Bug 120207 – [Qt] Support architecture specific test expectations in NRWT

• Component: Tools / Tests
• Status: RESOLVED INVALID

[Allan Sandfeld Jensen]

We have been missing support for adding qt-arm to the search path in NRWT for some time and since many tests also fail specifically on 32bit x86, it would be great if we could specifically skip tests on either of these platforms.

[Allan Sandfeld Jensen]

Created attachment 209455 [details]
Patch

[Csaba Osztrogonác]

*** Bug 67777 has been marked as a duplicate of this bug. ***

[Csaba Osztrogonác]

Comment on attachment 209455 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=209455&action=review

> Tools/Scripts/webkitpy/port/qt.py:146
> +        #              |
> +        #     (qt-x86|qt-x64|qt-arm)
>          #               \
>          #    (qt-linux|qt-mac|qt-win)
>          #                |

What if removing qt-linux, qt-mac and qt-mac at all?
- qt-linux: empty, haven't been used ever
- qt-win: empty, nobody tried to run layout tests on win in the past
- qt-mac: no bot, not maintained long long time ago (last skipping was on 2012.Jun 20. - r120808)

[Peter Gal]

View in context: https://bugs.webkit.org/attachment.cgi?id=209455&action=review

> Tools/Scripts/webkitpy/port/qt.py:133
> +        return py_machine

Shouldn't we give some kind of warning/error in case that there is no match to any of the known architecture? It should not be a really common case, but maybe it's better if we play it really safe.

[Allan Sandfeld Jensen]

(In reply to comment #4)
> View in context: https://bugs.webkit.org/attachment.cgi?id=209455&action=review
> 
> > Tools/Scripts/webkitpy/port/qt.py:133
> > +        return py_machine
> 
> Shouldn't we give some kind of warning/error in case that there is no match to any of the known architecture? It should not be a really common case, but maybe it's better if we play it really safe.

It could return empty string that would make sure the architecture search path doesn't get added. Though this version means anyone running a sh4 or mips architecture could add their own expectations.

[Allan Sandfeld Jensen]

Created attachment 210337 [details]
Patch

Do not add search paths for unknown architectures

[Allan Sandfeld Jensen]

(In reply to comment #3)
> (From update of attachment 209455 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=209455&action=review
> 
> > Tools/Scripts/webkitpy/port/qt.py:146
> > +        #              |
> > +        #     (qt-x86|qt-x64|qt-arm)
> >          #               \
> >          #    (qt-linux|qt-mac|qt-win)
> >          #                |
> 
> What if removing qt-linux, qt-mac and qt-mac at all?
> - qt-linux: empty, haven't been used ever
> - qt-win: empty, nobody tried to run layout tests on win in the past
> - qt-mac: no bot, not maintained long long time ago (last skipping was on 2012.Jun 20. - r120808)

I doubt it will ever be used in trunk, but I might try to run the test on other platforms in the branch (if possible). Anyway, we can remove the other operating systems in another patch.

[Anders Carlsson]

Comment on attachment 210337 [details]
Patch

Qt has been removed, clearing review flags.

[Jocelyn Turcotte]

=== Bulk closing of Qt bugs ===

If you believe that this bug report is still relevant for a non-Qt port of webkit.org, please re-open it and remove [Qt] from the summary.

If you believe that this is still an important QtWebKit bug, please fill a new report at https://bugreports.qt-project.org and add a link to this issue. See http://qt-project.org/wiki/ReportingBugsInQt for additional guidelines.