Bug 142622 – Unreviewed GTK+ gardening on 12th Mar.

• Component: WebKitGTK
• Status: RESOLVED INVALID

[Hunseop Jeong]

Mark attachment-foo test and update the missing tests to existing bugs.

[Hunseop Jeong]

Created attachment 248515 [details]
Patch

[Carlos Garcia Campos]

Comment on attachment 248515 [details]
Patch

Shouldn't we use Skip instead of Missing in these cases?

[Hunseop Jeong]

Created attachment 248575 [details]
Patch

[Hunseop Jeong]

I changed Missing to Skip.

[Carlos Alberto Lopez Perez]

Comment on attachment 248575 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=248575&action=review

> LayoutTests/platform/gtk/TestExpectations:2376
> +# GTK port doesn't support attachment feature yet.
> +Bug(GTK) fast/attachment/attachment-progress.html [ Skip ]
> +Bug(GTK) fast/attachment/attachment-select-on-click-inside-user-select-all.html [ Skip ]
> +Bug(GTK) fast/attachment/attachment-select-on-click.html [ Skip ]
> +Bug(GTK) fast/attachment/attachment-subtitle.html [ Skip ]
> +Bug(GTK) fast/attachment/attachment-title.html [ Skip ]

I think is beter if you can open a bug number telling that we still dont support the feature and then use that bug number here instead of just using "Bug(GTK)".

[Carlos Alberto Lopez Perez]

Regarding the usage of Missing vs Skip.

When there is a test that only has an expectation for another port (Mac for example) and there isn't a general expected. There are two options:

1) The test is specific to that port (Mac), so just use "Skip".

2) The test can be used also on our port, but still nobody generated a valid expected for our port, then use "Missing"

In the second case, if you know that the expected generated for the GTK port is valid you can add it. If you have doubts, better open a bug report CC'ing the author of the test, and mark the test as Missing.

[Hunseop Jeong]

Thanks for your guide, but I closed this bug because it is old one.