| Summary: | fast/loader/input-element-page-cache-crash.html fails on Chromium | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Mihai Parparita <mihaip> | ||||||
| Component: | New Bugs | Assignee: | Mihai Parparita <mihaip> | ||||||
| Status: | RESOLVED FIXED | ||||||||
| Severity: | Normal | CC: | commit-queue, tony | ||||||
| Priority: | P2 | ||||||||
| Version: | 528+ (Nightly build) | ||||||||
| Hardware: | Other | ||||||||
| OS: | OS X 10.5 | ||||||||
| Attachments: |
|
||||||||
|
Description
Mihai Parparita
2010-08-27 19:28:42 PDT
Created attachment 65804 [details]
Patch
Tony, since you're the gardener, can you take a look at this? Comment on attachment 65804 [details]
Patch
Should the test initialize sessionStorage.finishTest to false? As in, if I run the test twice in a row, will sessionStorage.finishTest be confused?
Alternately, if another test used the same variable, could that confuse this test? Created attachment 65860 [details]
Patch
I had assumed that sessionStorage gets reset between tests, and I even tried using --iterations 2 when running this test to see if it passes if run twice. However, it turned out that the test output was the same whether it navigated to the page and back or just finished immediately. I added some extra logging to make sure that it actually goes through the steps, and sure enough it failed. I added a cleanup step at the end which removes the sessionStorage variable, and now it passes when run twice (and still passes with the Chromium DRT). Comment on attachment 65860 [details]
Patch
LGTM
Comment on attachment 65860 [details] Patch Clearing flags on attachment: 65860 Committed r66340: <http://trac.webkit.org/changeset/66340> All reviewed patches have been landed. Closing bug. |