Bug 111058 – [EFL][Qt] plugins/plugin-clip-subframe.html is failing

• Status: RESOLVED
• Resolution: WONTFIX
• Component: WebKitGTK

[Zan Dobersek]

The plugins/plugin-clip-subframe.html layout test is failing, introduced in r144236.
http://trac.webkit.org/changeset/144236
http://test-results.appspot.com/dashboards/flakiness_dashboard.html#group=%40ToT%20-%20webkit.org&tests=plugins%2Fplugin-clip-subframe.html

Diff:
--- /home/slave/webkitgtk/gtk-linux-64-release/build/layout-test-results/plugins/plugin-clip-subframe-expected.txt
+++ /home/slave/webkitgtk/gtk-linux-64-release/build/layout-test-results/plugins/plugin-clip-subframe-actual.txt
@@ -1,3 +1,2 @@
-CONSOLE MESSAGE: line 3: NPP_SetWindow: non-NULL window, Rect {18, 218, 300, 150}, Clip Rect {65518, 65318, 65518, 65318}, Type 1
-CONSOLE MESSAGE: line 3: NPP_SetWindow: non-NULL window, Rect {18, 218, 300, 150}, Clip Rect {0, 0, 90, 150}, Type 1
+CONSOLE MESSAGE: PLUGIN: NPP_New: Could not find a test named "log-npp-set-window", maybe its .cpp file wasn't added to the build system?

[Ádám Kallai]

This test is failing on Qt too. The diff is same.

http://build.webkit.sed.hu/results/x86-64%20Linux%20Qt%20Release/r144298%20(48665)/plugins/plugin-clip-subframe-pretty-diff.html

[Zan Dobersek]

The test is run properly after the Tools/DumpRenderTree/TestNetscapePlugin/Tests/LogNPPSetWindow.cpp file is added to the build, though the results vary across platforms. For instance, this is the  output on GTK:
CONSOLE MESSAGE: line 3: NPP_SetWindow: non-NULL window, Rect {0, 0, 300, 150}, Clip Rect {0, 0, 0, 0}, Type 1
CONSOLE MESSAGE: line 3: NPP_SetWindow: non-NULL window, Rect {0, 0, 300, 150}, Clip Rect {0, 0, 0, 0}, Type 1
CONSOLE MESSAGE: line 3: NPP_SetWindow: non-NULL window, Rect {0, 0, 300, 150}, Clip Rect {0, 0, 90, 150}, Type 1

[Sudarsana Nagineni (babu)]

Failing on EFL too.

http://build.webkit.org/results/EFL%20Linux%2064-bit%20Debug%20WK2/r144439%20%289980%29/plugins/plugin-clip-subframe-diff.txt

[John Bauman]

Ah, looks like it just needs baselines added. The important part is that the clip rect is empty at the beginning and not empty at the end.

[Zan Dobersek]

Fixing the issue on GTK in bug #112570.

[Michael Catanzaro]

Closing this bug because the EFL port has been removed from trunk.

If you feel this bug applies to a different upstream WebKit port and was closed in error, please either update the title and reopen the bug, or leave a comment to request this.