Bug 67356 - REGRESSION (r94093): Private browsing doesn't start with a clean state on Snow Leopard
Summary: REGRESSION (r94093): Private browsing doesn't start with a clean state on Sno...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Page Loading (show other bugs)
Version: 528+ (Nightly build)
Hardware: Mac OS X 10.6
: P1 Normal
Assignee: Alexey Proskuryakov
URL:
Keywords: Regression
Depends on:
Blocks:
 
Reported: 2011-08-31 21:13 PDT by Khoo Yit Phang
Modified: 2011-09-02 12:38 PDT (History)
4 users (show)

See Also:


Attachments
proposed fix (1.92 KB, patch)
2011-09-01 16:43 PDT, Alexey Proskuryakov
darin: review+
webkit.review.bot: commit-queue-
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Khoo Yit Phang 2011-08-31 21:13:04 PDT
There appears to have been a change in Safari's Private Browsing behavior between the r94061 and r94145 nightlies in the following scenario:
1) log in to some website (e.g., facebook.com or gmail.com);
2) turn on Private Browsing;
3) reload the website in step 1.

In r94061, reloading in step 3 be logged out as if cookies were reset; whereas in r94145, reloading in step 3 would remain logged in.

Looking at the SVN commit logs between these revisions, there doesn't seem to be any that explicitly calls out such a change, though r94093 appears to be related to cookies. Is this an expected change? I much rather prefer the old behavior as it is more thorough in ensuring privacy.
Comment 1 Alexey Proskuryakov 2011-09-01 10:20:32 PDT
Per your analysis, it seems likely that I broke something, will investigate ASAP.
Comment 2 Alexey Proskuryakov 2011-09-01 11:45:18 PDT
This only happens on Snow Leopard, not on Lion. Investigating...
Comment 3 Alexey Proskuryakov 2011-09-01 16:43:55 PDT
Created attachment 106062 [details]
proposed fix
Comment 4 WebKit Review Bot 2011-09-01 17:22:16 PDT
Comment on attachment 106062 [details]
proposed fix

Rejecting attachment 106062 [details] from commit-queue.

Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '-..." exit_code: 1

Last 500 characters of output:
f0d9405ba69418331dcdfeb2b468b45aa2d1040f
r94371 = 8b34cc661db8cceb5d89e4be6eecf8ad466a75fc
Done rebuilding .git/svn/refs/remotes/origin/master/.rev_map.268f45cc-cd09-0410-ab3c-d52691b4dbfc
First, rewinding head to replay your work on top of it...
Fast-forwarded master to refs/remotes/origin/master.
Updating chromium port dependencies using gclient...

________ running '/usr/bin/python gyp_webkit' in '/mnt/git/webkit-commit-queue/Source/WebKit/chromium'
Updating webkit projects from gyp files...

Full output: http://queues.webkit.org/results/9586217
Comment 5 Alexey Proskuryakov 2011-09-01 21:48:52 PDT
Comment on attachment 106062 [details]
proposed fix

What did that mean?
Comment 6 WebKit Review Bot 2011-09-01 22:48:04 PDT
Comment on attachment 106062 [details]
proposed fix

Rejecting attachment 106062 [details] from commit-queue.

Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '-..." exit_code: 1

Last 500 characters of output:
d546a50615913604e5f2b269d0f7ca6ccf93b070
r94386 = 5e8e385bd2e33d41d28bd86612dafef1390d00bd
Done rebuilding .git/svn/refs/remotes/origin/master/.rev_map.268f45cc-cd09-0410-ab3c-d52691b4dbfc
First, rewinding head to replay your work on top of it...
Fast-forwarded master to refs/remotes/origin/master.
Updating chromium port dependencies using gclient...

________ running '/usr/bin/python gyp_webkit' in '/mnt/git/webkit-commit-queue/Source/WebKit/chromium'
Updating webkit projects from gyp files...

Full output: http://queues.webkit.org/results/9581597
Comment 7 Alexey Proskuryakov 2011-09-01 23:25:56 PDT
Ah right, I accidentally removed reviewer line. Will fix and land manually.
Comment 8 Alexey Proskuryakov 2011-09-02 12:38:54 PDT
Committed <trac.webkit.org/r94426>.