Bug 98183 – Remove MSYS-related changes to the http testing infrastructure.

• Component: Tools / Tests
• Status: RESOLVED FIXED

[Raphael Kubo da Costa (:rakuco)]

Remove MSYS-related changes to the http testing infrastructure.

[Raphael Kubo da Costa (:rakuco)]

Created attachment 166725 [details]
Patch

[Eric Seidel (no email)]

Comment on attachment 166725 [details]
Patch

LGTM.

[Dirk Pranke]

Roger, is this okay with you? Do you ever use msys?

[Raphael Kubo da Costa (:rakuco)]

To make it clear: these changes were added in r77161 (bug 50036) by RIM. I talked to a few RIM guys today, and it looks like msys and mingw were used in older products and support for this in trunk is not needed anymore. This patch basically reverts that commit, modulo the copyright changes.

Perhaps we could rid of all the msys-related code (after this change it will consist of one check in o-r-w-t and one in commit-log-editor), but I'm not sure, since I see some commits related to that from aroben in his Apple days.

[Raphael Kubo da Costa (:rakuco)]

(In reply to comment #4)
> Perhaps we could rid

Sorry, s/rid/get rid/.

[Dirk Pranke]

I do consider msys-on-win a supported configuration for the python code, for what it's worth (I use it from time-to-time with chromium). That said, there's very little msys-specific python code.

[Raphael Kubo da Costa (:rakuco)]

(In reply to comment #6)
> I do consider msys-on-win a supported configuration for the python code, for what it's worth (I use it from time-to-time with chromium). That said, there's very little msys-specific python code.

I see. Things look better in the webkitpy world anwyay; sys.platform does not seem to have a separate identifier for mingw, and no code tries to probe that and follow a different code path/use different files.

[Dirk Pranke]

I'm pretty sure sys.platform returns 'msys'? I'm in the middle of rebuilding my win machine so I can't confirm this at the moment ...

[Raphael Kubo da Costa (:rakuco)]

(In reply to comment #8)
> I'm pretty sure sys.platform returns 'msys'? I'm in the middle of rebuilding my win machine so I can't confirm this at the moment ...

It does not seem so according to http://docs.python.org/library/sys.html#sys.platform

[Eric Seidel (no email)]

Unless RIM is still actively developing something on MinGW/MYSYS, we should shoot it all in the head.  Dead code should be actively culled. :)

[Raphael Kubo da Costa (:rakuco)]

(In reply to comment #10)
> Unless RIM is still actively developing something on MinGW/MYSYS, we should shoot it all in the head.  Dead code should be actively culled. :)

I concur :-) My worry is that some other vendor such as Apple (given aroben's commits at the time) might need some sort of MinGW/MSYS support; if that is not the case, I can finish removing the other isMsys() bits in another patch.

[Eric Seidel (no email)]

Apple will scream if this affects them.  The chances of that are astronomically small.

[Dirk Pranke]

That's why I copied Roger on this and asked in comment #3 :).

[Raphael Kubo da Costa (:rakuco)]

(In reply to comment #13)
> That's why I copied Roger on this and asked in comment #3 :).

Hmm, I still don't see any Roger on the CC list here...

[Dirk Pranke]

whoops :).

Roger, is this change okay with you?

[Roger Fong]

Yeah we don't use that anywhere, cull away!

[WebKit Review Bot]

Comment on attachment 166725 [details]
Patch

Clearing flags on attachment: 166725

Committed r130202: <http://trac.webkit.org/changeset/130202>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[Raphael Kubo da Costa (:rakuco)]

(In reply to comment #16)
> Yeah we don't use that anywhere, cull away!

Thanks everyone, I've filed bug 98195 for that.