From ArcRiley at pysoy.org Tue Mar 11 13:53:00 2008 From: ArcRiley at pysoy.org (Arc Riley) Date: Tue, 11 Mar 2008 13:53:00 -0400 Subject: [PySoy-Dev] digging in for Beta-3 Message-ID: We're ten days away now from the planned Beta-3 release date and a large number of critical tickets remaining (and even more un-filed). We also have Summer of Code gearing up and a number of students will be looking at the codebase for project proposals soon, so meeting this deadline is important. If anyone who's not already helping is interested, join us in #PySoy -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080311/93528828/attachment.html From physci at gmail.com Tue Mar 11 14:03:15 2008 From: physci at gmail.com (Derek Rhodes) Date: Tue, 11 Mar 2008 14:03:15 -0400 Subject: [PySoy-Dev] test3 Message-ID: test3 From ArcRiley at pysoy.org Tue Mar 11 14:15:46 2008 From: ArcRiley at pysoy.org (Arc Riley) Date: Tue, 11 Mar 2008 14:15:46 -0400 Subject: [PySoy-Dev] test3 In-Reply-To: References: Message-ID: test received, the lists are back online. On Tue, Mar 11, 2008 at 2:03 PM, Derek Rhodes wrote: > test3 > _______________________________________________ > PySoy-Dev mailing list > PySoy-Dev at pysoy.org > http://www.pysoy.org/mailman/listinfo/pysoy-dev > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080311/c650ad81/attachment.htm From ArcRiley at pysoy.org Fri Mar 14 01:02:42 2008 From: ArcRiley at pysoy.org (Arc Riley) Date: Fri, 14 Mar 2008 01:02:42 -0400 Subject: [PySoy-Dev] batteries, er, build system included? Message-ID: We were talking about pulling Cython into PySoy's trunk, patching it to work for our needs, and thus releasing it with source releases. This means that, whatever we call this Cython branch, will be located at svn.pysoy.org/trunk/pysoy/X and imported by setup.py for building PySoy, where currently we import (and thus depend on) Pyrex from site-packages This is open to discussion both here and on IRC, nothing has been done along these lines yet. If there's no strong arguments to not do this, or better ideas, this may be started as soon as this weekend. Here's some quick info: *Pros* - one less dependency - we wouldn't have our community project written in a language maintained by Greg's personal project [1] anymore - we'll be able to use language enhancements *immediately *[2]* * - based on Cython, we'll have some really nice features over Pyrex right away - we should be able to merge new Cython enhancements over time fairly easily [3] - once the Cython.Compiler is patched we shouldn't need to modify any of our existing codebase - no more Pyrex version problems! - we'll be able to move a great deal of the code in setup.py [4] - our source releases will be much smaller even with the added [5] - since Cython is pure Python it can be used directly from the source tree [6] *Cons * - it's a bit more work, when we're already going to be a bit late for our Beta-3 milestone - this will launch us into the politics of SageX aka Cython vs Pyrex, something we've done well to avoid so far - it draws some focus away from the engine development [1] I'm sure Greg's a great guy, but PySoy was founded to grow beyond Jiba's personal project hosted on his personal website with himself listed as "the author", one look at the Pyrex website and where it's located says it all - Cython was founded to grow Pyrex beyond Greg's personal project, too. [2] We currently have to wait for new features to accepted by Greg, often rewritten by him, and several months pass before released in a new Pyrex version [3] and of course send them upstream. Who knows, maybe someday they'll be merged and Sage+PySoy can join efforts on Cython? [4] our setup.py replaces much of Python.Distutils with our own code as a workaround, it's kludgy and we'd be better off with this in the build environment [5] we've been releasing post-processed source tarballs, with all the .c files included - these total 2.8megs as of r1146. Cython 0.9.6.10 is only 2.2megs uncompressed. The resulting source tarball, not considering variables such as compression and our own Cython patches, would be .4megs smaller [6] this means it will not need to be installed to site-packages, or anywhere else on the system, to be used -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080314/dacbffda/attachment.html From eastein at WPI.EDU Fri Mar 14 01:25:00 2008 From: eastein at WPI.EDU (Eric Stein) Date: Fri, 14 Mar 2008 01:25:00 -0400 Subject: [PySoy-Dev] batteries, er, build system included? In-Reply-To: References: Message-ID: <47DA0C2C.4060304@wpi.edu> Arc Riley wrote: > We were talking about pulling Cython into PySoy's trunk, patching it > to work for our needs, and thus releasing it with source releases. > This means that, whatever we call this Cython branch, will be located > at svn.pysoy.org/trunk/pysoy/X > and imported by setup.py for building PySoy, where currently we import > (and thus depend on) Pyrex from site-packages > > This is open to discussion both here and on IRC, nothing has been done > along these lines yet. If there's no strong arguments to not do this, > or better ideas, this may be started as soon as this weekend. > > Here's some quick info: > > *Pros* > > * one less dependency > * we wouldn't have our community project written in a language > maintained by Greg's personal project [1] anymore > * we'll be able to use language enhancements /immediately /[2]/ > / > * based on Cython, we'll have some really nice features over Pyrex > right away > * we should be able to merge new Cython enhancements over time > fairly easily [3] > * once the Cython.Compiler is patched we shouldn't need to modify > any of our existing codebase > * no more Pyrex version problems! > * we'll be able to move a great deal of the code in setup.py [4] > * our source releases will be much smaller even with the added [5] > * since Cython is pure Python it can be used directly from the > source tree [6] > > *Cons > * > > * it's a bit more work, when we're already going to be a bit late > for our Beta-3 milestone > * this will launch us into the politics of SageX aka Cython vs > Pyrex, something we've done well to avoid so far > * it draws some focus away from the engine development > > > > [1] I'm sure Greg's a great guy, but PySoy was founded to grow beyond > Jiba's personal project hosted on his personal website with himself > listed as "the author", one look at the Pyrex website and where it's > located says it all - Cython was founded to grow Pyrex beyond Greg's > personal project, too. This is entirely true. Pyrex doesn't even have a public SVN or wiki. It stagnates. > > [2] We currently have to wait for new features to accepted by Greg, > often rewritten by him, and several months pass before released in a > new Pyrex version Also true. > > [3] and of course send them upstream. Who knows, maybe someday > they'll be merged and Sage+PySoy can join efforts on Cython? Yes, that could be useful. > > [4] our setup.py replaces much of Python.Distutils with our own code > as a workaround, it's kludgy and we'd be better off with this in the > build environment I agree that our build environment is a kludge. It would be good to make it clean and consistent. > > [5] we've been releasing post-processed source tarballs, with all the > .c files included - these total 2.8megs as of r1146. Cython 0.9.6.10 > is only 2.2megs uncompressed. The resulting source > tarball, not considering variables such as compression and our own > Cython patches, would be .4megs smaller Honestly, I could care less about file size on scales this small. I am, however uncomfortable having our releases of our free software seem obfuscated and difficult to verify for others - the generated c files are nearly impossible to read. It would be best if everyone worked from the highest level code, even those who don't plan to do development themselves. > > [6] this means it will not need to be installed to site-packages, or > anywhere else on the system, to be used I like the idea of switching to Cython, but I have some hesitation about including it in our repository. I believe it makes more sense to locate things like so on the svn server: /trunk/pysoy - pysoy itself /trunk/soython - our branch of Cython I guess it makes plenty of sense to include soython within the pysoy directory structure for releases, but mixing it in the revision control system seems sloppy and unneeded. Developers can even symlink pysoy/soython to ../soython or wherever they keep their soython working copy if they don't want to install it. It may not seem like a major issue to some, but I think it's important that if we are going to branch another package, we avoid needlessly intermixing it with our primary project. Eric From ArcRiley at pysoy.org Fri Mar 14 02:11:07 2008 From: ArcRiley at pysoy.org (Arc Riley) Date: Fri, 14 Mar 2008 02:11:07 -0400 Subject: [PySoy-Dev] batteries, er, build system included? In-Reply-To: <47DA0C2C.4060304@wpi.edu> References: <47DA0C2C.4060304@wpi.edu> Message-ID: > however uncomfortable having our releases of our free software seem > obfuscated and difficult to verify for others - the generated c files > are nearly impossible to read. It would be best if everyone worked from > the highest level code, even those who don't plan to do development > themselves. Another problem is we're using #DEF's for OS-specific code which means we'd need separate source downloads for each OS. Regardless of how this discussion comes out, I think we have consensus that Beta-3 needs to be released as .pyx sources. The code in setup.py which only built from Pyrex when Version=="Trunk" was recently removed. I like the idea of switching to Cython, but I have some hesitation about > including it in our repository. I believe it makes more sense to locate > things like so on the svn server Keep in mind, if the build environment isn't under trunk/pysoy/ then it must be intended to be installed to site-packages One argument in favor of having it install to site-packages is we'll eventually have PySoy "plugin" modules which may need a similar build system. One argument against is this means it's a separate package from PySoy and, thus, replaces a Pyrex as a dependency with this Cython variant. Piet expressed a similar concern than one of mine, that all these projects (PySoy, Sage, etc) keep building Pyrex based branches for the specific needs of their project, when we should be working on one project together. I don't know how pragmatic that is, but that's something to think about. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080314/d2ccd2a7/attachment.htm From eastein at WPI.EDU Fri Mar 14 21:17:43 2008 From: eastein at WPI.EDU (Eric Stein) Date: Fri, 14 Mar 2008 21:17:43 -0400 Subject: [PySoy-Dev] batteries, er, build system included? In-Reply-To: References: <47DA0C2C.4060304@wpi.edu> Message-ID: <47DB23B7.9060209@wpi.edu> Arc Riley wrote: > > however uncomfortable having our releases of our free software seem > obfuscated and difficult to verify for others - the generated c files > are nearly impossible to read. It would be best if everyone > worked from > the highest level code, even those who don't plan to do development > themselves. > > > Another problem is we're using #DEF's for OS-specific code which means > we'd need separate source downloads for each OS. > > Regardless of how this discussion comes out, I think we have consensus > that Beta-3 needs to be released as .pyx sources. The code in > setup.py which only built from Pyrex when Version=="Trunk" was > recently removed. > > > I like the idea of switching to Cython, but I have some hesitation > about > including it in our repository. I believe it makes more sense to > locate > things like so on the svn server > > > Keep in mind, if the build environment isn't under trunk/pysoy/ then > it must be intended to be installed to site-packages > > One argument in favor of having it install to site-packages is we'll > eventually have PySoy "plugin" modules which may need a similar build > system. > > One argument against is this means it's a separate package from PySoy > and, thus, replaces a Pyrex as a dependency with this Cython variant. > > Piet expressed a similar concern than one of mine, that all these > projects (PySoy, Sage, etc) keep building Pyrex based branches for the > specific needs of their project, when we should be working on one > project together. I don't know how pragmatic that is, but that's > something to think about. > > I propose we talk to Piet about this and see if we can get him to be more responsitve and transparent in his development so we can work with him without hampering our efforts. Perhaps he will be open to setting up an SVN repository and other team tools - it could go somewhere, as he clearly understands the issue. Eric From ArcRiley at pysoy.org Sun Mar 16 18:01:47 2008 From: ArcRiley at pysoy.org (Arc Riley) Date: Sun, 16 Mar 2008 18:01:47 -0400 Subject: [PySoy-Dev] batteries, er, build system included? In-Reply-To: References: <47DA0C2C.4060304@wpi.edu> <47DB23B7.9060209@wpi.edu> Message-ID: > I propose we talk to Piet about this and see if we can get him to be > more responsitve and transparent in his development so we can work with > him without hampering our efforts. Piet Delport is very interested in setting up a SVN repository with Trac and the whole works. He is just as frustrated by all these project-specific Pyrex builds as everyone else and wants to see a viable community project form that can pool all these efforts in a mutually beneficial way. Piet, however, is not the maintainer of Pyrex. He has no authority or control over the source or releases, nor has he ever talked to the maintainer. He provides patches and does a wonderful job at helping people at IRC - we owe him a lot of gratitude - but he has no pull. The maintainer is a man named Greg Ewing, Pyrex is hosted on his personal website, you can find his contact info there. The Cython guys have already tried to work with him, as have other groups, this is why there have been so many project-specific branches. Honestly, learning from our mistake of getting involved with Soya3d and trying to turn it into a community project, I think it'd be easier and time efficient to do our own thing, collaborating with Cython where it makes sense to, than working with Greg. Especially with so much energy being devoted to the copyleft games group -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080316/76d17429/attachment.htm From nyad55 at gmail.com Tue Mar 18 12:12:27 2008 From: nyad55 at gmail.com (Jason Ward) Date: Tue, 18 Mar 2008 18:12:27 +0200 Subject: [PySoy-Dev] Google summer of Code Message-ID: <49f16abf0803180912g69b67c30rf08be331c9bce4f8@mail.gmail.com> Hi I am very interested in PySoy for Google Summer of Code. My name is Jason This is some of what I can do: I have been coding my own engine in assembly language which uses opengl and contains frustum culling, collision detection and soon to be physics, a model loader from a modified blender python export script. But I have only done windows coding. I know python, can read C/C++ but I would like to use this project as a chance to practice these. The wiki says that PySoy only contains basic physics. Doesn't ODE give it advanced physics or has that not yet been put into PySoy? Anyway from the ideas that you have I like the networking one and the animation one. My only experience with network programming is some stuff I did with the socket library in python. But I do know and have used 3D programming for a while so I may also be able to dabble in animation even though I haven't done that specifically. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080318/2b59061a/attachment.htm From ArcRiley at pysoy.org Tue Mar 18 16:00:26 2008 From: ArcRiley at pysoy.org (Arc Riley) Date: Tue, 18 Mar 2008 16:00:26 -0400 Subject: [PySoy-Dev] relicensing Message-ID: We've discussed this on IRC a few times, I'm putting this out on the list for anyone who wasn't part of those discussions. The AGPLv3 was recently released which is identical to the GPLv3 except for this one clause: < 13. Use with the GNU Affero General Public License. > 13. Remote Network Interaction; Use with the GNU General Public License. ''' Notwithstanding any other provision of this License, if you modify the Program, your modified version must prominently offer all users interacting with it remotely through a computer network (if your version supports such interaction) an opportunity to receive the Corresponding Source of your version by providing access to the Corresponding Source from a network server at no charge, through some standard or customary means of facilitating copying of software. This Corresponding Source shall include the Corresponding Source for any work covered by version 3 of the GNU General Public License that is incorporated pursuant to the following paragraph. Notwithstanding any other provision of this License, you have permission to link or combine any covered work with a work licensed under version 3 of the GNU General Public License into a single combined work, and to convey the resulting work. The terms of this License will continue to apply to the part which is the covered work, but the work with which it is combined will remain governed by version 3 of the GNU General Public License. ''' In laymen's terms, if PySoy is used on a network by others (ie, as a server, in a p2p arrangement, etc), any modifications or additions including game code and content would need to be available. The purpose of this is: - prevent "secret server code" models from taking root, thus keeping games copyleft - ensure that game code running remotely as a Firefox plugin will have the GPLv3 terms apply as well - since game clients will provide services to each other, mods/hacks must (legally) be shared We'll still be able to link to anything we could with the GPLv3 license, and the AGPLv3 terms only go into effect when networking is enabled. While we have no integrated network code right now, switching now vs closer to 1.0_beta4 release would prevent forking to avoid the AGPLv3 terms back to the current revision. Now to be clear, a 3rd party could release PySoy under the AGPLv3 without our permission. The GPLv3 allows this. Because this is a community project, and this does reflect a change of licensing, I'm putting this out to receive feedback from all developers. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080318/ef140e33/attachment.htm From ArcRiley at pysoy.org Thu Mar 20 03:56:18 2008 From: ArcRiley at pysoy.org (Arc Riley) Date: Thu, 20 Mar 2008 03:56:18 -0400 Subject: [PySoy-Dev] bumpmapping textures Message-ID: r1188 should fix bump mapping with the provided examples. One of the problems was the bumpmap example, which I've yet to resolve why it's not working correctly but have moved it to broken/ for now. Much more work is needed both in this example and to clean up bump mapping in general. We could use someone to focus on it for a few days and get the rest of the wrinkles ironed out. Specifically when a material changes it's .normal status the MaterialLists which use it are not updated (for _hasBumps) I believe everyone has had a chance to look at the AGPL issue by now and is either in favor of it or ambivalent, so I'm going to start updating the license. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080320/d8159eaf/attachment.htm From ArcRiley at pysoy.org Fri Mar 21 17:08:09 2008 From: ArcRiley at pysoy.org (Arc Riley) Date: Fri, 21 Mar 2008 17:08:09 -0400 Subject: [PySoy-Dev] Sprint this weekend (SoC students especially) Message-ID: In case you haven't heard we're doing a sprint this weekend. Roll up your sleeves and grab a ticket! We're all hanging out in our normal IRC channel (#PySoy on irc.freenode.net) to coordinate. Gobby is available for pair programming, sessions with a more experienced developer are available on request. If you're intending to apply for SoC with us this year, you should participate. We want to see whether applicants can work in a group and are competent working on the project, especially as last year we had one student get approved who quickly showed he didn't actually know how to code. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080321/fc430b49/attachment.htm From ArcRiley at pysoy.org Wed Mar 26 16:13:21 2008 From: ArcRiley at pysoy.org (Arc Riley) Date: Wed, 26 Mar 2008 16:13:21 -0400 Subject: [PySoy-Dev] Summer of Code Applications Message-ID: A number of people have expressed an interest in PySoy for Summer of Code, some have joined this list, however we're not seeing many actual applications being submitted. Let me re-iterate, the key to applying for Summer of Code is to do so *early *. If you do not have a solid plan in mind, you should be working with us to develop one. Last minute applications are rarely approved, and almost half we get each year are received in the last 24 hours. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.pysoy.org/pipermail/pysoy-dev/attachments/20080326/5d95ea9d/attachment.htm