Re: [Scons-dev] modernize 0.4 is released for Python 3 porting experiments
Am 26.10.2014 um 22:40 schrieb anatoly techtonik: On Fri, Oct 24, 2014 at 12:47 AM, Russel Winder rus...@winder.org.uk wrote: On Thu, 2014-10-23 at 12:38 -0400, Gary Oberbrunner wrote: On Thu, Oct 23, 2014 at 12:07 PM, Russel Winder rus...@winder.org.uk wrote: I think we need to resync or abandon the current Python 3 branch, currently it is so far behind default/tip that it is not worth working on as is. Really? I merged default into it not too long ago. About 2 months according to my repository! (Actually I though it was more than that, I missed the latest merge until I read your email and checked. I will try and play with things next week.) Mercurial developers are very skeptical about Python 3 porting for system level utilities, so I am not alone: http://mercurial.selenic.com/wiki/Python3 I would say the situation for Scons is slightly different than the one for Mercurial: - In order to use Scons you have to have a python installation of your own - Mercurial installs its own one. - Using Scons means you have to write your build script (SConstruct) in that version of the Python language, that is also used by Scons. I am currently working on a project, where Python is used for several things and the team decided to go for Python 3.3. So in order to need just a single Python installation for being able to contribute to this project, i took the branch https://bitbucket.org/userzimmermann/scons, built a scons-local package from that and integrated that into our repository. That works quite well with PYthon 3.3, although we sometime experience problem when doing parallel builds (e.g -j4) - Scons is trying to access a build target while a custom builder still has a handle open for that file. But since we are using a 'hacked' version of Scons, i will surely not complain. So i guess from my situation you can see, that it would be much to my regret if support for Python 3 in Scons would be considered subordinate. On the contrary: I find Python 2 old-fashioned and i am waiting for Scons' Python 3 support wishfully. Regards, Rocco ___ Scons-dev mailing list Scons-dev@scons.org https://pairlist2.pair.net/mailman/listinfo/scons-dev
Re: [Scons-dev] modernize 0.4 is released for Python 3 porting experiments
On Mon, 2014-10-27 at 16:41 +0100, getspammed wrote: […] I am currently working on a project, where Python is used for several things and the team decided to go for Python 3.3. So in order to need just a single Python installation for being able to contribute to this project, i took the branch https://bitbucket.org/userzimmermann/scons, built a scons-local package from that and integrated that into our repository. That works quite well with PYthon 3.3, although we sometime experience problem when doing parallel builds (e.g -j4) - Scons is trying to access a build target while a custom builder still has a handle open for that file. But since we are using a 'hacked' version of Scons, i will surely not complain. So i guess from my situation you can see, that it would be much to my regret if support for Python 3 in Scons would be considered subordinate. On the contrary: I find Python 2 old-fashioned and i am waiting for Scons' Python 3 support wishfully. Is there any way we can integrate your work into the SCons repository so that current default/tip can be merged into the python3 branch with your work so we have an official releasable version of SCons for Python 3 that is harmonious with the current SCons release? -- Russel. = Dr Russel Winder t: +44 20 7585 2200 voip: sip:russel.win...@ekiga.net 41 Buckmaster Roadm: +44 7770 465 077 xmpp: rus...@winder.org.uk London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder signature.asc Description: This is a digitally signed message part ___ Scons-dev mailing list Scons-dev@scons.org https://pairlist2.pair.net/mailman/listinfo/scons-dev
Re: [Scons-dev] modernize 0.4 is released for Python 3 porting experiments
On Mon, 2014-10-27 at 00:40 +0300, anatoly techtonik wrote: […] Mercurial developers are very skeptical about Python 3 porting for system level utilities, so I am not alone: http://mercurial.selenic.com/wiki/Python3 […] Mercurial has been built on the fundamental assumption that string is a sequence of bytes. They therefore cannot easily switch from Python 2 to Python 3. Their problem. This issue doesn't affect SCons as far as I am aware. -- Russel. = Dr Russel Winder t: +44 20 7585 2200 voip: sip:russel.win...@ekiga.net 41 Buckmaster Roadm: +44 7770 465 077 xmpp: rus...@winder.org.uk London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder signature.asc Description: This is a digitally signed message part ___ Scons-dev mailing list Scons-dev@scons.org https://pairlist2.pair.net/mailman/listinfo/scons-dev
Re: [Scons-dev] modernize 0.4 is released for Python 3 porting experiments
On Fri, Oct 24, 2014 at 12:47 AM, Russel Winder rus...@winder.org.uk wrote: On Thu, 2014-10-23 at 12:38 -0400, Gary Oberbrunner wrote: On Thu, Oct 23, 2014 at 12:07 PM, Russel Winder rus...@winder.org.uk wrote: I think we need to resync or abandon the current Python 3 branch, currently it is so far behind default/tip that it is not worth working on as is. Really? I merged default into it not too long ago. About 2 months according to my repository! (Actually I though it was more than that, I missed the latest merge until I read your email and checked. I will try and play with things next week.) Mercurial developers are very skeptical about Python 3 porting for system level utilities, so I am not alone: http://mercurial.selenic.com/wiki/Python3 ___ Scons-dev mailing list Scons-dev@scons.org https://pairlist2.pair.net/mailman/listinfo/scons-dev
Re: [Scons-dev] modernize 0.4 is released for Python 3 porting experiments
On Thu, Oct 23, 2014 at 12:07 PM, Russel Winder rus...@winder.org.uk wrote: I think we need to resync or abandon the current Python 3 branch, currently it is so far behind default/tip that it is not worth working on as is. Really? I merged default into it not too long ago. -- Gary ___ Scons-dev mailing list Scons-dev@scons.org https://pairlist2.pair.net/mailman/listinfo/scons-dev
Re: [Scons-dev] modernize 0.4 is released for Python 3 porting experiments
On Thu, 2014-10-23 at 12:38 -0400, Gary Oberbrunner wrote: On Thu, Oct 23, 2014 at 12:07 PM, Russel Winder rus...@winder.org.uk wrote: I think we need to resync or abandon the current Python 3 branch, currently it is so far behind default/tip that it is not worth working on as is. Really? I merged default into it not too long ago. About 2 months according to my repository! (Actually I though it was more than that, I missed the latest merge until I read your email and checked. I will try and play with things next week.) -- Russel. = Dr Russel Winder t: +44 20 7585 2200 voip: sip:russel.win...@ekiga.net 41 Buckmaster Roadm: +44 7770 465 077 xmpp: rus...@winder.org.uk London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder signature.asc Description: This is a digitally signed message part ___ Scons-dev mailing list Scons-dev@scons.org https://pairlist2.pair.net/mailman/listinfo/scons-dev
[Scons-dev] modernize 0.4 is released for Python 3 porting experiments
Hi, Just a quick note for those who want to make SCons run with Python 3. New version of modernize is released, which is more idempotent that previous stuff and can produce better 2/3 compatible code: https://pypi.python.org/pypi/modernize It uses six. -- anatoly t. ___ Scons-dev mailing list Scons-dev@scons.org https://pairlist2.pair.net/mailman/listinfo/scons-dev
