Bug 203366 – REGRESSION: [ iOS ] ( r251015 ) Layout Test fast/text/whitespace/pre-wrap-overflow-selection.html is flaky

• Status: RESOLVED
• Resolution: FIXED
• Component: Tools / Tests

[Truitt Savell]

The following layout test is flaky on iOS

fast/text/whitespace/pre-wrap-overflow-selection.html

Probable cause:

This was caused by r251015. It is easily reproducible by just running the test in iterations.

Flakiness Dashboard:

https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=fast%2Ftext%2Fwhitespace%2Fpre-wrap-overflow-selection.html

Diff:
--- /Volumes/Data/slave/ios-simulator-13-release-tests-wk2/build/layout-test-results/fast/text/whitespace/pre-wrap-overflow-selection-expected.txt
+++ /Volumes/Data/slave/ios-simulator-13-release-tests-wk2/build/layout-test-results/fast/text/whitespace/pre-wrap-overflow-selection-actual.txt
@@ -22,12 +22,10 @@
       RenderBlock {HR} at (0,198) size 784x2 [border: (1px inset #000000)]
       RenderBlock {PRE} at (0,213) size 108x78 [border: (4px solid #0000FF)]
         RenderText {#text} at (4,4) size 100x70
-          text run at (4,4) width 86: "This   text"
-          text run at (89,4) width 15: "     "
+          text run at (4,4) width 100: "This   text     "
           text run at (4,18) width 71: "will wrap"
           text run at (74,18) width 1: " "
-          text run at (4,32) width 71: "and   fit"
-          text run at (74,32) width 24: "   "
+          text run at (4,32) width 94: "and   fit   "
           text run at (4,46) width 79: "within the"
           text run at (82,46) width 1: " "
           text run at (4,60) width 32: "box."

[Radar WebKit Bug Importer]

<rdar://problem/56580680>

[Truitt Savell]

*** Bug 203363 has been marked as a duplicate of this bug. ***

[Truitt Savell]

marked this as a failure in https://trac.webkit.org/changeset/251538/webkit
while it is investigated and to avoid slowing down iOS EWS.

[Antti Koivisto]

Created attachment 382844 [details]
patch

[Antti Koivisto]

Created attachment 382941 [details]
patch that fixes the underlying flakiness

[Simon Fraser (smfr)]

Comment on attachment 382941 [details]
patch that fixes the underlying flakiness

View in context: https://bugs.webkit.org/attachment.cgi?id=382941&action=review

> Source/WebKit/ChangeLog:13
> +        In this case the randomess is switch between simple and complex line layout triggered by editor state flush. The results

"randomess is switch"

> Source/WebKit/ChangeLog:14
> +        between line layout paths are obsevably identical but produce slightly different render tree dumps.

"obsevably"

> Source/WebKit/WebProcess/InjectedBundle/API/c/WKBundlePage.cpp:518
> +    WebKit::toImpl(page)->flushPendingEditorStateUpdate();

Seems like this is something that we should to at "update the rendering" time? (In a later fix).

[Antti Koivisto]

Created attachment 382976 [details]
patch

[WebKit Commit Bot]

Comment on attachment 382976 [details]
patch

Rejecting attachment 382976 [details] from commit-queue.

New failing tests:
webgl/1.0.3/conformance/extensions/webgl-draw-buffers.html
Full output: https://webkit-queues.webkit.org/results/13220879

[WebKit Commit Bot]

Created attachment 383000 [details]
Archive of layout-test-results from webkit-cq-01 for mac-highsierra

The attached test failures were seen while running run-webkit-tests on the commit-queue.
Bot: webkit-cq-01  Port: mac-highsierra  Platform: Mac OS X 10.13.6

[WebKit Commit Bot]

The commit-queue encountered the following flaky tests while processing attachment 382976 [details]:

imported/w3c/web-platform-tests/websockets/bufferedAmount-unchanged-by-sync-xhr.any.worker.html bug 202003 (author: [e-mail])
The commit-queue is continuing to process your patch.

[WebKit Commit Bot]

Comment on attachment 382976 [details]
patch

Clearing flags on attachment: 382976

Committed r252172: <https://trac.webkit.org/changeset/252172>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.