From transfire at gmail.com Sat Mar 3 20:06:09 2007 From: transfire at gmail.com (Trans) Date: Sun, 04 Mar 2007 01:06:09 -0000 Subject: [Facets] On subproject's for Facets In-Reply-To: <1172166568.983148.317630@j27g2000cwj.googlegroups.com> References: <1172166568.983148.317630@j27g2000cwj.googlegroups.com> Message-ID: <1172970369.628699.121320@8g2000cwh.googlegroups.com> On Feb 22, 12:49 pm, "Trans" wrote: > For Facets 2.0 I've been thinking a log time about offering the More > library in individual pieces. That way people can download and install > only the pieces they need or want --if that suits them better, rather > then install the whole More library. The parts will remain compatible > with the whole too i.e they install to the same system locations. And > I emphasis that the whole lib will still be installable in one pass > just as it is now --I don't want to get rid of that, but I think it > would be beneficial to offer the option of smaller pieces for a couple > of reasons. Namely, have each piece will give greater exposure to > these libs in the Gem namespace and Rubyforge Download logs. It will > also make it easier to version control each lib since they can have > there own subprojects. > > So my first query is, what do others think about this. Is it worth the > effort? > > My next question is how to break them up? Should I do so at the finest > level, where by nearly every More file is it's own project (with > exceptions like ann.rb & ann_attr.rb of course). Or should I try to > break it up into categories, for instance all the console related libs > in one subproject? > > My biggest concern about this however is dependencies. If one of these > libs depends on another, then which lib should it depend on in the > spec: Facets as a whole or the particular subproject? Is there a way > to specify either/or in gem dependencies? (I don't think there is). > Thankfully there are very few dependencies, with the exception of the > core extensions. So it shouldn't be too much of an issue, but it is a > problem nonetheless. > > The last issue with this is the naming, I suppose Core will be one > subproject called facets-core, and the other subproject will have > names like facets-annotations, facets-heap, etc. ? Though I point out > that Facets itself used to be ONLY the core lib and the More libs were > parts in another project called Calibre (you can still see vestiges of > this if you look at list of available gems, i've been holding off on > deleting those until I worked this out for Facets). > > Please (with surgar on top) let me know what opinions y'all have on > this matter. Well, I've been working on this, and mostly have it all in place. One question I have is whether to release the subpakages with names prefixed by 'facets' or not. For example, should I release facets' annotations lib as just 'annotations' or should it be 'facets- annotations' (or 'facets_annotations'). On one hand hand, I sort-of prefer it without the prefix, it's shorter and other projects like seattle.rb don't preifx their libs either. On the other hand, the prefix clearly designates it a part of facets. And if the former, another issue is what to do if a name is already taken? What do you think? T. From john at oxyliquit.de Mon Mar 5 02:27:16 2007 From: john at oxyliquit.de (Jonathan Buch) Date: Mon, 05 Mar 2007 09:27:16 +0200 Subject: [Facets] On subproject's for Facets In-Reply-To: <1172970369.628699.121320@8g2000cwh.googlegroups.com> References: <1172166568.983148.317630@j27g2000cwj.googlegroups.com> <1172970369.628699.121320@8g2000cwh.googlegroups.com> Message-ID: Hi, > On one hand hand, I sort-of > prefer it without the prefix, it's shorter and other projects like > seattle.rb don't prefix their libs either. On the other hand, the > prefix clearly designates it a part of facets. And if the former, > another issue is what to do if a name is already taken? > > What do you think? I think it's "natural" to prefer ones own libs without prefixes, as they put all libs (kinda like right now) in a single 'bucket' which doesn't really help to promote the single lib. But, I right now I think I favour the prefix approach, mainly due to facets being THE support lib (did I say I love .humanize?) for me and facets just being a cool name. :) Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From transfire at gmail.com Thu Mar 8 18:10:28 2007 From: transfire at gmail.com (Trans) Date: Thu, 08 Mar 2007 23:10:28 -0000 Subject: [Facets] [ANN] Facets / Multiton 2.0.0 Message-ID: <1173395428.885556.123030@p10g2000cwp.googlegroups.com> Facets / Multiton 2.0.0 Implementation of the Multiton Design Pattern by Ara T. Howard and Trans http://facets.rubyforge.org [ S Y N O P S I S ] Multiton is similar to the Singleton, but rather than a single object it pools a cached set corresponding to the initialization parameters, ie. initialize the class again with the same parameters and get the same object. [ I N S T A L L ] Please note Facets/Multiton is already included in the complete Facets package. You do not need this if you already have Facets installed. You can install this library via gem or tarball. Eg. gem install facets_multition or wget http://rubyforge.org/frs/download.php/18259/facets_multiton-2.0.0.tgz tar -xvzf facets_multiton-2.0.0.tgz cd facets_multiton-2.0.0 sudo ruby setup.rb [ C R E D I T ] Ara T. Howard (Original Code) Trans (Improvements) [ L I C E N S E ] Facets/Multiton Copyright (c) 2005-2007 Ara T. Howard, Trans All Right Reserved Distributed under the terms of the Ruby/GPL license. From transfire at gmail.com Thu Mar 8 18:11:24 2007 From: transfire at gmail.com (Trans) Date: Thu, 08 Mar 2007 23:11:24 -0000 Subject: [Facets] On subproject's for Facets In-Reply-To: References: <1172166568.983148.317630@j27g2000cwj.googlegroups.com> <1172970369.628699.121320@8g2000cwh.googlegroups.com> Message-ID: <1173395484.972404.312740@v33g2000cwv.googlegroups.com> On Mar 5, 2:27 am, "Jonathan Buch" wrote: > Hi, > > > On one hand hand, I sort-of > > prefer it without the prefix, it's shorter and other projects like > > seattle.rb don't prefix their libs either. On the other hand, the > > prefix clearly designates it a part of facets. And if the former, > > another issue is what to do if a name is already taken? > > > What do you think? > > I think it's "natural" to prefer ones own libs without prefixes, as > they put all libs (kinda like right now) in a single 'bucket' which > doesn't really help to promote the single lib. > > But, I right now I think I favour the prefix approach, mainly due to > facets being THE support lib (did I say I love .humanize?) for me > and facets just being a cool name. :) > > Jo As you will soon notice. I took your advice. Thanks, T. From transfire at gmail.com Sat Mar 10 17:37:26 2007 From: transfire at gmail.com (Trans) Date: Sat, 10 Mar 2007 22:37:26 -0000 Subject: [Facets] [ANN] Facets / Annotations 2.0.0 Message-ID: <1173566246.379726.231270@v33g2000cwv.googlegroups.com> = Facets/Annotations A Complete In-code Annotations System by Trans and George Moschovitis http://facets.rubyforge.org == Synopsis Annotations provides a complete and easy to use system of annotating methods and arbitrary objects. == What's New This is Facets/Annotations first release as an independent project. As with all Facets' subprojects it is included in the complete facets package. So if you 'facets' installed you do not need this. Version 2.0+ is a major simplification over 1.0 series. Overall speed is much faster and the notation more consistent. Where before one would use "ann.attr_name.ann_name", now it's ann(:attr_name, :ann_name). This change was inspired by Ara T. Howard. == Example class X attr :foo, String, :default => "bar" end X.ann :foo, :class #=> String X.ann :foo, :default #=> "bar" X.ann :foo, :default => "baz" X.ann :foo, :default #=> "baz" == Installation IMPORTANT! This library is included in the complete facets package. You do not this library if you already have facets installed. You can install either via gem or tarball. gem install facets_annotations or wget http://rubyforge.org/frs/download.php/18317/facets_annotations-2.0.0.tgz cd facets_annotations sudo ruby setup.rb == Documentation Please see doc/ directory for additional documentation. == Credit Trans [Implementation] George Moschovitis [Original Concept] Ara T. Howard [Suggestions] == License Copyright (c) 2004-2007 Thomas Sawyer, George Moschovitis Distributed under the Ruby/GPL dual-license. From transfire at gmail.com Fri Mar 16 00:27:51 2007 From: transfire at gmail.com (Trans) Date: Fri, 16 Mar 2007 04:27:51 -0000 Subject: [Facets] Goldilocks Message-ID: <1174019271.069536.97440@l75g2000hse.googlegroups.com> My primary goal for Facets 2.0 is to divide the "More" set of libraries into useful groups. The hope of this is two fold. It should make development a bit easier and it should make the libs more accessible to potential users. At first I set out simply to separate each and every lib into it's own mini-project, but this turned out to be too fine. So I came at it from the opposite direction, and tried separating out just the items I was sure should have their own sub- project. This turned out to be too course --I still had a rather large motley set remaining. So here I am with my Goldilocks dilemma, figuring out sub-projects that are "just right". I've done pretty well so far, but there's still some play I haven't yet been able to bring to conclusion. So I would like to give a rundown and see what ideas others might have for putting Goldilocks to bed. If you look at my subproject directory currently you'll see this: (I've added some notes of my considerations too) annotations aop bbcode # too versatile to be in markup? binaryreader # ??? IO-releated? bitmask # ??? buildingblock cache cargo # container classes like PQueue and LinkedList console core # core extensions coroutine # ??? crypt # also includes doublemetaphone dictionary # move to cargo? enumerable2 # ??? filesystem httpaccess ioredirect # maybe create IO-related subproject? json # okay to consider this a microformat? lazy markup # markup libs like HtmlHelper math meta # metaprogramming like overload.rb microformats # xoxo, hcard, etc. minis # odd assortment of tiny libs, eg. expirable multiton net # network related libs (need better name?) openobject # maybe in cargo but has three libs recorder reflection rtals script # rename import, move to meta? semaphore # ??? snapshot stateparser system tagiterator # move to markup? timer # ??? tkxml # move to markup? typecast units version_number As you can see I still have a few (???) that I have no clear idea how to group, and on a few others I'm debating. General considerations of the whole affair are welcome too, of course. T. From transfire at gmail.com Fri Mar 16 00:58:46 2007 From: transfire at gmail.com (Trans) Date: Fri, 16 Mar 2007 04:58:46 -0000 Subject: [Facets] Goldilocks In-Reply-To: <1174019271.069536.97440@l75g2000hse.googlegroups.com> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> Message-ID: <1174021126.515405.145240@l75g2000hse.googlegroups.com> I should add one last thing. Some of these sub-projects have "cute" names which other have very technical names. Opinions on which is best? T. From transfire at gmail.com Fri Mar 16 11:36:29 2007 From: transfire at gmail.com (Trans) Date: Fri, 16 Mar 2007 15:36:29 -0000 Subject: [Facets] Goldilocks In-Reply-To: <1174021126.515405.145240@l75g2000hse.googlegroups.com> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <1174021126.515405.145240@l75g2000hse.googlegroups.com> Message-ID: <1174059389.429486.215810@o5g2000hsb.googlegroups.com> This process is starting to take a very different tact. As I work on it, the libs start to speak for themselves. It's becoming clear to me what really needs to happen. And it's very nice I think, but it does mean a big change. Essentially certain libs "cry out" for there own package, while others are more at home in a collection. There will be two main collection projects, one for the extension methods and one for small/common libs. Then there will be an number of other islet projects for those that are significant enough. So rather then all of the libs in a single package they will each have there own packages, each can be installed independently if you want, and I will use a package aggregator (which I will write) that will take separate packages and merge them into a single package, so we can install everything in one go as we do now too. I think this is the best way to go. Although it is a big change, it will make it easier handle the core extensions and the "more" additions and support external libs just as easily --the whole being a sort of "best-of-breed" collection, containing the core lib, more lib, and the other "islet" projects. This change will require a modified naming scheme. The separated packages don't have a single package to name, but in general I think I'll refer to them as the Islets. As for the others, it was tempting to renaming the whole unified package "Ruby Stars", and just calling the core lib Facets, but it's probably best to stick with Facets for the whole. So instead I came up with core => Corsets more => Garnets Of course I suspect I'll hear the suggestion: core => Core more => More But I don't think I want a project that's just called "More". It has no semantic value in itself. Plus what would I call the Islets? EvenMore? ;) In any case, that's the upshot of what's in store for "Facets 2.O". T. From transfire at gmail.com Sat Mar 17 10:19:43 2007 From: transfire at gmail.com (Trans) Date: Sat, 17 Mar 2007 14:19:43 -0000 Subject: [Facets] Goldilocks In-Reply-To: <1174059389.429486.215810@o5g2000hsb.googlegroups.com> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <1174021126.515405.145240@l75g2000hse.googlegroups.com> <1174059389.429486.215810@o5g2000hsb.googlegroups.com> Message-ID: <1174141183.609508.57510@o5g2000hsb.googlegroups.com> Anyone got a gun? I'm going to shoot Goldilocks and than myself. From john at oxyliquit.de Sat Mar 17 15:21:22 2007 From: john at oxyliquit.de (Jonathan Buch) Date: Sat, 17 Mar 2007 21:21:22 +0200 Subject: [Facets] Goldilocks In-Reply-To: <1174141183.609508.57510@o5g2000hsb.googlegroups.com> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <1174021126.515405.145240@l75g2000hse.googlegroups.com> <1174059389.429486.215810@o5g2000hsb.googlegroups.com> <1174141183.609508.57510@o5g2000hsb.googlegroups.com> Message-ID: Hi, > Anyone got a gun? I'm going to shoot Goldilocks and than myself. uhm... don't... please? ;) About Corsets and Garnets... I don't really like those names, and like you suspected the first thing that came to my mind is 'Core'. :P How about using 'CoreFacets'. :P But now I get the feeling that the packaging is actually coming along better now and that it's just the naming problem again. So, keep up the good work! Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From transfire at gmail.com Sat Mar 17 16:54:19 2007 From: transfire at gmail.com (Trans) Date: Sat, 17 Mar 2007 20:54:19 -0000 Subject: [Facets] Goldilocks In-Reply-To: References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <1174021126.515405.145240@l75g2000hse.googlegroups.com> <1174059389.429486.215810@o5g2000hsb.googlegroups.com> <1174141183.609508.57510@o5g2000hsb.googlegroups.com> Message-ID: <1174164859.223677.165940@b75g2000hsg.googlegroups.com> On Mar 17, 3:21 pm, "Jonathan Buch" wrote: > Hi, > > > Anyone got a gun? I'm going to shoot Goldilocks and than myself. > > uhm... don't... please? ;) > > About Corsets and Garnets... I don't really like those names, > and like you suspected the first thing that came to my mind is > 'Core'. :P > How about using 'CoreFacets'. :P > > But now I get the feeling that the packaging is actually coming > along better now and that it's just the naming problem again. > So, keep up the good work! :) It seems you have gotten to know me well. And you are right, it is coming along. As much as I hate to say it, guns can be good motivators! ;) I've actually narrowed it down quite nicely. There will be no more "more", so facets/more/multiton -> facets/multiton As for the core, I am still a bit in debate. I may call the package "atomix" (or something like that, perhaps, "atom" or "atomics" or maybe even "atomic_core". Not sure. While I could just call it "core", I noticed there's already a copycat project out there called "core_ex". Don't think I want to be perceived as copying them, nor do I want any confusion between similar projects. In any case, the require will be either facets/core/string/plural -> facets/core/string/plural (no change) facets/core/string/plural -> facets/atom/string/plural or facets/core/string/plural -> facets/string/plural The later is certainly more convenient. Though the first options would mean few people will have to change their current require statements. Thanks for the encouragement, T. From transfire at gmail.com Tue Mar 20 00:54:23 2007 From: transfire at gmail.com (Trans) Date: Tue, 20 Mar 2007 04:54:23 -0000 Subject: [Facets] Goldilocks In-Reply-To: <1174164859.223677.165940@b75g2000hsg.googlegroups.com> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <1174021126.515405.145240@l75g2000hse.googlegroups.com> <1174059389.429486.215810@o5g2000hsb.googlegroups.com> <1174141183.609508.57510@o5g2000hsb.googlegroups.com> <1174164859.223677.165940@b75g2000hsg.googlegroups.com> Message-ID: <1174366463.046006.171120@l77g2000hsb.googlegroups.com> Well, I think I almost have it all worked out. I came to the conclusion that libs "begging" for there own projects need to have just that and I moved them out of Facets altogether. When I did that it became very clear that Facets consisted of the core extensions plus a set of very basic classes, modules and meta-libs. And the requires would be okay as: require "facets/string/plural" # core require "facets/multiton" # more Though I still wonder if I should use singular --after all some people are using this notation with the current version of facets. require "facet/string/plural" require "facet/multiton" The bigger issue however is that while I could have both "core" and "more" in a single package, it becomes harder to manage and document. It would be more convenient to have 2 packages. One for the extensions and another for the additions. One nice thing about this is that you can depend on the core extensions with out the additions if you don't need them. But what do I call the second package? Keep in mind, since the second depends on the first, it would also be the goto package for installing all of facets in one go. So I was thinking maybe calling it "facets+". I tested it out and it looks like the "+" character works on all file systems. Another option is "multifacets", though that does seem a bit much (reminds me of "Leeloo multi-pass" from the 5th Element :) Alternatively the first could be called "corefacets" and the later "facets". My only hesitation here is that I consider "facet" as basically synonymous with "extension method". So the addition classes and modules aren't "facets" per se, but, well bonus material. So the names seem almost backwards to me. But I suppose I'm just being a stickler for definition here and should give over it. Anyhow, I'm pretty close. Just these last details to iron out. Oh, you might be wondering what libs I took out of facets to give there own projects?... The console tools, rtals, tkxml, the aop- related libs, httpaccess, units, bbcode, the markup helpers w/ builders, and a few others. These all had either very selective use cases or were developing into more complex systems --hence the "begging". T. From noreply at rubyforge.org Wed Mar 21 09:15:39 2007 From: noreply at rubyforge.org (noreply at rubyforge.org) Date: Wed, 21 Mar 2007 09:15:39 -0400 (EDT) Subject: [Facets] [ facets-Bugs-9427 ] Test Tracker #2 Message-ID: <20070321131539.5FBE752409A9@rubyforge.org> Bugs item #9427, was opened at 2007-03-21 08:15 You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9427&group_id=804 Category: None Group: None Status: Open Resolution: None Priority: 3 Submitted By: Trans Onoma (transami) Assigned to: Nobody (None) Summary: Test Tracker #2 Initial Comment: This is the second test of the tracker system. ---------------------------------------------------------------------- You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9427&group_id=804 From noreply at rubyforge.org Wed Mar 21 09:44:22 2007 From: noreply at rubyforge.org (noreply at rubyforge.org) Date: Wed, 21 Mar 2007 09:44:22 -0400 (EDT) Subject: [Facets] [ facets-Bugs-9424 ] Testing Tracker Message-ID: <20070321134422.26A8E5240986@rubyforge.org> Bugs item #9424, was opened at 2007-03-21 07:44 You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9424&group_id=804 Category: None Group: None Status: Open Resolution: Accepted Priority: 3 Submitted By: Trans Onoma (transami) Assigned to: Nobody (None) Summary: Testing Tracker Initial Comment: This is a test of trackers email system. ---------------------------------------------------------------------- >Comment By: Trans Onoma (transami) Date: 2007-03-21 08:44 Message: Dear readers, I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. ---------------------------------------------------------------------- Comment By: Trans Onoma (transami) Date: 2007-03-21 08:12 Message: Dear readers, I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. ---------------------------------------------------------------------- Comment By: Trans Onoma (transami) Date: 2007-03-21 07:57 Message: Dear readers, I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. ---------------------------------------------------------------------- You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9424&group_id=804 From transfire at gmail.com Wed Mar 21 10:07:20 2007 From: transfire at gmail.com (Trans) Date: Wed, 21 Mar 2007 14:07:20 -0000 Subject: [Facets] [ facets-Bugs-9424 ] Testing Tracker In-Reply-To: <20070321134422.26A8E5240986@rubyforge.org> References: <20070321134422.26A8E5240986@rubyforge.org> Message-ID: <1174486040.788231.298250@e1g2000hsg.googlegroups.com> > I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. As you can see I finally got it working. T. From noreply at rubyforge.org Thu Mar 22 13:29:38 2007 From: noreply at rubyforge.org (noreply at rubyforge.org) Date: Thu, 22 Mar 2007 13:29:38 -0400 (EDT) Subject: [Facets] [ facets-Bugs-9424 ] Testing Tracker Message-ID: <20070322172938.309B1A970003@rubyforge.org> Bugs item #9424, was opened at 2007-03-21 07:44 You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9424&group_id=804 Category: None Group: None Status: Open Resolution: Accepted Priority: 3 Submitted By: Trans Onoma (transami) Assigned to: Nobody (None) Summary: Testing Tracker Initial Comment: This is a test of trackers email system. ---------------------------------------------------------------------- >Comment By: Trans Onoma (transami) Date: 2007-03-22 12:29 Message: Still trying to get tracker to list working properly... ---------------------------------------------------------------------- Comment By: Trans Onoma (transami) Date: 2007-03-22 11:53 Message: This is a further test of the bug tracker to mailing list forwarding. Something still not right... ---------------------------------------------------------------------- Comment By: Trans Onoma (transami) Date: 2007-03-21 08:44 Message: Dear readers, I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. ---------------------------------------------------------------------- Comment By: Trans Onoma (transami) Date: 2007-03-21 08:12 Message: Dear readers, I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. ---------------------------------------------------------------------- Comment By: Trans Onoma (transami) Date: 2007-03-21 07:57 Message: Dear readers, I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. ---------------------------------------------------------------------- You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9424&group_id=804 From noreply at rubyforge.org Thu Mar 22 08:50:29 2007 From: noreply at rubyforge.org (noreply at rubyforge.org) Date: Thu, 22 Mar 2007 08:50:29 -0400 (EDT) Subject: [Facets] [ facets-Bugs-9466 ] String margin error Message-ID: <20070322125029.81F8D5240960@rubyforge.org> Bugs item #9466, was opened at 2007-03-22 07:50 You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9466&group_id=804 Category: None Group: None Status: Open Resolution: None Priority: 3 Submitted By: Trans Onoma (transami) Assigned to: Nobody (None) Summary: String margin error Initial Comment: I have been getting errors from Facets when I use the margin facility: brains hgs 52 %> ./class_generator.rb /usr/local/lib/ruby/gems/1.8/gems/facets-1.8.54/lib/facets/core/string/margin.rb:16:in `margin': undefined method `[]' for nil:NilClass (NoMethodError) from ./class_generator.rb:353:in `process_arg' from ./class_generator.rb:309:in `each_pair' from ./class_generator.rb:309:in `process_arg' from ./class_generator.rb:191:in `initialize' from ./class_generator.rb:132:in `open' from ./class_generator.rb:132:in `initialize' from ./class_generator.rb:2079:in `new' from ./class_generator.rb:2079 brains hgs 53 %> cat -n /usr/local/lib/ruby/gems/1.8/gems/facets-1.8.54/lib/facets/core/string/margin.rb | body 10 20 10 # }.margin 11 # 12 #-- 13 # This may still need a bit of tweaking. 14 #++ 15 def margin(n=0) 16 d = /\A.*\n\s*(.)/.match( self )[1] 17 d = /\A\s*(.)/.match( self)[1] unless d 18 return '' unless d 19 if n == 0 20 gsub(/\n\s*\Z/,'').gsub(/^\s*[#{d}]/, '') brains hgs 54 %> It seems to me that match is only ever used if it will fail (it is a test). Line 17 expects possible failure of line 16 ("unless d"). When match fails, it returns nil. ----------------------------------------------------------- Regexp#match rxp.match(str) => matchdata or nil ------------------------------------------------------------------------ Returns a +MatchData+ object describing the match, or +nil+ if there was no match. This is equivalent to retrieving the value of the special variable +$~+ following a normal match. /(.)(.)(.)/.match("abc")[2] #=> "b" Nil does not have the [] method, so arguably the example in the docs is buggy as well. I suggest the following change in the patch below. However, I have a problem. Running setup will attempt a full install. Running setup.rb config setup.rb setup setup.rb test tells me that it has run 0 tests, with 0 failures and 0 errors. How do I get tests to run to be sure that this change doesn't break things before I install? --Hugh --- facets-1.8.54/lib/facets/core/string/margin.rb.orig 2007-02-20 15:39:38.000000000 +0000 +++ facets-1.8.54/lib/facets/core/string/margin.rb 2007-03-22 11:37:53.927850000 +0000 @@ -13,8 +13,8 @@ # This may still need a bit of tweaking. #++ def margin(n=0) - d = /\A.*\n\s*(.)/.match( self )[1] - d = /\A\s*(.)/.match( self)[1] unless d + d = ((/\A.*\n\s*(.)/.match( self )) || + (/\A\s*(.)/.match( self)))[1] return '' unless d if n == 0 gsub(/\n\s*\Z/,'').gsub(/^\s*[#{d}]/, '') ---------------------------------------------------------------------- You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9466&group_id=804 From noreply at rubyforge.org Thu Mar 22 12:53:16 2007 From: noreply at rubyforge.org (noreply at rubyforge.org) Date: Thu, 22 Mar 2007 12:53:16 -0400 (EDT) Subject: [Facets] [ facets-Bugs-9424 ] Testing Tracker Message-ID: <20070322165316.98C015240A53@rubyforge.org> Bugs item #9424, was opened at 2007-03-21 07:44 You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9424&group_id=804 Category: None Group: None Status: Open Resolution: Accepted Priority: 3 Submitted By: Trans Onoma (transami) Assigned to: Nobody (None) Summary: Testing Tracker Initial Comment: This is a test of trackers email system. ---------------------------------------------------------------------- >Comment By: Trans Onoma (transami) Date: 2007-03-22 11:53 Message: This is a further test of the bug tracker to mailing list forwarding. Something still not right... ---------------------------------------------------------------------- Comment By: Trans Onoma (transami) Date: 2007-03-21 08:44 Message: Dear readers, I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. ---------------------------------------------------------------------- Comment By: Trans Onoma (transami) Date: 2007-03-21 08:12 Message: Dear readers, I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. ---------------------------------------------------------------------- Comment By: Trans Onoma (transami) Date: 2007-03-21 07:57 Message: Dear readers, I've just tied Rubyforge's ticket tracker into the facets-universal mailing list. Up until November of last year this system was hardly ever used. So I rarely checked it. The pace has picked up since then, so I want to make sure all tickets come to my/our attention. ---------------------------------------------------------------------- You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9424&group_id=804 From transfire at gmail.com Sun Mar 25 12:58:44 2007 From: transfire at gmail.com (Trans) Date: Sun, 25 Mar 2007 16:58:44 -0000 Subject: [Facets] Goldilocks In-Reply-To: <1174366463.046006.171120@l77g2000hsb.googlegroups.com> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <1174021126.515405.145240@l75g2000hse.googlegroups.com> <1174059389.429486.215810@o5g2000hsb.googlegroups.com> <1174141183.609508.57510@o5g2000hsb.googlegroups.com> <1174164859.223677.165940@b75g2000hsg.googlegroups.com> <1174366463.046006.171120@l77g2000hsb.googlegroups.com> Message-ID: <1174841924.881592.297020@n59g2000hsh.googlegroups.com> I conclude that I have three options. In no particular order: Option A. Two projects. One is Facets which has two packages, the extension methods and the basic/general-purpose components. Both can be installed via one "multi"-package 'facets'. The second project would contain libs that stand on their own. Unsure of name for this second project though (Garnets?) Facets ace_extensions ace_components Garnets (?) annotations (better in ace_components?) filesys (better in ace_components?) markers (xml/xhtml helpers) tapistry (aop lib) microfiche (microformat libs) comrade (commandline lib) units flexcache rtals bbcode tkxml Option B. One Project, Facets, which contains all the libs. The core extensions package itself would also be called "facets" and the add-on components would be a different package, though not sure of name for that (adrets?). Facets facets (extensions lib) adrets (?) (components lib) annotations (better as component?) filesys (better as component?) markers (xml/xhtml helpers) microfiche (microformat libs) tapistry (aop lib) comrade (commandline lib) units flexcache rtals bbcode tkxml Option C. One project (Facets) where the packages are separate but more tightly coupled name wise. A few packages would need to go in another project though. Facets ace_extensions ace_components ace_annotations ace_filesystem ace_markups ace_microformats ace_aspects ace_commandline ace_units ace_cache Move to another project rtals bbcode tkxml Any preference. Any reason to prefer one solution over another? Thanks, T. From o.renaud at laposte.net Sun Mar 25 13:23:24 2007 From: o.renaud at laposte.net (Olivier Renaud) Date: Sun, 25 Mar 2007 19:23:24 +0200 Subject: [Facets] Goldilocks In-Reply-To: <1174841924.881592.297020@n59g2000hsh.googlegroups.com> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <1174366463.046006.171120@l77g2000hsb.googlegroups.com> <1174841924.881592.297020@n59g2000hsh.googlegroups.com> Message-ID: <200703251923.24843.o.renaud@laposte.net> Le dimanche 25 mars 2007 18:58, Trans a ?crit?: > I conclude that I have three options. In no particular order: > > Option A. Two projects. One is Facets which has two packages, the > extension methods and the basic/general-purpose components. Both can > be installed via one "multi"-package 'facets'. The second project > would contain libs that stand on their own. Unsure of name for this > second project though (Garnets?) > [...] Hi, For your interrogation, I'm afraid I don't know enough the components of Facets, to help you in your decision. But the name you are thinking about (Garnet) reminded me of a blog post I read recently about Rubinius. In short, the Rubinius guys will use the name Garnet to name the dialect of Ruby that generates C code (previously known as Cuby). Not a good thing to have two Ruby-related projects with the same name :s Here is the post : http://on-ruby.blogspot.com/2007/01/serial-rubinius-interview-episode-v.html -- Olivier Renaud From transfire at gmail.com Sun Mar 25 14:16:58 2007 From: transfire at gmail.com (Trans) Date: Sun, 25 Mar 2007 18:16:58 -0000 Subject: [Facets] Goldilocks In-Reply-To: <200703251923.24843.o.renaud@laposte.net> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <1174366463.046006.171120@l77g2000hsb.googlegroups.com> <1174841924.881592.297020@n59g2000hsh.googlegroups.com> <200703251923.24843.o.renaud@laposte.net> Message-ID: <1174846618.740683.168030@d57g2000hsg.googlegroups.com> On Mar 25, 1:23 pm, Olivier Renaud wrote: > Le dimanche 25 mars 2007 18:58, Trans a ?crit : > > > I conclude that I have three options. In no particular order: > > > Option A. Two projects. One is Facets which has two packages, the > > extension methods and the basic/general-purpose components. Both can > > be installed via one "multi"-package 'facets'. The second project > > would contain libs that stand on their own. Unsure of name for this > > second project though (Garnets?) > > [...] > > Hi, > > For your interrogation, I'm afraid I don't know enough the components of > Facets, to help you in your decision. > > But the name you are thinking about (Garnet) reminded me of a blog post I read > recently about Rubinius. In short, the Rubinius guys will use the name Garnet > to name the dialect of Ruby that generates C code (previously known as Cuby). > > Not a good thing to have two Ruby-related projects with the same name :s > > Here is the post : > > http://on-ruby.blogspot.com/2007/01/serial-rubinius-interview-episode... Thanks. I won't use that term for sure then, and instead offer up the term "Adret". Which is French word (but used sometimes in English) that means the sunny-side of a mountain (a very big facet ;) Do you have any opinion on which of these three options would be best in general / work best for you? Thanks, T. From o.renaud at laposte.net Mon Mar 26 18:38:23 2007 From: o.renaud at laposte.net (Olivier Renaud) Date: Tue, 27 Mar 2007 00:38:23 +0200 Subject: [Facets] Goldilocks In-Reply-To: <1174846618.740683.168030@d57g2000hsg.googlegroups.com> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <200703251923.24843.o.renaud@laposte.net> <1174846618.740683.168030@d57g2000hsg.googlegroups.com> Message-ID: <200703270038.24129.o.renaud@laposte.net> Le dimanche 25 mars 2007 20:16, Trans a ?crit?: > Thanks. I won't use that term for sure then, and instead offer up the > term "Adret". Which is French word (but used sometimes in English) > that means the sunny-side of a mountain (a very big facet ;) > > Do you have any opinion on which of these three options would be best > in general / work best for you? > > Thanks, > T. Yes I know the word Adret, since I am french :) It's not a very common word, though. Does the "t" is pronounced in english ? In fench the pronounciation is "adr?", if the accent makes sense for you. I have one question : what does the prefix "ace_" mean ? Also, with the name ace_extensions, there is another gem called extensions (http://extensions.rubyforge.org), which is somehow similar to what facets core does. I just discovered it, Hmm that doesn't answer your question at all ! But I don't really know. About whether splitting the project in two or not, I don't care. It's more a matter of simplicity of organization for you, I guess. For the two main libs, the extensions and the components, I think they should keep a reference to "facets" in their names. Also, I prefer simple names over multiples different names (the names Facets/core Facets/more were great for that, both explicit and simple). I think the option B is the closest of what I am expecting. -- Olivier Renaud From transfire at gmail.com Mon Mar 26 22:10:47 2007 From: transfire at gmail.com (Trans) Date: Tue, 27 Mar 2007 02:10:47 -0000 Subject: [Facets] Goldilocks In-Reply-To: <200703270038.24129.o.renaud@laposte.net> References: <1174019271.069536.97440@l75g2000hse.googlegroups.com> <200703251923.24843.o.renaud@laposte.net> <1174846618.740683.168030@d57g2000hsg.googlegroups.com> <200703270038.24129.o.renaud@laposte.net> Message-ID: <1174961447.062833.107680@r56g2000hsd.googlegroups.com> On Mar 26, 6:38 pm, Olivier Renaud wrote: > Le dimanche 25 mars 2007 20:16, Trans a ?crit : > > > Thanks. I won't use that term for sure then, and instead offer up the > > term "Adret". Which is French word (but used sometimes in English) > > that means the sunny-side of a mountain (a very big facet ;) > > > Do you have any opinion on which of these three options would be best > > in general / work best for you? > > > Thanks, > > T. > > Yes I know the word Adret, since I am french :) It's not a very common word, > though. Does the "t" is pronounced in english ? In fench the pronounciation > is "adr?", if the accent makes sense for you. How coincidental!? :) As for Adret in English, I've seen both pronunciations. And yes it is a very obscure word. > I have one question : what does the prefix "ace_" mean ? The idea is really just to give the different packages cohesion without over doing it, in the way that "active" and "action" are used for the parts of Rails. While "facets_" could be used as a prefix, that obscures the capability of the parts to be used independently of one another. I thought of "ace" because it can mean small/atomic units. > Also, with the name ace_extensions, there is another gem called extensions > (http://extensions.rubyforge.org), which is somehow similar to what facets > core does. I just discovered it, Yes, I've known of the project since it's inception. In fact a number of facets come from there. A well done project, but much more conservative than Facets. I've talked to Gavin (the maintainer) on a number of occasions. It's complimented by his project "additions". But development trailed off a couple of years ago. Actually you may not know this. Originally Facets was just the method extensions. The additional class/modules were in another project called "Calibre". > Hmm that doesn't answer your question at all ! But I don't really know. About > whether splitting the project in two or not, I don't care. It's more a matter > of simplicity of organization for you, I guess. > > For the two main libs, the extensions and the components, I think they should > keep a reference to "facets" in their names. Also, I prefer simple names over > multiples different names (the names Facets/core Facets/more were great for > that, both explicit and simple). I agree with the simplicity at least. The names are nice but really say very little. Only the word "core" indicates anything at all about what it does. In wanting to provide some independent usage to some of the "more" libs. "more" as a package name starts to fall flat. > I think the option B is the closest of what I am expecting. Interesting b/c that seems to be in opposition to your last statement. Did you actually mean option C? Or am I misunderstanding "simple names over multiples different names". I greatly appreciate your input, btw. I'm just "mentally frozen" on this. So any outside discourse really helps. T. From transfire at gmail.com Tue Mar 27 09:17:00 2007 From: transfire at gmail.com (transfire at gmail.com) Date: Tue, 27 Mar 2007 13:17:00 -0000 Subject: [Facets] Facets 2.0 and the World of Bundles Message-ID: <1175001420.049497.27470@n59g2000hsh.googlegroups.com> Hi-- Time to reveal my surprise... I have been working on a small project called GemBundles. Gembundles are ordinary gems but they have an extra section for bundling dependencies. This allows one to make multi-packages and avoid downloading dependencies. So what's this have to do with Nitro? For Facets 2.0, I want to spin-off parts of the library into their own packages. Now I haven't decided exactly how I want to do this. Either the packages will be categorizations of Facets' libs --which will still keep the whole rather tight knit, while allowing parts to still be used independently if need be. Or the packages will be on their own, with their own unique names and development tracks --in which case Facets itself becomes just the core extensions and the most general of the "more" components. Now, if the former choice, little changes for Nitro, except that it will probably want to ultimately depend on the "facetsbundle" package in order to avoid directly handling a dozen or so dependencies at installation time. But if the later, well, things get interesting here. Nitro will want to create it's own _special bundle_ containing the various libs it requires. This will of course include Facet's core extensions and the general components, and some of the independent spun-off packages. But, unlike now, Nitro need not include any it doesn't use, AND Nitro is at greater liberty to choose dependencies from outside sources other than Facets. Going this route, Nitro could in fact name this bundle "Glue". How's that for coming full circle? At this point I have to makes a choice: Should Facets remain the super uber-lib, containing nearly any common library under the sun, but neatly categorized? Or should I refine Facets to it's most general features and spin-off the more specialized parts? There are trade-offs to each of course. On the one hand, we have a large comprehensive "one- stop-shop" library, well organized, and with high name recognition. On the other, we have a variety of libs from which to pick and choose, each specialized to a particular need, with independent development tracks and announcements. In essence it is a choice of greater Uniformity vs greater Flexibility. It's a difficult choice, and I am torn between them. For Nitro that decision basically boils down to whether Nitro prefers to depend on one comprehensive mega-library, or would it rather have the added flexibility to bundle together a variety of smaller libs at it's discretion? T. From noreply at rubyforge.org Tue Mar 27 14:50:40 2007 From: noreply at rubyforge.org (noreply at rubyforge.org) Date: Tue, 27 Mar 2007 14:50:40 -0400 (EDT) Subject: [Facets] [ facets-Bugs-9624 ] Lots of typos in documentation Message-ID: <20070327185040.244AA5240B3D@rubyforge.org> Bugs item #9624, was opened at 2007-03-27 11:50 You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9624&group_id=804 Category: None Group: None Status: Open Resolution: None Priority: 3 Submitted By: Tyler Rick (tylerrick) Assigned to: Nobody (None) Summary: Lots of typos in documentation Initial Comment: A lot of them could probably be found and fixed just by going through all the docs with a spell-checker, but some of them can't be detected with a spell-checker... "show always have a reader and a writer" => "*should* always have a reader and a writer"? there vs. their etc. Anyway, I'd be willing to help fix them if you can suggest an efficient way to submit them (as a patch file? a tar containing the changed files?) so I can do them all at once... ---------------------------------------------------------------------- You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9624&group_id=804 From noreply at rubyforge.org Tue Mar 27 14:58:21 2007 From: noreply at rubyforge.org (noreply at rubyforge.org) Date: Tue, 27 Mar 2007 14:58:21 -0400 (EDT) Subject: [Facets] [ facets-Bugs-9625 ] Many methods are missing documentation Message-ID: <20070327185821.963225240B3D@rubyforge.org> Bugs item #9625, was opened at 2007-03-27 11:58 You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9625&group_id=804 Category: None Group: None Status: Open Resolution: None Priority: 3 Submitted By: Tyler Rick (tylerrick) Assigned to: Nobody (None) Summary: Many methods are missing documentation Initial Comment: Module#integrate, Module#nesting, String#entropy, ... and I could name many others. A lot of methods I would have *no idea* what they're for based on the name alone. Even looking at the source often doesn't help me figure it out. I would like to see every method by documented so that you can tell just by reading the description/examples what it does, what it's for, *why* you'd want to use it, and how to use it... ---------------------------------------------------------------------- You can respond by visiting: http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9625&group_id=804 From transfire at gmail.com Tue Mar 27 20:36:36 2007 From: transfire at gmail.com (Trans) Date: Wed, 28 Mar 2007 00:36:36 -0000 Subject: [Facets] [ facets-Bugs-9624 ] Lots of typos in documentation In-Reply-To: <20070327185040.244AA5240B3D@rubyforge.org> References: <20070327185040.244AA5240B3D@rubyforge.org> Message-ID: <1175042196.375945.34350@n76g2000hsh.googlegroups.com> On Mar 27, 2:50 pm, wrote: > Bugs item #9624, was opened at 2007-03-27 11:50 > You can respond by visiting:http://rubyforge.org/tracker/?func=detail&atid=3169&aid=9624&group_id... > > Category: None > Group: None > Status: Open > Resolution: None > Priority: 3 > Submitted By: Tyler Rick (tylerrick) > Assigned to: Nobody (None) > Summary: Lots of typos in documentation > > Initial Comment: > A lot of them could probably be found and fixed just by going through all the docs with a spell-checker, but some of them can't be detected with a spell-checker... > > "show always have a reader and a writer" => "*should* always have a reader and a writer"? > > there vs. their > > etc. > > Anyway, I'd be willing to help fix them if you can suggest an efficient way to submit them (as a patch file? a tar containing the changed files?) so I can do them all at once... Oh, you discoverd how bad of a typist I am have you? ;) Your in luck. This is going to get very easy for you (and others) to fix soon. But it has to wait until I work out this Facets 2.0 organization. T.