Bug 209000

Summary: Resource load statistics data summary should return all third party data
Product: WebKit Reporter: Kate Cheney <katherine_cheney>
Component: WebKit Misc.Assignee: Kate Cheney <katherine_cheney>
Status: RESOLVED FIXED    
Severity: Normal CC: bfulgham, commit-queue, webkit-bug-importer, wilander
Priority: P2 Keywords: InRadar
Version: WebKit Nightly Build   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Patch
none
Patch none

Kate Cheney
Reported 2020-03-12 09:19:08 PDT
It would be more informative to return all third party data, not just prevalent domains.
Attachments
Patch (8.96 KB, patch)
2020-03-12 09:24 PDT, Kate Cheney 		no flags
DetailsFormatted DiffDiff
Patch (12.44 KB, patch)
2020-03-12 13:13 PDT, Kate Cheney 		no flags
DetailsFormatted DiffDiff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.
Kate Cheney
Comment 1 2020-03-12 09:19:35 PDT
<rdar://problem/60348306>
Kate Cheney
Comment 2 2020-03-12 09:24:56 PDT
Created attachment 393383 [details] Patch
John Wilander
Comment 3 2020-03-12 10:33:38 PDT
Comment on attachment 393383 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=393383&action=review r=me. Please address my comment on the cookie blocking mode. > Source/WebKit/NetworkProcess/Classifier/ResourceLoadStatisticsDatabaseStore.cpp:844 > + || sortedStatistics.bindText(1, "%") I think this should be made conditional on the configured WebCore::ThirdPartyCookieBlockingMode. If == WebCore::ThirdPartyCookieBlockingMode::All, send back all third party data, otherwise keep the old behavior. You already have a thirdPartyCookieBlockingMode() getter in the ResourceLoadStatisticsStore base class. Then your API tests might have to enable WebCore::ThirdPartyCookieBlockingMode::All if they aren't already running under that configuration. > Tools/TestWebKitAPI/Tests/WebKitCocoa/ResourceLoadStatistics.mm:986 > + // Set two third parties to be prevalent, leave evil1.com as non-prevalent to ensure Personally, I'd prefer not to call prevalent domains "evil" since there's nothing inherently evil about them. But I know it's existing code.
Kate Cheney
Comment 4 2020-03-12 13:13:50 PDT
Created attachment 393405 [details] Patch
Kate Cheney
Comment 5 2020-03-12 13:14:26 PDT
(In reply to katherine_cheney from comment #4) > Created attachment 393405 [details] > Patch Thanks John! Made these changes. Waiting on EWS before cq+
WebKit Commit Bot
Comment 6 2020-03-12 14:22:53 PDT
Comment on attachment 393405 [details] Patch Clearing flags on attachment: 393405 Committed r258357: <https://trac.webkit.org/changeset/258357>
WebKit Commit Bot
Comment 7 2020-03-12 14:22:55 PDT
All reviewed patches have been landed. Closing bug.
Radar WebKit Bug Importer
Comment 8 2020-03-12 14:23:17 PDT
<rdar://problem/60387086>
Note You need to log in before you can comment on or make changes to this bug.