vs toolchain: Move toolchain env computation into its own function.

vs toolchain: Move toolchain env computation into its own function.

Also replace a few ';'s with os.path.pathsep and a few \s with os.path.sep.
No behavior change on Windows; the motivation is that this might make it
possible to do this environment computation on non-Windows hosts too some day.

BUG=495204
Committed: https://crrev.com/44a40f8429c25eaf7bf8577cafbcf3a8df100840
Cr-Commit-Position: refs/heads/master@{#375476}

[Nico, 2016-02-13 22:30:21]

thakis@chromium.org changed reviewers:
+ scottmg@chromium.org

[Nico, 2016-02-13 22:30:22]

Rough ideas:

* Make SetEnv.cmd a hybrid .bat / .sh script and run it under bash on non-Win
* Write a very simple .bat interpreter in python and interpret SetEnv.cmd using
that – SetEnv.cmd uses ~no .bat features

I think the second option is probably more appealing, but let me know if you
have an opinion here. (Motivation is that my manual setting of INCLUDES is
error-prone as me screwing it up a few days ago shows.)

[scottmg, 2016-02-15 17:52:42]

lgtm

I guess it depends how long option 2 gets. We shouldn't really need to add more
fancy features to SetEnv so it's probably ok to pseudo-run it.

Maybe there's some way elegant way we could store it the settings as ~data to
have both OSs "run" it, maybe only to avoid the if/goto for the /x64 flag or
something.

There's also the crazy thing perl does with its wrapper batch files that might
work (or maybe that's what you meant by hybrid).

https://codereview.chromium.org/1697023002/diff/1/build/toolchain/win/setup_t...
File build/toolchain/win/setup_toolchain.py (right):

https://codereview.chromium.org/1697023002/diff/1/build/toolchain/win/setup_t...
build/toolchain/win/setup_toolchain.py:151: env['PATH'] = runtime_dirs +
os.path.pathsep + env['PATH']
I've always used os.pathsep for this ... looks like that's more common from cs,
but not a big deal either way if this seems clearer.

[Nico, 2016-02-15 17:55:24]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2016-02-15 17:55:50]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1697023002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1697023002/1

[commit-bot: I haz the power, 2016-02-15 18:18:09]

Committed patchset #1 (id:1)

[Nico, 2016-02-16 03:25:51]

On 2016/02/15 17:52:42, scottmg wrote:
> lgtm
> 
> I guess it depends how long option 2 gets. We shouldn't really need to add
more
> fancy features to SetEnv so it's probably ok to pseudo-run it.

Looks like it's a bit under 60 lines (without tests):
https://codereview.chromium.org/1183633003/diff/140001/build/toolchain/win/se...
Pursue that for a bit, or abort abort abort?

[commit-bot: I haz the power, 2016-02-16 22:50:29]

Description was changed from

==========
vs toolchain: Move toolchain env computation into its own function.

Also replace a few ';'s with os.path.pathsep and a few \s with os.path.sep.
No behavior change on Windows; the motivation is that this might make it
possible to do this environment computation on non-Windows hosts too some day.

BUG=495204
==========

to

==========
vs toolchain: Move toolchain env computation into its own function.

Also replace a few ';'s with os.path.pathsep and a few \s with os.path.sep.
No behavior change on Windows; the motivation is that this might make it
possible to do this environment computation on non-Windows hosts too some day.

BUG=495204

Committed: https://crrev.com/44a40f8429c25eaf7bf8577cafbcf3a8df100840
Cr-Commit-Position: refs/heads/master@{#375476}
==========

[commit-bot: I haz the power, 2016-02-16 22:50:29]

Patchset 1 (id:??) landed as
https://crrev.com/44a40f8429c25eaf7bf8577cafbcf3a8df100840
Cr-Commit-Position: refs/heads/master@{#375476}