[Scons-dev] Why we need to release separate SCons3 for Python 3

[Bill Deegan]

All,

The plan (as Gary stated) is to move to 2.7 (and maybe 2.6), so the release
will start warning if using (soon to be) deprecated python < 2.7.
Similar to the process when we moved forward from python 1.5.2 to 2.4...

:)


I'm working on making those changes today. (version checking and some docs).
I've also updated the wiki FAQ to explicitly state that python >= 3.0 is
not currently supported.
I've seen many emails and IRC queries about this issue, so best to state
that explicitly wherever we refer to the python version.

One we have a plan in place and some movement we can update the docs/web to
indicate a work in progress to support python 3.x..

-Bill


On Fri, Feb 22, 2013 at 11:56 AM, Kenny, Jason L <jason.l.kenny at intel.com>wrote:


>  So as I recall, the issue is that Scons still does not use subprocess,

> it used other stuff to do the spawn, and these apis did not like Unicode

> strings. There subprocess has also had a number of bugs that needed to be

> fixed. I don’t think this is a major issue. I have this all working for me

> in Parts, as I provide a custom spawner for Scons. Given that the registry

> functions, environment, etc on windows return Unicode strings I feel that

> this should not be a major issue. I believe I was arguing that at the time

> we should try to use a api in SCons for all output, to better supporting

> logging and warning , error messages as well as cross-platform coloring

> like I have in parts. This allow for dealing with the console encoding

> issues ( if any exist for the system) very easily and in a targeted way .*

> ***

>

> ** **

>

> Going forward I really think we need to make a branch that targets 2.7 and

> 3.x and start fixing it to work.****

>

> ** **

>

> Jason****

>

> ** **

>

> *From:* scons-dev-bounces at scons.org [mailto:scons-dev-bounces at scons.org] *On

> Behalf Of *Gary Oberbrunner

> *Sent:* Friday, February 22, 2013 1:35 PM

>

> *To:* SCons developer list

> *Subject:* Re: [Scons-dev] Why we need to release separate SCons3 for

> Python 3****

>

> ** **

>

> ** **

>

> On Fri, Feb 22, 2013 at 2:20 PM, Kenny, Jason L <jason.l.kenny at intel.com>

> wrote:****

>

> On the Unicode side, I am not sure what the issue is. We have Unicode

> issue with python on windows today already, and these seem to be working

> fine. Given that I deal with the Unicode for our product here, I happy to

> look at this more, or help out. However there should be a major issue I

> believe with scons and Unicode. ****

>

> ** **

>

> I assume you mean there should NOT be a major issue.****

>

> ** **

>

> We have in fact had issues with Unicode before, especially in SConf.  What

> encoding should we assume for stdout of a subprocess?  Sometimes processes

> check the locale and encode strings specifically for the locale, assuming

> they are writing to a console (terminal) which will display them (and

> sometimes they even check for a tty and do something different).  If we

> guess wrong for the locale of any given string from a subprocess, we may

> get a decoding error which will throw an exception, or at least we will

> mangle the text.  This happened quite a bit a few years ago in SCons, so I

> and others added some hacky code to work around it.  But it never seemed

> right.  For one example, search the dev archives for " Is SCons really this

> broken" (

> http://thread.gmane.org/gmane.comp.programming.tools.scons.devel/9873).

> Jason, you were part of that discussion at the time.****

>

>  ****

>

> --

> Gary ****

>

> _______________________________________________

> Scons-dev mailing list

> Scons-dev at scons.org

> http://two.pairlist.net/mailman/listinfo/scons-dev

>

>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/scons-dev/attachments/20130222/0e3c8111/attachment.htm>