Use `file --dereference' instead of `readlink -f' + `file'.

Use `file --dereference' instead of `readlink -f' + `file'.

`readlink -f' is not standard and may not be present on all systems (such as
some BSDs).

Instead of the readlink+file combo, just ask file(1) to dereference the file
it is given.

R=mark@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=211783

[Mark Mentovai, 2013-07-15 16:30:20]

You say you’re doing this because readlink is nonstandard, but so is the
--dereference argument to file.

Why don’t we just ask Python to give us the information we want? Something in
the platform or sys modules is likely to fill this need. The right information
can even be gleaned from sys.path.

[Raphael Kubo da Costa (rakuco), 2013-07-15 18:41:23]

On 2013/07/15 16:30:20, Mark Mentovai wrote:
> You say you’re doing this because readlink is nonstandard, but so is the
> --dereference argument to file.

That's true, it's just more widespread than readlink -f.

> Why don’t we just ask Python to give us the information we want? Something in
> the platform or sys modules is likely to fill this need. The right information
> can even be gleaned from sys.path.

Would using the information present in the host's Python libraries work if we're
cross-compiling? Right now, we inspect the libpython.so present in <(sysroot).

[Mark Mentovai, 2013-07-15 19:11:28]

Cross-compiling, right. LGTM.

[commit-bot: I haz the power, 2013-07-15 19:25:03]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/raphael.kubo.da.costa@intel.com/192210...

[commit-bot: I haz the power, 2013-07-15 21:12:28]

Retried try job too often on linux_aura for step(s) unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_aura...

[commit-bot: I haz the power, 2013-07-15 21:16:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/raphael.kubo.da.costa@intel.com/192210...

[commit-bot: I haz the power, 2013-07-15 22:51:23]

Retried try job too often on linux_aura for step(s) unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_aura...

[commit-bot: I haz the power, 2013-07-15 22:57:12]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/raphael.kubo.da.costa@intel.com/192210...

[commit-bot: I haz the power, 2013-07-15 23:59:39]

Retried try job too often on linux_aura for step(s) unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_aura...

[commit-bot: I haz the power, 2013-07-16 08:54:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/raphael.kubo.da.costa@intel.com/192210...

[commit-bot: I haz the power, 2013-07-16 10:34:10]

Retried try job too often on linux_aura for step(s) unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_aura...

[Raphael Kubo da Costa (rakuco), 2013-07-16 11:49:29]

Committed patchset #1 manually as r211783 (presubmit successful).