Bug 104862

Summary: "loader_unittests should reset state between runs"
Product: WebKit Reporter: Julie Parent <jparent>
Component: New BugsAssignee: Julie Parent <jparent>
Status: RESOLVED FIXED    
Severity: Normal CC: dpranke, ojan, tony, webkit.review.bot
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Patch
none
Patch for landing none

Julie Parent
Reported 2012-12-12 16:50:42 PST
"loader_unittests should reset state between runs"
Attachments
Patch (2.62 KB, patch)
2012-12-12 16:54 PST, Julie Parent 		no flags
DetailsFormatted DiffDiff
Patch for landing (2.65 KB, patch)
2012-12-13 08:55 PST, Julie Parent 		no flags
DetailsFormatted DiffDiff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.
Julie Parent
Comment 1 2012-12-12 16:54:20 PST
Created attachment 179152 [details] Patch
Dirk Pranke
Comment 2 2012-12-12 17:01:52 PST
Comment on attachment 179152 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=179152&action=review seems reasonable. > Tools/TestResultServer/static-dashboards/loader_unittests.js:91 > + parseCrossDashboardParameters(); was the parseCrossDashboardParameters() line meant to be part of this?
Julie Parent
Comment 3 2012-12-12 17:15:17 PST
Yeah, parseCrossDashboardParameters() was intentional. One of the other tests calls it, and this test was depending on the modifications to global variables that parseCrossDashboardParameters made (and of course, to that other test running before this one!!!). Once I added the call to resetGlobals(), that state was getting reset, so this test needs to explicitly call it
Dirk Pranke
Comment 4 2012-12-12 17:22:16 PST
(In reply to comment #3) > Yeah, parseCrossDashboardParameters() was intentional. One of the other tests calls it, and this test was depending on the modifications to global variables that parseCrossDashboardParameters made (and of course, to that other test running before this one!!!). Once I added the call to resetGlobals(), that state was getting reset, so this test needs to explicitly call it Yeah, sorry, I figured that out after looking at the code but forgot to delete the comment.
WebKit Review Bot
Comment 5 2012-12-12 17:37:13 PST
Comment on attachment 179152 [details] Patch Rejecting attachment 179152 [details] from commit-queue. Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '-..." exit_code: 1 /mnt/git/webkit-commit-queue/Tools/ChangeLog neither lists a valid reviewer nor contains the string "Unreviewed" or "Rubber stamp" (case insensitive). Full output: http://queues.webkit.org/results/15312045
Julie Parent
Comment 6 2012-12-13 08:55:13 PST
Created attachment 179279 [details] Patch for landing
WebKit Review Bot
Comment 7 2012-12-13 09:09:46 PST
Comment on attachment 179279 [details] Patch for landing Clearing flags on attachment: 179279 Committed r137599: <http://trac.webkit.org/changeset/137599>
WebKit Review Bot
Comment 8 2012-12-13 09:09:50 PST
All reviewed patches have been landed. Closing bug.
Ojan Vafai
Comment 9 2012-12-14 14:59:28 PST
Comment on attachment 179279 [details] Patch for landing Hopefully we can get rid of this once Loader no longer uses any globals?
Note You need to log in before you can comment on or make changes to this bug.