[PATCH] python2.6: use subprocess if available

[TK Soh]

On Tue, Oct 7, 2008 at 6:17 AM, Benoit Boissinot <bboissin at gmail.com> wrote:
> On Tue, Oct 7, 2008 at 2:31 AM, TK Soh <teekaysoh at gmail.com> wrote:
>> On Sun, Oct 5, 2008 at 7:44 PM, Dirkjan Ochtman <dirkjan at ochtman.nl> wrote:
>>> Dirkjan Ochtman <dirkjan <at> ochtman.nl> writes:
>>>> # HG changeset patch
>>>> # User Dirkjan Ochtman <dirkjan <at> ochtman.nl>
>>>> # Date 1223235326 -7200
>>>> # Node ID 070d038a50c52cedcdc8209895237744a045ce13
>>>> # Parent  8b874f8cd29f46f9412cf97352cc5ba5528f10f3
>>>> python2.6: use subprocess if available
>>>
>>> This is kind of a proof concept. It seems to pass the test suite, but I'm a
>>> little suspicious about it, I think some of these cases might not be covered.
>>>
>>> I haven't yet fixed the usage in run-tests.py, because I don't think we want it
>>> to depend on mercurial.util.
>>>
>>> So, comments would be very welcome!
>>
>> I am just curious about the benefit of using subprocess, if the
>> existing implementation already meet the goals. Does hg behave/perform
>> differently with/without subprocess?
>
> The benefit is that you don't get a deprecation warning all over the place when
> using os.popen* in python2.6.

Thanks for the clarification. Never have a chance to try python 2.6 yet.