>   :user => "root",
>   :password => ",sql51e",
>   :destroy => true,
>   :evolve_schema => :full
>
> )
>
> # At the moment this is needed. Hopefuly will not be needed in
> # the final release.
>
> Aspects.setup
>
>
> r = Root.create_with :name => "George", :dirty => true
> r0 = Root.all.first

Prior to requiring the Root file,

require 'og'
Aspects.setup #=> ./lib/engine/cyclone.rb:9: undefined method `setup' for 
Aspects:Module (NoMethodError)

Using og (0.41.0)

This help at all?

Thanks again for all the assistance,
Matt

From george.moschovitis at gmail.com  Sun Aug 19 04:58:18 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 19 Aug 2007 11:58:18 +0300
Subject: [Nitro] Problem with Og post method
In-Reply-To: <200708190353.05044.weather@speakeasy.net>
References: <200708160224.56992.weather@speakeasy.net>
	<200708181205.32571.weather@speakeasy.net>
	<afd80b540708190031w7f117150k6c65b40201468388@mail.gmail.com>
	<200708190353.05044.weather@speakeasy.net>
Message-ID: <afd80b540708190158p5d084dal1b74ebd80d6e15df@mail.gmail.com>

Please,

get the latest version (either from darcs or from the repository gems:
http://www.nitroproject.org/repo/gems

0.41.0 is about a year old and I cannot easily help you.

thanks,
George.



On 8/19/07, Matthew B Gardner <weather at speakeasy.net> wrote:
>
> Hello -
>
> On Sunday 19 August 2007 03:31, George Moschovitis wrote:
> > The following works as expected in the Repository version of Og. If you
> > have further questions please let me
> > know.
> >
> >
> > require "og"
> >
> > class Root
> >   attr_accessor :name, Og::VarChar(100)
> >
> >   after :og_read, :call => :to_world
> >
> >   # more verbose (and less efficient) alternative
> >   # after :og_read do
> >   #   to_world()
> >   # end
> >
> >   def initialize(name = nil)
> >     @name   = name
> >     @dirty  = false
> >     @remove = false
> >
> >     to_world
> >   end
> >
> >   def to_world
> >     puts "----- POST ASPECT CALLED -----"
> >   end
> >
> >   def dirty?
> >     @dirty
> >   end
> >
> >   def dirty!
> >     @dirty = true
> >   end
> >
> >   def clean!
> >     @dirty = false
> >   end
> >
> >   def remove!
> >     @remove = true
> >   end
> >
> >   def remove?
> >     @remove
> >   end
> > end
> >
> > Og.start(
> >
> >   :name => "test",
> >   :adapter => :mysql,
> >   :user => "root",
> >   :password => ",sql51e",
> >   :destroy => true,
> >   :evolve_schema => :full
> >
> > )
> >
> > # At the moment this is needed. Hopefuly will not be needed in
> > # the final release.
> >
> > Aspects.setup
> >
> >
> > r = Root.create_with :name => "George", :dirty => true
> > r0 = Root.all.first
>
> Prior to requiring the Root file,
>
> require 'og'
> Aspects.setup #=> ./lib/engine/cyclone.rb:9: undefined method `setup' for
> Aspects:Module (NoMethodError)
>
> Using og (0.41.0)
>
> This help at all?
>
> Thanks again for all the assistance,
> Matt
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070819/4bf4c839/attachment.html 

From mvyver at gmail.com  Sun Aug 19 08:22:50 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Sun, 19 Aug 2007 22:22:50 +1000
Subject: [Nitro] Nitro>>Developers Center page: some errors
In-Reply-To: <afd80b540708190011o2d32204fxa8da3fc12bb8c88f@mail.gmail.com>
References: <389c43e40708182159w5b31fec4r9e7264febc6720ec@mail.gmail.com>
	<afd80b540708190011o2d32204fxa8da3fc12bb8c88f@mail.gmail.com>
Message-ID: <389c43e40708190522s40a8cf74oa0c223dd475e4c8@mail.gmail.com>

Hi George,
Thanks for the reply
> > Version Information: Microsoft .NET Framework Version:2.0.50727.832;
> > ASP.NET Version:2.0.50727.832
>
>
> Microsoft.NET ?? This looks like an error in your openid provider.

This si probably off-topic:

I'm new to openID, the provider I used was

http://www.openid.org/start.aspx

This was the one linked to in the openid text in the phrase: 'If you
don't have an OpenID you can grab one here' on the page

http://www.nitroproject.org/users/signin#

I'll take this up with that sites' admin/author.
I assume that either of www.myopenid.com or www.lifejournal.com are
known to be good? Is one 'better'?

Regards
Mark


>
> -g.
>
> --
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
>  http://cull.gr
> http://www.joy.gr
> http://nitroproject.org

From transfire at gmail.com  Sun Aug 19 13:59:08 2007
From: transfire at gmail.com (Trans)
Date: Sun, 19 Aug 2007 17:59:08 -0000
Subject: [Nitro] OG vs Active Record
In-Reply-To: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
Message-ID: <1187546348.768139.237450@k79g2000hse.googlegroups.com>



On Aug 16, 7:49 pm, "Mark Van De Vyver" <mvy... at gmail.com> wrote:
> Hi,
> Thanks for all the effort that has gone into making Nitro/OG available.
>
> I'm relatively new to Ruby and am trying to decide between diving into
> ActiveRecord and OG.... I'd appreciate any advice.
> The useage case/scenario is:  Single 'amateur' developer, large
> numbers of databases, large databases (10'sGB/table, 10's tables/DB),
> multiple ruby scripts accessing the dbserver directly (no web
> interface - at this stage.)
>
> I understand that this is a little like asking a parent if their child
> is beautiful, but was hoping for some frank feedback on the following.
>
> 1) Am I right that OG largely overcomes the issues discussed in this
> (long) AR thread (I summarize a key issue below):http://rubyurl.com/xWb
> and here:http://aralbalkan.com/764
>
> A summary of a key AR issue (for me) is:
> <quote>
> Whether its in migrations or in the DB schema, I have to specify all
> the columns as well as their lengths and unique constraints to have
> the database created.  Then I have to specify the validates_* rules
> in the model to match.  Finally I have to keep both of them in sync
> over time.
> Why not do it all inside the model and have that set up all the
> validates_* rules in a single step?
> </quote>
>
> 2) Has anyone compared the performance of AR vs OG. The following
> thread about AR's cpu useage, surprised me:http://rubyurl.com/vFz
>
> I see the same performance on my system (script attached).  Has anyone
> run this test under OG?
>
> 3) How well is MySQL supported (bearing in mind that OG is at 0.5), or
> is most development done with Postgresql?  On MySQL, how about the
> MyISAM and Memory Engine/tables?

Objectively speaking. I would say Og is a better choice for ORM if you
are specifically NOT doing Rails development, but really it's a matter
of preference. The ActiveRecord model requires you to create the
Database first and foremost, Og starts with the model. While
ActiveRecord has model <-> db sync issues to manage, it's not usually
that bad. It's elaborate migration system helps a good bit. With Og
it's a lot easier, but you still have to deal with migrations, but
they aren't critical to functionality as with ARs and Og helps to make
it simple enough. However, FYI, there is a solution that makes
ActiveRecord easier to use by more intelligently using DB metadata:
http://magicmodels.rubyforge.org/. So there are options.

I'm sure you know this too, but just in case there was any doubt:
While Og is part of Nitro, it can be used completely independently.
I've actually been trying to convince George to free Og form the Nitro
repo for a while now! ;-)

Finally I should mention, if speed is your number one criteria, you
may want to look as DBI too. DBI is a straight forward adapter system
which simply allows you to work with the database directly in SQL. No
"magic" conveniences, but also no meta-code to slow anything down.

T.


From john at oxyliquit.de  Sun Aug 19 14:42:10 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sun, 19 Aug 2007 20:42:10 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708181736.13354.weather@speakeasy.net>
References: <200708181159.06393.weather@speakeasy.net>
	<op.tw9unazvxegpx6@jo.local>
	<200708181736.13354.weather@speakeasy.net>
Message-ID: <op.txbjplddxegpx6@jo.local>

Hi,

> Is Character#account_oid not saved until I call Account#update? I can't say
> for sure, but I think the field was NULL in the db after Character#create,
> but was then set in the db after Account#update. Just trying to clarify how
> this works.

Ah, no, nothing gets 'autosaved' just by setting something (would be quite
inefficient, saving 10 values in a row would mean 10 queries, unacceptable).
By doing a .save or .update you then send changes to the DB.

> I may have misspoken -- what I meant is that the has_many objects were loaded
> and attached when I loaded the object they belong to. The Character object is
> now also loading and attaching when I load the corresponding Account object.
> I'm assuming, from what you said, that this loading and attachment happens
> during Og#allocate?

Ah yes, 'loaded and attached', that'd be a good description on what it does.  :)

On the internals... well, that always depends on how the object is loaded.

`MyModel[oid]` for example would only make a SELECT and hand it over to
`read_one()`  (sql.rb) which then creates a new object (MyModel.allocate, which
only, if you didn't know that, prevents ruby from calling the `initialize()`).
The newly allocated object then is filled by `read_row()`, which just sets
instance variables (one for each field) on that object.
Conceptually pretty simple I think.  :)

> As far as I can tell, it's working -- no bug squashing needed on this front.

Good good.  :)

> I'm still feeling this relationship out though -- I'm actually curious about
> why I can refer to Character#account by "account", but Account#character only
> by self.character and why it treats "character" as a local variable (but
> not "account" in the Character class).

Well... simple Ruby:

class A
   attr_accessor :a

   def set(o)
     a = o # this will set an instance variable
     self.a = o # this will use `def a=()` which gets created by attr_acc
   end
end

A.new.set 'foo'
A.new.a = 'foo' # will do the same

The same applies for both Character and Account, not sure how your exact code
looks like of course.

Hope that helps,

Jo

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

From john at oxyliquit.de  Sun Aug 19 14:42:08 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sun, 19 Aug 2007 20:42:08 +0200
Subject: [Nitro] OG vs Active Record
In-Reply-To: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
Message-ID: <op.txbixtaxxegpx6@jo.local>

Hi,

> I'm relatively new to Ruby and am trying to decide between diving into
> ActiveRecord and OG.... I'd appreciate any advice.
> The useage case/scenario is:  Single 'amateur' developer, large
> numbers of databases, large databases (10'sGB/table, 10's tables/DB),
> multiple ruby scripts accessing the dbserver directly (no web
> interface - at this stage.)

Multi-database is not as thoroughly tested as single, but the number
(and also the sizes of the tables) do not make Og slow.  It's just how
much you need to transfer between Og/DB.
On startup time, if you stay below a 100 tables, it'll still be quite
ok on a halfway capable computer.  There are a few tweaks you can do,
but doing them means your app and db design is 'finished'.

> I understand that this is a little like asking a parent if their child
> is beautiful, but was hoping for some frank feedback on the following.
>
> 1) Am I right that OG largely overcomes the issues discussed in this
> (long) AR thread (I summarize a key issue below):
> http://rubyurl.com/xWb

Yes, by using Og basically _all_ your information is in the model file.
Even the DB layout is there (which is what differenciates Og from AR).

> and here:
> http://aralbalkan.com/764

Yep, this is exactly what Og does.  :)

> A summary of a key AR issue (for me) is:
> <quote>
> Whether its in migrations or in the DB schema, I have to specify all
> the columns as well as their lengths and unique constraints to have
> the database created.  Then I have to specify the validates_* rules
> in the model to match.  Finally I have to keep both of them in sync
> over time.
> Why not do it all inside the model and have that set up all the
> validates_* rules in a single step?
> </quote>

This certainly could be done and it has been asked for already on irc
some year ago.  I wasn't too sure about that back then, and I still
hesitate on that.  We could make it an option, `Og.auto_validations`
which then does the trick....  G?

Personally, I wouldn't use it largly because I tend to see some of the
validations as useless and time consuming (`uniq` for example, which is
again done on DB level, where it actually belongs to.
Some of the validations like 'belongs_to' (meaning it shouldn't exist
when the 'main' object disappears) can only be done via certain DBs
(like psql) with foreign key constraints.....  There's just so much
stuff the DB does better, that's what I think, anyway.  :P).

Anyway, like I said, the idea isn't all too bad, I'm just not sure
if it's worth it, as the models/validations don't get changed so
often (I draw stuff on real paper before doing DB work :P).

> 2) Has anyone compared the performance of AR vs OG. The following
> thread about AR's cpu useage, surprised me:
> http://rubyurl.com/vFz
>
> I see the same performance on my system (script attached).  Has anyone
> run this test under OG?

I'd love if you could adapt (I haven't had a look at it yet) that script
and run it on Og?

I would think that the performance of AR and Og could be quite comparable.
(Though, I could be horribly wrong, when I started with Nitro I thought
we were slower than rails, which turned out quite a wrong assumption ;))

But, I think you'll see the same 'behaviour' with Og and MySQL.  MySQL
is just too old, too optimized and there's too much movement behind it
for Og to really stress it.  At least that's my guess.  ;)

> 3) How well is MySQL supported (bearing in mind that OG is at 0.5), or
> is most development done with Postgresql?  On MySQL, how about the
> MyISAM and Memory Engine/tables?

Og is at 0.50, so about 50 revisions (actually I think more like 35
public ones..), not the version number '0.5'.

MySQL is (and always has been) actually the 'premier' DB used by the head
maintainer.   It's just always those weird other people like me who just
use PostgreSQL.  ;)  (Which means, that I optimize Og for my psql usage,
only run the specs with it etc, which means that sometimes the psql part
is better supported than the mysql/sqlite one).
In-memory SQLite is actually quite nicely supported, and I know one from
IRC who used it to speed up Og quite a bit (synchronizing the mem with
a real backend now and then).
I know nothing about MySQL memory engines, so can't comment.

Hope that helps,

Jo

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

From john at oxyliquit.de  Sun Aug 19 14:42:11 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Sun, 19 Aug 2007 20:42:11 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708181803.45176.weather@speakeasy.net>
References: <200708181159.06393.weather@speakeasy.net>
	<op.tw9unazvxegpx6@jo.local>
	<200708181803.45176.weather@speakeasy.net>
Message-ID: <op.txbku8jixegpx6@jo.local>

Hi,

> I may have spoken a little prematurely on this -- I am encountering a problem.
> When I call Character#create, or when a Character object is loaded (via it's
> corresponding Account being loaded), it's creating it's own account. To
> illustrate, this is what's happening:
>
> a1 = Account.new
> a1.character = Character.create self
> a1.equal?(a1.character.account) #=> false
>
> The same thing is happening when loaded from the database, so the account is
> pointing to the character, but the character is pointing to a different
> account (an account with the same attributes, but a different identity).

Yes, a different 'identity', but not 'equality', depending on which method is
called.  ;)
Ruby knows .eql? .equal? .== .===
.equal? is for testing if the object ids are the same, which they are not in
this case.

a1 == a1.character.account # true

Og overrides ==() to react on the primary key of that model.

> I'm assuming I'm doing this to myself, here is how I'm initializing Character:

Looks all good.  :)

> Let me know if that isn't enough information, or if it's unclear.

> Thanks for any help (getting closer!),

You're welcome.  :)

Jo

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

From weather at speakeasy.net  Sun Aug 19 16:42:31 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sun, 19 Aug 2007 16:42:31 -0400
Subject: [Nitro] Problem with Og post method
In-Reply-To: <afd80b540708190158p5d084dal1b74ebd80d6e15df@mail.gmail.com>
References: <200708160224.56992.weather@speakeasy.net>
	<200708190353.05044.weather@speakeasy.net>
	<afd80b540708190158p5d084dal1b74ebd80d6e15df@mail.gmail.com>
Message-ID: <200708191642.31292.weather@speakeasy.net>

Hello -

On Sunday 19 August 2007 04:58, George Moschovitis wrote:
> Please,
>
> get the latest version (either from darcs or from the repository gems:
> http://www.nitroproject.org/repo/gems
>
> 0.41.0 is about a year old and I cannot easily help you.
>
> thanks,
> George.
>

Sorry about that -- I've now upgraded to 0.50.0 for both glue and og.

class Root

	property :name, VarChar(100)
  
  	after :og_read, :call => :to_world

	def initialize name = "_error"
		@name   = name
		@dirty  = false
    		@remove = false
	end

  	def to_world
    		p "--WE GOT HERE--"
    		world.objects << self
  	end

	...
end

Aspects.setup
-end of file-


Root#to_world is never getting called -- I tried placing the call in my 
Account < Root class, but that didn't work either.

Thanks again for the help,
Matt

From weather at speakeasy.net  Sun Aug 19 16:57:32 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sun, 19 Aug 2007 16:57:32 -0400
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <op.txbku8jixegpx6@jo.local>
References: <200708181159.06393.weather@speakeasy.net>
	<200708181803.45176.weather@speakeasy.net>
	<op.txbku8jixegpx6@jo.local>
Message-ID: <200708191657.32254.weather@speakeasy.net>

Hello -

>On Sunday 19 August 2007 14:42, Jonathan Buch wrote:
> Hi,
>
> > I may have spoken a little prematurely on this -- I am encountering a
> > problem. When I call Character#create, or when a Character object is
> > loaded (via it's corresponding Account being loaded), it's creating it's
> > own account. To illustrate, this is what's happening:
> >
> > a1 = Account.new
> > a1.character = Character.create self
> > a1.equal?(a1.character.account) #=> false
> >
> > The same thing is happening when loaded from the database, so the account
> > is pointing to the character, but the character is pointing to a
> > different account (an account with the same attributes, but a different
> > identity).
>
> Yes, a different 'identity', but not 'equality', depending on which method
> is called.  ;)
> Ruby knows .eql? .equal? .== .===
> .equal? is for testing if the object ids are the same, which they are not
> in this case.
>

Well, my issue is that I need my Account object to reference my Character 
object and my Character object to refer back to that same Account object. I 
assumed that the relationship would work like this:

load Account
load Account#character_oid
Account <-> Character

But, it's doing this, I think:
load Account
load Account#character
load Character#account
Account -> Character -> Account (different instance)

Is this how it works then, or did I perhaps implement something incorrectly? 
Since it's just a one_on_one relationship, I can always reference the Account 
myself in the Character class, but I was trying to be Ogish and have it load 
and attach itself. However, it's essential that the created objects reference 
eachother -- do I need to do the loading myself then 
(Character.find_by_account_oid(Account#oid), where Character#account_oid is a 
field created and set by me)?

> Og overrides ==() to react on the primary key of that model.
>

Oh, this probably explains some unexpected results from Array#delete(self) in 
my Character class -- I assumed it was Og-related, but good to know for sure 
and why.

Thanks again,
Matt

From weather at speakeasy.net  Sun Aug 19 17:24:17 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sun, 19 Aug 2007 17:24:17 -0400
Subject: [Nitro] Problem with Og post method
In-Reply-To: <200708191642.31292.weather@speakeasy.net>
References: <200708160224.56992.weather@speakeasy.net>
	<afd80b540708190158p5d084dal1b74ebd80d6e15df@mail.gmail.com>
	<200708191642.31292.weather@speakeasy.net>
Message-ID: <200708191724.17450.weather@speakeasy.net>

Hopefully this catches any repliers before responding to my previous post -- I 
got this working. I realized (from the example -- thanks again George) that I 
need to call Aspects.setup inside my program after Og.setup and it looks to 
be working fine.

Thanks again,
Matt

On Sunday 19 August 2007 16:42, Matthew B Gardner wrote:
> Hello -
>
> On Sunday 19 August 2007 04:58, George Moschovitis wrote:
> > Please,
> >
> > get the latest version (either from darcs or from the repository gems:
> > http://www.nitroproject.org/repo/gems
> >
> > 0.41.0 is about a year old and I cannot easily help you.
> >
> > thanks,
> > George.
>
> Sorry about that -- I've now upgraded to 0.50.0 for both glue and og.
>
> class Root
>
> 	property :name, VarChar(100)
>
>   	after :og_read, :call => :to_world
>
> 	def initialize name = "_error"
> 		@name   = name
> 		@dirty  = false
>     		@remove = false
> 	end
>
>   	def to_world
>     		p "--WE GOT HERE--"
>     		world.objects << self
>   	end
>
> 	...
> end
>
> Aspects.setup
> -end of file-
>
>
> Root#to_world is never getting called -- I tried placing the call in my
> Account < Root class, but that didn't work either.
>
> Thanks again for the help,
> Matt
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

From george.moschovitis at gmail.com  Sun Aug 19 17:40:18 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 20 Aug 2007 00:40:18 +0300
Subject: [Nitro] Problem with Og post method
In-Reply-To: <200708191724.17450.weather@speakeasy.net>
References: <200708160224.56992.weather@speakeasy.net>
	<afd80b540708190158p5d084dal1b74ebd80d6e15df@mail.gmail.com>
	<200708191642.31292.weather@speakeasy.net>
	<200708191724.17450.weather@speakeasy.net>
Message-ID: <afd80b540708191440p4a80dbb2y6938c1e381cb8bf0@mail.gmail.com>

>
> got this working. I realized (from the example -- thanks again George)
> that I
> need to call Aspects.setup inside my program after Og.setup and it looks
> to
> be working fine.
>

Great :)

Hopefully, we will find some time to work together  with Tom to finalize the
aspects design so that Aspects.setup is no longer needed.

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070820/89a74c42/attachment.html 

From george.moschovitis at gmail.com  Sun Aug 19 17:43:31 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 20 Aug 2007 00:43:31 +0300
Subject: [Nitro] OG vs Active Record
In-Reply-To: <op.txbixtaxxegpx6@jo.local>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local>
Message-ID: <afd80b540708191443k2a1c23d7y63e9fe39b877e2c0@mail.gmail.com>

>
> This certainly could be done and it has been asked for already on irc
> some year ago.  I wasn't too sure about that back then, and I still
> hesitate on that.  We could make it an option, `Og.auto_validations`
> which then does the trick....  G?
>

I don't understand what you mean here, can you elaborate?

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070820/0067fea6/attachment.html 

From weather at speakeasy.net  Sun Aug 19 19:10:43 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Sun, 19 Aug 2007 19:10:43 -0400
Subject: [Nitro] Og create method
Message-ID: <200708191910.43128.weather@speakeasy.net>

I'm getting a database error when using Class#create -- this is an example of 
the error:

ERROR: DB error Duplicate entry '6' for key 1, [INSERT INTO ogcharacter 
(`oid`, `aliases`, `name`, `account_oid`) VALUES (6, '--- []\n\n', 'Tester', 
4)]

The code is:

self.character = Character.create msg.capitalize, self

This problem doesn't occur if I use:

self.character = Character.new msg.capitalize, self
self.character.save!

I'm guessing that the first code is trying to insert itself twice? Am I doing 
something wrong w/ #create?

Thanks for any help,
Matt

From weather at speakeasy.net  Mon Aug 20 01:00:37 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Mon, 20 Aug 2007 01:00:37 -0400
Subject: [Nitro] Og create method
In-Reply-To: <200708191910.43128.weather@speakeasy.net>
References: <200708191910.43128.weather@speakeasy.net>
Message-ID: <200708200100.37353.weather@speakeasy.net>

Hello -

On Sunday 19 August 2007 19:10, Matthew B Gardner wrote:
> I'm getting a database error when using Class#create -- this is an example
> of the error:
>
> ERROR: DB error Duplicate entry '6' for key 1, [INSERT INTO ogcharacter
> (`oid`, `aliases`, `name`, `account_oid`) VALUES (6, '--- []\n\n',
> 'Tester', 4)]
>
> The code is:
>
> self.character = Character.create msg.capitalize, self
>
> This problem doesn't occur if I use:
>
> self.character = Character.new msg.capitalize, self
> self.character.save!
>
> I'm guessing that the first code is trying to insert itself twice? Am I
> doing something wrong w/ #create?
>
> Thanks for any help,
> Matt

It seems to be that line calling #create, mixed with this initialization:

belongs_to :account, Account

class Character
	def initialize(name = nil, acc = nil)
		super(name)
		self.account = acc # this is what's producing the error
	end
end


Hopefully that helps.
-Matt

From george.moschovitis at gmail.com  Mon Aug 20 01:11:39 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 20 Aug 2007 08:11:39 +0300
Subject: [Nitro] Repository Gems
In-Reply-To: <op.tw8wrtaoxegpx6@jo.local>
References: <389c43e40708171043w7124fe1euc4a64231e7aa34f@mail.gmail.com>
	<op.tw8wrtaoxegpx6@jo.local>
Message-ID: <afd80b540708192211k36ef36fdj3a15cac8256e12be@mail.gmail.com>

>
> Btw, George, didn't we say we wanted to relase an 0.49 first?  I'm still
> for that, as to make some kinda 'release candidate' so we can make a
> bugfix release out of 0.50 and not have to support a buggy release
> until the next one.
>

I say lets release 0.50 as is and we can release a bugfix release as 0.51.0
What do you think?

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070820/203c2eab/attachment.html 

From john at oxyliquit.de  Mon Aug 20 06:52:08 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 20 Aug 2007 12:52:08 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708191657.32254.weather@speakeasy.net>
References: <200708181159.06393.weather@speakeasy.net>
	<200708181803.45176.weather@speakeasy.net>
	<op.txbku8jixegpx6@jo.local>
	<200708191657.32254.weather@speakeasy.net>
Message-ID: <op.txcs5kw6xegpx6@jo.local>

Hi,

> Well, my issue is that I need my Account object to reference my Character
> object and my Character object to refer back to that same Account object. I
> assumed that the relationship would work like this:
>load Account
> load Account#character_oid
> Account <-> Character
>But, it's doing this, I think:
> load Account
> load Account#character
> load Character#account
> Account -> Character -> Account (different instance)

>>> a1 = Account.new
>>> a1.character = Character.create self
>>> a1.equal?(a1.character.account) #=> false

This sums it up I guess, you call a1.character.account so it is
"Account[25] -> Character[17] -> Account[25]".  Where the first and last
are the 'same' object, but not identical.

The identify of objects is (imo) mostly irrelevant and hard to 'get to'
anyway.  When Og wanted to keep the identity for identical (meaning same
primary key) elements, it'd really have to jump through hoops.  It has
several other problems then too, like having to keep track of every single
loaded element (this 'cache' has to be cleaned, when?), making Og non-thread-
safe (you always work on the identical copy, if you change it in one thread
it gets changed for all), and the implementation would just be so complex and
would make Og even slower.

Og theoretically would be able to save the Account in the Character right
after saving (which I asked George about) which would then point to the
identical object as the Account.

> Is this how it works then, or did I perhaps implement something incorrectly?
> Since it's just a one_on_one relationship, I can always reference the Account
> myself in the Character class, but I was trying to be Ogish and have it load
> and attach itself. However, it's essential that the created objects reference
> eachother -- do I need to do the loading myself then
> (Character.find_by_account_oid(Account#oid), where Character#account_oid is a
> field created and set by me)?

Well, why would you want to do that by yourself, if Og does it already?  When
you load it yourself, the object_id you will get will differ in the exact
same way.

But, to answer your question:  yes, that's how it works.

a1 = Account.create
c1 = Character.create
a1.character = c1
a1.save
c1.blah # do other stuff

I do this, so if I need the same character in the same scope again, I just
make it to a local variable, that is imo nicer to read, makes shorter code
and even is more efficient.

> However, it's essential that the created objects reference eachother

They do, just not to the object with the same object_id, is that really
a problem?

>> Og overrides ==() to react on the primary key of that model.
>>
>Oh, this probably explains some unexpected results from Array#delete(self) in
> my Character class -- I assumed it was Og-related, but good to know for sure
> and why.

Glad it cleared some things up.  :)

Jo

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

From john at oxyliquit.de  Mon Aug 20 06:52:05 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 20 Aug 2007 12:52:05 +0200
Subject: [Nitro] OG vs Active Record
In-Reply-To: <afd80b540708191443k2a1c23d7y63e9fe39b877e2c0@mail.gmail.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local>
	<afd80b540708191443k2a1c23d7y63e9fe39b877e2c0@mail.gmail.com>
Message-ID: <op.txcrx5elxegpx6@jo.local>

On Sun, 19 Aug 2007 23:43:31 +0200, George Moschovitis <george.moschovitis at gmail.com> wrote:

>>
>> This certainly could be done and it has been asked for already on irc
>> some year ago.  I wasn't too sure about that back then, and I still
>> hesitate on that.  We could make it an option, `Og.auto_validations`
>> which then does the trick....  G?
>>
>
> I don't understand what you mean here, can you elaborate?

Og.auto_validations = true

class User
   attr_accessor :name, String, :uniq => true
end

This would then automatically create a `validate_unique :name`.
Similar with other annotations then, :primary_key would do that too..
It could just 'walk' through some filter in the enchanting,
which just reacts on certain annotations and autocreates validations..
Just an idea though.  :)

Jo

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

From john at oxyliquit.de  Mon Aug 20 06:52:07 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 20 Aug 2007 12:52:07 +0200
Subject: [Nitro] Repository Gems
In-Reply-To: <afd80b540708192211k36ef36fdj3a15cac8256e12be@mail.gmail.com>
References: <389c43e40708171043w7124fe1euc4a64231e7aa34f@mail.gmail.com>
	<op.tw8wrtaoxegpx6@jo.local>
	<afd80b540708192211k36ef36fdj3a15cac8256e12be@mail.gmail.com>
Message-ID: <op.txcry9thxegpx6@jo.local>

Hi,

> I say lets release 0.50 as is and we can release a bugfix release as 0.51.0
> What do you think?

what made you change your mind.  :)

Jo

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

From john at oxyliquit.de  Mon Aug 20 06:52:08 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 20 Aug 2007 12:52:08 +0200
Subject: [Nitro] Og create method
In-Reply-To: <200708200100.37353.weather@speakeasy.net>
References: <200708191910.43128.weather@speakeasy.net>
	<200708200100.37353.weather@speakeasy.net>
Message-ID: <op.txctnursxegpx6@jo.local>

Hi,

> It seems to be that line calling #create, mixed with this initialization:
>
> belongs_to :account, Account
>
> class Character
> 	def initialize(name = nil, acc = nil)
> 		super(name)
> 		self.account = acc # this is what's producing the error
> 	end
> end

ahhh yes, you actually can't do this exactly.  This is setting a relation.
Setting a relation means (DB wise) that there must already be objects present
too relate to (meaning the character must have a primary key already which
also will go into acc.character_oid.

Now, this is the point where I'm supposed to make a suggestion on how else
to do this.  ;)

     class Account
       has_one Character
     end

     class Character
       belongs_to Account

       def initilize(name)
         super(name)
       end
     end

     a1 = Account.new
     c1 = (a1.character = Character.new)

With the current Og autosaving scheme, this will create 2 objects refering
to each other and saved to the DB already.  (If not, then the current
official 'og/relation/refers_to.rb' differs from mine in that account).

Hope that helps,

Jo

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

From george.moschovitis at gmail.com  Mon Aug 20 07:10:29 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 20 Aug 2007 14:10:29 +0300
Subject: [Nitro] Repository Gems
In-Reply-To: <op.txcry9thxegpx6@jo.local>
References: <389c43e40708171043w7124fe1euc4a64231e7aa34f@mail.gmail.com>
	<op.tw8wrtaoxegpx6@jo.local>
	<afd80b540708192211k36ef36fdj3a15cac8256e12be@mail.gmail.com>
	<op.txcry9thxegpx6@jo.local>
Message-ID: <afd80b540708200410l241a5035q91921ac5eb6ea0a9@mail.gmail.com>

>
> what made you change your mind.  :)


0.50.0 better denotes the tons of new features (and changes) i think.
0.5xcould be a new series of development.

-g.

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.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070820/c9ecc720/attachment.html 

From weather at speakeasy.net  Mon Aug 20 08:59:40 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Mon, 20 Aug 2007 08:59:40 -0400
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <op.txcs5kw6xegpx6@jo.local>
References: <200708181159.06393.weather@speakeasy.net>
	<200708191657.32254.weather@speakeasy.net>
	<op.txcs5kw6xegpx6@jo.local>
Message-ID: <200708200859.40255.weather@speakeasy.net>

Hello -

>On Monday 20 August 2007 06:52, Jonathan Buch wrote:
> Hi,
>
> > Well, my issue is that I need my Account object to reference my Character
> > object and my Character object to refer back to that same Account object.
> > I assumed that the relationship would work like this:
> >load Account
> > load Account#character_oid
> > Account <-> Character
> >But, it's doing this, I think:
> > load Account
> > load Account#character
> > load Character#account
> > Account -> Character -> Account (different instance)
> >
> >>> a1 = Account.new
> >>> a1.character = Character.create self
> >>> a1.equal?(a1.character.account) #=> false
>
> This sums it up I guess, you call a1.character.account so it is
> "Account[25] -> Character[17] -> Account[25]".  Where the first and last
> are the 'same' object, but not identical.
>
> The identify of objects is (imo) mostly irrelevant and hard to 'get to'
> anyway.  When Og wanted to keep the identity for identical (meaning same
> primary key) elements, it'd really have to jump through hoops.  It has
> several other problems then too, like having to keep track of every single
> loaded element (this 'cache' has to be cleaned, when?), making Og
> non-thread- safe (you always work on the identical copy, if you change it
> in one thread it gets changed for all), and the implementation would just
> be so complex and would make Og even slower.
>
> Og theoretically would be able to save the Account in the Character right
> after saving (which I asked George about) which would then point to the
> identical object as the Account.
>
> > Is this how it works then, or did I perhaps implement something
> > incorrectly? Since it's just a one_on_one relationship, I can always
> > reference the Account myself in the Character class, but I was trying to
> > be Ogish and have it load and attach itself. However, it's essential that
> > the created objects reference eachother -- do I need to do the loading
> > myself then
> > (Character.find_by_account_oid(Account#oid), where Character#account_oid
> > is a field created and set by me)?
>
> Well, why would you want to do that by yourself, if Og does it already? 
> When you load it yourself, the object_id you will get will differ in the
> exact same way.
>
> But, to answer your question:  yes, that's how it works.
>
> a1 = Account.create
> c1 = Character.create
> a1.character = c1
> a1.save
> c1.blah # do other stuff
>
> I do this, so if I need the same character in the same scope again, I just
> make it to a local variable, that is imo nicer to read, makes shorter code
> and even is more efficient.
>
> > However, it's essential that the created objects reference eachother
>
> They do, just not to the object with the same object_id, is that really
> a problem?
>

Well, my project is a text-based game in Ruby (a MUD, if you're familiar) -- 
each connection has a <-> reference to an Account object. So, I need the 
Character object to reference the same Account instance that references it -- 
what's happening is that user input is being sent to the Character 
(connection->account->character), but output from the Character is being sent 
to an account with a non-existent connection (character->account->nil). I 
just got rid of the relationship and now keep track of the Account a 
Character belongs to using the method I wrote above, which seems to work 
fine.

> >> Og overrides ==() to react on the primary key of that model.
> >
> >Oh, this probably explains some unexpected results from Array#delete(self)
> > in my Character class -- I assumed it was Og-related, but good to know
> > for sure and why.
>
> Glad it cleared some things up.  :)
>
> Jo

Thanks so much for the help -- I do have one more question, when you get the 
time. I have a Node class that has_many comments (which belong_to a Node) -- 
when I load the Node table, it loads the nodes and loads and attaches the 
comments (which I still find really cool, btw) -- that's all that's happening 
there, correct? What I mean is, the comments aren't also loading their 
respective Nodes, as the Character was loading its Account, are they? Just 
wanted to double check that I don't have a bunch of extra Node instances 
running around.

Thanks again -- I really appreciate all the help you and George have offered.
-Matt

From weather at speakeasy.net  Mon Aug 20 09:23:32 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Mon, 20 Aug 2007 09:23:32 -0400
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708200859.40255.weather@speakeasy.net>
References: <200708181159.06393.weather@speakeasy.net>
	<op.txcs5kw6xegpx6@jo.local>
	<200708200859.40255.weather@speakeasy.net>
Message-ID: <200708200923.32484.weather@speakeasy.net>

> Thanks so much for the help -- I do have one more question, when you get
> the time. I have a Node class that has_many comments (which belong_to a
> Node) -- when I load the Node table, it loads the nodes and loads and
> attaches the comments (which I still find really cool, btw) -- that's all
> that's happening there, correct? What I mean is, the comments aren't also
> loading their respective Nodes, as the Character was loading its Account,
> are they? Just wanted to double check that I don't have a bunch of extra
> Node instances running around.

Just ran a test and it looks like they are running around (or maybe they're 
running around b/c the test included referencing comment->node). At any rate, 
it looks like relationships may not be a great idea for my project.

From rob at robmela.com  Mon Aug 20 13:15:56 2007
From: rob at robmela.com (Robert Mela)
Date: Mon, 20 Aug 2007 13:15:56 -0400
Subject: [Nitro] OG vs Active Record
In-Reply-To: <op.txbixtaxxegpx6@jo.local>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local>
Message-ID: <46C9CC4C.6070702@robmela.com>

I'm finding an inherent trap in ORMs.

Consider the case where a Country has Provice/Region/State, and each of 
those has Cities

Processing all cities in a country in the old days was a three-way join 
with control-break processing.  One query, one result set.

ORMs have some inherent traps which are only partially solved by preload.

In the simplest case a preload two-levels deep can have the same 
efficiency as a single query with control-break processing.  For example


countries.each do |c|
   c.provinces.each do |p|
       p.cities.each do |c|
      end
   end
end

Without using preload,  100 countries of 100 provincies of 100 cities 
would result in 1 fetch of 100 countries.  For each country, one fetch 
of 100 provinces.   That's 100 fetches.    For each province, one fetch 
of 100 cities.   We've now got 10,000 fetches.    That's 10,000 
round-trips to the database.

With preloads, of course, it's a single round trip.

Unfortunately, preloads don't work when there's a lot of application 
logic.   For example:

countries.each do |c|
    c.roundish_provinces_with_lots_of_lakes.each do |p|
          p.cities_with_more_than_ten_macdonalds.eac do |c|
                ....
         end
    end
end

There is no way to tie a preload to this.    If 
"roundish_provinces_with_lots_of_lakes" involves addtional joines, and 
so does "cities_with_more_than_ten_macdonalds" requires a lookup into a 
MacDonald's count table for each entry, you could be looking at 
thousands of round-trips to the database with no opportunity for 
precaching or preload.

If you want speed you need to be very careful about how you use ORMs 
with large datasets.

I recently found an ActiveRecord retrieval that made 8 thousand 
individual database calls totally 4 megabytes of generated SQL.   All 
calls were to the same, self-referential table.

The table only has 6,000 rows...

I wound up yanking acts_as_tree and a homegrown acts_as_hierarchy, 
fetching the entire table at once -- all 6,000 rows -- and building my 
own non-Object Oriented tree inside the Model.   Fetching 6,000 rows 
across a network *and* building the internal tree structure takes 3 
milliseconds.   Navigating the structure is microseconds.    Before that 
the operation took anywhere from 10 seconds to two minutes....

I've been thinking about that experience and realized that, as systems 
get large, it makes a lot of sense to have the models be as simple as 
possible -- doing little more than enforcing simple constraints -- and 
have most business and application logic at the controller level. 

If that's not possible, then there needs to be a ModelLogic layer... MVC 
becomes MMVC... which despite initial appearances does not parse as a 
Roman Numeral.




Jonathan Buch wrote:
> Hi,
>
>   
>> I'm relatively new to Ruby and am trying to decide between diving into
>> ActiveRecord and OG.... I'd appreciate any advice.
>> The useage case/scenario is:  Single 'amateur' developer, large
>> numbers of databases, large databases (10'sGB/table, 10's tables/DB),
>> multiple ruby scripts accessing the dbserver directly (no web
>> interface - at this stage.)
>>     
>
> Multi-database is not as thoroughly tested as single, but the number
> (and also the sizes of the tables) do not make Og slow.  It's just how
> much you need to transfer between Og/DB.
> On startup time, if you stay below a 100 tables, it'll still be quite
> ok on a halfway capable computer.  There are a few tweaks you can do,
> but doing them means your app and db design is 'finished'.
>
>   
>> I understand that this is a little like asking a parent if their child
>> is beautiful, but was hoping for some frank feedback on the following.
>>
>> 1) Am I right that OG largely overcomes the issues discussed in this
>> (long) AR thread (I summarize a key issue below):
>> http://rubyurl.com/xWb
>>     
>
> Yes, by using Og basically _all_ your information is in the model file.
> Even the DB layout is there (which is what differenciates Og from AR).
>
>   
>> and here:
>> http://aralbalkan.com/764
>>     
>
> Yep, this is exactly what Og does.  :)
>
>   
>> A summary of a key AR issue (for me) is:
>> <quote>
>> Whether its in migrations or in the DB schema, I have to specify all
>> the columns as well as their lengths and unique constraints to have
>> the database created.  Then I have to specify the validates_* rules
>> in the model to match.  Finally I have to keep both of them in sync
>> over time.
>> Why not do it all inside the model and have that set up all the
>> validates_* rules in a single step?
>> </quote>
>>     
>
> This certainly could be done and it has been asked for already on irc
> some year ago.  I wasn't too sure about that back then, and I still
> hesitate on that.  We could make it an option, `Og.auto_validations`
> which then does the trick....  G?
>
> Personally, I wouldn't use it largly because I tend to see some of the
> validations as useless and time consuming (`uniq` for example, which is
> again done on DB level, where it actually belongs to.
> Some of the validations like 'belongs_to' (meaning it shouldn't exist
> when the 'main' object disappears) can only be done via certain DBs
> (like psql) with foreign key constraints.....  There's just so much
> stuff the DB does better, that's what I think, anyway.  :P).
>
> Anyway, like I said, the idea isn't all too bad, I'm just not sure
> if it's worth it, as the models/validations don't get changed so
> often (I draw stuff on real paper before doing DB work :P).
>
>   
>> 2) Has anyone compared the performance of AR vs OG. The following
>> thread about AR's cpu useage, surprised me:
>> http://rubyurl.com/vFz
>>
>> I see the same performance on my system (script attached).  Has anyone
>> run this test under OG?
>>     
>
> I'd love if you could adapt (I haven't had a look at it yet) that script
> and run it on Og?
>
> I would think that the performance of AR and Og could be quite comparable.
> (Though, I could be horribly wrong, when I started with Nitro I thought
> we were slower than rails, which turned out quite a wrong assumption ;))
>
> But, I think you'll see the same 'behaviour' with Og and MySQL.  MySQL
> is just too old, too optimized and there's too much movement behind it
> for Og to really stress it.  At least that's my guess.  ;)
>
>   
>> 3) How well is MySQL supported (bearing in mind that OG is at 0.5), or
>> is most development done with Postgresql?  On MySQL, how about the
>> MyISAM and Memory Engine/tables?
>>     
>
> Og is at 0.50, so about 50 revisions (actually I think more like 35
> public ones..), not the version number '0.5'.
>
> MySQL is (and always has been) actually the 'premier' DB used by the head
> maintainer.   It's just always those weird other people like me who just
> use PostgreSQL.  ;)  (Which means, that I optimize Og for my psql usage,
> only run the specs with it etc, which means that sometimes the psql part
> is better supported than the mysql/sqlite one).
> In-memory SQLite is actually quite nicely supported, and I know one from
> IRC who used it to speed up Og quite a bit (synchronizing the mem with
> a real backend now and then).
> I know nothing about MySQL memory engines, so can't comment.
>
> Hope that helps,
>
> Jo
>
>   

-------------- next part --------------
A non-text attachment was scrubbed...
Name: rob.vcf
Type: text/x-vcard
Size: 116 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070820/2d608371/attachment.vcf 

From billk at cts.com  Mon Aug 20 14:16:15 2007
From: billk at cts.com (Bill Kelly)
Date: Mon, 20 Aug 2007 11:16:15 -0700
Subject: [Nitro] OG vs Active Record
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com><op.txbixtaxxegpx6@jo.local>
	<46C9CC4C.6070702@robmela.com>
Message-ID: <023001c7e356$3293d430$6442a8c0@musicbox>


From: "Robert Mela" <rob at robmela.com>
>
> If you want speed you need to be very careful about how you use ORMs 
> with large datasets.
> 
> I recently found an ActiveRecord retrieval that made 8 thousand 
> individual database calls totally 4 megabytes of generated SQL.   All 
> calls were to the same, self-referential table.

When I've run into issues like this with Og, I've just written a
custom SQL query.

This particular example doesn't exactly parallel your problem, but
it does show using refers_to to keep the data normalized; and shows
how to ask Og for particular table names and such needed to construct
the custom query:


class IPHost
  property :ip, String, :unique => true
  property :hostname, String
end

class Playername
  property :playername, String, :unique => true
end

class Servername
  property :servername, String, :unique => true
end

class SuicidesAllTime
  property :method, String      # ex: mg, cg, trap, phalanx, lava, squished, cratered, drowned
  property :date, Date
  property :count, Integer
  refers_to :servername, Servername
  refers_to :victim, Playername

  def self.top_suicides_list(victim_name_str, method_str, servername_str, date=Date.today, limit=10)
    date = get_insert_date(date)
    __total__ = STATS_TOTAL_NAME
    
    @playername_total ||= Playername.find_or_create_by_playername(__total__)
    @servername_total ||= Servername.find_or_create_by_servername(__total__)
    date_str = "#{date.year}-#{date.month}-#{date.mday}"

    if victim_name_str.nil?
      victim = nil
    elsif victim_name_str == __total__
      victim = @playername_total
    else
      victim = PlayerSeen.find_closest_playername(victim_name_str)
      return [] unless victim
    end
    
    if servername_str.nil?
      servername = nil
    elsif servername_str == __total__
      servername = @servername_total
    else
      servername = Servername.find_by_servername(servername_str)
      return [] unless servername
    end

    ptotal = @playername_total
    stotal = @servername_total

    sql =
      "SELECT P1.playername AS victim, " +
             "FRAG.method, SV.servername AS server, FRAG.count " +
        "FROM #{self.table} FRAG, #{Playername.table} P1, " +
             "#{Servername.table} SV " +
       "WHERE " +
         (victim ? "FRAG.victim_oid = #{victim.oid}" : "FRAG.victim_oid <> #{ptotal.oid}") +
        " AND " + (servername ? "FRAG.servername_oid = #{servername.oid}" : "FRAG.servername_oid <> #{stotal.oid}") +
        " AND " + (method_str ? "FRAG.method = '#{method_str}'" : "FRAG.method <> '#{__total__}'") +
        " AND FRAG.date = '#{date_str}' " +
         "AND P1.oid = FRAG.victim_oid " +
         "AND SV.oid = FRAG.servername_oid " +
         "ORDER BY FRAG.count DESC"
    sql << " LIMIT #{limit}" if limit
    res = ogstore.query(sql)
    rows = []
    res.each_row {|row,dummy| rows << row}
    rows
  end

  # ...

end


Dunno if this is useful to anyone; but so far it's been possible for 
me to use Og to do most of the work, and just write custom queries as
I need them.


Regards,

Bill



From transfire at gmail.com  Mon Aug 20 14:41:38 2007
From: transfire at gmail.com (Trans)
Date: Mon, 20 Aug 2007 18:41:38 -0000
Subject: [Nitro] OG vs Active Record
In-Reply-To: <46C9CC4C.6070702@robmela.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
Message-ID: <1187635298.554791.248090@w3g2000hsg.googlegroups.com>

Nice post Robert. Thanks for the interesting analysis.


From rob at robmela.com  Mon Aug 20 16:07:03 2007
From: rob at robmela.com (Robert Mela)
Date: Mon, 20 Aug 2007 16:07:03 -0400
Subject: [Nitro] OG vs Active Record
In-Reply-To: <023001c7e356$3293d430$6442a8c0@musicbox>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com><op.txbixtaxxegpx6@jo.local>	<46C9CC4C.6070702@robmela.com>
	<023001c7e356$3293d430$6442a8c0@musicbox>
Message-ID: <46C9F467.1020107@robmela.com>

That's a very good approach and I'm considering it for some situations 
if I can ever unravel the business logic that's stretched across several 
models.

In other cases I would perform per-class selects into hashes or arrays 
and loop through them at the controller level.   Separating the logic 
makes it a lot more transparent, since often times much of the logic 
only applies in a particular area, and there's little sense in 
generalizing it by embedding it in the model.

It's all OK, and it's a tough call as to where it should be done.

What I would love is an ORM that could cache records locally, perhaps 
indexed in memory on certain criteria, so that it would query the local 
cache based on criteria rather than generating round-trip SQL.

ActiveRecord has a :cacheable option somewhere, but it caches using the 
SQL query string as a key.   The cache is useless if you need to 
retrieve the same objects as part of a different query.

For tables of just a few thousand records a local, in-memory cache would 
be fantastic.  Nitro is much better suited to that than Rails because 
you only need one in-process cache can be read by many concurrent threads.

Bill Kelly wrote:
>
>   def self.top_suicides_list(victim_name_str, method_str, servername_str, date=Date.today, limit=10)
>     date = get_insert_date(date)
>     __total__ = STATS_TOTAL_NAME
>     
>     @playername_total ||= Playername.find_or_create_by_playername(__total__)
>     @servername_total ||= Servername.find_or_create_by_servername(__total__)
>     date_str = "#{date.year}-#{date.month}-#{date.mday}"
>
>     if victim_name_str.nil?
>       victim = nil
>     elsif victim_name_str == __total__
>       victim = @playername_total
>     else
>       victim = PlayerSeen.find_closest_playername(victim_name_str)
>       return [] unless victim
>     end
>     
>     if servername_str.nil?
>       servername = nil
>     elsif servername_str == __total__
>       servername = @servername_total
>     else
>       servername = Servername.find_by_servername(servername_str)
>       return [] unless servername
>     end
>
>     ptotal = @playername_total
>     stotal = @servername_total
>
>     sql =
>       "SELECT P1.playername AS victim, " +
>              "FRAG.method, SV.servername AS server, FRAG.count " +
>         "FROM #{self.table} FRAG, #{Playername.table} P1, " +
>              "#{Servername.table} SV " +
>        "WHERE " +
>          (victim ? "FRAG.victim_oid = #{victim.oid}" : "FRAG.victim_oid <> #{ptotal.oid}") +
>         " AND " + (servername ? "FRAG.servername_oid = #{servername.oid}" : "FRAG.servername_oid <> #{stotal.oid}") +
>         " AND " + (method_str ? "FRAG.method = '#{method_str}'" : "FRAG.method <> '#{__total__}'") +
>         " AND FRAG.date = '#{date_str}' " +
>          "AND P1.oid = FRAG.victim_oid " +
>          "AND SV.oid = FRAG.servername_oid " +
>          "ORDER BY FRAG.count DESC"
>     sql << " LIMIT #{limit}" if limit
>     res = ogstore.query(sql)
>     rows = []
>     res.each_row {|row,dummy| rows << row}
>     rows
>   end
>
>   

-------------- next part --------------
A non-text attachment was scrubbed...
Name: rob.vcf
Type: text/x-vcard
Size: 116 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070820/df47bd18/attachment.vcf 

From john at oxyliquit.de  Mon Aug 20 20:07:09 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 21 Aug 2007 02:07:09 +0200
Subject: [Nitro] OG vs Active Record
In-Reply-To: <46C9CC4C.6070702@robmela.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
Message-ID: <op.txdtph0mxegpx6@jo.local>

Hi,

> If you want speed you need to be very careful about how you use
> ORMs with large datasets.

true words.  :P

> I recently found an ActiveRecord retrieval that made 8 thousand
> individual database calls totally 4 megabytes of generated SQL.
> All calls were to the same, self-referential table.
>The table only has 6,000 rows...

Sounds like someone is trying to fetch a tree recursively.  ;)

> I've been thinking about that experience and realized that, as
> systems get large, it makes a lot of sense to have the models be
> as simple as possible -- doing little more than enforcing simple
> constraints -- and have most business and application logic at
> the controller level.
> If that's not possible, then there needs to be a ModelLogic
> layer... MVC becomes MMVC... which despite initial appearances
> does not parse as a Roman Numeral.

This is how I like to do that stuff, it comes more natural to me
than to put everything except cooking coffee into the models.
I like my models clean, with a modest amount of convenience
functions.
I for example cringe when I have to move controller stuff down
into the model for convenience' sake and HTML inside the models
make me want to cry for my mommie.

Acually (from long discussions) I'd crank that up to MMMVC, the
new M is for ModelPresentationLogic.  Templating is such a hard
problem.... </sidenote>

Btw, on the matter of 8 thousand queries, let me once more present
my (almost) infamous age old tip on oxyliquit:

http://www.oxyliquit.de/tip/31

End is, don't redo something the DB does better anyway.

Jo
PS:  ltree (a indexed tree implentation for psql) and Og work
      together beautifully (with a few enhancements).

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

From john at oxyliquit.de  Mon Aug 20 20:07:11 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 21 Aug 2007 02:07:11 +0200
Subject: [Nitro] OG vs Active Record
In-Reply-To: <023001c7e356$3293d430$6442a8c0@musicbox>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<023001c7e356$3293d430$6442a8c0@musicbox>
Message-ID: <op.txdtyddixegpx6@jo.local>

Hi,

> Dunno if this is useful to anyone; but so far it's been possible for
> me to use Og to do most of the work, and just write custom queries as
> I need them.

you're correct in that you need custom queries sometimes, but that is
quite horrible there within the model.  ;)

May I say that VIEWs are really useful?  Please, also consider my Tip
on http://www.oxyliquit.de/tip/31 , life can only go better.  :P

I agree on that it only 'moves' the problem, but...  I think it's
just a whole lot cleaner.

Jo

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

From john at oxyliquit.de  Mon Aug 20 20:07:11 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 21 Aug 2007 02:07:11 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708200923.32484.weather@speakeasy.net>
References: <200708181159.06393.weather@speakeasy.net>
	<op.txcs5kw6xegpx6@jo.local>
	<200708200859.40255.weather@speakeasy.net>
	<200708200923.32484.weather@speakeasy.net>
Message-ID: <op.txduk0wfxegpx6@jo.local>

Hi,

>> Thanks so much for the help -- I do have one more question, when you get
>> the time. I have a Node class that has_many comments (which belong_to a
>> Node) -- when I load the Node table, it loads the nodes and loads and
>> attaches the comments (which I still find really cool, btw) -- that's all
>> that's happening there, correct?

>> What I mean is, the comments aren't also
>> loading their respective Nodes, as the Character was loading its Account,
>> are they? Just wanted to double check that I don't have a bunch of extra
>> Node instances running around.

Yes, that's all there is to it.  When you don't call .node you won't have
extra node copies around.

> Just ran a test and it looks like they are running around (or maybe they're
> running around b/c the test included referencing comment->node). At any rate,
> it looks like relationships may not be a great idea for my project.

class Node
   has_many Comment
end
class Comment
   belongs_to Node
end

n = Node.new
n.comments << Comment.create
n.comments << Comment.create

c1 = n.comments.first
c2 = n.comments.last

n # original node
c1.node # node copy #2
c2.node # node copy #3

You now have 3 copies of the original node lying about.  Unecessarily.

Relationships _are_ a good idea, think about normalizing of databases.  :)

It's really just about being a little careful on where the objects come from,
use available local or instance variables to store object you are reusing,
don't do unnecessary methods calls (which .node is).

What Og can't do is update every relationship object and set the referral to
the parent object to the identical (object_id) object.

Hope that helps,

Jo

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

From weather at speakeasy.net  Mon Aug 20 20:21:48 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Mon, 20 Aug 2007 20:21:48 -0400
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708200923.32484.weather@speakeasy.net>
References: <200708181159.06393.weather@speakeasy.net>
	<200708200859.40255.weather@speakeasy.net>
	<200708200923.32484.weather@speakeasy.net>
Message-ID: <200708202021.48593.weather@speakeasy.net>

On Monday 20 August 2007 09:23, Matthew B Gardner wrote:
> > Thanks so much for the help -- I do have one more question, when you get
> > the time. I have a Node class that has_many comments (which belong_to a
> > Node) -- when I load the Node table, it loads the nodes and loads and
> > attaches the comments (which I still find really cool, btw) -- that's all
> > that's happening there, correct? What I mean is, the comments aren't also
> > loading their respective Nodes, as the Character was loading its Account,
> > are they? Just wanted to double check that I don't have a bunch of extra
> > Node instances running around.
>
> Just ran a test and it looks like they are running around (or maybe they're
> running around b/c the test included referencing comment->node). At any
> rate, it looks like relationships may not be a great idea for my project.

Just wanted to acknowledge that the comments were loading their respective 
nodes when I referenced Comment#node -- so I'll just avoid doing that (I was 
only doing it to test). This thread has definitely helped me understand 
better how Og works with my objects -- thanks to both George and Jonathan for 
the help and explanations.

From billk at cts.com  Mon Aug 20 23:12:28 2007
From: billk at cts.com (Bill Kelly)
Date: Mon, 20 Aug 2007 20:12:28 -0700
Subject: [Nitro] confused about app.dispatcher.root
References: <00fb01c7e038$d63f6de0$6442a8c0@musicbox>
	<afd80b540708180836w6f206e29w3b1f795faa2138e1@mail.gmail.com>
Message-ID: <033001c7e3a1$1b07e310$6442a8c0@musicbox>


From: George Moschovitis
> 
> > class StoreController < Nitro::Controller
> 
> no need to extend Nitro::Controller (in fact there is no Nitro::Controller
> class, it is Raw::Controller now)

Ah, OK thanks.


> > ...the error I'm getting is:
> > 
> > 
> > DEBUG: Rendering '/status_500'
> > ERROR: Error while handling StoreController#index(status_500)
> > ERROR: wrong number of arguments (1 for 0)
> 
> please show as the  error BEFORE 'Rendering /status_500'.

That was the first error printed in the logfile.  The error that
happened before, was actually printed after.  :)

Turned out my old model code was doing ogstore.transaction where
now the ogstore must be omitted.


Thanks,

Bill



From billk at cts.com  Tue Aug 21 00:17:28 2007
From: billk at cts.com (Bill Kelly)
Date: Mon, 20 Aug 2007 21:17:28 -0700
Subject: [Nitro] OG vs Active Record
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com><op.txbixtaxxegpx6@jo.local>
	<46C9CC4C.6070702@robmela.com><023001c7e356$3293d430$6442a8c0@musicbox>
	<op.txdtyddixegpx6@jo.local>
Message-ID: <033e01c7e3aa$301d71d0$6442a8c0@musicbox>


From: "Jonathan Buch" <john at oxyliquit.de>
> 
>> Dunno if this is useful to anyone; but so far it's been possible for
>> me to use Og to do most of the work, and just write custom queries as
>> I need them.
> 
> you're correct in that you need custom queries sometimes, but that is
> quite horrible there within the model.  ;)

If you thought that was ugly, you should have seen the "optimized"
queries.  <grin>


> May I say that VIEWs are really useful?  Please, also consider my Tip
> on http://www.oxyliquit.de/tip/31 , life can only go better.  :P
> 
> I agree on that it only 'moves' the problem, but...  I think it's
> just a whole lot cleaner.

Thanks for the tip.  I had considered using VIEWs at one point, but
it's nice to see an example.


Ultimately, despite my best attempts at optimizing my queries, I was
not to achieve enough performance, once the tables started getting
millions of rows.  (I studied postgresql's analysis; it was using
the indexes when it should have been.)

Many of my queries started out as being very simple statements in
Og... just a few lines of ruby.  Some of them grew to 50 or 100 lines
for the fastest versions I came up with.  I got about an order of
magnitude speedup over Og... but it still wasn't enough.

Grisly details: http://rubyurl.com/XsB


Regards,

Bill



From george.moschovitis at gmail.com  Tue Aug 21 01:34:14 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 21 Aug 2007 08:34:14 +0300
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708202021.48593.weather@speakeasy.net>
References: <200708181159.06393.weather@speakeasy.net>
	<200708200859.40255.weather@speakeasy.net>
	<200708200923.32484.weather@speakeasy.net>
	<200708202021.48593.weather@speakeasy.net>
Message-ID: <afd80b540708202234v4316c03bj4d33fdac6947c636@mail.gmail.com>

>
> This thread has definitely helped me understand
> better how Og works with my objects -- thanks to both George and Jonathan
> for
> the help and explanations.
>

np.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070821/bdd197ce/attachment.html 

From weather at speakeasy.net  Tue Aug 21 01:40:48 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Tue, 21 Aug 2007 01:40:48 -0400
Subject: [Nitro] Aspects after method
Message-ID: <200708210140.49252.weather@speakeasy.net>

Just a simple question -- I'm trying to get my class to call foo after both 
og_read and initialize and I tried the following:

after [:og_read, :initialize], :call => :foo

It's calling foo on og_read, but not on initialize. The class, at least for my 
purposes at the moment, is being created via og_create -- I tried replacing 
initialize with that and achieved the same results. Any suggestions?

Thanks,
Matt

From weather at speakeasy.net  Tue Aug 21 07:06:17 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Tue, 21 Aug 2007 07:06:17 -0400
Subject: [Nitro] update() conflict
Message-ID: <200708210706.17327.weather@speakeasy.net>

One of my classes uses a Publisher module that defines an update() method, 
making it so I can't use the Og update(). I've tried defining an og_update 
method by Og::SqlStore.update() and some others, but that's not working. I'm 
coming from C, so I'm not entirely sure if that's the correct syntax, and I 
can't tell from the Og docs which class the update() method available belongs 
to. Is it something this easy and I just need a different syntax, or do I 
need to utilize a more complicated solution?

Thanks for any help,
Matt

From george.moschovitis at gmail.com  Tue Aug 21 08:20:22 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 21 Aug 2007 15:20:22 +0300
Subject: [Nitro] Aspects after method
In-Reply-To: <200708210140.49252.weather@speakeasy.net>
References: <200708210140.49252.weather@speakeasy.net>
Message-ID: <afd80b540708210520t123d5featdded3360a3ef441b@mail.gmail.com>

Let me investigate this and get back to you.

-g.

On 8/21/07, Matthew B Gardner <weather at speakeasy.net> wrote:
>
> Just a simple question -- I'm trying to get my class to call foo after
> both
> og_read and initialize and I tried the following:
>
> after [:og_read, :initialize], :call => :foo
>
> It's calling foo on og_read, but not on initialize. The class, at least
> for my
> purposes at the moment, is being created via og_create -- I tried
> replacing
> initialize with that and achieved the same results. Any suggestions?
>
> Thanks,
> Matt
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070821/c9e0a508/attachment.html 

From george.moschovitis at gmail.com  Tue Aug 21 08:21:42 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 21 Aug 2007 15:21:42 +0300
Subject: [Nitro] update() conflict
In-Reply-To: <200708210706.17327.weather@speakeasy.net>
References: <200708210706.17327.weather@speakeasy.net>
Message-ID: <afd80b540708210521s7c23e038x568353d53dbcc0b0@mail.gmail.com>

you have to define an alias in lib/og/model.rb

-g.

On 8/21/07, Matthew B Gardner <weather at speakeasy.net> wrote:
>
> One of my classes uses a Publisher module that defines an update() method,
> making it so I can't use the Og update(). I've tried defining an og_update
> method by Og::SqlStore.update() and some others, but that's not working.
> I'm
> coming from C, so I'm not entirely sure if that's the correct syntax, and
> I
> can't tell from the Og docs which class the update() method available
> belongs
> to. Is it something this easy and I just need a different syntax, or do I
> need to utilize a more complicated solution?
>
> Thanks for any help,
> Matt
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070821/9158ab0d/attachment.html 

From john at oxyliquit.de  Tue Aug 21 08:46:07 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 21 Aug 2007 14:46:07 +0200
Subject: [Nitro] The has_one/belongs_to relationship
In-Reply-To: <200708202021.48593.weather@speakeasy.net>
References: <200708181159.06393.weather@speakeasy.net>
	<200708200859.40255.weather@speakeasy.net>
	<200708200923.32484.weather@speakeasy.net>
	<200708202021.48593.weather@speakeasy.net>
Message-ID: <op.txetq5sqxegpx6@jo.local>

Hi,

> Just wanted to acknowledge that the comments were loading their respective
> nodes when I referenced Comment#node -- so I'll just avoid doing that (I was
> only doing it to test). This thread has definitely helped me understand
> better how Og works with my objects -- thanks to both George and Jonathan for
> the help and explanations.

you're very welcome.  :)   In fact, that really gave a boost to the ML-post-
count.   ^____^

So, also thank you.  :)

Jo

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

From rob at robmela.com  Tue Aug 21 22:12:56 2007
From: rob at robmela.com (Robert Mela)
Date: Tue, 21 Aug 2007 22:12:56 -0400
Subject: [Nitro] OG vs Active Record
In-Reply-To: <op.txdtph0mxegpx6@jo.local>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<op.txdtph0mxegpx6@jo.local>
Message-ID: <46CB9BA8.5030702@robmela.com>

Jonathan Buch wrote:
> Sounds like someone is trying to fetch a tree recursively.  ;)
>
>   

I hope whoever implements an acts_as_tree for Og has the hindsight and 
foresight to implement a cached tree structure option -- perhaps with 
update timestamps to allow simple, fast, partial cache refresh.




-------------- next part --------------
A non-text attachment was scrubbed...
Name: rob.vcf
Type: text/x-vcard
Size: 116 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070821/13251df3/attachment.vcf 

From george.moschovitis at gmail.com  Wed Aug 22 01:38:13 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 22 Aug 2007 08:38:13 +0300
Subject: [Nitro] OG vs Active Record
In-Reply-To: <46CB9BA8.5030702@robmela.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<op.txdtph0mxegpx6@jo.local> <46CB9BA8.5030702@robmela.com>
Message-ID: <afd80b540708212238o32c572afs304d196067cff96f@mail.gmail.com>

>
> I hope whoever implements an acts_as_tree for Og has the hindsight and
> foresight to implement a cached tree structure option -- perhaps with
> update timestamps to allow simple, fast, partial cache refresh.
>


I am attaching an Og Mixin I use for hierarchical queries in my projects. I
think
this greatly demonstrates the power of Og when encapsulating complex
patterns.
Consider this example when reviewing Og vs AvtiveRecord.

I think this generates fairly optimized queries and caching is not needed.

-g.



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070822/5000ff6d/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nested_intervals.rb
Type: application/x-ruby
Size: 4552 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070822/5000ff6d/attachment.bin 

From john at oxyliquit.de  Wed Aug 22 10:18:28 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Wed, 22 Aug 2007 16:18:28 +0200
Subject: [Nitro] OG vs Active Record
In-Reply-To: <afd80b540708212238o32c572afs304d196067cff96f@mail.gmail.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<op.txdtph0mxegpx6@jo.local> <46CB9BA8.5030702@robmela.com>
	<afd80b540708212238o32c572afs304d196067cff96f@mail.gmail.com>
Message-ID: <op.txgse6duxegpx6@jo.local>

Hi,

>> I hope whoever implements an acts_as_tree for Og has the hindsight and
>> foresight to implement a cached tree structure option -- perhaps with
>> update timestamps to allow simple, fast, partial cache refresh.

Og standard has a hierarchical (Nested Sets), which generates a tree out
of a model comparable to acts_as_tree (I think).

Why I dumped it:  complexity.

There is no way you can update such a structure (same with Nested Intervals)
by hand, moving children to a new parent is hell and inserting is 6 queries
(in the case of Og) instead of 1 and still can fail when using multithreading.

> I am attaching an Og Mixin I use for hierarchical queries in my projects. I
> think this greatly demonstrates the power of Og when encapsulating complex
> patterns.  Consider this example when reviewing Og vs AvtiveRecord.
>
> I think this generates fairly optimized queries and caching is not needed.

Same as Nested Sets, Nested Intervals are hell with inserting/updating.  They
additionally have a limitation on the depth/size of the tree.  While this is
ok for small sets of data, bigger ones it rules out.
Again, the pure complexity makes it very much impractical to do stuff
'by hand'.  Updating here is quicker than with NestedSets it seems, although
right now the code from G looks like it'd also use ~4 queries (only a quick
glance, haven't looked at actual queries).

Nontheless, both, and NestedIntervals in particular, are incredibly
interesting and very nice for many usecases.

On the topic of caching, caching is a hard problem....   I'm now more for
caching the end result and not a single query (which is Georges' stance as
well I think).

Jo

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

From mvyver at gmail.com  Wed Aug 22 21:16:36 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Thu, 23 Aug 2007 11:16:36 +1000
Subject: [Nitro] OG vs Active Record
In-Reply-To: <46C9CC4C.6070702@robmela.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
Message-ID: <389c43e40708221816t5821a92cwc34a6c554bbe0f44@mail.gmail.com>

Hi Robert,
Thanks for the additional insights.

On 8/21/07, Robert Mela <rob at robmela.com> wrote:
> I'm finding an inherent trap in ORMs.

I'm wondering if these types of issues are inherenit to ORM's or
specific to how sophisticated an ORM is with the SQL it constructs?
I'm not an SQL or Ruby guru, but if this is inherent to ORMs then it
makes me think that it can't be overcome by more sophisticated use of
SQL.

<big snip>

> If you want speed you need to be very careful about how you use ORMs
> with large datasets.

Thanks for the heads-up - this is closer to my use-case.

> I recently found an ActiveRecord retrieval that made 8 thousand
> individual database calls totally 4 megabytes of generated SQL.   All
> calls were to the same, self-referential table.
>
> The table only has 6,000 rows...
>
> I wound up yanking acts_as_tree and a homegrown acts_as_hierarchy,
> fetching the entire table at once -- all 6,000 rows -- and building my
> own non-Object Oriented tree inside the Model.   Fetching 6,000 rows
> across a network *and* building the internal tree structure takes 3
> milliseconds.   Navigating the structure is microseconds.    Before that
> the operation took anywhere from 10 seconds to two minutes....
>
> I've been thinking about that experience and realized that, as systems
> get large, it makes a lot of sense to have the models be as simple as
> possible -- doing little more than enforcing simple constraints -- and
> have most business and application logic at the controller level.

Thanks
Mark

> If that's not possible, then there needs to be a ModelLogic layer... MVC
> becomes MMVC... which despite initial appearances does not parse as a
> Roman Numeral.
>
>
>
>
> Jonathan Buch wrote:
> > Hi,
> >
> >
> >> I'm relatively new to Ruby and am trying to decide between diving into
> >> ActiveRecord and OG.... I'd appreciate any advice.
> >> The useage case/scenario is:  Single 'amateur' developer, large
> >> numbers of databases, large databases (10'sGB/table, 10's tables/DB),
> >> multiple ruby scripts accessing the dbserver directly (no web
> >> interface - at this stage.)
> >>
> >
> > Multi-database is not as thoroughly tested as single, but the number
> > (and also the sizes of the tables) do not make Og slow.  It's just how
> > much you need to transfer between Og/DB.
> > On startup time, if you stay below a 100 tables, it'll still be quite
> > ok on a halfway capable computer.  There are a few tweaks you can do,
> > but doing them means your app and db design is 'finished'.
> >
> >
> >> I understand that this is a little like asking a parent if their child
> >> is beautiful, but was hoping for some frank feedback on the following.
> >>
> >> 1) Am I right that OG largely overcomes the issues discussed in this
> >> (long) AR thread (I summarize a key issue below):
> >> http://rubyurl.com/xWb
> >>
> >
> > Yes, by using Og basically _all_ your information is in the model file.
> > Even the DB layout is there (which is what differenciates Og from AR).
> >
> >
> >> and here:
> >> http://aralbalkan.com/764
> >>
> >
> > Yep, this is exactly what Og does.  :)
> >
> >
> >> A summary of a key AR issue (for me) is:
> >> <quote>
> >> Whether its in migrations or in the DB schema, I have to specify all
> >> the columns as well as their lengths and unique constraints to have
> >> the database created.  Then I have to specify the validates_* rules
> >> in the model to match.  Finally I have to keep both of them in sync
> >> over time.
> >> Why not do it all inside the model and have that set up all the
> >> validates_* rules in a single step?
> >> </quote>
> >>
> >
> > This certainly could be done and it has been asked for already on irc
> > some year ago.  I wasn't too sure about that back then, and I still
> > hesitate on that.  We could make it an option, `Og.auto_validations`
> > which then does the trick....  G?
> >
> > Personally, I wouldn't use it largly because I tend to see some of the
> > validations as useless and time consuming (`uniq` for example, which is
> > again done on DB level, where it actually belongs to.
> > Some of the validations like 'belongs_to' (meaning it shouldn't exist
> > when the 'main' object disappears) can only be done via certain DBs
> > (like psql) with foreign key constraints.....  There's just so much
> > stuff the DB does better, that's what I think, anyway.  :P).
> >
> > Anyway, like I said, the idea isn't all too bad, I'm just not sure
> > if it's worth it, as the models/validations don't get changed so
> > often (I draw stuff on real paper before doing DB work :P).
> >
> >
> >> 2) Has anyone compared the performance of AR vs OG. The following
> >> thread about AR's cpu useage, surprised me:
> >> http://rubyurl.com/vFz
> >>
> >> I see the same performance on my system (script attached).  Has anyone
> >> run this test under OG?
> >>
> >
> > I'd love if you could adapt (I haven't had a look at it yet) that script
> > and run it on Og?
> >
> > I would think that the performance of AR and Og could be quite comparable.
> > (Though, I could be horribly wrong, when I started with Nitro I thought
> > we were slower than rails, which turned out quite a wrong assumption ;))
> >
> > But, I think you'll see the same 'behaviour' with Og and MySQL.  MySQL
> > is just too old, too optimized and there's too much movement behind it
> > for Og to really stress it.  At least that's my guess.  ;)
> >
> >
> >> 3) How well is MySQL supported (bearing in mind that OG is at 0.5), or
> >> is most development done with Postgresql?  On MySQL, how about the
> >> MyISAM and Memory Engine/tables?
> >>
> >
> > Og is at 0.50, so about 50 revisions (actually I think more like 35
> > public ones..), not the version number '0.5'.
> >
> > MySQL is (and always has been) actually the 'premier' DB used by the head
> > maintainer.   It's just always those weird other people like me who just
> > use PostgreSQL.  ;)  (Which means, that I optimize Og for my psql usage,
> > only run the specs with it etc, which means that sometimes the psql part
> > is better supported than the mysql/sqlite one).
> > In-memory SQLite is actually quite nicely supported, and I know one from
> > IRC who used it to speed up Og quite a bit (synchronizing the mem with
> > a real backend now and then).
> > I know nothing about MySQL memory engines, so can't comment.
> >
> > Hope that helps,
> >
> > Jo
> >
> >
>
>
>

From mvyver at gmail.com  Wed Aug 22 21:42:44 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Thu, 23 Aug 2007 11:42:44 +1000
Subject: [Nitro] OG vs Active Record
In-Reply-To: <op.txdtyddixegpx6@jo.local>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<023001c7e356$3293d430$6442a8c0@musicbox> <op.txdtyddixegpx6@jo.local>
Message-ID: <389c43e40708221842h3ba11f04yd2249d12e6e8563a@mail.gmail.com>

Hi Jo,

> > Dunno if this is useful to anyone; but so far it's been possible for
> > me to use Og to do most of the work, and just write custom queries as
> > I need them.

Is my understanding of the OG<->db relationship way off the mark to
observe that how much custom code is required will depend on how
sophisticated OG is in generating/processing SQL? This of course is a
function of how much development is done.
Or is there something about mapping objects to db's that means it'll
never be possible to (efficiently) do the more sophisticated things?

> you're correct in that you need custom queries sometimes, but that is
> quite horrible there within the model.  ;)
>
> May I say that VIEWs are really useful?  Please, also consider my Tip
> on http://www.oxyliquit.de/tip/31 , life can only go better.  :P

Thanks for pointing out your tip.
I've a question about defining an object that OG recognizes as a view
rather than a table, I'll post in another thread since it starts to
get off topic.

> I agree on that it only 'moves' the problem, but...  I think it's
> just a whole lot cleaner.

If OG could recognise a 'view' object then it'd all be back in the model?

Cheers
Mark

> 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
>

From rob at robmela.com  Thu Aug 23 11:49:27 2007
From: rob at robmela.com (Robert Mela)
Date: Thu, 23 Aug 2007 11:49:27 -0400
Subject: [Nitro] OG vs Active Record
In-Reply-To: <op.txgse6duxegpx6@jo.local>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>	<op.txbixtaxxegpx6@jo.local>
	<46C9CC4C.6070702@robmela.com>	<op.txdtph0mxegpx6@jo.local>
	<46CB9BA8.5030702@robmela.com>	<afd80b540708212238o32c572afs304d196067cff96f@mail.gmail.com>
	<op.txgse6duxegpx6@jo.local>
Message-ID: <46CDAC87.8020201@robmela.com>

I can appreciate the convinience of using adjacencies.   But there are 
as you suggest other criteria.

Where I work there's a tree of about 38,000 nodes changing a few times a 
day.  The amount of hand-editing that goes on is near nil, though the 
simplicity of adjacencies has been useful in debugging.

The simplicity of adjacencies is a much larger factor if your tree is 
small and there are good reasons to take the short-cut of manual edits 
or manual generation.

The broadest solution would be maintaining the adjacency list 
*alongside* the nested set.   Simply add a parent_id field to each 
node.  You can hand-insert a node, and run a script to have the system 
place it in the nested set.

Alternatively, you can have a utility script that regenerates the 
adjacency list after an insert.

Jonathan Buch wrote:
> Same as Nested Sets, Nested Intervals are hell with inserting/updating.  They
> additionally have a limitation on the depth/size of the tree.  While this is
> ok for small sets of data, bigger ones it rules out.
> Again, the pure complexity makes it very much impractical to do stuff
> 'by hand'.  Updating here is quicker than with NestedSets it seems, although
> right now the code from G looks like it'd also use ~4 queries (only a quick
> glance, haven't looked at actual queries).
>
> Nontheless, both, and NestedIntervals in particular, are incredibly
> interesting and very nice for many usecases.
>
> On the topic of caching, caching is a hard problem....   I'm now more for
> caching the end result and not a single query (which is Georges' stance as
> well I think).
>
> Jo
>
>   

-------------- next part --------------
A non-text attachment was scrubbed...
Name: rob.vcf
Type: text/x-vcard
Size: 116 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070823/5c88df7d/attachment.vcf 

From john at oxyliquit.de  Thu Aug 23 12:22:39 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Thu, 23 Aug 2007 18:22:39 +0200
Subject: [Nitro] OG vs Active Record
In-Reply-To: <389c43e40708221842h3ba11f04yd2249d12e6e8563a@mail.gmail.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<023001c7e356$3293d430$6442a8c0@musicbox>
	<op.txdtyddixegpx6@jo.local>
	<389c43e40708221842h3ba11f04yd2249d12e6e8563a@mail.gmail.com>
Message-ID: <op.txis31ujxegpx6@jo.local>

Hi,

> Is my understanding of the OG<->db relationship way off the mark to
> observe that how much custom code is required will depend on how
> sophisticated OG is in generating/processing SQL? This of course is a
> function of how much development is done.

I tend to see the 'write custom queries' as an optimization.  When I
need more speed, optimization is done.

> Or is there something about mapping objects to db's that means it'll
> never be possible to (efficiently) do the more sophisticated things?

Well, the ORM can't go around 'guessing' your intentions.  There must
be a clever way to tell the ORM what exactly you need.  Best is, for
one purpose only 1 query, not 250.

The way on how to 'specify your needs' varies of course.  I'm sure
someone very clever will come up with a way to create a 'auto vew
creator' using a special DSL and thus make it much more comfortable
to 'specify your needs'.

"I want this, then please join that table, add some information from
"that, aggregrate some information and finally do some calculations
"based on the results in a subquery."

For more sophisticated SQL there is just more information needed, and
where does that information come from if not from the programmer.

Anyway, that is my reason to write my VIEWs myself instead of letting
some machine do horribly hacked together queries.

>> you're correct in that you need custom queries sometimes, but that is
>> quite horrible there within the model.  ;)
>>
>> May I say that VIEWs are really useful?  Please, also consider my Tip
>> on http://www.oxyliquit.de/tip/31 , life can only go better.  :P
>
> Thanks for pointing out your tip.
> I've a question about defining an object that OG recognizes as a view
> rather than a table, I'll post in another thread since it starts to
> get off topic.

Og doesn't recognize views at all, it doesn't think that far atm.

Idea:

class MyUserView
   is Og::VIEW(MyUser)

   att_accessor :a, :b, String

   set_view <<-SQL
     SELECT foo.a,blah.b FROM foo, blah
   SQL
end

MyUserView.all

>> I agree on that it only 'moves' the problem, but...  I think it's
>> just a whole lot cleaner.
>
> If OG could recognise a 'view' object then it'd all be back in the model?

MyUserView.view_source # => MyUser


I have a slightly similar implementation here (which I shared with someone
on IRC already who implemented it more like my current idea) which I needed
for work.
It has a few glitches (and thus I won't push to Og main), but it's actually
very simple to implement, like 4 lines added.  It also autogenerates the
view itself (unlike my earlier tip where you have to do that yourself).

But, coupling a VIEW to another object is sometimes not that clever.  Views
might be completely detached, not having to do much with the original
purpose of the used models....  *hrm*

Sorry for the unhelpfulness of this post, it's really just ideas blabbered
out loud.

Jo

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

From mvyver at gmail.com  Thu Aug 23 19:42:57 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Fri, 24 Aug 2007 09:42:57 +1000
Subject: [Nitro] OG vs Active Record
In-Reply-To: <op.txis31ujxegpx6@jo.local>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<023001c7e356$3293d430$6442a8c0@musicbox> <op.txdtyddixegpx6@jo.local>
	<389c43e40708221842h3ba11f04yd2249d12e6e8563a@mail.gmail.com>
	<op.txis31ujxegpx6@jo.local>
Message-ID: <389c43e40708231642n162cbfc9k114630093b0ea81c@mail.gmail.com>

Hi,

> > Is my understanding of the OG<->db relationship way off the mark to
> > observe that how much custom code is required will depend on how
> > sophisticated OG is in generating/processing SQL? This of course is a
> > function of how much development is done.
>
> I tend to see the 'write custom queries' as an optimization.  When I
> need more speed, optimization is done.

Definitely.  I've only dealt with a one simple type of directed graph,
some recent reading makes me think that SQL can be mapped quite well
(but not always perfectly/optimally) to OO concepts, then I saw the
OG::NestedSets, which I don't understand but am satisfied that OGs
approach makes it a bandwagon worth jumping on :)

> > Or is there something about mapping objects to db's that means it'll
> > never be possible to (efficiently) do the more sophisticated things?
>
> Well, the ORM can't go around 'guessing' your intentions.  There must
> be a clever way to tell the ORM what exactly you need.  Best is, for
> one purpose only 1 query, not 250.

Sure, by adopting some conventions the chances that OG 'gets it right'
are better.  Correct? - I'm assuming Nitro/OG adopt the preference for
convention over configuration?

> The way on how to 'specify your needs' varies of course.  I'm sure
> someone very clever will come up with a way to create a 'auto vew
> creator' using a special DSL and thus make it much more comfortable
> to 'specify your needs'.
>
> "I want this, then please join that table, add some information from
> "that, aggregrate some information and finally do some calculations
> "based on the results in a subquery."
>
> For more sophisticated SQL there is just more information needed, and
> where does that information come from if not from the programmer.

I may be wrong but my current understanding is that OG could
eventually 'acquire' some of that information from the database/or OGs
managed objects and their inter-relations.  It wouldn't always be
optimal, but it could get quite sophisticated - right?

< snip >

> Sorry for the unhelpfulness of this post, it's really just ideas blabbered
> out loud.

No, this has been very helpful.  I'll think a bout views a little more.

Thanks
Mark

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

From george.moschovitis at gmail.com  Fri Aug 24 03:05:28 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Aug 2007 10:05:28 +0300
Subject: [Nitro] OG vs Active Record
In-Reply-To: <op.txis31ujxegpx6@jo.local>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<023001c7e356$3293d430$6442a8c0@musicbox> <op.txdtyddixegpx6@jo.local>
	<389c43e40708221842h3ba11f04yd2249d12e6e8563a@mail.gmail.com>
	<op.txis31ujxegpx6@jo.local>
Message-ID: <afd80b540708240005o462ede83h7807178855c5921b@mail.gmail.com>

>
> I have a slightly similar implementation here (which I shared with someone
> on IRC already who implemented it more like my current idea) which I
> needed
> for work.
> It has a few glitches (and thus I won't push to Og main), but it's
> actually
> very simple to implement, like 4 lines added.  It also autogenerates the
> view itself (unlike my earlier tip where you have to do that yourself).
>


I would like to see the code you are using.

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070824/e71b5ca7/attachment.html 

From george.moschovitis at gmail.com  Fri Aug 24 03:43:49 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Aug 2007 10:43:49 +0300
Subject: [Nitro] Nitro
Message-ID: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>

Out of couriosity...

is anyone using (or plans to use) Nitro? Or are you all just interested in
Og?

thanks,
-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070824/25ea265c/attachment-0001.html 

From arne at arnebrasseur.net  Fri Aug 24 04:07:59 2007
From: arne at arnebrasseur.net (arne at arnebrasseur.net)
Date: Fri, 24 Aug 2007 01:07:59 -0700 (PDT)
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <3387.208.97.187.133.1187942879.squirrel@webmail.arnebrasseur.net>

> Out of couriosity...
>
> is anyone using (or plans to use) Nitro? Or are you all just interested in
> Og?

I'm very interested in Nitro. I have at least one project boiling inside
my head that I want to work at the coming year.

(ab)

> thanks,
> -g.
>
> --
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From ravage at fragmentized.net  Fri Aug 24 06:27:44 2007
From: ravage at fragmentized.net (Rui Miguel)
Date: Fri, 24 Aug 2007 11:27:44 +0100
Subject: [Nitro] Nitro
In-Reply-To: <3387.208.97.187.133.1187942879.squirrel@webmail.arnebrasseur.net>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<3387.208.97.187.133.1187942879.squirrel@webmail.arnebrasseur.net>
Message-ID: <0A508BFC-DDC5-47CD-B0FE-701266E2C45C@fragmentized.net>

Also interested in Nitro and trying to use it on a project, more to  
come i'm sure as i get more confortable with it.
Intranet stuff mostly.

Regards,
Rui Miguel

On Aug 24, 2007, at Aug 24 - 09:07, arne at arnebrasseur.net wrote:

>> Out of couriosity...
>>
>> is anyone using (or plans to use) Nitro? Or are you all just  
>> interested in
>> Og?
>
> I'm very interested in Nitro. I have at least one project boiling  
> inside
> my head that I want to work at the coming year.
>
> (ab)
>
>> thanks,
>> -g.
>>
>> --
>> http://www.me.gr
>> http://phidz.com
>> http://blog.gmosx.com
>> http://cull.gr
>> http://www.joy.gr
>> http://nitroproject.org
>> _______________________________________________
>> 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

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070824/46b2fa21/attachment.bin 

From weather at speakeasy.net  Fri Aug 24 07:05:16 2007
From: weather at speakeasy.net (Matthew B Gardner)
Date: Fri, 24 Aug 2007 07:05:16 -0400
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <200708240705.16630.weather@speakeasy.net>

I've been pondering using it for the web-portion of my project, but I'm 
curious about two things mostly: how does it compare with Rails (if that's a 
reasonable question to address), and what's the state of the documentation -- 
just the rdocs?

-Matt

On Friday 24 August 2007 03:43, George Moschovitis wrote:
> Out of couriosity...
>
> is anyone using (or plans to use) Nitro? Or are you all just interested in
> Og?
>
> thanks,
> -g.

From george.moschovitis at gmail.com  Fri Aug 24 08:26:13 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Aug 2007 15:26:13 +0300
Subject: [Nitro] Og save
Message-ID: <afd80b540708240526q3df2f9e5gcf1586ae1abc0749@mail.gmail.com>

Dear devs,

Please note that the implementation of the og save method in the repository
version is changed from earlier versions. Ie it now uses Exceptions instead
of the return value to notify possible (verification) problems. The change
is quite old but I will add some comments in the code to make it more
visible (even I have forgotten about it ;-)).

regards,
-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070824/fa07cad6/attachment.html 

From billk at cts.com  Fri Aug 24 10:56:24 2007
From: billk at cts.com (Bill Kelly)
Date: Fri, 24 Aug 2007 07:56:24 -0700
Subject: [Nitro] Nitro
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <042901c7e65e$f1665bc0$6442a8c0@musicbox>

From: George Moschovitis
>
> Out of couriosity...
> 
> is anyone using (or plans to use) Nitro? Or are you all just interested in Og?

I'm using Nitro for an online store.

I also use Og separately from Nitro.

P.S. thank you for Nitro and Og.  :)


Regards,

Bill



From transfire at gmail.com  Fri Aug 24 11:29:21 2007
From: transfire at gmail.com (Trans)
Date: Fri, 24 Aug 2007 15:29:21 -0000
Subject: [Nitro] Aspects
Message-ID: <1187969361.601995.195500@q5g2000prf.googlegroups.com>

Hi--

I having a look at the aspects.rb code today. I understand there are
some issues with the code. I'd like to review it's functionality and
the issues involved with it.

Concerning basic design, the objective is of course to create pre,
post and around advice for methods. Now that's easy enough if methods
already exist, basically:

  alias _methodname, methodname

  def methodname
    ...
    _methodname
    ...
  end

The difficulty comes when the methods do not already exist. For
instance, if we want to encapsulate a set of advice for reuse --which
technically is what an aspect is. It is further complicated when we
consider that methods can be dynamically added to a class/module after
an aspect is applied. So those are the basic issues that make general
implementation quite difficult.

I've been thinking of implementing this on top of cut.rb. Cuts already
basically handles this stuff, so it should be easy to build aspects.rb
with it.  However I'm not confident in the overall robustness of
cut.rb. Among other things it overrides Module#included and
Class#method_added, so I expect corner cases to arise where these
methods have been overridden elsewhere. I'm sure it could be improved,
but it's not easy --it's complex meta-programming. So I've been
thinking about how to build a better implementation of cut.rb using
delegation instead of dynamic methods. That implementation would
easily be more robust, but in this case, I'm not sure about
performance characteristics since it would be heavily reliant on
method_missing. (Basically imagine every method call being routed via
method_missing).

OTOH. maybe a whole different approach is in order.

Thoughts?

T.


From transfire at gmail.com  Fri Aug 24 11:30:16 2007
From: transfire at gmail.com (Trans)
Date: Fri, 24 Aug 2007 15:30:16 -0000
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <1187969416.113329.140600@z24g2000prh.googlegroups.com>



On Aug 24, 12:43 am, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> Out of couriosity...
>
> is anyone using (or plans to use) Nitro? Or are you all just interested in
> Og?

What spurred that question?

T.


From Reid.Thompson at ateb.com  Fri Aug 24 13:38:20 2007
From: Reid.Thompson at ateb.com (Reid Thompson)
Date: Fri, 24 Aug 2007 13:38:20 -0400
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <1187977103.6443.7.camel@jhereg>


On Fri, 2007-08-24 at 10:43 +0300, George Moschovitis wrote:
> Out of couriosity...
> 
> is anyone using (or plans to use) Nitro? Or are you all just
> interested in Og?
> 
> thanks,
> -g.

I'd like to use nitro -- I've just had a hard time getting started due
to lack of time and documentation/tutorials ( lack of time could be
offset by docs/tutorials )

From john at oxyliquit.de  Fri Aug 24 14:00:39 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Fri, 24 Aug 2007 20:00:39 +0200
Subject: [Nitro] OG vs Active Record
In-Reply-To: <afd80b540708240005o462ede83h7807178855c5921b@mail.gmail.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<023001c7e356$3293d430$6442a8c0@musicbox>
	<op.txdtyddixegpx6@jo.local>
	<389c43e40708221842h3ba11f04yd2249d12e6e8563a@mail.gmail.com>
	<op.txis31ujxegpx6@jo.local>
	<afd80b540708240005o462ede83h7807178855c5921b@mail.gmail.com>
Message-ID: <op.txj54xpoxegpx6@jo.local>

Hi,

> I would like to see the code you are using.

the meat of all it is:

sql.rb:

   def og_create_schema(store)
     return unless Og.create_schema

     if self.class.const_defined?(:OG_SCHEMA)
       store.create_custom_schema(self.class.const_get(:OG_SCHEMA))
     else
       unless self.class.schema_inheritance_child?
         store.create_table(self.class)
         store.create_table_indices(self.class)
         store.create_table_joins(self.class)
       end
       store.evolve_schema(self.class)
     end
   end

This can already be used like this:

class MyUserView
   attr_accessor :a, String

   OG_SCHEMA = <<-SQL
     CREATE VIEW ogmyuserview AS SELECT a FROM foo
   SQL
end

Another bit is quite necessary though:

manager.rb:

   def manage_classes(*args)
     # ...
     classes = classes.sort_by {|x| x.const_defined?(:OG_SCHEMA) ? 1 : 0 }
     classes.each { |c| manage(c) }
   end

Which moves the enchanting of 'custom' classes to the back
because they might depend on other tables which have to be
created/evolved first.  (might fail due to interdependencies
between views, this is one of the 'glitches')

That's basically it, my very own auto-view-creator by adding
5 lines.  :P

Jo

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

From john at oxyliquit.de  Fri Aug 24 14:00:39 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Fri, 24 Aug 2007 20:00:39 +0200
Subject: [Nitro] OG vs Active Record
In-Reply-To: <46CDAC87.8020201@robmela.com>
References: <389c43e40708161949u2781f25x11d916169056f82f@mail.gmail.com>
	<op.txbixtaxxegpx6@jo.local> <46C9CC4C.6070702@robmela.com>
	<op.txdtph0mxegpx6@jo.local> <46CB9BA8.5030702@robmela.com>
	<afd80b540708212238o32c572afs304d196067cff96f@mail.gmail.com>
	<op.txgse6duxegpx6@jo.local> <46CDAC87.8020201@robmela.com>
Message-ID: <op.txj5p0ulxegpx6@jo.local>

Hi,

> I can appreciate the convinience of using adjacencies.   But
> there are as you suggest other criteria.

I never said I'd be using adjacency lists.  ;)

Right now I'm more on the track on materialized path, which
(with help from postgres) has really good performance for all
operations I need and at the same time is conceptually easy.

> Where I work there's a tree of about 38,000 nodes changing a
> few times a day.  The amount of hand-editing that goes on is
> near nil, though the simplicity of adjacencies has been useful
> in debugging.

Yeah, debugging, good example.  Although an extensive test suite
covering all that might be enough of a safety net for relations
and can provide good enough information when stuff goes wrong...

> The simplicity of adjacencies is a much larger factor if your
> tree is small and there are good reasons to take the short-cut
> of manual edits or manual generation.
>The broadest solution would be maintaining the adjacency list
> *alongside* the nested set.   Simply add a parent_id field to
> each node.  You can hand-insert a node, and run a script to
> have the system place it in the nested set.

I think the NestedSets implementation within Og has the parent
saved too.  That could be a starting place for someone who wants
to investigate on that further.

> Alternatively, you can have a utility script that regenerates
> the adjacency list after an insert.

I heard some professionals use that two layer approach.  They
insert/update only the adjacency list, which is quite quick for
some operations.  Then they just regenerate a nested-sets table
with the new information for fast selects.

Too bad I don't have too much experience yet with real life
high-performance stuff with huge databases.  Thanks Robert for
also replying here.  :)

Jo

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

From john at oxyliquit.de  Fri Aug 24 14:00:40 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Fri, 24 Aug 2007 20:00:40 +0200
Subject: [Nitro] Nitro
In-Reply-To: <200708240705.16630.weather@speakeasy.net>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<200708240705.16630.weather@speakeasy.net>
Message-ID: <op.txkmpso1xegpx6@jo.local>

Hi,

> I've been pondering using it for the web-portion of my project, but I'm
> curious about two things mostly: how does it compare with Rails (if that's a
> reasonable question to address), and what's the state of the documentation --
> just the rdocs?

the state of documentation is... just the rdocs I guess.  :)   Well, that is
if you don't count oxyliquit, which is only ever as good as people want it
to be.  :P

But, there have been benchmarks showing a clear speed advantage of Nitro
over Rails.   Kirk Haines for example did one, classifying us somewhere
in the middle of IOWA and Rails (where IOWA of course shines, Kirk does a
lot of things for speed and memory usage there which Nitro doesn't).

Jo

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

From john at oxyliquit.de  Fri Aug 24 14:00:40 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Fri, 24 Aug 2007 20:00:40 +0200
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <op.txkr9gwhxegpx6@jo.local>

Hi,

> Out of couriosity...
>
> is anyone using (or plans to use) Nitro? Or are you all just interested in
> Og?

tough question.  I guess the question comes from the recent discussions
evolving around Og?  I love that the Nitro ML is picking up a little steam
right now.  And any exposure of Og makes Nitro more appealing.

As for me, some of the recent features I don't really use (REST) and I
moved (also due to my recent internship) to something else which happens
to fit my mind slightly better.  It 'evolved' and is quite close to
the nitro way and smells as good as Nitro.

What can I say.  :)

Jo

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

From john at oxyliquit.de  Fri Aug 24 14:00:40 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Fri, 24 Aug 2007 20:00:40 +0200
Subject: [Nitro] Og save
In-Reply-To: <afd80b540708240526q3df2f9e5gcf1586ae1abc0749@mail.gmail.com>
References: <afd80b540708240526q3df2f9e5gcf1586ae1abc0749@mail.gmail.com>
Message-ID: <op.txkm0vjyxegpx6@jo.local>

Hi,

> Please note that the implementation of the og save method in the repository
> version is changed from earlier versions. Ie it now uses Exceptions instead
> of the return value to notify possible (verification) problems. The change
> is quite old but I will add some comments in the code to make it more
> visible (even I have forgotten about it ;-)).

I'm not really that fond of that change, as I see the occurance of a validation
error as not something exceptional but as an 'expected' route and thus see
a exception as overkill.

begin
   obj.save
rescue Og::ValidationError
   # do stuff
end

if !obj.save
   # do stuff
end

Of course I also read the discussion where that change stems from, and I
agree that it might confuse some people who do `obj.save` and nothing
happens in the db....

Just my piece of mind,

Jo

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

From george.moschovitis at gmail.com  Fri Aug 24 14:04:32 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Aug 2007 21:04:32 +0300
Subject: [Nitro] Nitro
In-Reply-To: <1187969416.113329.140600@z24g2000prh.googlegroups.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<1187969416.113329.140600@z24g2000prh.googlegroups.com>
Message-ID: <afd80b540708241104mbe1b304mf168f11b5217a0ca@mail.gmail.com>

>
> What spurred that question?


I will use Nitro as the underlying platform for some important projects. As
it is hard to find good Ruby developers in Greece (and almost impossible to
find Nitro developers ;-)) I was wondering if anyone is using (or is
planning to use) Nitro professionally.  I may even have some small
outsourcing jobs at a later date. So it would be nice to know that there
*are* some people that will be actively working with Nitro this year (I most
certainly will be much more active!)

regards,
George.




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



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070824/ea45430c/attachment.html 

From george.moschovitis at gmail.com  Fri Aug 24 14:06:46 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Aug 2007 21:06:46 +0300
Subject: [Nitro] Nitro
In-Reply-To: <op.txkr9gwhxegpx6@jo.local>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<op.txkr9gwhxegpx6@jo.local>
Message-ID: <afd80b540708241106x53487fccy9c9b17e3a3231391@mail.gmail.com>

>
> the nitro way and smells as good as Nitro.
>

But this smells to me like the situation with Compiz + Beryl: unnecessary
duplication of efforts.

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070824/b7ff4126/attachment.html 

From rob at robmela.com  Fri Aug 24 16:41:40 2007
From: rob at robmela.com (Robert Mela)
Date: Fri, 24 Aug 2007 16:41:40 -0400
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <46CF4284.5040808@robmela.com>

Nitro.

It's just that databases are a pain in the neck and require more 
discussion ;)

George Moschovitis wrote:
> Out of couriosity...
>
> is anyone using (or plans to use) Nitro? Or are you all just 
> interested in Og?
>
> thanks,
> -g.
>
> -- 
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org <http://nitroproject.org>
> ------------------------------------------------------------------------
>
> _______________________________________________
> 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: rob.vcf
Type: text/x-vcard
Size: 116 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070824/21a1fc05/attachment.vcf 

From rob at robmela.com  Fri Aug 24 16:45:44 2007
From: rob at robmela.com (Robert Mela)
Date: Fri, 24 Aug 2007 16:45:44 -0400
Subject: [Nitro] Nitro
In-Reply-To: <3387.208.97.187.133.1187942879.squirrel@webmail.arnebrasseur.net>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<3387.208.97.187.133.1187942879.squirrel@webmail.arnebrasseur.net>
Message-ID: <46CF4378.9020902@robmela.com>

At this point I would not choose Rails for a new project.  I would only 
choose Nitro.

I too have several projects, at least one of which will be 
implemented.   Nitro will be the framework.

I'm also willing to evangelize.  I live in Boston and there's a decent 
sized community using Rails.     I'd like to give a presentation and 
demo of Nitro to a couple of local Ruby users groups, probably in early 
November, once one of my projects has some traction.

arne at arnebrasseur.net wrote:
>> Out of couriosity...
>>
>> is anyone using (or plans to use) Nitro? Or are you all just interested in
>> Og?
>>     
>
> I'm very interested in Nitro. I have at least one project boiling inside
> my head that I want to work at the coming year.
>
> (ab)
>
>   
>> thanks,
>> -g.
>>
>> --
>> http://www.me.gr
>> http://phidz.com
>> http://blog.gmosx.com
>> http://cull.gr
>> http://www.joy.gr
>> http://nitroproject.org
>> _______________________________________________
>> 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
>
>
>   

-------------- next part --------------
A non-text attachment was scrubbed...
Name: rob.vcf
Type: text/x-vcard
Size: 116 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070824/be3a18e4/attachment.vcf 

From rob at robmela.com  Fri Aug 24 17:01:40 2007
From: rob at robmela.com (Robert Mela)
Date: Fri, 24 Aug 2007 17:01:40 -0400
Subject: [Nitro] Nitro on rubyforge
Message-ID: <46CF4734.2080307@robmela.com>

1) Sorry if I missed this -- but are the 0.50 gems ready?   They're not 
on rubyforge

2) Seems the developer list on rubyforge ( 
http://rubyforge.org/project/memberlist.php?group_id=418 ) is out of 
date.  One of the things I look at when considering technologies for a 
project is the size and viability of the development community around 
it.  I'm sure many people include that criterium in their decision making.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: rob.vcf
Type: text/x-vcard
Size: 116 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070824/5d86485c/attachment.vcf 

From jan.provaznik at solnet.cz  Fri Aug 24 17:25:44 2007
From: jan.provaznik at solnet.cz (=?UTF-8?Q?Jan_Provazn=C3=ADk?=)
Date: Fri, 24 Aug 2007 23:25:44 +0200
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <jnariw.z1tzkr.bin8vk80w5kmtewkvw7ajotno.rubymail@solnet.cz>


George Moschovitis:
> Out of couriosity...
>
> is anyone using (or plans to use) Nitro? Or are you all just interested in
> Og?
>
> thanks,
> -g.
>


Yes. We are moving our product (WebIS - enterprise groupware, currently 
running on +-150 servers, tens-hundreds users per server) on Nitro 
framework.

Currently, I'm (quite successfully) testing WebIS with Nitro pre0.50, 
but for the present it doesn't use much of Nitro power yet (because of 
compatibility with older WebIS version).

I will send some small patches soon - pre0.50 is currently hardcoded to 
use CookieSession, and also some of session classes (e.g. 
raw/context/session/file.rb) are not prepared for 0.50, but that are 
oddments.

Regards, Jenik.


From george.moschovitis at gmail.com  Sat Aug 25 02:36:17 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 25 Aug 2007 09:36:17 +0300
Subject: [Nitro] Nitro on rubyforge
In-Reply-To: <46CF4734.2080307@robmela.com>
References: <46CF4734.2080307@robmela.com>
Message-ID: <afd80b540708242336n2bef11c9ke66cc96aa8cc36cd@mail.gmail.com>

On 8/25/07, Robert Mela <rob at robmela.com> wrote:
>
> 1) Sorry if I missed this -- but are the 0.50 gems ready?   They're not
> on rubyforge


nope, you can find experimental gems here:

http://www.nitroproject.org/repo/gems

I will automate the process of generating there glycerin gems.


2) Seems the developer list on rubyforge (
> http://rubyforge.org/project/memberlist.php?group_id=418 ) is out of
> date.  One of the things I look at when considering technologies for a
> project is the size and viability of the development community around
> it.  I'm sure many people include that criterium in their decision making.


we use this mailing list not the developer list. We should probably make
this clearer though.

-g.




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


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070825/5e3d5e0f/attachment-0001.html 

From george.moschovitis at gmail.com  Sat Aug 25 02:38:00 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 25 Aug 2007 09:38:00 +0300
Subject: [Nitro] Nitro
In-Reply-To: <jnariw.z1tzkr.bin8vk80w5kmtewkvw7ajotno.rubymail@solnet.cz>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<jnariw.z1tzkr.bin8vk80w5kmtewkvw7ajotno.rubymail@solnet.cz>
Message-ID: <afd80b540708242338u208ccb65ibd0d2ffd033d7037@mail.gmail.com>

>
> I will send some small patches soon - pre0.50 is currently hardcoded to
> use CookieSession, and also some of session classes (e.g.
> raw/context/session/file.rb) are not prepared for 0.50, but that are
> oddments.


Excellent, I will love to get some patches or even
comments/remarks/suggestions about Nitro too, not just Og.

thank you,
George.



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



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070825/fb435f34/attachment.html 

From george.moschovitis at gmail.com  Sat Aug 25 02:39:18 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 25 Aug 2007 09:39:18 +0300
Subject: [Nitro] Nitro
In-Reply-To: <46CF4378.9020902@robmela.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<3387.208.97.187.133.1187942879.squirrel@webmail.arnebrasseur.net>
	<46CF4378.9020902@robmela.com>
Message-ID: <afd80b540708242339n5055e70bxb1a7efead12b8316@mail.gmail.com>

>
> I'm also willing to evangelize.  I live in Boston and there's a decent
> sized community using Rails.     I'd like to give a presentation and
> demo of Nitro to a couple of local Ruby users groups, probably in early
> November, once one of my projects has some traction.


That sounds great. Perhaps you could add some Nitro related posts to your
blog or something in the mean time.
Let's make Nitro more widely known!

-g.


http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070825/beed31c0/attachment.html 

From dan at tastapod.com  Sat Aug 25 07:38:09 2007
From: dan at tastapod.com (Dan North)
Date: Sat, 25 Aug 2007 12:38:09 +0100
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <46D014A1.1020808@tastapod.com>

I've said this before and I still stand by it: I /get/ nitro. It works 
the way I want to work. Rails just doesn't. I want to start small and 
grow into an application. For instance, I don't want to even think about 
the database until I have a feature that requires persistence. I can't 
even /start/ a rails app without a fully-configured mysql or sqlite data 
store. That's just dumb.

The thing that really won me over was the little movies on the original 
nitro site, where you start with a single class that is the entire app 
and break out the model and view as you need to (and not before). Then 
the view code got cleaner, and cleaner, until there wasn't even any ruby 
in it. Just some very intuitive html-ish tags. Lovely!

I'm not writing ruby web apps in anger right now (I'm stuck in 
investment banking Java land) but I can guarantee that when I do, I'll 
be using nitro to do it.

Keep up the great work George,
Dan

George Moschovitis wrote:
> Out of couriosity...
>
> is anyone using (or plans to use) Nitro? Or are you all just 
> interested in Og?
>
> thanks,
> -g.
>
> -- 
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org <http://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/20070825/66b8f294/attachment.html 

From george.moschovitis at gmail.com  Sat Aug 25 14:01:16 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 25 Aug 2007 21:01:16 +0300
Subject: [Nitro] Aspects
In-Reply-To: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
Message-ID: <afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>

>
> The difficulty comes when the methods do not already exist.


yeap, this is the problem. My version (included in the repository code)
handles this
my using a mandatory call to Aspects.setup after all dynamic methods are
defined.
But I think there are still some bugs (for example on bug reported earlier
in this week).

I tried to use the version from facets but I had some problems. Tomorrow I
will review
my version and the code from facets, and I will be in better position to
help you with this.


> method_missing. (Basically imagine every method call being routed via
> method_missing).


this sounds *awfull* to me!

Thoughts?


more thoughts (hopefully) tomorrow.

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070825/20b00749/attachment.html 

From george.moschovitis at gmail.com  Sat Aug 25 14:03:55 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 25 Aug 2007 21:03:55 +0300
Subject: [Nitro] Nitro
In-Reply-To: <46D014A1.1020808@tastapod.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<46D014A1.1020808@tastapod.com>
Message-ID: <afd80b540708251103o4ae8cbfcrcec6a177c354c387@mail.gmail.com>

>
> I've said this before and I still stand by it: I *get* nitro. It works the
> way I want to work. Rails just doesn't. I want to start small and grow into
> an ..Keep up the great work George,
> Dan
>

Thank you, this is very encouraging. I am kind of putting my ass in the line
by using Nitro on such important projects and it is nice to hear that other
people are planning to use it and (hopefully) help to make it as robust as
possible.

thanks,
George.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070825/2e11c215/attachment.html 

From transfire at gmail.com  Sat Aug 25 15:55:54 2007
From: transfire at gmail.com (Trans)
Date: Sat, 25 Aug 2007 19:55:54 -0000
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
Message-ID: <1188071754.562666.319290@r34g2000hsd.googlegroups.com>



On Aug 25, 11:01 am, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> > The difficulty comes when the methods do not already exist.
>
> yeap, this is the problem. My version (included in the repository code)
> handles this
> my using a mandatory call to Aspects.setup after all dynamic methods are
> defined.
> But I think there are still some bugs (for example on bug reported earlier
> in this week).

You're using this for Og models, right? Are you using it for Nitro
controllers too? Anywhere else?

I don't 100% understand the design btw. Could you explain
what :on, :where, :only, :except, and :join options do?

> I tried to use the version from facets but I had some problems. Tomorrow I
> will review
> my version and the code from facets, and I will be in better position to
> help you with this.
>
> > method_missing. (Basically imagine every method call being routed via
> > method_missing).
>
> this sounds *awfull* to me!

He he. Well, it's not *that* bad! The nice things is that it's 100%
dynamic. No fuss with included or method_added, which can be easy to
break.

In any case, for the time being it's probably best just to fix and
improve your lib. It doesn't needs to be an all or nothing affair. I
think there's room for both approaches in the library. I can add mine
as a separate lib, and over time we can see how they fair/work
together/cross-pollinate/etc.

I have to go now, but let compare notes tomorrow.

T.


From transfire at gmail.com  Sat Aug 25 22:04:23 2007
From: transfire at gmail.com (Trans)
Date: Sun, 26 Aug 2007 02:04:23 -0000
Subject: [Nitro] Aspects
In-Reply-To: <1188071754.562666.319290@r34g2000hsd.googlegroups.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
Message-ID: <1188093863.651665.98980@r34g2000hsd.googlegroups.com>



On Aug 25, 12:55 pm, Trans <transf... at gmail.com> wrote:
> On Aug 25, 11:01 am, "George Moschovitis"
>
> <george.moschovi... at gmail.com> wrote:
> > > The difficulty comes when the methods do not already exist.
>
> > yeap, this is the problem. My version (included in the repository code)
> > handles this
> > my using a mandatory call to Aspects.setup after all dynamic methods are
> > defined.
> > But I think there are still some bugs (for example on bug reported earlier
> > in this week).
>
> You're using this for Og models, right? Are you using it for Nitro
> controllers too? Anywhere else?
>
> I don't 100% understand the design btw. Could you explain
> what :on, :where, :only, :except, and :join options do?
>
> > I tried to use the version from facets but I had some problems. Tomorrow I
> > will review
> > my version and the code from facets, and I will be in better position to
> > help you with this.
>
> > > method_missing. (Basically imagine every method call being routed via
> > > method_missing).
>
> > this sounds *awfull* to me!
>
> He he. Well, it's not *that* bad! The nice things is that it's 100%
> dynamic. No fuss with included or method_added, which can be easy to
> break.
>
> In any case, for the time being it's probably best just to fix and
> improve your lib. It doesn't needs to be an all or nothing affair. I
> think there's room for both approaches in the library. I can add mine
> as a separate lib, and over time we can see how they fair/work
> together/cross-pollinate/etc.
>
> I have to go now, but let compare notes tomorrow.

I thought about this some more and I think you are right that dynamic
dispatch, as I suggested it, is too inefficient in itself. But I
realized that it would be possible to add an advice cache, and trigger
the appropriate parts of the cache be recalculated when certain
changes occur (method added, aspect added, etc). So then it would be
quite efficient, in fact it amounts effectively that same basic design
as yours.

There's also another part of this. I have a joinpoint model I really
like, which is completely flexible. I'd like to use that, and then
layer your pre, post, around methods on top of that. I pretty
confident that would be a great design. And the we can have one lib
that satisfies us both.

T.


From mvyver at gmail.com  Sun Aug 26 02:41:55 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Sun, 26 Aug 2007 16:41:55 +1000
Subject: [Nitro] OG vs some other ORM packages
Message-ID: <389c43e40708252341r1c3f7870y6e587bbb8a353913@mail.gmail.com>

Hi,
I'd appreciate any comments on my understanding of the approaches used
to implement ORM, and whether my understanding of which Ruby packages
do what is correct.  Also, please shout if I've missed some relevant
package(s).

What would be really interesting is for anyone familiar enough with a
package to say: "Their approach to XYZ is great/bad".
For example:
I think the OG approach of using graph theoretic concepts to 'map'
Ruby objects to dbs is 'great' and offers enormous amounts of head
room, this might be enhanced by exploiting the Gratr package.... ;)

I wanted to see what approaches different packages took and see if any
had any really 'neat'
ideas.  In some cases I've shown in parentheses somethings I found
interesting (+) from a very superficial look at the packages.

The list of Ruby packages I looked at came from the links below.  Not
all of these packages are listed by me, so I may have missed some
important one(s).

http://rubyforge.org/softwaremap/trove_list.php?form_cat=66&page=5
http://raa.ruby-lang.org/cat.rhtml?category_major=Library;category_minor=Database

It seems that two ORM approaches are to:
 - "Wrap" Ruby objects around existing database elements
 - "Map" Ruby objects to databases elements.

Often it seems that 'Map' is the term used to describe both approaches.  Anyway:

Wrapper:
=======
ActiveRecord
Momomoto (PostgreSQL, +InformationSchema, +stored procedures)
Kansas (DBI)
sequel

Mapper:
=======
Lafcadio
Vapor (postgresql only)
ObjectDB (dbi)
odba (dbi)
sqlyzer (dbi)
sop (dbi)

Mapper and wrapper:
=================
OG
Lafcadio (?)

Misc/Uncertain:
=============
tapkit
Merbs' DataMapper
sqldsl
Ruby/OpenBase
dbdiff
Enum Column Plugin For Rails
sql-serialize
ActiveSearch
Rural
ActiveRecord::Extensions
SqlStatement
Visualize Models


Regards
Mark

From william.full.moon at gmail.com  Sun Aug 26 06:04:28 2007
From: william.full.moon at gmail.com (* William)
Date: Sun, 26 Aug 2007 20:04:28 +1000
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708251103o4ae8cbfcrcec6a177c354c387@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com><46D014A1.1020808@tastapod.com>
	<afd80b540708251103o4ae8cbfcrcec6a177c354c387@mail.gmail.com>
Message-ID: <008001c7e7c8$80951d80$0401a8c0@ghostgum>

Ditto dan's comment

   _____  

From: nitro-general-bounces at rubyforge.org
[mailto:nitro-general-bounces at rubyforge.org] On Behalf Of George Moschovitis
Sent: Sunday, 26 August 2007 04:04
To: General discussion about Nitro
Subject: Re: [Nitro] Nitro
Importance: Low



I've said this before and I still stand by it: I get nitro. It works the way
I want to work. Rails just doesn't. I want to start small and grow into an
..Keep up the great work George,
Dan



Thank you, this is very encouraging. I am kind of putting my ass in the line
by using Nitro on such important projects and it is nice to hear that other
people are planning to use it and (hopefully) help to make it as robust as
possible. 

thanks,
George.


-- 
HYPERLINK "http://www.me.gr"http://www.me.gr
HYPERLINK "http://phidz.com"http://phidz.com
HYPERLINK "http://blog.gmosx.com"http://blog.gmosx.com
HYPERLINK "http://cull.gr"http://cull.gr
HYPERLINK "http://www.joy.gr"http://www.joy.gr
HYPERLINK "http://nitroproject.org"http://nitroproject.org 

No virus found in this outgoing message.
Checked by AVG Free Edition. 
Version: 7.5.484 / Virus Database: 269.12.8/973 - Release Date: 25-Aug-2007
17:00
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070826/51a396f0/attachment.html 

From wyhaines at gmail.com  Sun Aug 26 10:12:02 2007
From: wyhaines at gmail.com (Kirk Haines)
Date: Sun, 26 Aug 2007 07:12:02 -0700
Subject: [Nitro] OG vs some other ORM packages
In-Reply-To: <389c43e40708252341r1c3f7870y6e587bbb8a353913@mail.gmail.com>
References: <389c43e40708252341r1c3f7870y6e587bbb8a353913@mail.gmail.com>
Message-ID: <f4cd26df0708260712r4bcc2acfr72ec5f9b29dabfa1@mail.gmail.com>

On 8/25/07, Mark Van De Vyver <mvyver at gmail.com> wrote:

> Kansas (DBI)

Kansas is my baby.  It's been around for a long time and is basically
quite stable.  It uses DBI for the underlying db connectivity instead
of reinventing the wheel there.

It takes the approach of creating Ruby classes that correspond to
database tables, and it tries to be fairly lightweight.  The code is
not particularly large.

It uses Ruby syntax to describe queries:

sodiums = dbh.select(:Chemicals) {|c| c.name =~ '%sodium%'}
critical_issues = dbh.select(:Roofs) {|r| r.sort_by(r.severity =>
'DESC'); r.severity > 3}

It has transaction support that is not dependent on the database
layer.  Queries can return results in normal ruby types such as an
Array or Hash instead of the db-coupled Kansas result types.  The
db-coupled objects can also be marked as read-only, so that changes to
them do not get propagated back to the db.  It supports creating
relationships between tables, for conveniences like:

school = dbh.select(:Schools) {|s| s.name = school_we_are_looking_for }.first
school_cyanides = school.inventory.select {|i| i.name =~ /cyanide/}
art_inventory = school.departments.select {|d| d.name == 'Art'}.chemicals

That's a basic high level view.  I have used it on dozens and dozens
of sites over the years, and it just works.  I have a small community
of users, and insufficient documentation, but it is in daily, active
use, if not daily, active development.


Kirk Haines

From john at oxyliquit.de  Sun Aug 26 20:20:09 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Mon, 27 Aug 2007 02:20:09 +0200
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708241106x53487fccy9c9b17e3a3231391@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<op.txkr9gwhxegpx6@jo.local>
	<afd80b540708241106x53487fccy9c9b17e3a3231391@mail.gmail.com>
Message-ID: <op.txl77lwtxegpx6@jo.local>

Hi,

>> the nitro way and smells as good as Nitro.
>>
> But this smells to me like the situation with Compiz + Beryl: unnecessary
> duplication of efforts.

Ah no, imo it has enough differences to have a reason to exist.

Nitro was allways a little more.. 'opaque' to me opposed to Og, also
because Og produced more problems and so needed more code reading.  ;)

Hackability is a big plus for me, I have commit rights to the public repo,
structure is a little less opaque, some things are way more modular
(like pluggable Logger classes/caches/cache backends), it's smaller,
the test/spec-suite is always green (there's even a tutorial which gets
tested by RSpec, that gets geek points).

There's stuff which is impractical/impossible with it, like streaming
data.  Caching is a little lacking too.  If one wanted that, REST is
not included, feature wise it only has stuff which the current users
care about (which are very few) and there's no support for any ORM.

Nitro is feature wise still on top.  In fact, I use the FormHelper
 from Nitro to power Og, even within my current projects.

So, I still wholeheartedly recommend Nitro to anyone I meet, it always
depends on what you need.  I would recommend Rails to someone who is
searching for something which has most business backing and IOWA to
someone caring about the highest speed.

Hope that helps explaining my state of mind.  :)

Jo

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

From george.moschovitis at gmail.com  Mon Aug 27 11:29:53 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Aug 2007 18:29:53 +0300
Subject: [Nitro] Nitro
In-Reply-To: <op.txl77lwtxegpx6@jo.local>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<op.txkr9gwhxegpx6@jo.local>
	<afd80b540708241106x53487fccy9c9b17e3a3231391@mail.gmail.com>
	<op.txl77lwtxegpx6@jo.local>
Message-ID: <afd80b540708270829m299677abuce35433afef8d34f@mail.gmail.com>

>
> structure is a little less opaque, some things are way more modular
> (like pluggable Logger classes/caches/cache backends),


Nitro is totally modular in this respect too...

but anw, I get your point.

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070827/903c662e/attachment.html 

From transfire at gmail.com  Mon Aug 27 11:31:20 2007
From: transfire at gmail.com (Trans)
Date: Mon, 27 Aug 2007 15:31:20 -0000
Subject: [Nitro] Aspects
In-Reply-To: <1188093863.651665.98980@r34g2000hsd.googlegroups.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
Message-ID: <1188228680.122408.39990@19g2000hsx.googlegroups.com>



On Aug 25, 7:04 pm, Trans <transf... at gmail.com> wrote:
> On Aug 25, 12:55 pm, Trans <transf... at gmail.com> wrote:
>
>
>
> > On Aug 25, 11:01 am, "George Moschovitis"
>
> > <george.moschovi... at gmail.com> wrote:
> > > > The difficulty comes when the methods do not already exist.
>
> > > yeap, this is the problem. My version (included in the repository code)
> > > handles this
> > > my using a mandatory call to Aspects.setup after all dynamic methods are
> > > defined.
> > > But I think there are still some bugs (for example on bug reported earlier
> > > in this week).
>
> > You're using this for Og models, right? Are you using it for Nitro
> > controllers too? Anywhere else?
>
> > I don't 100% understand the design btw. Could you explain
> > what :on, :where, :only, :except, and :join options do?
>
> > > I tried to use the version from facets but I had some problems. Tomorrow I
> > > will review
> > > my version and the code from facets, and I will be in better position to
> > > help you with this.
>
> > > > method_missing. (Basically imagine every method call being routed via
> > > > method_missing).
>
> > > this sounds *awfull* to me!
>
> > He he. Well, it's not *that* bad! The nice things is that it's 100%
> > dynamic. No fuss with included or method_added, which can be easy to
> > break.
>
> > In any case, for the time being it's probably best just to fix and
> > improve your lib. It doesn't needs to be an all or nothing affair. I
> > think there's room for both approaches in the library. I can add mine
> > as a separate lib, and over time we can see how they fair/work
> > together/cross-pollinate/etc.
>
> > I have to go now, but let compare notes tomorrow.
>
> I thought about this some more and I think you are right that dynamic
> dispatch, as I suggested it, is too inefficient in itself. But I
> realized that it would be possible to add an advice cache, and trigger
> the appropriate parts of the cache be recalculated when certain
> changes occur (method added, aspect added, etc). So then it would be
> quite efficient, in fact it amounts effectively that same basic design
> as yours.
>
> There's also another part of this. I have a joinpoint model I really
> like, which is completely flexible. I'd like to use that, and then
> layer your pre, post, around methods on top of that. I pretty
> confident that would be a great design. And the we can have one lib
> that satisfies us both.

Reconsidered. My cut based AOP is looking good. But I think pre, post
and around method filters don't need to be strapped to such a "major"
AOP system. So I'd like to focus on your lib. Do you have any time to
work on this today, George?

T.


From james.britt at gmail.com  Mon Aug 27 13:45:49 2007
From: james.britt at gmail.com (James Britt)
Date: Mon, 27 Aug 2007 10:45:49 -0700
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
Message-ID: <46D30DCD.6080509@gmail.com>

George Moschovitis wrote:
> Out of couriosity...
> 
> is anyone using (or plans to use) Nitro? Or are you all just interested 
> in Og?

Ruby-doc.org is still powered by Nitro.

I'm slowly planning a re-do to fix assorted annoyances.

I'll probably want to use 0.50 for it.

I'd really like to change who the API docs are served up, something that 
allows better searching and commenting and cleaner presentation of core 
and stdlib and mixins while not making folks know in advance ht to look for.


-- 
James Britt

http://www.rubyaz.org         - Hacking in the Desert
http://www.jamesbritt.com     - Playing with Better Toys

From billk at cts.com  Mon Aug 27 14:46:59 2007
From: billk at cts.com (Bill Kelly)
Date: Mon, 27 Aug 2007 11:46:59 -0700
Subject: [Nitro] Latest darcs pull gives errors on postgresql
Message-ID: <038801c7e8da$a714eef0$6442a8c0@musicbox>

Hi,

After doing a darcs pull, my Nitro app now gets the following error:

DEBUG: SELECT nextval('"ogproduct"_oid_seq')
ERROR: DB error ERROR:  invalid name syntax
, [SELECT nextval('"ogproduct"_oid_seq')]


It's happening in: og/lib/og/adapter/postgresql.rb

  def insert(klass, inserts)
    next_oid = nil

    if !inserts[klass.primary_key] || inserts[klass.primary_key] == "NULL"
      if seq = klass.ann(klass.primary_key, :sequence)

 # ^^^^^^^^^^^^^^^^ seq looks like `"ogproduct"_oid_seq'` here

        next_oid = Integer(query("SELECT nextval('#{seq}')").first_value)
        inserts[klass.primary_key] = write_attr(next_oid, :class => Integer)
      end
    end
    super

    return next_oid
  end


I saw a patch related to table name quoting, so I guess this
is related?


Regards,

Bill



From george.moschovitis at gmail.com  Mon Aug 27 15:39:15 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Aug 2007 22:39:15 +0300
Subject: [Nitro] Aspects
In-Reply-To: <1188228680.122408.39990@19g2000hsx.googlegroups.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
Message-ID: <afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>

>
>  So I'd like to focus on your lib. Do you have any time to
> work on this today, George?



I can find some time. You want to start from the code in glue/aspects.rb ??

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070827/b299c47a/attachment.html 

From george.moschovitis at gmail.com  Mon Aug 27 15:40:03 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Aug 2007 22:40:03 +0300
Subject: [Nitro] Latest darcs pull gives errors on postgresql
In-Reply-To: <038801c7e8da$a714eef0$6442a8c0@musicbox>
References: <038801c7e8da$a714eef0$6442a8c0@musicbox>
Message-ID: <afd80b540708271240q5fa5eb6die2b7663e62310374@mail.gmail.com>

>
> I saw a patch related to table name quoting, so I guess this
> is related?
>

Yeap, this is related. Jonathan, is it possible that you have a look at
this?

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070827/92e43722/attachment-0001.html 

From transfire at gmail.com  Mon Aug 27 15:49:51 2007
From: transfire at gmail.com (Trans)
Date: Mon, 27 Aug 2007 19:49:51 -0000
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
Message-ID: <1188244191.623248.112470@57g2000hsv.googlegroups.com>



On Aug 27, 12:39 pm, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> >  So I'd like to focus on your lib. Do you have any time to
> > work on this today, George?
>
> I can find some time. You want to start from the code in glue/aspects.rb ??

Okay. I'll have a look.

t.


From transfire at gmail.com  Mon Aug 27 17:27:23 2007
From: transfire at gmail.com (Trans)
Date: Mon, 27 Aug 2007 21:27:23 -0000
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
Message-ID: <1188250043.795585.43290@o80g2000hse.googlegroups.com>



On Aug 27, 12:39 pm, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> >  So I'd like to focus on your lib. Do you have any time to
> > work on this today, George?
>
> I can find some time. You want to start from the code in glue/aspects.rb ??

I took a look at the code. Good thing is that it's very neatly written
and easy to understand. Koodos there. On the downside though this
implementation is a less efficient. For one thing, iteration over all
of ObjectSpace should never be needed. I don't mind keeping it in
there as a just in case fallback, but nothing more. But the main thing
is that, wrapping by means of instance_method(name).bind(self).call is
very slow. Also, as you surely know, it's too static and won't catch
some dynamic class changes.

I'd suggest instead of using the bind technique, that we just define
special name:original, name:before, name:after and name:around methods
(literally). Which look something like:

  alias name:original, name

  def name(args)
    name:before(args)
    name:original(args)
    name:after(args)
  end

  def name:before(args)
    advices_before[name].each do |advice|
      advice.call(args)
    end
  end

  def name:after(args)
    advices_after[name].each do |advice|
      advice.call(args)
    end
  end

The around advice is little more difficult (or we can just use a
before and an after together?).

Doing it that way, we can update advice, on a per method basis, and
not worry about re-wrapping methods, and whatnot.

Sound good? Shall I implement?

T.


From billk at cts.com  Mon Aug 27 18:25:34 2007
From: billk at cts.com (Bill Kelly)
Date: Mon, 27 Aug 2007 15:25:34 -0700
Subject: [Nitro] Latest darcs pull gives errors on postgresql
References: <038801c7e8da$a714eef0$6442a8c0@musicbox>
	<afd80b540708271240q5fa5eb6die2b7663e62310374@mail.gmail.com>
Message-ID: <006a01c7e8f9$2fcfd980$6442a8c0@musicbox>


From: George Moschovitis
> 
> > I saw a patch related to table name quoting, so I guess this
> > is related?
> 
> Yeap, this is related. Jonathan, is it possible that you have a look at this?

For what it's worth, I was able to kludge around the problem by adding 
the following line to #enchant in og/lib/og/adapter/postgresql.rb:


  def enchant(klass, manager)
    super

    pk = klass.primary_key

    seq = "#{klass::OGTABLE}_#{pk}_seq"
    seq.tr!('"', '')  ####### <-- ADDED THIS LINE to strip quotes on table name

    pkann = klass.ann(pk)

    unless pkann[:sequence]
      if pkann[:sql] =~ /SERIAL/i
        klass.ann(pk, {:sequence => seq})
      else
        klass.ann(pk, {:sequence => false})
      end
    end
  end


Regards,

Bill



From george.moschovitis at gmail.com  Tue Aug 28 01:28:03 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 28 Aug 2007 08:28:03 +0300
Subject: [Nitro] Aspects
In-Reply-To: <1188250043.795585.43290@o80g2000hse.googlegroups.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
Message-ID: <afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>

>From a quick look, your suggestion seems easy to understand, I cannot see
any problem either. I agree with removing the ObjectSpace traversal (I hear
it has problems with JRuby as well)

The around advice is little more difficult (or we can just use a
> before and an after together?).


using before/after together is difficult to pass state from before to
after.


Sound good? Shall I implement?


yeap, when you have something ready, post to this list, and I will review
and make suggestions and/or improvements.

thanks,
-g.



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070828/59c0f1d3/attachment.html 

From george.moschovitis at gmail.com  Tue Aug 28 01:28:35 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 28 Aug 2007 08:28:35 +0300
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
Message-ID: <afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>

Btw, how will we handle wrapping of dynamically generated methods?

-g.

On 8/28/07, George Moschovitis <george.moschovitis at gmail.com> wrote:
>
>
> From a quick look, your suggestion seems easy to understand, I cannot see
> any problem either. I agree with removing the ObjectSpace traversal (I hear
> it has problems with JRuby as well)
>
> The around advice is little more difficult (or we can just use a
> > before and an after together?).
>
>
> using before/after together is difficult to pass state from before to
> after.
>
>
> Sound good? Shall I implement?
>
>
> yeap, when you have something ready, post to this list, and I will review
> and make suggestions and/or improvements.
>
> thanks,
> -g.
>
>
>
> --
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070828/c21b34ab/attachment.html 

From mvyver at gmail.com  Tue Aug 28 02:50:29 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Tue, 28 Aug 2007 16:50:29 +1000
Subject: [Nitro] OG and ruby-dbi
Message-ID: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>

Hi,
As a exercise in 'cutting-my-teeth', I thought to try and implement a
DBI adapter for OG.  Googling the nitro mail-list archive showed one
previous question about this and it didn't suggest that this is
'impossible'/fraught with difficulty - is this still the case?
Of course this may be redundant, but my main goal is to get to know
OG/Ruby/ORM etc.

I noticed the test folder in OG is relatively sparse, and I was
intending to try and use R-spec as I went along....

I'll be looking at how other projects have used DBI (and have already
seen some tips), but I'd appreciate an opportunity to ask
questions/get advice.

A couple of questions.
Which is the most active/preferred communication forum, the mail-list
or the freenode irc channel? (I'm in Sydney,Oz so that might rule irc
out, but I have ome flexibility in my work hours)
Any objections/gotchas to using R-spec to do the specs/BDD with OG?

I'd appreciate any assistance/advice :)

Regards
Mark

From transfire at gmail.com  Tue Aug 28 11:26:27 2007
From: transfire at gmail.com (Trans)
Date: Tue, 28 Aug 2007 15:26:27 -0000
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
Message-ID: <1188314787.091049.162930@w3g2000hsg.googlegroups.com>



On Aug 27, 10:28 pm, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> Btw, how will we handle wrapping of dynamically generated methods?

Nice thing about this design is that we don't need to do any re-
wrapping, so we can easily trigger off the method_added callback and
generate the advice specific to the added method without difficulty.
We will also need to watch the included callback too btw -- ideally we
would generate the advice based solely on the methods defined in the
module.

And very cool, if we want we can define method_added such that it is
advised too, basically:

  def method_added(m)
    send(:"method_added:before", m)
    send(:"method_added:original", m)
    send(:"method_added:after", m)
  end

One would never have to worry about overwriting someone else
method_added (and forgetting to call super), if they just used advice
for it instead.

T.


From transfire at gmail.com  Tue Aug 28 14:26:27 2007
From: transfire at gmail.com (Trans)
Date: Tue, 28 Aug 2007 18:26:27 -0000
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
Message-ID: <1188325587.907262.313400@k79g2000hse.googlegroups.com>

What should this do?

   class X
     before :x do
       print "BEFORE:"
     end

     def x
       puts "x"
     end
   end

   X.new.x  #=> BEFORE:x

   class Y < X
     def x
       puts "y"
     end
   end

   Y.new.x  #=>  ?

Should it be 'BEFORE:y' or just 'y' b/c Y#x overrides everything in X?

T.


From dan at tastapod.com  Tue Aug 28 16:06:52 2007
From: dan at tastapod.com (Dan North)
Date: Tue, 28 Aug 2007 21:06:52 +0100
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
Message-ID: <46D4805C.4000902@tastapod.com>

Hi Mark.

I'm a committer on rspec, and I'll be happy to help you to use rspec to 
describe nitro/og behaviour (either via this list or directly).

Also, the rspec community itself is very helpful and responsive (mail 
lists at http://rubyforge.org/mail/?group_id=797).

Cheers,
Dan

Mark Van De Vyver wrote:
> Hi,
> As a exercise in 'cutting-my-teeth', I thought to try and implement a
> DBI adapter for OG.  Googling the nitro mail-list archive showed one
> previous question about this and it didn't suggest that this is
> 'impossible'/fraught with difficulty - is this still the case?
> Of course this may be redundant, but my main goal is to get to know
> OG/Ruby/ORM etc.
>
> I noticed the test folder in OG is relatively sparse, and I was
> intending to try and use R-spec as I went along....
>
> I'll be looking at how other projects have used DBI (and have already
> seen some tips), but I'd appreciate an opportunity to ask
> questions/get advice.
>
> A couple of questions.
> Which is the most active/preferred communication forum, the mail-list
> or the freenode irc channel? (I'm in Sydney,Oz so that might rule irc
> out, but I have ome flexibility in my work hours)
> Any objections/gotchas to using R-spec to do the specs/BDD with OG?
>
> I'd appreciate any assistance/advice :)
>
> Regards
> Mark
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>   


From john at oxyliquit.de  Tue Aug 28 17:49:39 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 28 Aug 2007 23:49:39 +0200
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
Message-ID: <op.txsgfrvdxegpx6@jo.local>

Hi,

> As a exercise in 'cutting-my-teeth', I thought to try and implement a
> DBI adapter for OG.  Googling the nitro mail-list archive showed one
> previous question about this and it didn't suggest that this is
> 'impossible'/fraught with difficulty - is this still the case?
> Of course this may be redundant, but my main goal is to get to know
> OG/Ruby/ORM etc.

a worthy goal, nice.  ^_^

It is quite doable, and without that much effort actually, the 'needed'
interface is quite easy.  Best would be to look inside either the
Postgresql or Oracle adapters, those have the most 'overrides'.

> I noticed the test folder in OG is relatively sparse, and I was
> intending to try and use R-spec as I went along....

The test folder should even be non-existant.  The spec/ folder is the
new 'home'.  :)
Do `rake test og` inside the nitro project dir to test the og rspecs.

> I'll be looking at how other projects have used DBI (and have already
> seen some tips), but I'd appreciate an opportunity to ask
> questions/get advice.

Sure, I'll sure try to help if you run into troubles implementing that.
Don't hesitate to ask.  :)

> A couple of questions.
> Which is the most active/preferred communication forum, the mail-list
> or the freenode irc channel? (I'm in Sydney,Oz so that might rule irc
> out, but I have ome flexibility in my work hours)

Both?  If I'm present on IRC, this might be faster and thus preferred,
but the ML is good of course.  :)
I'm 'Kashia' or 'Jo[work]' on IRC btw.

> Any objections/gotchas to using R-spec to do the specs/BDD with OG?

Just have a look inside one or two other specs for og, it's quite easy
to use.  Though, some specs are still failing, haven't gotten around to
fix all of them yet.

> I'd appreciate any assistance/advice :)

Like I said, as a startingpoint I'd see the oracle/psql adapters.  I
would love the specs to be fixed... but a well.  If you could run it
once with mysql/psql and see what fails there, you should still be
able to use the testsuite for validation purposes.

Jo

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

From john at oxyliquit.de  Tue Aug 28 17:49:38 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 28 Aug 2007 23:49:38 +0200
Subject: [Nitro] Latest darcs pull gives errors on postgresql
In-Reply-To: <006a01c7e8f9$2fcfd980$6442a8c0@musicbox>
References: <038801c7e8da$a714eef0$6442a8c0@musicbox>
	<afd80b540708271240q5fa5eb6die2b7663e62310374@mail.gmail.com>
	<006a01c7e8f9$2fcfd980$6442a8c0@musicbox>
Message-ID: <op.txsg72tixegpx6@jo.local>

Hi,

>> > I saw a patch related to table name quoting, so I guess this
>> > is related?
>>
>> Yeap, this is related. Jonathan, is it possible that you have a look at this?

I haven't looked at the quoting at all, so I'd have to read the whole
lot first.

> For what it's worth, I was able to kludge around the problem by adding
> the following line to #enchant in og/lib/og/adapter/postgresql.rb:
>
>     seq.tr!('"', '')  ####### <-- ADDED THIS LINE to strip quotes on table name

This is smelly.  :/

Is it assured that the klass::OGTABLE _always_ has quotes G?  If yes,
just do klass::OGTABLE[1,-2] and not the .tr!

Would be great if you could make a patch for that, thanks!

Jo

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

From john at oxyliquit.de  Tue Aug 28 17:49:40 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 28 Aug 2007 23:49:40 +0200
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <46D4805C.4000902@tastapod.com>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
	<46D4805C.4000902@tastapod.com>
Message-ID: <op.txsgtmy2xegpx6@jo.local>

Hi,

> I'm a committer on rspec, and I'll be happy to help you to use rspec to
> describe nitro/og behaviour (either via this list or directly).
>
> Also, the rspec community itself is very helpful and responsive (mail
> lists at http://rubyforge.org/mail/?group_id=797).

this is interesting.  :D

If I could give you a little task which might look a little daunting at
first?
I once wrote this humongous parameter handling Unit::Test for Nitro
(which is sadly still incomplete though.  :P)  This should be converted
to RSpec, but I used special/custom Test::Unit 'assertions' to make
life easier there.
Now my question, could you have a look inside that one and say if one
could convert these assertions to valid and 'true to' rspec?
If you could provide a 'buffer' so to speak, I could then just convert
that huge file using your 'facilities'.  :)

Would be sweet if you'd be up for that, no pressure though.  :)

Jo

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

From john at oxyliquit.de  Tue Aug 28 17:49:41 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Tue, 28 Aug 2007 23:49:41 +0200
Subject: [Nitro] Nitro
In-Reply-To: <afd80b540708270829m299677abuce35433afef8d34f@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<op.txkr9gwhxegpx6@jo.local>
	<afd80b540708241106x53487fccy9c9b17e3a3231391@mail.gmail.com>
	<op.txl77lwtxegpx6@jo.local>
	<afd80b540708270829m299677abuce35433afef8d34f@mail.gmail.com>
Message-ID: <op.txsgyqzaxegpx6@jo.local>

Hi,

>> structure is a little less opaque, some things are way more modular
>> (like pluggable Logger classes/caches/cache backends),
>
> Nitro is totally modular in this respect too...
>
> but anw, I get your point.

yes, modular, but very different.  :)

Jo

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

From dan at tastapod.com  Tue Aug 28 18:34:18 2007
From: dan at tastapod.com (Dan North)
Date: Tue, 28 Aug 2007 23:34:18 +0100
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <op.txsgtmy2xegpx6@jo.local>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>	<46D4805C.4000902@tastapod.com>
	<op.txsgtmy2xegpx6@jo.local>
Message-ID: <46D4A2EA.1090907@tastapod.com>

I'd prefer to answer specific questions to be honest, because I'm a bit 
swamped at the moment. Can you post fragments of the humungous testcase 
and I can see if I can help?

Cheers,
Dan

Jonathan Buch wrote:
> Hi,
>
>   
>> I'm a committer on rspec, and I'll be happy to help you to use rspec to
>> describe nitro/og behaviour (either via this list or directly).
>>
>> Also, the rspec community itself is very helpful and responsive (mail
>> lists at http://rubyforge.org/mail/?group_id=797).
>>     
>
> this is interesting.  :D
>
> If I could give you a little task which might look a little daunting at
> first?
> I once wrote this humongous parameter handling Unit::Test for Nitro
> (which is sadly still incomplete though.  :P)  This should be converted
> to RSpec, but I used special/custom Test::Unit 'assertions' to make
> life easier there.
> Now my question, could you have a look inside that one and say if one
> could convert these assertions to valid and 'true to' rspec?
> If you could provide a 'buffer' so to speak, I could then just convert
> that huge file using your 'facilities'.  :)
>
> Would be sweet if you'd be up for that, no pressure though.  :)
>
> Jo
>
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070828/3f3804d6/attachment.html 

From billk at cts.com  Tue Aug 28 19:24:44 2007
From: billk at cts.com (Bill Kelly)
Date: Tue, 28 Aug 2007 16:24:44 -0700
Subject: [Nitro] Latest darcs pull gives errors on postgresql
References: <038801c7e8da$a714eef0$6442a8c0@musicbox><afd80b540708271240q5fa5eb6die2b7663e62310374@mail.gmail.com><006a01c7e8f9$2fcfd980$6442a8c0@musicbox>
	<op.txsg72tixegpx6@jo.local>
Message-ID: <028401c7e9ca$9e43af50$6442a8c0@musicbox>


From: "Jonathan Buch" <john at oxyliquit.de>
> 
>>> > I saw a patch related to table name quoting, so I guess this
>>> > is related?
>>>
>>> Yeap, this is related. Jonathan, is it possible that you have a look at this?
> 
> I haven't looked at the quoting at all, so I'd have to read the whole
> lot first.
> 
>> For what it's worth, I was able to kludge around the problem by adding
>> the following line to #enchant in og/lib/og/adapter/postgresql.rb:
>>
>>     seq.tr!('"', '')  ####### <-- ADDED THIS LINE to strip quotes on table name
> 
> This is smelly.  :/

Duh :)


That's why I didn't submit it in patch/diff form. ;)


By "for what it's worth", i meant, on the chance that it may help
to illustrate a particular point in the code where the issue
manifested.


Regards,

Bill



From mvyver at gmail.com  Tue Aug 28 19:39:56 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Wed, 29 Aug 2007 09:39:56 +1000
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <46D4805C.4000902@tastapod.com>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
	<46D4805C.4000902@tastapod.com>
Message-ID: <389c43e40708281639h5ff09fd0r7a1902818e16fe30@mail.gmail.com>

Hi Dan,
>
> I'm a committer on rspec, and I'll be happy to help you to use rspec to
> describe nitro/og behaviour (either via this list or directly).
>
> Also, the rspec community itself is very helpful and responsive (mail
> lists at http://rubyforge.org/mail/?group_id=797).

Thanks for the kind offer, no doubt I'll take you up on it at some point.
On irc I'm hedge-hog

Cheers
Mark

> Cheers,
> Dan
>
> Mark Van De Vyver wrote:
> > Hi,
> > As a exercise in 'cutting-my-teeth', I thought to try and implement a
> > DBI adapter for OG.  Googling the nitro mail-list archive showed one
> > previous question about this and it didn't suggest that this is
> > 'impossible'/fraught with difficulty - is this still the case?
> > Of course this may be redundant, but my main goal is to get to know
> > OG/Ruby/ORM etc.
> >
> > I noticed the test folder in OG is relatively sparse, and I was
> > intending to try and use R-spec as I went along....
> >
> > I'll be looking at how other projects have used DBI (and have already
> > seen some tips), but I'd appreciate an opportunity to ask
> > questions/get advice.
> >
> > A couple of questions.
> > Which is the most active/preferred communication forum, the mail-list
> > or the freenode irc channel? (I'm in Sydney,Oz so that might rule irc
> > out, but I have ome flexibility in my work hours)
> > Any objections/gotchas to using R-spec to do the specs/BDD with OG?
> >
> > I'd appreciate any assistance/advice :)
> >
> > Regards
> > Mark
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>

From billk at cts.com  Tue Aug 28 20:05:03 2007
From: billk at cts.com (Bill Kelly)
Date: Tue, 28 Aug 2007 17:05:03 -0700
Subject: [Nitro] Latest darcs pull gives errors on postgresql
References: <038801c7e8da$a714eef0$6442a8c0@musicbox><afd80b540708271240q5fa5eb6die2b7663e62310374@mail.gmail.com><006a01c7e8f9$2fcfd980$6442a8c0@musicbox><op.txsg72tixegpx6@jo.local>
	<028401c7e9ca$9e43af50$6442a8c0@musicbox>
Message-ID: <02a501c7e9d0$3fe54d50$6442a8c0@musicbox>


From: "Bill Kelly" <billk at cts.com>
> 
> From: "Jonathan Buch" <john at oxyliquit.de>
>> 
>>>> > I saw a patch related to table name quoting, so I guess this
>>>> > is related?
>>>>
>>>> Yeap, this is related. Jonathan, is it possible that you have a look at this?
>> 
>> I haven't looked at the quoting at all, so I'd have to read the whole
>> lot first.
>> 
>>> For what it's worth, I was able to kludge around the problem by adding
>>> the following line to #enchant in og/lib/og/adapter/postgresql.rb:
>>>
>>>     seq.tr!('"', '')  ####### <-- ADDED THIS LINE to strip quotes on table name
>> 
>> This is smelly.  :/
> 
> Duh :)
> 
> 
> That's why I didn't submit it in patch/diff form. ;)
> 
> 
> By "for what it's worth", i meant, on the chance that it may help
> to illustrate a particular point in the code where the issue
> manifested.

Sorry, didn't mean to get defensive.  O:-)



Regards,

Bill



From john at oxyliquit.de  Tue Aug 28 21:11:23 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Wed, 29 Aug 2007 03:11:23 +0200
Subject: [Nitro] Latest darcs pull gives errors on postgresql
In-Reply-To: <028401c7e9ca$9e43af50$6442a8c0@musicbox>
References: <038801c7e8da$a714eef0$6442a8c0@musicbox>
	<afd80b540708271240q5fa5eb6die2b7663e62310374@mail.gmail.com>
	<006a01c7e8f9$2fcfd980$6442a8c0@musicbox>
	<op.txsg72tixegpx6@jo.local>
	<028401c7e9ca$9e43af50$6442a8c0@musicbox>
Message-ID: <op.txsqgnp1xegpx6@jo.local>

Hi,

>> This is smelly.  :/
>
> Duh :)
>
> That's why I didn't submit it in patch/diff form. ;)
>
>
> By "for what it's worth", i meant, on the chance that it may help
> to illustrate a particular point in the code where the issue
> manifested.

ah of course, my full 'train of thought' went somhow along these
lines:  huh... yes, problem, how fix, hrm, yes worked around, works,
hrm, smells awful, how can do better, ask for specifics, present
preliminary slightly less smelly alternative.

My posting is a little 'impulsive' sometimes, I will however apologize
if I realize I offended someone gravely.  ;)

Jo

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

From john at oxyliquit.de  Tue Aug 28 21:11:21 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Wed, 29 Aug 2007 03:11:21 +0200
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <46D4A2EA.1090907@tastapod.com>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
	<46D4805C.4000902@tastapod.com> <op.txsgtmy2xegpx6@jo.local>
	<46D4A2EA.1090907@tastapod.com>
Message-ID: <op.txsqqmyixegpx6@jo.local>

Hi,

> I'd prefer to answer specific questions to be honest, because I'm a bit
> swamped at the moment. Can you post fragments of the humungous testcase
> and I can see if I can help?

ah of course, sorry.  :)

Example assertion:

   def assert_request_good(url, value, message = nil)
     result = handle_request(url, value)

     message = build_message(message, "<?> returned error. \nResult: ?\nInstead of: ?", url, result, value)

     assert_block message do
       if result =~ /error/
         false
       else
         cmp(value, eval(result))
       end
     end
   end


Example usage:

   def test_mixed
     tests = {
       '/single/foo?first=one&second=two' => {
         :action => :single,
         :params => {'first' => 'one', 'second' => 'two'},
         :arguments => ['foo']
       }
     }

     tests.each do |url, value|
       assert_request_good url, value
     end
   end

That's basically it, it's a huge amount of tests, all using the
assert_request_good / assert_request_bad methods.  Now those would
somehow have to be 'specified' or removed, depending on how that
stuff would be best handled by RSpec.


I hope that is more 'answerable'.  :)

Jo

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

From billk at cts.com  Tue Aug 28 23:20:35 2007
From: billk at cts.com (Bill Kelly)
Date: Tue, 28 Aug 2007 20:20:35 -0700
Subject: [Nitro] Latest darcs pull gives errors on postgresql
References: <038801c7e8da$a714eef0$6442a8c0@musicbox><afd80b540708271240q5fa5eb6die2b7663e62310374@mail.gmail.com><006a01c7e8f9$2fcfd980$6442a8c0@musicbox><op.txsg72tixegpx6@jo.local><028401c7e9ca$9e43af50$6442a8c0@musicbox>
	<op.txsqgnp1xegpx6@jo.local>
Message-ID: <02f701c7e9eb$90a35c80$6442a8c0@musicbox>


From: "Jonathan Buch" <john at oxyliquit.de>
>
> My posting is a little 'impulsive' sometimes, I will however apologize
> if I realize I offended someone gravely.  ;)

:D


Regards,

Bill



From george.moschovitis at gmail.com  Wed Aug 29 00:23:40 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Aug 2007 07:23:40 +0300
Subject: [Nitro] Aspects
In-Reply-To: <1188325587.907262.313400@k79g2000hse.googlegroups.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
	<1188325587.907262.313400@k79g2000hse.googlegroups.com>
Message-ID: <afd80b540708282123y61026d4ficd5443ff717f70e1@mail.gmail.com>

>
>
> Should it be 'BEFORE:y' or just 'y' b/c Y#x overrides everything in X?
>

I would say 'BEFORE:y' but I am not really sure. Which is easier to
implement?

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070829/f0d397f5/attachment.html 

From mvyver at gmail.com  Wed Aug 29 00:27:37 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Wed, 29 Aug 2007 14:27:37 +1000
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <op.txsgfrvdxegpx6@jo.local>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
	<op.txsgfrvdxegpx6@jo.local>
Message-ID: <389c43e40708282127n3381f33l3e66d8b9c377f81f@mail.gmail.com>

OK.  Apologies in advance for the newbie issues. Here we go....

> The test folder should even be non-existant.  The spec/ folder is the
> new 'home'.  :)
> Do `rake test og` inside the nitro project dir to test the og rspecs.

>From the following it may be that 'systemu' might be a required
package - maybe not?
Anyway, I can't get the rake test og to run.

I've set up as described here:
http://www.arnebrasseur.net/2007/07/01/the-daily-nitro/en/

 # echo ${RUBYOPT}
-r/usr/src/nitro-repo/script/lib/glycerin -rubygems

 #  pushd /usr/src/nitro-repo/
/usr/src/nitro-repo ~
mv-laptop:/usr/src/nitro-repo # rake test og
(in /usr/src/nitro-repo)
Please install systemu for better-looking results
rake aborted!
Don't know how to build task 'test'

(See full trace by running task with --trace)

 # rake --trace test og
(in /usr/src/nitro-repo)
Please install systemu for better-looking results
rake aborted!
Don't know how to build task 'test'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1472:in `[]'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1739:in `top_level'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1739:in `each'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1739:in `top_level'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1761:in
`standard_exception_handling'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1733:in `top_level'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1711:in `run'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1761:in
`standard_exception_handling'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1708:in `run'
/usr/lib/ruby/gems/1.8/gems/rake-0.7.3/bin/rake:7
/usr/bin/rake:16:in `load'
/usr/bin/rake:16


Cheers
MV

> > I'll be looking at how other projects have used DBI (and have already
> > seen some tips), but I'd appreciate an opportunity to ask
> > questions/get advice.
>
> Sure, I'll sure try to help if you run into troubles implementing that.
> Don't hesitate to ask.  :)
>
> > A couple of questions.
> > Which is the most active/preferred communication forum, the mail-list
> > or the freenode irc channel? (I'm in Sydney,Oz so that might rule irc
> > out, but I have ome flexibility in my work hours)
>
> Both?  If I'm present on IRC, this might be faster and thus preferred,
> but the ML is good of course.  :)
> I'm 'Kashia' or 'Jo[work]' on IRC btw.
>
> > Any objections/gotchas to using R-spec to do the specs/BDD with OG?
>
> Just have a look inside one or two other specs for og, it's quite easy
> to use.  Though, some specs are still failing, haven't gotten around to
> fix all of them yet.
>
> > I'd appreciate any assistance/advice :)
>
> Like I said, as a startingpoint I'd see the oracle/psql adapters.  I
> would love the specs to be fixed... but a well.  If you could run it
> once with mysql/psql and see what fails there, you should still be
> able to use the testsuite for validation purposes.
>
> Jo
>
> --
> Feel the love
> http://pinkjuice.com/pics/ruby.png
>

From mvyver at gmail.com  Wed Aug 29 01:36:07 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Wed, 29 Aug 2007 15:36:07 +1000
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <389c43e40708282127n3381f33l3e66d8b9c377f81f@mail.gmail.com>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
	<op.txsgfrvdxegpx6@jo.local>
	<389c43e40708282127n3381f33l3e66d8b9c377f81f@mail.gmail.com>
Message-ID: <389c43e40708282236l23a66afelc5652f86a3bb96a3@mail.gmail.com>

Ok, got it:

# rake test:og
(in /usr/src/nitro-repo)
Running sql...         [   1 specs - all passed ]
1 examples, 0 failures

No failing examples, let's add some tests!

:)

On 8/29/07, Mark Van De Vyver <mvyver at gmail.com> wrote:
> OK.  Apologies in advance for the newbie issues. Here we go....
>
> > The test folder should even be non-existant.  The spec/ folder is the
> > new 'home'.  :)
> > Do `rake test og` inside the nitro project dir to test the og rspecs.
>
> From the following it may be that 'systemu' might be a required
> package - maybe not?
> Anyway, I can't get the rake test og to run.
>
> I've set up as described here:
> http://www.arnebrasseur.net/2007/07/01/the-daily-nitro/en/
>
>  # echo ${RUBYOPT}
> -r/usr/src/nitro-repo/script/lib/glycerin -rubygems
>
>  #  pushd /usr/src/nitro-repo/
> /usr/src/nitro-repo ~
> mv-laptop:/usr/src/nitro-repo # rake test og
> (in /usr/src/nitro-repo)
> Please install systemu for better-looking results
> rake aborted!
> Don't know how to build task 'test'
>
> (See full trace by running task with --trace)
>
>  # rake --trace test og
> (in /usr/src/nitro-repo)
> Please install systemu for better-looking results
> rake aborted!
> Don't know how to build task 'test'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1472:in `[]'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1739:in `top_level'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1739:in `each'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1739:in `top_level'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1761:in
> `standard_exception_handling'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1733:in `top_level'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1711:in `run'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1761:in
> `standard_exception_handling'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/lib/rake.rb:1708:in `run'
> /usr/lib/ruby/gems/1.8/gems/rake-0.7.3/bin/rake:7
> /usr/bin/rake:16:in `load'
> /usr/bin/rake:16
>
>
> Cheers
> MV
>
> > > I'll be looking at how other projects have used DBI (and have already
> > > seen some tips), but I'd appreciate an opportunity to ask
> > > questions/get advice.
> >
> > Sure, I'll sure try to help if you run into troubles implementing that.
> > Don't hesitate to ask.  :)
> >
> > > A couple of questions.
> > > Which is the most active/preferred communication forum, the mail-list
> > > or the freenode irc channel? (I'm in Sydney,Oz so that might rule irc
> > > out, but I have ome flexibility in my work hours)
> >
> > Both?  If I'm present on IRC, this might be faster and thus preferred,
> > but the ML is good of course.  :)
> > I'm 'Kashia' or 'Jo[work]' on IRC btw.
> >
> > > Any objections/gotchas to using R-spec to do the specs/BDD with OG?
> >
> > Just have a look inside one or two other specs for og, it's quite easy
> > to use.  Though, some specs are still failing, haven't gotten around to
> > fix all of them yet.
> >
> > > I'd appreciate any assistance/advice :)
> >
> > Like I said, as a startingpoint I'd see the oracle/psql adapters.  I
> > would love the specs to be fixed... but a well.  If you could run it
> > once with mysql/psql and see what fails there, you should still be
> > able to use the testsuite for validation purposes.
> >
> > Jo
> >
> > --
> > Feel the love
> > http://pinkjuice.com/pics/ruby.png
> >
>

From william.full.moon at gmail.com  Wed Aug 29 02:19:12 2007
From: william.full.moon at gmail.com (* William)
Date: Wed, 29 Aug 2007 16:19:12 +1000
Subject: [Nitro] Nitro
In-Reply-To: <op.txsgyqzaxegpx6@jo.local>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<op.txkr9gwhxegpx6@jo.local>
	<afd80b540708241106x53487fccy9c9b17e3a3231391@mail.gmail.com>
	<op.txl77lwtxegpx6@jo.local>
	<afd80b540708270829m299677abuce35433afef8d34f@mail.gmail.com>
	<op.txsgyqzaxegpx6@jo.local>
Message-ID: <9e03c3c60708282319i17741d0cv6f8e96f27d71fca@mail.gmail.com>

Hmmm ...

Honestly I don't "believe" that modular it the distinguishing factor.  Clear
and Concise would be better terms.

Lots of things can be "modular"  with no functional benefit.

Discrete and easy to understand would not require "modules".  It is just
nicer to package things in units (or containers).

I think we see this a lot, a "buzz word" makes a statement and we tend to
make it a description.  Really, it is a label for a string of attributes
that fit-well-with modular (or the particular buzz word).

*grin*

On 29/08/07, Jonathan Buch <john at oxyliquit.de> wrote:
>
> Hi,
>
> >> structure is a little less opaque, some things are way more modular
> >> (like pluggable Logger classes/caches/cache backends),
> >
> > Nitro is totally modular in this respect too...
> >
> > but anw, I get your point.
>
> yes, modular, but very different.  :)
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070829/e1e94e63/attachment.html 

From mvyver at gmail.com  Wed Aug 29 04:28:58 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Wed, 29 Aug 2007 18:28:58 +1000
Subject: [Nitro] OG and DBI: driver option name
Message-ID: <389c43e40708290128t6261c48ajdf839c703403d87@mail.gmail.com>

Hi,
I'd appreciate any suggestions/comments on the following issue.

In OG, adapters are specializations of a SQL store.
DBI supports many database drivers.
I suppose my issue is the blurring/overlap of adapters and drivers.

So is DBI the adapter, or is DBIMySQL to be the adapter.  A problem
with the later approach is there will be as many Dbi<vendor>Adapter
classes as dbi supports.... 13 from a current count.

If I use a class DbiAdapter then it's initialize method would need to
handle which 'flavor'/'backend'/'driver' (anyone suggest a 'correct'
term to use) of dbi.  This could not come from the current
options[:adapter], because that should be 'dbi':
That is in
Adapter.for_name(:dbi)

This suggests an additional option is needed if DBI is used.  Any
suggestion to improve on using :dbi_driver?

Cheers
Mark

From george.moschovitis at gmail.com  Wed Aug 29 08:49:13 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Aug 2007 15:49:13 +0300
Subject: [Nitro] OG and DBI: driver option name
In-Reply-To: <389c43e40708290128t6261c48ajdf839c703403d87@mail.gmail.com>
References: <389c43e40708290128t6261c48ajdf839c703403d87@mail.gmail.com>
Message-ID: <afd80b540708290549r64ad30bq1103f18dd96bf7e3@mail.gmail.com>

>
> So is DBI the adapter, or is DBIMySQL to be the adapter.  A problem
> with the later approach is there will be as many Dbi<vendor>Adapter
> classes as dbi supports.... 13 from a current count.


I would say that DBI is the adapter (a specialization of SQL store). However
this specific adapter
will be able to handle multiple rdbms.

suggestion to improve on using :dbi_driver?
>

I think :dbi_driver is ok as a parameter name:

  Og.start(
    :name => "myapp",
    :adapter => :dbi,
    :dbi_driver => :mysql,
    :user => "root",
    :evolve_schema => :full # true
  )

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070829/99e2dcec/attachment.html 

From transfire at gmail.com  Wed Aug 29 08:56:40 2007
From: transfire at gmail.com (Trans)
Date: Wed, 29 Aug 2007 12:56:40 -0000
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708282123y61026d4ficd5443ff717f70e1@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708251101gc5b2109u1297c3ee1ad17835@mail.gmail.com>
	<1188071754.562666.319290@r34g2000hsd.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
	<1188325587.907262.313400@k79g2000hse.googlegroups.com>
	<afd80b540708282123y61026d4ficd5443ff717f70e1@mail.gmail.com>
Message-ID: <1188392200.409137.173050@g4g2000hsf.googlegroups.com>



On Aug 28, 9:23 pm, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> > Should it be 'BEFORE:y' or just 'y' b/c Y#x overrides everything in X?
>
> I would say 'BEFORE:y' but I am not really sure. Which is easier to
> implement?

The later is easier. The difficulty with the 'BEFORE:y' is if #super
is called. The method in class Y runs the advice and then the method
in the superclass runs it again. So to do it, I have to figure out a
way for it to only run once.

T.


From george.moschovitis at gmail.com  Wed Aug 29 10:14:24 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Aug 2007 17:14:24 +0300
Subject: [Nitro] Aspects
In-Reply-To: <1188392200.409137.173050@g4g2000hsf.googlegroups.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
	<1188325587.907262.313400@k79g2000hse.googlegroups.com>
	<afd80b540708282123y61026d4ficd5443ff717f70e1@mail.gmail.com>
	<1188392200.409137.173050@g4g2000hsf.googlegroups.com>
Message-ID: <afd80b540708290714r327a270bu5d1a284850fcff1d@mail.gmail.com>

Lets go for the later then...

-g.

On 8/29/07, Trans <transfire at gmail.com> wrote:
>
>
>
> On Aug 28, 9:23 pm, "George Moschovitis"
> <george.moschovi... at gmail.com> wrote:
> > > Should it be 'BEFORE:y' or just 'y' b/c Y#x overrides everything in X?
> >
> > I would say 'BEFORE:y' but I am not really sure. Which is easier to
> > implement?
>
> The later is easier. The difficulty with the 'BEFORE:y' is if #super
> is called. The method in class Y runs the advice and then the method
> in the superclass runs it again. So to do it, I have to figure out a
> way for it to only run once.
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070829/7dfd81b9/attachment.html 

From transfire at gmail.com  Wed Aug 29 13:28:48 2007
From: transfire at gmail.com (Trans)
Date: Wed, 29 Aug 2007 17:28:48 -0000
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708290714r327a270bu5d1a284850fcff1d@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<1188093863.651665.98980@r34g2000hsd.googlegroups.com>
	<1188228680.122408.39990@19g2000hsx.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
	<1188325587.907262.313400@k79g2000hse.googlegroups.com>
	<afd80b540708282123y61026d4ficd5443ff717f70e1@mail.gmail.com>
	<1188392200.409137.173050@g4g2000hsf.googlegroups.com>
	<afd80b540708290714r327a270bu5d1a284850fcff1d@mail.gmail.com>
Message-ID: <1188408528.318463.192760@r34g2000hsd.googlegroups.com>



On Aug 29, 7:14 am, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> Lets go for the later then...

I got it to work either way. So lets decide based on what is the most
desirable behavior.

Consider carefully the difference.

  BEFORE:Y#x
  Y#x
  BEFORE:X#x
  X#x
  AFTER:X#x
  AFTER:Y#x

vs.

  BEFORE:Y#x
  BEFORE:X#x
  Y#x
  X#x
  AFTER:X#x
  AFTER:Y#x

The 1st is analogous to sub-classing, so it's perhaps the most
obvious. However, is what Matz seems to have in mind for his :before
and :after decorators which are modeled after CLOS. See
http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-talk/85464.

T.


From billk at cts.com  Wed Aug 29 20:14:37 2007
From: billk at cts.com (Bill Kelly)
Date: Wed, 29 Aug 2007 17:14:37 -0700
Subject: [Nitro] changing SessionStore
Message-ID: <048901c7ea9a$c09dd1e0$6442a8c0@musicbox>

Hi,

I was thinking of trying the MemorySessionStore, but I noticed
nitro/lib/nitro/application.rb unconditionally sets:
  @session_store = Raw::CookieSessionStore.new
in Application#start.

If I wanted to try the MemorySessionStore, at what point should
I initialize app.session_store ?


Thanks,

Bill



From billk at cts.com  Wed Aug 29 21:34:34 2007
From: billk at cts.com (Bill Kelly)
Date: Wed, 29 Aug 2007 18:34:34 -0700
Subject: [Nitro] vcr_record, vcr_playback
Message-ID: <07f701c7eaa5$eba570e0$6442a8c0@musicbox>

Hi,

I'm interested in trying the vcr_record and vcr_playback
routines in raw/lib/raw/adapter/webrick/vcr.rb.

I remember some previous mention of these, but nothing
turned up in my google, oxyliquit, or mailing list 
archives searches.

Could anyone provide a small example?


Thanks,

Bill



From george.moschovitis at gmail.com  Thu Aug 30 01:08:22 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 30 Aug 2007 08:08:22 +0300
Subject: [Nitro] vcr_record, vcr_playback
In-Reply-To: <07f701c7eaa5$eba570e0$6442a8c0@musicbox>
References: <07f701c7eaa5$eba570e0$6442a8c0@musicbox>
Message-ID: <afd80b540708292208i5a05916du9fd5e796c7cb10a6@mail.gmail.com>

Bill,

let me verify that this functionality works in the repo version and I will
get back to you later today.

-g.

On 8/30/07, Bill Kelly <billk at cts.com> wrote:
>
> Hi,
>
> I'm interested in trying the vcr_record and vcr_playback
> routines in raw/lib/raw/adapter/webrick/vcr.rb.
>
> I remember some previous mention of these, but nothing
> turned up in my google, oxyliquit, or mailing list
> archives searches.
>
> Could anyone provide a small example?
>
>
> Thanks,
>
> Bill
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070830/b6bd936d/attachment.html 

From billk at cts.com  Thu Aug 30 02:13:24 2007
From: billk at cts.com (Bill Kelly)
Date: Wed, 29 Aug 2007 23:13:24 -0700
Subject: [Nitro] Testing the whole website with GET/POST
Message-ID: <367301c7eacc$df9b18a0$6442a8c0@musicbox>

Hi,

Including techniques such as vcr_record / vcr_playback, I was 
wondering if anyone might be able to share tips for testing
a site "externally"... ?

For instance, I have a Nitro/Og site that has a reasonable
amount of tests at the model level.  But I don't have anything
testing it yet from the "submit forms" level.

I'd kind of like to be able to exercise the site logic from
the point of view of a "browser".


Anyone have any tips on that?  (Or alternative approaches?)


Thx!

Bill



From mvyver at gmail.com  Thu Aug 30 02:55:34 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Thu, 30 Aug 2007 16:55:34 +1000
Subject: [Nitro] OG: Model#instance_attribute_set silently updates Model#new
	instances using WHERE oid=NULL
Message-ID: <389c43e40708292355n435a9d5ck882e2b3148c7e06f@mail.gmail.com>

Hi,
I've been working through some examples and came across the following behavior.
I'm posting this here since there is an error page instead of the
ticket system at:
http://www.nitroproject.org/tickets

Without the '$DBG = true' setting there would be no hint that the last
statement was unsuccessful.
I'm not sure if an exception should be thrown, or the record should
just be saved (when saved? -> nil) before updating.

HTH
Mark

#-----------------------------
require 'rubygems'
require 'og'

# Full debug information.
$DBG = true

mysql_config = {
  :name => 'test',
  :store => 'mysql',
  :user => 'root',
  :password => '',
  :destroy=>false
}

# keeping it simple
class SimplerMember
  attr_accessor :first_name, :last_name, :address1, :city, :state,
:email, String
end

  # define some data.
  FIXTURE = { :first_name => 'Fred', :last_name => 'Flintstone',
              :address1 => '123 High St.', :city => 'Reykjavik',
              :state => 'Michigan', :email => 'fred at flintstone.com' }

  # Initialize Og.
  mysql = Og.connect(mysql_config)

  # This will INSERT an empty row, then make an UPDATE for each
  # attribute/field using WHERE oid=<some_number>
  simmem_rec = SimplerMember.create
  FIXTURE.each { |key, value|
simmem_rec.instance_attribute_set("#{key}", value) }

  #This does an UPDATE for each field using WHERE oid=NULL
  simmem_rec = SimplerMember.new
  FIXTURE.each { |key, value|
simmem_rec.instance_attribute_set("#{key}", value) }

#-----------------------------

From george.moschovitis at gmail.com  Thu Aug 30 03:59:16 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 30 Aug 2007 10:59:16 +0300
Subject: [Nitro] vcr_record, vcr_playback
In-Reply-To: <afd80b540708292208i5a05916du9fd5e796c7cb10a6@mail.gmail.com>
References: <07f701c7eaa5$eba570e0$6442a8c0@musicbox>
	<afd80b540708292208i5a05916du9fd5e796c7cb10a6@mail.gmail.com>
Message-ID: <afd80b540708300059gfda57bds88d53b91ec6ffabe@mail.gmail.com>

Bill,

I added the missing flags --record / --playback, to be used like this:

nitro --record session.rec
nitro --playback session.rec

however it does not work, i get the following error:

/usr/lib/ruby/1.8/yaml/rubytypes.rb:6:in `to_yaml': can't dump anonymous
class Class (TypeError)

I think the problem are the IO/File objects in the Webrick request/response
classes. Dunno how this used to work. If anyone can have a look at
raw/adapter/webrick/vcr.rb and offer some help, I would appreciate it.


-g.


On 8/30/07, George Moschovitis <george.moschovitis at gmail.com> wrote:
>
> Bill,
>
> let me verify that this functionality works in the repo version and I will
> get back to you later today.
>
> -g.
>
> On 8/30/07, Bill Kelly < billk at cts.com> wrote:
> >
> > Hi,
> >
> > I'm interested in trying the vcr_record and vcr_playback
> > routines in raw/lib/raw/adapter/webrick/vcr.rb.
> >
> > I remember some previous mention of these, but nothing
> > turned up in my google, oxyliquit, or mailing list
> > archives searches.
> >
> > Could anyone provide a small example?
> >
> >
> > Thanks,
> >
> > Bill
> >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
>
> --
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org




-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070830/df8aacc7/attachment.html 

From george.moschovitis at gmail.com  Thu Aug 30 04:01:43 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 30 Aug 2007 11:01:43 +0300
Subject: [Nitro] Testing the whole website with GET/POST
In-Reply-To: <367301c7eacc$df9b18a0$6442a8c0@musicbox>
References: <367301c7eacc$df9b18a0$6442a8c0@musicbox>
Message-ID: <afd80b540708300101k56f32ba9he14a305c4d07e550@mail.gmail.com>

A quick tip:

have a look at raw/adapter/script.rb

this is used when running nitro --console. You could use this infrastructure
in test scripts. This is similar to Rail's integration testing. I plan on
working on improved testing utilities real soon.

hope tjhis helps,
-g.

On 8/30/07, Bill Kelly <billk at cts.com> wrote:
>
> Hi,
>
> Including techniques such as vcr_record / vcr_playback, I was
> wondering if anyone might be able to share tips for testing
> a site "externally"... ?
>
> For instance, I have a Nitro/Og site that has a reasonable
> amount of tests at the model level.  But I don't have anything
> testing it yet from the "submit forms" level.
>
> I'd kind of like to be able to exercise the site logic from
> the point of view of a "browser".
>
>
> Anyone have any tips on that?  (Or alternative approaches?)
>
>
> Thx!
>
> Bill
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070830/602e1401/attachment-0001.html 

From george.moschovitis at gmail.com  Thu Aug 30 04:05:16 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 30 Aug 2007 11:05:16 +0300
Subject: [Nitro] OG: Model#instance_attribute_set silently updates
	Model#new instances using WHERE oid=NULL
In-Reply-To: <389c43e40708292355n435a9d5ck882e2b3148c7e06f@mail.gmail.com>
References: <389c43e40708292355n435a9d5ck882e2b3148c7e06f@mail.gmail.com>
Message-ID: <afd80b540708300105u395fdb1y57d9d9a8d143611a@mail.gmail.com>

thansk for the report, I will investigate this issue later today.

-g.

On 8/30/07, Mark Van De Vyver <mvyver at gmail.com> wrote:
>
> Hi,
> I've been working through some examples and came across the following
> behavior.
> I'm posting this here since there is an error page instead of the
> ticket system at:
> http://www.nitroproject.org/tickets
>
> Without the '$DBG = true' setting there would be no hint that the last
> statement was unsuccessful.
> I'm not sure if an exception should be thrown, or the record should
> just be saved (when saved? -> nil) before updating.
>
> HTH
> Mark
>
> #-----------------------------
> require 'rubygems'
> require 'og'
>
> # Full debug information.
> $DBG = true
>
> mysql_config = {
>   :name => 'test',
>   :store => 'mysql',
>   :user => 'root',
>   :password => '',
>   :destroy=>false
> }
>
> # keeping it simple
> class SimplerMember
>   attr_accessor :first_name, :last_name, :address1, :city, :state,
> :email, String
> end
>
>   # define some data.
>   FIXTURE = { :first_name => 'Fred', :last_name => 'Flintstone',
>               :address1 => '123 High St.', :city => 'Reykjavik',
>               :state => 'Michigan', :email => 'fred at flintstone.com' }
>
>   # Initialize Og.
>   mysql = Og.connect(mysql_config)
>
>   # This will INSERT an empty row, then make an UPDATE for each
>   # attribute/field using WHERE oid=<some_number>
>   simmem_rec = SimplerMember.create
>   FIXTURE.each { |key, value|
> simmem_rec.instance_attribute_set("#{key}", value) }
>
>   #This does an UPDATE for each field using WHERE oid=NULL
>   simmem_rec = SimplerMember.new
>   FIXTURE.each { |key, value|
> simmem_rec.instance_attribute_set("#{key}", value) }
>
> #-----------------------------
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070830/899e211d/attachment.html 

From mvyver at gmail.com  Thu Aug 30 04:10:55 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Thu, 30 Aug 2007 18:10:55 +1000
Subject: [Nitro] OG: Model#instance_attribute_set silently updates
	Model#new instances using WHERE oid=NULL
In-Reply-To: <afd80b540708300105u395fdb1y57d9d9a8d143611a@mail.gmail.com>
References: <389c43e40708292355n435a9d5ck882e2b3148c7e06f@mail.gmail.com>
	<afd80b540708300105u395fdb1y57d9d9a8d143611a@mail.gmail.com>
Message-ID: <389c43e40708300110w28ccf68bw1f07d940ab672945@mail.gmail.com>

Hi George,
Thank you and the other developers! - the more I see of Og the more
I'm impressed by it.
Related to that example:

Using irb:

## Works
mysql.store.save( Member.new(FIXTURE) )

## Fails
# mysql.store.insert( Member.new(FIXTURE) )
ArgumentError: wrong number of arguments (1 for 2)
        from (irb):113:in `insert'
        from (irb):113
        from :0

In the ri documents it seems just one argument is required:
Og::Store#insert insert(obj)
Insert an object in the store.

HTH
Mark

On 8/30/07, George Moschovitis <george.moschovitis at gmail.com> wrote:
> thansk for the report, I will investigate this issue later today.
>
> -g.
>
>
> On 8/30/07, Mark Van De Vyver <mvyver at gmail.com > wrote:
> >
> > Hi,
> > I've been working through some examples and came across the following
> behavior.
> > I'm posting this here since there is an error page instead of the
> > ticket system at:
> > http://www.nitroproject.org/tickets
> >
> > Without the '$DBG = true' setting there would be no hint that the last
> > statement was unsuccessful.
> > I'm not sure if an exception should be thrown, or the record should
> > just be saved (when saved? -> nil) before updating.
> >
> > HTH
> > Mark
> >
> > #-----------------------------
> > require 'rubygems'
> > require 'og'
> >
> > # Full debug information.
> > $DBG = true
> >
> > mysql_config = {
> >   :name => 'test',
> >   :store => 'mysql',
> >   :user => 'root',
> >   :password => '',
> >   :destroy=>false
> > }
> >
> > # keeping it simple
> > class SimplerMember
> >   attr_accessor :first_name, :last_name, :address1, :city, :state,
> > :email, String
> > end
> >
> >   # define some data.
> >   FIXTURE = { :first_name => 'Fred', :last_name => 'Flintstone',
> >               :address1 => '123 High St.', :city => 'Reykjavik',
> >               :state => 'Michigan', :email => ' fred at flintstone.com' }
> >
> >   # Initialize Og.
> >   mysql = Og.connect(mysql_config)
> >
> >   # This will INSERT an empty row, then make an UPDATE for each
> >   # attribute/field using WHERE oid=<some_number>
> >   simmem_rec = SimplerMember.create
> >   FIXTURE.each { |key, value|
> > simmem_rec.instance_attribute_set("#{key}", value) }
> >
> >   #This does an UPDATE for each field using WHERE oid=NULL
> >   simmem_rec = SimplerMember.new
> >   FIXTURE.each { |key, value|
> > simmem_rec.instance_attribute_set("#{key}", value) }
> >
> > #-----------------------------
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
>
> --
> http://www.me.gr
> http://phidz.com
> http://blog.gmosx.com
> http://cull.gr
> http://www.joy.gr
> http://nitroproject.org

From john at oxyliquit.de  Thu Aug 30 04:54:12 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Thu, 30 Aug 2007 10:54:12 +0200
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <389c43e40708282236l23a66afelc5652f86a3bb96a3@mail.gmail.com>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
	<op.txsgfrvdxegpx6@jo.local>
	<389c43e40708282127n3381f33l3e66d8b9c377f81f@mail.gmail.com>
	<389c43e40708282236l23a66afelc5652f86a3bb96a3@mail.gmail.com>
Message-ID: <op.txu6g9hwxegpx6@jo.local>

Hi,

> Ok, got it:
>
> # rake test:og
> (in /usr/src/nitro-repo)
> Running sql...         [   1 specs - all passed ]
> 1 examples, 0 failures
>
> No failing examples, let's add some tests!

......  there should be dozens of specs... not only one.

GEORGE?

[Create initial testing infrastructure in og, nitro, glue.
George Moschovitis <george.moschovitis at gmail.com>**20070810075936] {
move ./rakefile ./Rakefile
adddir ./glue/.old_test
move ./glue/spec ./glue/.old_test/spec

...

Please care to explain?

Jo

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

From george.moschovitis at gmail.com  Thu Aug 30 05:55:36 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 30 Aug 2007 12:55:36 +0300
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <op.txu6g9hwxegpx6@jo.local>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
	<op.txsgfrvdxegpx6@jo.local>
	<389c43e40708282127n3381f33l3e66d8b9c377f81f@mail.gmail.com>
	<389c43e40708282236l23a66afelc5652f86a3bb96a3@mail.gmail.com>
	<op.txu6g9hwxegpx6@jo.local>
Message-ID: <afd80b540708300255y7ccaf6f4i6f2005043fab2e6d@mail.gmail.com>

Don't worry, they are in a hidden directory .old_tests. I will copy them.

-g.

On 8/30/07, Jonathan Buch <john at oxyliquit.de> wrote:
>
> Hi,
>
> > Ok, got it:
> >
> > # rake test:og
> > (in /usr/src/nitro-repo)
> > Running sql...         [   1 specs - all passed ]
> > 1 examples, 0 failures
> >
> > No failing examples, let's add some tests!
>
> ......  there should be dozens of specs... not only one.
>
> GEORGE?
>
> [Create initial testing infrastructure in og, nitro, glue.
> George Moschovitis <george.moschovitis at gmail.com>**20070810075936] {
> move ./rakefile ./Rakefile
> adddir ./glue/.old_test
> move ./glue/spec ./glue/.old_test/spec
>
> ...
>
> Please care to explain?
>
> 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.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070830/4941cfac/attachment.html 

From george.moschovitis at gmail.com  Thu Aug 30 07:36:58 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 30 Aug 2007 14:36:58 +0300
Subject: [Nitro] Aspects
In-Reply-To: <1188408528.318463.192760@r34g2000hsd.googlegroups.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
	<1188325587.907262.313400@k79g2000hse.googlegroups.com>
	<afd80b540708282123y61026d4ficd5443ff717f70e1@mail.gmail.com>
	<1188392200.409137.173050@g4g2000hsf.googlegroups.com>
	<afd80b540708290714r327a270bu5d1a284850fcff1d@mail.gmail.com>
	<1188408528.318463.192760@r34g2000hsd.googlegroups.com>
Message-ID: <afd80b540708300436s1f4dbb15p81f329118b9b53e5@mail.gmail.com>

I find the second solution more natural (what matz descibes in the provided
link) more natural.
This is the behaviour I would expect. But, even the first solution may be ok
in 90% of the cases.
Unless theres is a considerable performance/complexity penalty, lets go for
the second solution. Then we can switch to the official solution when Ruby
1.9 is finally released.

regards,
-g.


On 8/29/07, Trans <transfire at gmail.com> wrote:
>
>
>
> On Aug 29, 7:14 am, "George Moschovitis"
> <george.moschovi... at gmail.com> wrote:
> > Lets go for the later then...
>
> I got it to work either way. So lets decide based on what is the most
> desirable behavior.
>
> Consider carefully the difference.
>
>   BEFORE:Y#x
>   Y#x
>   BEFORE:X#x
>   X#x
>   AFTER:X#x
>   AFTER:Y#x
>
> vs.
>
>   BEFORE:Y#x
>   BEFORE:X#x
>   Y#x
>   X#x
>   AFTER:X#x
>   AFTER:Y#x
>
> The 1st is analogous to sub-classing, so it's perhaps the most
> obvious. However, is what Matz seems to have in mind for his :before
> and :after decorators which are modeled after CLOS. See
> http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-talk/85464.
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070830/2d1ea2bf/attachment.html 

From mvyver at gmail.com  Thu Aug 30 08:21:13 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Thu, 30 Aug 2007 22:21:13 +1000
Subject: [Nitro] OG/Ruby/MySQLAdapter create performance
Message-ID: <389c43e40708300521w70d0d841reae16a87c995b462@mail.gmail.com>

Hi,
Following up on my earlier post (which contained the AR benchmark code
and reference), I've 'converted/ported' the AR benchmark code.
My main purpose was to checkout the different OG ways of creating
records - the times are an interesting bonus.

The previous post:
http://rubyurl.com/w27

A couple of points.  I'm a Ruby newbie, so there are probably some
clangers in here that make the results meaningless - please let me
know if there is anything that should be changed.
I've only looked at ways that OG provides to create data records.

I'd like to convert this into an './examples/og_create_examples.rb'
and write a spec for each example of these creates - time permitting,
so please post any interesting 'ways to create' (in addition the the
following specific request).  I suppose it'd be useful to have a
benchmark/example/spec for each or C.R.U.D.....

In the commented out section you'll see some AR code that creates using SQL.
Can someone provide/point me to how this would be done in OG.

On my machine OG and AR where 'about' the same.
Using the hash initialized model the cleanest/simplest OG create
syntax was the fastest - but this might be improved on?

Cheers
Mark
-------------- next part --------------
A non-text attachment was scrubbed...
Name: og_create_benchmark.rb
Type: application/x-ruby
Size: 10053 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070830/751f25ab/attachment-0001.bin 

From billk at cts.com  Thu Aug 30 21:28:43 2007
From: billk at cts.com (Bill Kelly)
Date: Thu, 30 Aug 2007 18:28:43 -0700
Subject: [Nitro] Testing the whole website with GET/POST
References: <367301c7eacc$df9b18a0$6442a8c0@musicbox>
	<afd80b540708300101k56f32ba9he14a305c4d07e550@mail.gmail.com>
Message-ID: <009401c7eb6e$4535c470$6442a8c0@musicbox>


From: George Moschovitis
> 
> A quick tip:
> 
> have a look at raw/adapter/script.rb
> 
> this is used when running nitro --console. You could use this
> infrastructure in test scripts. This is similar to Rail's integration
> testing. I plan on working on improved testing utilities real soon. 

Hi George,

The script adapter looks useful.

However, it looks like it may not be handling the session cookie
for me as I request subsequent links.

I use session to store a single value, the order_id.  But the
order_id is coming out nil each successive page I fetch. 
I presume this is because the ScriptAdaptor get/post methods
are probably not doing automatic cookie handling the way a
browser would.

I see that the Raw::Context returned by a get() contains cookie
data in @response_cookies.  However, I'm guessing that will
be server->browser cookie data, rather than browser->server
cookie data.

Is there a recommended approach for dealing with session
cookies via the ScriptAdapter?


Regards,

Bill





From george.moschovitis at gmail.com  Fri Aug 31 03:44:57 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 31 Aug 2007 10:44:57 +0300
Subject: [Nitro] Testing the whole website with GET/POST
In-Reply-To: <009401c7eb6e$4535c470$6442a8c0@musicbox>
References: <367301c7eacc$df9b18a0$6442a8c0@musicbox>
	<afd80b540708300101k56f32ba9he14a305c4d07e550@mail.gmail.com>
	<009401c7eb6e$4535c470$6442a8c0@musicbox>
Message-ID: <afd80b540708310044v18a85c9n9878e638e520723d@mail.gmail.com>

>
> Is there a recommended approach for dealing with session
> cookies via the ScriptAdapter?
>

I will have to revisit this. It is quite possible that I will work on this
over this weekend, and I may have some good news for you early next week.

-g.


-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070831/f5a22770/attachment.html 

From mvyver at gmail.com  Fri Aug 31 03:50:14 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Fri, 31 Aug 2007 17:50:14 +1000
Subject: [Nitro] OG and DBI: Oxyliquit draft tip
Message-ID: <389c43e40708310050o2adfb056s75ab532208934166@mail.gmail.com>

Hi,

To use DBI there is some work that needs to be done to install the
drivers.  I thought to make a Oxyliquit tip about this - I'd
appreciate feedback on the script below.

The draft below prepares for sqlite, mysql and odbc.
It'd be useful to know if it is possible to change DBNAME (used twice)
to install postgres , Oracle, any other drivers?

Before placing anything on Oxyliquit it would be useful to confirm
which distributions the following 'script' succeeds on - the location
of the odbc library might be an issue if it is not in /usr/lib....
other issues?
I've tested the following on openSUSE 10.2

My understanding is that DBI and ruby-ODBC are packaged in windows
one-click installer - can some one confirm that?

TIA
Mark

#-- start-bash ----------------------------
su
DBNAME=mysql
ODBCVER=0.9995
DBIVER=0.1.1
pushd /tmp
wget http://ch-werner.de/rubyodbc/ruby-odbc-${ODBCVER}.tar.gz
tar -xvf ruby-odbc-${ODBCVER}.tar.gz
pushd ruby-odbc-${ODBCVER}
ruby extconf.rb --enable-dlopen
make
make install
#Installation of utf8 version:
ruby -Cutf8 extconf.rb --enable-dlopen
make -C utf8
make -C utf8 install
popd

gem install ${DBNAME} sqlite
wget http://rubyforge.org/frs/download.php/12368/dbi-${DBIVER}.tar.gz
tar -xvf dbi-${DBIVER}.tar.gz
pushd ruby-dbi
ruby setup.rb config --with=dbi,dbd_sqlite,dbd_odbc,dbd_${DBNAME}
ruby setup.rb install
popd
popd
#-- end-bash ----------------------------

From mvyver at gmail.com  Fri Aug 31 05:32:36 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Fri, 31 Aug 2007 19:32:36 +1000
Subject: [Nitro] Nitro repo web page darcs instructions: Clarification
	suggestion.
Message-ID: <389c43e40708310232hdabdc8el8b80b4261a2a3935@mail.gmail.com>

On the nitro repo web site page:
http://www.nitroproject.org/repo

The current advice/hint is:

<hint>
 # darcs pull

You can go on and implemented your changes (while following the
project's coding conventions). Make sure that your change does not
break the the test cases. When you are ready with your changes it is
time to create your patch bundle. Give:

 # darcs send -o bundle
</hint>


In fact this should be:

<hint>
 # darcs pull

You can go on and implemented your changes (while following the
project's coding conventions). Make sure that your change does not
break the the test cases. When you are ready with your changes it is
time to create your patch bundle.  Then, in case you have added files
or directories run:

 # darcs add * -r

In all cases you must then run (and answer when prompted):

# darcs record
# darcs send -o bundle
</hint>

This now should be enough to allow anyone to send a quick patch
without banging their head on a darcs wall.

HTH
Mark

From mvyver at gmail.com  Fri Aug 31 08:09:08 2007
From: mvyver at gmail.com (Mark Van De Vyver)
Date: Fri, 31 Aug 2007 22:09:08 +1000
Subject: [Nitro] OG vs some other ORM packages
In-Reply-To: <f4cd26df0708260712r4bcc2acfr72ec5f9b29dabfa1@mail.gmail.com>
References: <389c43e40708252341r1c3f7870y6e587bbb8a353913@mail.gmail.com>
	<f4cd26df0708260712r4bcc2acfr72ec5f9b29dabfa1@mail.gmail.com>
Message-ID: <389c43e40708310509ocf9769bi9e43116d876101a2@mail.gmail.com>

Hi Kirk,

On 8/27/07, Kirk Haines <wyhaines at gmail.com> wrote:
> On 8/25/07, Mark Van De Vyver <mvyver at gmail.com> wrote:
>
> > Kansas (DBI)
>
> Kansas is my baby.  It's been around for a long time and is basically
> quite stable.  It uses DBI for the underlying db connectivity instead
> of reinventing the wheel there.

I've some questions about DBI and notice you're also listed as one of
its developers/maintainers, as well as the force behind DBI2!

Q1) Is there a definitive list of the DBD 'names' that are used to
make up DBIs' dsn?
On my system I have (/usr/lib/ruby/site_ruby/1.8/DBD/*) Mysql, ODBC,
but that is just 2/13 :)
I'd like to provide some documentation that is concrete for 'newbies',
i.e. if you've installed the ruby driver for db X, then to connect to
OG via the DBI adapter you should provide :dbi_driver => 'Xdrv' when
connecting to OG - the DBD (0.2.2 draft) documentation indicates this
'name' maybe case sensitive....?

Q2) Looking at DBIs project pages on Ruby-forge, there seem to be many
open bugs and patches, have all these really not been fixed/applied in
the latest release (August 12, 2006)?

Appreciate any light you can shed.

Regards
Mark


> It takes the approach of creating Ruby classes that correspond to
> database tables, and it tries to be fairly lightweight.  The code is
> not particularly large.
>
> It uses Ruby syntax to describe queries:
>
> sodiums = dbh.select(:Chemicals) {|c| c.name =~ '%sodium%'}
> critical_issues = dbh.select(:Roofs) {|r| r.sort_by(r.severity =>
> 'DESC'); r.severity > 3}
>
> It has transaction support that is not dependent on the database
> layer.  Queries can return results in normal ruby types such as an
> Array or Hash instead of the db-coupled Kansas result types.  The
> db-coupled objects can also be marked as read-only, so that changes to
> them do not get propagated back to the db.  It supports creating
> relationships between tables, for conveniences like:
>
> school = dbh.select(:Schools) {|s| s.name = school_we_are_looking_for }.first
> school_cyanides = school.inventory.select {|i| i.name =~ /cyanide/}
> art_inventory = school.departments.select {|d| d.name == 'Art'}.chemicals
>
> That's a basic high level view.  I have used it on dozens and dozens
> of sites over the years, and it just works.  I have a small community
> of users, and insufficient documentation, but it is in daily, active
> use, if not daily, active development.
>
>
> Kirk Haines
>

From transfire at gmail.com  Fri Aug 31 11:19:56 2007
From: transfire at gmail.com (Trans)
Date: Fri, 31 Aug 2007 15:19:56 -0000
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708300436s1f4dbb15p81f329118b9b53e5@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708271239q5c335ceaxa0f94947a403bde6@mail.gmail.com>
	<1188250043.795585.43290@o80g2000hse.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
	<1188325587.907262.313400@k79g2000hse.googlegroups.com>
	<afd80b540708282123y61026d4ficd5443ff717f70e1@mail.gmail.com>
	<1188392200.409137.173050@g4g2000hsf.googlegroups.com>
	<afd80b540708290714r327a270bu5d1a284850fcff1d@mail.gmail.com>
	<1188408528.318463.192760@r34g2000hsd.googlegroups.com>
	<afd80b540708300436s1f4dbb15p81f329118b9b53e5@mail.gmail.com>
Message-ID: <1188573596.260866.58190@19g2000hsx.googlegroups.com>



On Aug 30, 4:36 am, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> I find the second solution more natural (what matz descibes in the provided
> link) more natural.
> This is the behaviour I would expect. But, even the first solution may be ok
> in 90% of the cases.
> Unless theres is a considerable performance/complexity penalty, lets go for
> the second solution. Then we can switch to the official solution when Ruby
> 1.9 is finally released.

Okay, I have a good working solution for before and after advice.
Unfortunately, around advice are a real beatch. The problem is with
defining a special method (#super) inside the around advice's binding
(the advice is a Proc object). The idea being that one can just do:

  around :x do |*args|
    ...
    super
    ...
  end

Unfortunately I can't find a way to do this. And the only other
solution seems to be to pass in the "target":

  around :x do |target, *args|
    ...
    target.super
    ...
  end

But I don't like that --it's fine for a major aop framework, but not
for these light-weight advice wraps (at least IMHO).

T :(


From george.moschovitis at gmail.com  Fri Aug 31 12:40:10 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 31 Aug 2007 19:40:10 +0300
Subject: [Nitro] Aspects
In-Reply-To: <1188573596.260866.58190@19g2000hsx.googlegroups.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
	<1188325587.907262.313400@k79g2000hse.googlegroups.com>
	<afd80b540708282123y61026d4ficd5443ff717f70e1@mail.gmail.com>
	<1188392200.409137.173050@g4g2000hsf.googlegroups.com>
	<afd80b540708290714r327a270bu5d1a284850fcff1d@mail.gmail.com>
	<1188408528.318463.192760@r34g2000hsd.googlegroups.com>
	<afd80b540708300436s1f4dbb15p81f329118b9b53e5@mail.gmail.com>
	<1188573596.260866.58190@19g2000hsx.googlegroups.com>
Message-ID: <afd80b540708310940v1b20addfxbefd4714343180e9@mail.gmail.com>

>
>   around :x do |target, *args|
>     ...
>     target.super
>     ...
>   end
>


I don't find this too ugly. As long as the performance is "right". It seems
to be faster than the 'non-target' version.
W/o source code I cannot offer further  advices (no pun intended).

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070831/ddf15843/attachment.html 

From george.moschovitis at gmail.com  Fri Aug 31 12:41:41 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 31 Aug 2007 19:41:41 +0300
Subject: [Nitro] Nitro repo web page darcs instructions: Clarification
	suggestion.
In-Reply-To: <389c43e40708310232hdabdc8el8b80b4261a2a3935@mail.gmail.com>
References: <389c43e40708310232hdabdc8el8b80b4261a2a3935@mail.gmail.com>
Message-ID: <afd80b540708310941y395c5c12o861c3dd6a2272a85@mail.gmail.com>

Ok, thanks, I will update the text accordingly.

-g.


On 8/31/07, Mark Van De Vyver <mvyver at gmail.com> wrote:
>
> On the nitro repo web site page:
> http://www.nitroproject.org/repo
>
> The current advice/hint is:
>
> <hint>
> # darcs pull
>
> You can go on and implemented your changes (while following the
> project's coding conventions). Make sure that your change does not
> break the the test cases. When you are ready with your changes it is
> time to create your patch bundle. Give:
>
> # darcs send -o bundle
> </hint>
>
>
> In fact this should be:
>
> <hint>
> # darcs pull
>
> You can go on and implemented your changes (while following the
> project's coding conventions). Make sure that your change does not
> break the the test cases. When you are ready with your changes it is
> time to create your patch bundle.  Then, in case you have added files
> or directories run:
>
> # darcs add * -r
>
> In all cases you must then run (and answer when prompted):
>
> # darcs record
> # darcs send -o bundle
> </hint>
>
> This now should be enough to allow anyone to send a quick patch
> without banging their head on a darcs wall.
>
> HTH
> Mark
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070831/39fa0574/attachment.html 

From john at oxyliquit.de  Fri Aug 31 16:37:59 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Fri, 31 Aug 2007 22:37:59 +0200
Subject: [Nitro] OG and ruby-dbi
In-Reply-To: <afd80b540708300255y7ccaf6f4i6f2005043fab2e6d@mail.gmail.com>
References: <389c43e40708272350v10a060e8k7f6f9df886c45eec@mail.gmail.com>
	<op.txsgfrvdxegpx6@jo.local>
	<389c43e40708282127n3381f33l3e66d8b9c377f81f@mail.gmail.com>
	<389c43e40708282236l23a66afelc5652f86a3bb96a3@mail.gmail.com>
	<op.txu6g9hwxegpx6@jo.local>
	<afd80b540708300255y7ccaf6f4i6f2005043fab2e6d@mail.gmail.com>
Message-ID: <op.txxwtyhwxegpx6@jo.local>

Hi,

> Don't worry, they are in a hidden directory .old_tests. I will copy them.

I know, I only questioned the act itself.  What was wrong with the
structure built by Arne that this calls for such a movement?

Jo

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

From john at oxyliquit.de  Fri Aug 31 16:38:02 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Fri, 31 Aug 2007 22:38:02 +0200
Subject: [Nitro] Testing the whole website with GET/POST
In-Reply-To: <367301c7eacc$df9b18a0$6442a8c0@musicbox>
References: <367301c7eacc$df9b18a0$6442a8c0@musicbox>
Message-ID: <op.txxxljnfxegpx6@jo.local>

Hi,

> Including techniques such as vcr_record / vcr_playback, I was
> wondering if anyone might be able to share tips for testing
> a site "externally"... ?
>
> I'd kind of like to be able to exercise the site logic from
> the point of view of a "browser".
>
> Anyone have any tips on that?  (Or alternative approaches?)

I use a slightly alternative approach.  For external tests I
have recently used WWW::Mechanize.  It's really good at 'emulating'
a browser.  Testing like that makes you work on your actual
html output to produce good easy html, so it can be easily screen-
scraped.
I really liked it, it doesn't take too much work to recreate a
'browser visit'.  Another benefit is, it makes sure your app also
works with '1.0' type browsers, not having javascript.  ;)

Downsides, not sure yet, I've once run into a testcase which
mysteriously failed due to whatever, where a real browser
didn't react that way, there was no special behaviour there,
unsure how that could happen.

You might want to play with it.

Jo

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

From john at oxyliquit.de  Fri Aug 31 16:38:06 2007
From: john at oxyliquit.de (Jonathan Buch)
Date: Fri, 31 Aug 2007 22:38:06 +0200
Subject: [Nitro] Nitro
In-Reply-To: <9e03c3c60708282319i17741d0cv6f8e96f27d71fca@mail.gmail.com>
References: <afd80b540708240043xaba49beve9150d31841139af@mail.gmail.com>
	<op.txkr9gwhxegpx6@jo.local>
	<afd80b540708241106x53487fccy9c9b17e3a3231391@mail.gmail.com>
	<op.txl77lwtxegpx6@jo.local>
	<afd80b540708270829m299677abuce35433afef8d34f@mail.gmail.com>
	<op.txsgyqzaxegpx6@jo.local>
	<9e03c3c60708282319i17741d0cv6f8e96f27d71fca@mail.gmail.com>
Message-ID: <op.txxx8emkxegpx6@jo.local>

Hi,

> Honestly I don't "believe" that modular it the distinguishing factor.  Clear
> and Concise would be better terms.
>
> Lots of things can be "modular"  with no functional benefit.

I can certainly agree to this.

> Discrete and easy to understand would not require "modules".  It is just
> nicer to package things in units (or containers).
>
> I think we see this a lot, a "buzz word" makes a statement and we tend to
> make it a description.  Really, it is a label for a string of attributes
> that fit-well-with modular (or the particular buzz word).

Let me explain how the Ramaze logging looks like:

loggers:  Informer, LogHub, Analogger, Syslog, Growl (and one I forgot)
where all of them adhere to a certain interface.

these all can be set as the main logger, for example:

* Ramaze::Global.logger = Informer.new
* .. = LogHub.new(Informer.new, Growl.new)

Maybe you're right in that this might be better described as 'clear' instead
of modular.

Me being a non-native, I might have a mis-conception on if that is really
'modules' or not, or if it even classifies as 'modular'.
Do I misuse the word if I use it for 'single pieces can be ripped out and
completely replaced by something else'?

Maybe it's just allways been abused or I'm just generally confused.  ;)

Jo

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

From billk at cts.com  Fri Aug 31 16:50:34 2007
From: billk at cts.com (Bill Kelly)
Date: Fri, 31 Aug 2007 13:50:34 -0700
Subject: [Nitro] Testing the whole website with GET/POST
References: <367301c7eacc$df9b18a0$6442a8c0@musicbox>
	<op.txxxljnfxegpx6@jo.local>
Message-ID: <037301c7ec10$93ba9920$6442a8c0@musicbox>


From: "Jonathan Buch" <john at oxyliquit.de>
> 
>> Including techniques such as vcr_record / vcr_playback, I was
>> wondering if anyone might be able to share tips for testing
>> a site "externally"... ?
>>
>> I'd kind of like to be able to exercise the site logic from
>> the point of view of a "browser".
>>
>> Anyone have any tips on that?  (Or alternative approaches?)
> 
> I use a slightly alternative approach.  For external tests I
> have recently used WWW::Mechanize.  It's really good at 'emulating'
> a browser.  Testing like that makes you work on your actual
> html output to produce good easy html, so it can be easily screen-
> scraped.
> I really liked it, it doesn't take too much work to recreate a
> 'browser visit'.  Another benefit is, it makes sure your app also
> works with '1.0' type browsers, not having javascript.  ;)
> 
> Downsides, not sure yet, I've once run into a testcase which
> mysteriously failed due to whatever, where a real browser
> didn't react that way, there was no special behaviour there,
> unsure how that could happen.
> 
> You might want to play with it.

Cool, thanks, good to hear.

Indeed, I just installed WWW:Mechanize last night, and was
impressed that when I fetched the home page for my site,
it did all the cookie handling, and followed the redirects,
and wound up exactly at the page it should have.


Regards,

Bill




From george.moschovitis at gmail.com  Fri Aug 31 18:24:59 2007
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 1 Sep 2007 01:24:59 +0300
Subject: [Nitro] Testing the whole website with GET/POST
In-Reply-To: <op.txxxljnfxegpx6@jo.local>
References: <367301c7eacc$df9b18a0$6442a8c0@musicbox>
	<op.txxxljnfxegpx6@jo.local>
Message-ID: <afd80b540708311524o22927a00rc96cc19101abb634@mail.gmail.com>

>
> I use a slightly alternative approach.  For external tests I
> have recently used WWW::Mechanize.  It's really good at 'emulating'
> a browser.  Testing like that makes you work on your actual
> html output to produce good easy html, so it can be easily screen-
> scraped.
>


Very good idea! I will try this as well. This would be a greate candidate
for an oxyliquid tip.

-g.

-- 
http://www.me.gr
http://phidz.com
http://blog.gmosx.com
http://cull.gr
http://www.joy.gr
http://nitroproject.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20070901/e98b30ab/attachment.html 

From transfire at gmail.com  Fri Aug 31 20:15:49 2007
From: transfire at gmail.com (Trans)
Date: Sat, 01 Sep 2007 00:15:49 -0000
Subject: [Nitro] Aspects
In-Reply-To: <afd80b540708310940v1b20addfxbefd4714343180e9@mail.gmail.com>
References: <1187969361.601995.195500@q5g2000prf.googlegroups.com>
	<afd80b540708272228qbd94287tee2e99abe1136fba@mail.gmail.com>
	<afd80b540708272228s33a4e6b4k360ebc5ca29378e@mail.gmail.com>
	<1188325587.907262.313400@k79g2000hse.googlegroups.com>
	<afd80b540708282123y61026d4ficd5443ff717f70e1@mail.gmail.com>
	<1188392200.409137.173050@g4g2000hsf.googlegroups.com>
	<afd80b540708290714r327a270bu5d1a284850fcff1d@mail.gmail.com>
	<1188408528.318463.192760@r34g2000hsd.googlegroups.com>
	<afd80b540708300436s1f4dbb15p81f329118b9b53e5@mail.gmail.com>
	<1188573596.260866.58190@19g2000hsx.googlegroups.com>
	<afd80b540708310940v1b20addfxbefd4714343180e9@mail.gmail.com>
Message-ID: <1188605749.319240.254470@o80g2000hse.googlegroups.com>



On Aug 31, 9:40 am, "George Moschovitis"
<george.moschovi... at gmail.com> wrote:
> >   around :x do |target, *args|
> >     ...
> >     target.super
> >     ...
> >   end
>
> I don't find this too ugly. As long as the performance is "right". It seems
> to be faster than the 'non-target' version.
> W/o source code I cannot offer further  advices (no pun intended).

I wish there was another way, but I can't seem to find one, so I just
went ahead with the target solution. What irks me about it though is
1) it's not compatible with Matz version, 2) one starts to expect
target to offer more meta-information, like method name, arity,
__LINE__, binding, etc. and 3) then you think before and after advice
should passed the target object too. That's why I say it makes more
sense for a full-blown AOP deal, but not so much for a light-weight
advice solution. Well, Maybe I'll give in and just supply a full-blown
taget to all of them. In the mean time here the solution as it
currently stands (I'll attach it my next post).

T.


From transfire at gmail.com  Fri Aug 31 20:24:43 2007
From: transfire at gmail.com (TRANS)
Date: Fri, 31 Aug 2007 17:24:43 -0700
Subject: [Nitro] advice.rb
Message-ID: <4b6f054f0708311724n35b8ba91r42323cc630d86aa7@mail.gmail.com>

It's called it advice.rb (though maybe it should be renamed
advisable.rb) and you can find it in the current SVN repo.

I changed the name from aspects.rb b/c a) the Aspect/Aspects classes
are no longer needed and b) the name clashes with the class now in
aop.rb --which is a full-blown heavy-wieght cuts-based AOP solution
(work in progress).

For advice.rb to work you also need 1stclassmethod.rb and
unboundmethod/arguments. I've attached them, but it may be easier just
to pull down the SVN repo.

T.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: advice.rb
Type: application/x-ruby
Size: 3248 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070831/e4024133/attachment.bin 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: arguments.rb
Type: application/x-ruby
Size: 365 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070831/e4024133/attachment-0001.bin 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 1stclassmethod.rb
Type: application/x-ruby
Size: 3537 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20070831/e4024133/attachment-0002.bin