Py3 would break scripts, yes. Switching to lua probably would not be an option, though it might be possible to add support for it later depending on how things go.<br><br>Joe<br><br><div class="gmail_quote">On Fri, Dec 5, 2008 at 10:30 AM, Bob Holcomb <span dir="ltr"><<a href="mailto:bholcomb@mak.com">bholcomb@mak.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">If going to python 3 is going to break scripts, and that's ok, would<br>
switching to lua be a consideration?<br>
<br>
Cheers,<br>
<br>
Bob<br>
<br>
--<br>
Sent from my mobile device<br>
<br>
On Dec 5, 2008, at 12:09 PM, <a href="mailto:bf-committers-request@blender.org">bf-committers-request@blender.org</a> wrote:<br>
<br>
> Send Bf-committers mailing list submissions to<br>
> <a href="mailto:bf-committers@blender.org">bf-committers@blender.org</a><br>
><br>
> To subscribe or unsubscribe via the World Wide Web, visit<br>
> <a href="http://lists.blender.org/mailman/listinfo/bf-committers" target="_blank">http://lists.blender.org/mailman/listinfo/bf-committers</a><br>
> or, via email, send a message with subject or body 'help' to<br>
> <a href="mailto:bf-committers-request@blender.org">bf-committers-request@blender.org</a><br>
><br>
> You can reach the person managing the list at<br>
> <a href="mailto:bf-committers-owner@blender.org">bf-committers-owner@blender.org</a><br>
><br>
> When replying, please edit your Subject line so it is more specific<br>
> than "Re: Contents of Bf-committers digest..."<br>
><br>
><br>
> Today's Topics:<br>
><br>
> 1. Re: Bundle Python 3.0 for Blender 2.5 (Roland Hess)<br>
> 2. Re: Bundle Python 3.0 for Blender 2.5 (Campbell Barton)<br>
> 3. Re: Bundle Python 3.0 for Blender 2.5 (Campbell Barton)<br>
> 4. Re: Bundle Python 3.0 for Blender 2.5 (Brecht Van Lommel)<br>
> 5. Re: Bundle Python 3.0 for Blender 2.5 (Martin Poirier)<br>
> 6. Re: Bundle Python 3.0 for Blender 2.5 (Timothy Baldridge)<br>
> 7. Re: Bundle Python 3.0 for Blender 2.5 (Chris Want)<br>
> 8. Re: Bundle Python 3.0 for Blender 2.5 (Ken Hughes)<br>
> 9. Re: Bundle Python 3.0 for Blender 2.5 (Timothy Baldridge)<br>
><br>
><br>
> ----------------------------------------------------------------------<br>
><br>
> Message: 1<br>
> Date: Fri, 05 Dec 2008 07:46:49 -0500<br>
> From: Roland Hess <<a href="mailto:rolandh@reed-witting.com">rolandh@reed-witting.com</a>><br>
> Subject: Re: [Bf-committers] Bundle Python 3.0 for Blender 2.5<br>
> To: <a href="mailto:bf-committers@blender.org">bf-committers@blender.org</a><br>
> Message-ID: <<a href="mailto:493922B9.9030007@reed-witting.com">493922B9.9030007@reed-witting.com</a>><br>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed<br>
><br>
> Would the current python stick around as well for backward<br>
> compatibility? I'd read that Py 3.0 wasn't backward compatible in a<br>
> number of ways:<br>
><br>
> <a href="http://docs.python.org/dev/3.0/whatsnew/3.0.html" target="_blank">http://docs.python.org/dev/3.0/whatsnew/3.0.html</a><br>
><br>
> There is a converter available for that takes care of things like the<br>
> "print" statement, but I think changes like the ones noted in that doc<br>
> would break almost every existing script. While that's not a problem<br>
> for<br>
> me personally, that would be a serious frustration for people who<br>
> search<br>
> for a script, find it and have it not work. Not saying Blender 2.5<br>
> shouldn't go with Python 3.0, just that I'm surprised this wasn't<br>
> noted<br>
> on the "Cons" part of the proposal. This is a significant change, much<br>
> more than Py2.3->Py2.4->Py2.5.<br>
><br>
> --<br>
> Roland Hess<br>
> harkyman<br>
><br>
><br>
><br>
> ------------------------------<br>
><br>
> Message: 2<br>
> Date: Sat, 6 Dec 2008 00:41:20 +1100<br>
> From: "Campbell Barton" <<a href="mailto:ideasman42@gmail.com">ideasman42@gmail.com</a>><br>
> Subject: Re: [Bf-committers] Bundle Python 3.0 for Blender 2.5<br>
> To: "bf-blender developers" <<a href="mailto:bf-committers@blender.org">bf-committers@blender.org</a>><br>
> Message-ID:<br>
> <<a href="mailto:7c1ab96d0812050541u4f3bdc26m3053873848170fb0@mail.gmail.com">7c1ab96d0812050541u4f3bdc26m3053873848170fb0@mail.gmail.com</a>><br>
> Content-Type: text/plain; charset=ISO-8859-1<br>
><br>
> On Fri, Dec 5, 2008 at 11:46 PM, Roland Hess <rolandh@reed-<br>
> <a href="http://witting.com" target="_blank">witting.com</a>> wrote:<br>
>> Would the current python stick around as well for backward<br>
>> compatibility? I'd read that Py 3.0 wasn't backward compatible in a<br>
>> number of ways:<br>
>><br>
>> <a href="http://docs.python.org/dev/3.0/whatsnew/3.0.html" target="_blank">http://docs.python.org/dev/3.0/whatsnew/3.0.html</a><br>
>><br>
>> There is a converter available for that takes care of things like the<br>
>> "print" statement, but I think changes like the ones noted in that<br>
>> doc<br>
>> would break almost every existing script. While that's not a<br>
>> problem for<br>
>> me personally, that would be a serious frustration for people who<br>
>> search<br>
>> for a script, find it and have it not work. Not saying Blender 2.5<br>
>> shouldn't go with Python 3.0, just that I'm surprised this wasn't<br>
>> noted<br>
>> on the "Cons" part of the proposal. This is a significant change,<br>
>> much<br>
>> more than Py2.3->Py2.4->Py2.5.<br>
>><br>
>> --<br>
>> Roland Hess<br>
>> harkyman<br>
>><br>
>> _______________________________________________<br>
>> Bf-committers mailing list<br>
>> <a href="mailto:Bf-committers@blender.org">Bf-committers@blender.org</a><br>
>> <a href="http://lists.blender.org/mailman/listinfo/bf-committers" target="_blank">http://lists.blender.org/mailman/listinfo/bf-committers</a><br>
>><br>
><br>
> Its highly likely many/most? scripts will need to be modified to run<br>
> in Blender2.5 - so having to do edits for py3 is minimal effort if<br>
> your already updating your script for a new/modified api.<br>
> It is a con, but comparatively not a huge one IMHO. especially if you<br>
> consider we will move to py3 at some point anyway.<br>
><br>
> The proposals are in 2 parts, I only had pros/cons for including<br>
> python with blender.<br>
><br>
> I cross posted on blenderartist, where some scripters have replied.<br>
> <a href="http://blenderartists.org/forum/showthread.php?t=142544" target="_blank">http://blenderartists.org/forum/showthread.php?t=142544</a><br>
><br>
> --<br>
> - Campbell<br>
><br>
><br>
> ------------------------------<br>
><br>
> Message: 3<br>
> Date: Sat, 6 Dec 2008 01:12:50 +1100<br>
> From: "Campbell Barton" <<a href="mailto:ideasman42@gmail.com">ideasman42@gmail.com</a>><br>
> Subject: Re: [Bf-committers] Bundle Python 3.0 for Blender 2.5<br>
> To: "bf-blender developers" <<a href="mailto:bf-committers@blender.org">bf-committers@blender.org</a>><br>
> Message-ID:<br>
> <<a href="mailto:7c1ab96d0812050612r2b4a7eaaw3c598e7672e18dce@mail.gmail.com">7c1ab96d0812050612r2b4a7eaaw3c598e7672e18dce@mail.gmail.com</a>><br>
> Content-Type: text/plain; charset=ISO-8859-1<br>
><br>
> To clarify...<br>
> Including python with blender wont exclude system python modules<br>
> (PyGame/Numpy/PIL etc), Its just those modules will need to be built<br>
> with the same python as blender is. so blender can load them.<br>
><br>
> Basically - Do what we are doing on win32 on other OS's, and include<br>
> more modules so scripts dont complain for a full python install.<br>
><br>
><br>
> ------------------------------<br>
><br>
> Message: 4<br>
> Date: Fri, 05 Dec 2008 15:51:46 +0100<br>
> From: Brecht Van Lommel <<a href="mailto:brecht@blender.org">brecht@blender.org</a>><br>
> Subject: Re: [Bf-committers] Bundle Python 3.0 for Blender 2.5<br>
> To: bf-blender developers <<a href="mailto:bf-committers@blender.org">bf-committers@blender.org</a>><br>
> Message-ID: <<a href="mailto:49394002.30802@blender.org">49394002.30802@blender.org</a>><br>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed<br>
><br>
> Hi,<br>
><br>
>> Campbell Barton wrote:<br>
>>> Hey All, this proposal covers 2 topics<br>
>>> * Distribute Blender with python on all OS's (I'd suggest this<br>
>>> irrespective of py version used)<br>
>>> * Move to Python 3.0 for blender 2.5 (relies on a bundling python)<br>
><br>
> I think some sort of painful transition is inevitable. Python 3.0 will<br>
> inevitably break most scripts, just grep for "print" in<br>
> release/scripts and see how often it is used, and there are many more<br>
> incompatibilities, external scripts are probably no different.<br>
><br>
> On the other hand 2.5 is going to break the Draw module. For example<br>
> Draw.PupMenu is blocking, but this is not allowed in 2.50. Again more<br>
> than half of the scripts in release/scripts use that, so they will<br>
> break even if we try to preserve the 2.4x api.<br>
><br>
> I think we might as well get it over with in one go.<br>
><br>
> Further, I do not really understand the objection to bundling python<br>
> with Blender. If Blender is compiled with the same version as is<br>
> installed on the operating system, external modules using that version<br>
> will still be available.<br>
><br>
> If you are distributing a script with external modules you can<br>
> distribute them with the script even and it's easier because you know<br>
> the python version Blender uses. If this is for some custom script for<br>
> your own purposes I'm sure you are capable of compiling either Blender<br>
> or the external module so they use the same python version, if the<br>
> version is not the same already. Right now even, chances are there is<br>
> no Blender release with the python version you are using.<br>
><br>
> So, basically I agree with Campbell in that we should bundle 2.5 with<br>
> python 3.0, I don't see what would be the better alternative that is<br>
> feasible, as I don't think having two python versions compiled into<br>
> Blender is a practical possibility.<br>
><br>
> Brecht.<br>
><br>
><br>
><br>
><br>
> ------------------------------<br>
><br>
> Message: 5<br>
> Date: Fri, 5 Dec 2008 07:49:44 -0800 (PST)<br>
> From: Martin Poirier <<a href="mailto:theeth@yahoo.com">theeth@yahoo.com</a>><br>
> Subject: Re: [Bf-committers] Bundle Python 3.0 for Blender 2.5<br>
> To: bf-blender developers <<a href="mailto:bf-committers@blender.org">bf-committers@blender.org</a>><br>
> Message-ID: <<a href="mailto:289590.76304.qm@web51303.mail.re2.yahoo.com">289590.76304.qm@web51303.mail.re2.yahoo.com</a>><br>
> Content-Type: text/plain; charset=us-ascii<br>
><br>
><br>
><br>
><br>
> --- On Fri, 12/5/08, Campbell Barton <<a href="mailto:ideasman42@gmail.com">ideasman42@gmail.com</a>> wrote:<br>
><br>
>> Its highly likely many/most? scripts will need to be<br>
>> modified to run<br>
>> in Blender2.5 - so having to do edits for py3 is minimal<br>
>> effort if<br>
>> your already updating your script for a new/modified api.<br>
>> It is a con, but comparatively not a huge one IMHO.<br>
>> especially if you<br>
>> consider we will move to py3 at some point anyway.<br>
><br>
> True, but I really don't think that point should be now (as in this<br>
> very moment). For one thing, Py3K is just out of the door, as you<br>
> said yourself in the BA thread, the first version of a rewrite is<br>
> often not the most stable. ;)<br>
><br>
> 2.50 is not going to be released any time soon and you said<br>
> switching to Py3K means very little changes internally, so no need<br>
> to rush that decision (at least not until the design of the API is<br>
> fixed)<br>
><br>
> Regarding bundling, going over the *pro* list:<br>
><br>
> * We can include modules bundled scripts use so that users are not<br>
> faced with error messages when they try a python tool.<br>
><br>
> I guess that's a fix for users not reading the download page<br>
> properly? (or the page not being precise enough about external<br>
> dependency on Python)<br>
><br>
> If it's a problem with bundled script not having correct failsafe,<br>
> then that's different altogether.<br>
><br>
> * No need to distribute blender for versions py 2.3, 2.4, 2.5, 2.6....<br>
> and later 3.0 - We already are having to do both 32 and 64bit Blender<br>
> builds. Many users wont know what python version is installed on<br>
> their system.<br>
><br>
> We currently only have builds with 2.4 and 2.5 on the download page<br>
> (or py 2.3 for os X, but they don't have 2.4 builds), so I don't<br>
> quite see where you got your list. Moreover, if we decide to move to<br>
> 3K, there's no way in hell that we'll be maintaining two sets of<br>
> bundled script for backward compatibilities, so we can drop all 2.X<br>
> versions then.<br>
><br>
> Heck, we could stick to a single version too, regardless of bundling.<br>
><br>
> * A reliable python means we can use it for more important aspects<br>
> of Blender 2.5 (something Ton has suggested)<br>
><br>
> That, IMHO, should be the biggest argument.<br>
><br>
> * No conflicts with the system python - there have been bug reports<br>
> where the system python was crashing blender.<br>
><br>
> The one I remember (there might have been others mind you) was<br>
> caused by a faulty compile (of Python or Blender, it wasn't clear)<br>
> and couldn't be reproduced on other systems using the same distro.<br>
><br>
> * Blender already includes FFMpeg and some other significant libs on<br>
> most OS's<br>
><br>
> Comparing to ffmpeg is a bit loaded, their API is about as stable as<br>
> a pile of rumble and they don't have released versions we can<br>
> reliably link against (or did they finally bit that bullet?)<br>
><br>
> * Openoffice and some games include python, searched for python<br>
> problems with these apps and didnt find complaints, common with<br>
> Blender.<br>
><br>
> If you were thinking of EVE, they include Stackless Python, hardly<br>
> comparable (moreover, we already bundle on windows).<br>
><br>
> The most frequent "complain" about Blender's Python is the site-<br>
> package missing warning, if they disable import of external package,<br>
> of course they wouldn't see that.<br>
><br>
> I have a couple of technical questions though. By bundling the<br>
> interpreter and some libs but still leaving the possibility of<br>
> importing external libs, won't this also leave us open to the same<br>
> site-package incompatibilities/missing warning message that we<br>
> currently have and that you seem to identify as "the big problem<br>
> with not bundling python"? I mean, we already bundle Python on<br>
> windows and most of the reports are on that OS, so how is that a fix?<br>
><br>
> I find the long list of BA threads somewhat disingenuous (have you<br>
> even read them before posting?) as lots of them wouldn't be fixed<br>
> with bundling Python (permission problem on Vista, wrong libc<br>
> version, how to run multiple versions of blender, exception in a<br>
> script at runtime, ...), but I digress.<br>
><br>
> Martin<br>
><br>
><br>
><br>
><br>
><br>
> ------------------------------<br>
><br>
> Message: 6<br>
> Date: Fri, 5 Dec 2008 09:59:07 -0600<br>
> From: "Timothy Baldridge" <<a href="mailto:tbaldridge@gmail.com">tbaldridge@gmail.com</a>><br>
> Subject: Re: [Bf-committers] Bundle Python 3.0 for Blender 2.5<br>
> To: "bf-blender developers" <<a href="mailto:bf-committers@blender.org">bf-committers@blender.org</a>><br>
> Message-ID:<br>
> <<a href="mailto:b33fdb110812050759w590ecf1p981d5bda737ed7a6@mail.gmail.com">b33fdb110812050759w590ecf1p981d5bda737ed7a6@mail.gmail.com</a>><br>
> Content-Type: text/plain; charset=ISO-8859-1<br>
><br>
>> * Openoffice and some games include python, searched for python<br>
>> problems with these apps and didnt find complaints, common with<br>
>> Blender.<br>
><br>
> True, but the installs for most of these games stay constant, you're<br>
> not removing and adding modules on a regular basis. If we do bundle<br>
> Pythonr we should always check for a system Python first, as auto<br>
> installers won't know about the Blender Python. Or perhaps we've<br>
> solved this issue already?<br>
><br>
> If a user installs numpy into the system directory, does Blender see<br>
> it?<br>
><br>
><br>
> ------------------------------<br>
><br>
> Message: 7<br>
> Date: Fri, 05 Dec 2008 09:25:10 -0700<br>
> From: Chris Want <<a href="mailto:cwant@ualberta.ca">cwant@ualberta.ca</a>><br>
> Subject: Re: [Bf-committers] Bundle Python 3.0 for Blender 2.5<br>
> To: bf-blender developers <<a href="mailto:bf-committers@blender.org">bf-committers@blender.org</a>><br>
> Message-ID: <<a href="mailto:493955E6.5020706@ualberta.ca">493955E6.5020706@ualberta.ca</a>><br>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed<br>
><br>
><br>
> Reading this thread, I become more agreeable to the suggestion<br>
> that Early Ehlinger made earlier (no pun intended). He suggested<br>
> that python scripting should be handled as an external plugin<br>
> that loads at run time. Then the problem of having builds of<br>
> blender for different python versions doesn't matter: just<br>
> have one blender build and distribute it with plugins for<br>
> the different python versions. This, of course, would be a<br>
> lot easier with a well thought out C API, which I believe should<br>
> be the first step before tackling python (as I've mentioned<br>
> before).<br>
><br>
> I should also point out that bundling python with blender adds<br>
> extra complexity to Blender's plethora of build systems. The<br>
> example of FFMPEG is a poor one since not all build<br>
> systems support it, there are issues under windowa, the<br>
> scons system supports it by having the additional<br>
> requirement of having autoconf installed, and I have<br>
> recently seen mails on the list from our game engine developer<br>
> trying to change the FFMPEG sources in extern, to which<br>
> there are no replies from the official maintainer. This is<br>
> not a role model to follow.<br>
><br>
> Cheers,<br>
> Chris<br>
><br>
><br>
> ------------------------------<br>
><br>
> Message: 8<br>
> Date: Fri, 05 Dec 2008 08:46:38 -0800<br>
> From: Ken Hughes <<a href="mailto:khughes@pacific.edu">khughes@pacific.edu</a>><br>
> Subject: Re: [Bf-committers] Bundle Python 3.0 for Blender 2.5<br>
> To: bf-blender developers <<a href="mailto:bf-committers@blender.org">bf-committers@blender.org</a>><br>
> Message-ID: <<a href="mailto:49395AEE.2090608@pacific.edu">49395AEE.2090608@pacific.edu</a>><br>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed<br>
><br>
> Martin Poirier wrote:<br>
>> 2.50 is not going to be released any time soon and you said<br>
>> switching to Py3K means very little changes internally, so no need<br>
>> to rush that decision (at least not until the design of the API is<br>
>> fixed)<br>
>><br>
_______________________________________________<br>
Bf-committers mailing list<br>
<a href="mailto:Bf-committers@blender.org">Bf-committers@blender.org</a><br>
<a href="http://lists.blender.org/mailman/listinfo/bf-committers" target="_blank">http://lists.blender.org/mailman/listinfo/bf-committers</a><br>
</blockquote></div><br>