Closed Bug 440154 Opened 16 years ago Closed 16 years ago

about:config filter not applied after Firefox restarts.

Categories

(Firefox :: Session Restore, defect)

defect
Not set
minor

Tracking

()

VERIFIED FIXED
Firefox 3.1a1

People

(Reporter: alexander, Assigned: zeniko)

References

()

Details

Attachments

(1 file)

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9) Gecko/2008052906 Firefox/3.0
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9) Gecko/2008052906 Firefox/3.0

If you filter for a word in about:config, then close Firefox (saving the open tabs) and then you open Firefox, then Firefox will reopen the about:config tab and remember the word you used to filter, but will show all the preferences ignoring the filter.

Reproducible: Always

Steps to Reproduce:
1.Open a new tab and type "about:config" in the URL bar and hit enter.
2.Type something in the filter text box (for example "gfx").
3.Firefox correctly filters the preferences (in our example, you get 3 results).
4.Close Firefox (select to save tabs and quit in the menu that will appear).
5.Open Firefox again.

Actual Results:  
Firefox will open the tabs correctly, including the about:config tab, and will remember what you had typed in the filter text box ("gfx" in our example), but the filter will NOT be applied and ALL entries will be shown.

Expected Results:  
The filter should be applied, and only the preferences matching the filter should be shown.

Has the easy and obvious workaround: Change the text in the filter box and the filter will now be applied as it should.
Confirmed with Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1a1pre) Gecko/2008061822 Minefield/3.1a1pre

I don't see the problem after unchecking the "This might void your warranty" message.
Status: UNCONFIRMED → NEW
Component: Preferences → Session Restore
Ever confirmed: true
QA Contact: preferences → session.restore
Version: unspecified → Trunk
Probably because it uses onChange or something similar.
The filter is actually applied, the problem is just that the pref data isn't loaded until the warning page is clicked away. Appending the following two lines to ShowPrefs should fix the issue by forcing the filter to be applied again:

>   if (document.getElementById("textbox").value)
>     FilterPrefs();
(In reply to comment #3)
> The filter is actually applied, the problem is just that the pref data isn't
> loaded until the warning page is clicked away.
Attachment #326923 - Flags: review?(gavin.sharp)
Attachment #326923 - Flags: review?(gavin.sharp) → review+
Assignee: nobody → zeniko
Keywords: checkin-needed
OS: Windows XP → All
Hardware: PC → All
Target Milestone: --- → Firefox 3.1a1
http://hg.mozilla.org/index.cgi/mozilla-central/rev/3dd50dd527f4
Status: NEW → RESOLVED
Closed: 16 years ago
Keywords: checkin-needed
Resolution: --- → FIXED
Verified with Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2a1pre) Gecko/20090618 Minefield/3.6a1pre ID:20090618031329
Status: RESOLVED → VERIFIED
Flags: in-litmus?
in-litmus+
https://litmus.mozilla.org/show_test.cgi?id=7791

Is this a dupe of bug 350718?
Flags: in-litmus? → in-litmus+
(In reply to comment #7)
> Is this a dupe of bug 350718?

No, that was a different regression (due to XPCNativeWrapper whereas this bug was due to the new "This might void your warranty" warning).

Thanks for the tests, BTW. One detail (which might also apply to other Session Restore tests), though:

Make sure not to kill Firefox at once, as it only updates the session state every 10 or so seconds (unless you lower browser.sessionstore.interval to 1000 ms and restart before performing the tests). If our testers are too quick (and they will be once they get the hang of it), they might fail to verify what actually works.
Yeah, we actually have documentation about that: http://quality.mozilla.org/preparing-for-session-store-testing

Step 8 tells users to set it to 1 to ensure Session Store works.
(In reply to comment #9)
> Step 8 tells users to set it to 1 to ensure Session Store works.

Nit: That value is in _milli_seconds, so 1000 is a more reasonable value. ;-)
(In reply to comment #10)
> (In reply to comment #9)
> > Step 8 tells users to set it to 1 to ensure Session Store works.
> 
> Nit: That value is in _milli_seconds, so 1000 is a more reasonable value. ;-)

Actually, I feel more comfortable with 1 as some users are surprisingly quick and can perform an action in less than a second.
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: