[Piglit] [PATCH 1/1] Clean up shader_runner test scripts.

[Eric Anholt]

Stuart Abercrombie <sabercrombie at chromium.org> writes:

>>> These all sound like separate changes to me.
>
> What would you like to split out?
>
> I hit the incorrect shell problem because i needed to run
> variable-index-write.sh.  I needed to run variable-index-write.sh
> because it was outputting the wrong GLSL version and had to be fixed.
> I can't see a reason to fix variable-index-write.sh's version
> outputting but leave it in a state where it won't run on my machine.
> Is there some reason why you want all the other shell scripts to be
> temporarily left in a broken state?
>
> I hit the windows formatting problem because the script I used to
> parse the .shader_test files didn't like the CRs.  There is obviously
> no functional change.
>
> The rest is just making .shader_test files conform more closely with
> what we discussed would be a useful step towards making GLES work:
> GLSL >= always present in [require] with no explicit #version.  Given
> the irregularity of what was in the .shader_test files, what was
> required to reach this varied, but not in any complicated way.

Separate changes get separate commits.  You even explain the story here
like separate changes:

1) I neeeded to change the shell command to get it to even run.

2) I was then able to modify the script to fix the version stuff like I
originally wanted to.

3) After I did so, I found that one of my scripts couldn't parse one of
the files.

> So far I haven't had a very productive experience submitting small
> changes (see my last patch).

I couldn't understand why that last patch (unsigned->int) mattered.  Did
it fix a compile warning or something?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/piglit/attachments/20121204/9767c116/attachment.pgp>