Bug ID
int64 961
1.91M
| Comment ID
int64 3.98k
17.1M
| Author
stringlengths 8
48
| Comment Text
stringlengths 1
64.3k
⌀ |
|---|---|---|---|
484,848 | 4,028,852 |
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1b3) Gecko/20090307 Ubuntu/8.04 (hardy) Shiretoko/3.1b3
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1b3) Gecko/20090307 Ubuntu/8.04 (hardy) Shiretoko/3.1b3
The following scenario was tested using two new firefox profiles and a new user user created with
insert into users values ('test', md5('test'), null, 1, null);
and weave 2.117.
Reproducible: Always
Steps to Reproduce:
1. Create a new profile called test1
2. Install weave, change the server location. Furthermore, I disabled history and tab sync.
3. Go to a web app, enter your login and password and save them in firefox.
4. Sync weave and quit firefox.
5. Create a new profile 'test2'
6. Install weave
7. "Erase local data"
8. Configure weave as in step 2 and sync
9. Go to the web app, username and password were synced and are filled into the form.
10. Remove the password from the input box and enter a new one, log in.
11. Click on "change" in the appearing bar at the top to save the new password in firefox.
12. Sync and quit firefox.
13. Start firefox again with profile "test1".
14. Sync
15. Visit the web app login page and verify that there still is the old password - the password change was saved in test2 but not synced.
I'm attaching the weave logs of test1 and test2.
Actual Results:
Passwords not in sync
Expected Results:
Passwords should be in sync.
|
|
485,277 | 4,032,571 |
User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1b4pre) Gecko/20090325 Firefox/3.5b4
Build Identifier: 0.2.117
When Weave is enabled, new windows (either opened manually or as a popup) open noticeably slower.
On a 2.8GHz MacBook Pro with Shiretoko as of today and javascript.options.jit.chrome=true:
* no add-on vs. weave as the only add-on: almost instantly vs. about 2 seconds
* many add-ons vs. many add-ons + weave: 1 second vs. about 4 seconds
In either case the delay until the new window shows is too long.
Reproducible: Always
Steps to Reproduce:
1. Start Firefox with Weave enabled
2. Open new window and stop the time until it shows
3. Repeat 1 and 2 with Weave disabled
Actual Results:
When Weave is enabled, windows open slowly. Even if Weave isn't configured to actually sync anything.
Expected Results:
New windows show open only marginally slower than with Weave disabled.
Tested on OS X but I suspect it effects all platforms.
|
|
477,186 | 3,962,159 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; fr; rv:1.9.0.6) Gecko/2009011913 Firefox/3.0.6
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 6.0; fr; rv:1.9.0.6) Gecko/2009011913 Firefox/3.0.6
Firefox 3.0.6 was not responding when it tried to display details of a security certificate from COMODO CA Limited after it failed previously to obtain certificate identity.
Reproducible: Always
Steps to Reproduce:
1. Pre-step: clean your security certificates provided by COMODO CA Limited.
2. Go to this webpage (https) asking to accept a certificate from COMODO CA Limited.
Webpage from famous French hosting provider OVH (to order a server e.g.):
https://www.ovh.com/fr/commande/kimsufi.cgi?hard=083sk0
3. Error code: sec_error_unknown_issuer (invalid certificate used by OVH)
4. Click on Add the exception
5. Click on Obtain the certificate
6. Message: Unknown identity (invalid informations when the website is attempting to identify itself)
7. Click on Show
Actual Results:
Firefox is not responding.
Forced to abort it by using Vista.
Expected Results:
Displaying the details of the secutity certificate.
This crash can be reproduced when the certificate has been confirmed (without showing details).
Just go to certificate manager in Firefox (Server tab) and click on "Show" to display the details of COMODO CA Limited certificate relative to OVH webpage.
|
|
478,338 | 3,972,012 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-GB; rv:1.9.1b2) Gecko/20081201 Firefox/3.1b2
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-GB; rv:1.9.1b2) Gecko/20081201 Firefox/3.1b2
Search toolbar have an error called assertion failed when managing search engines.
After clinking manage search engines. Firefox will give an error.
ASSERT: *** Search: _installLocation: engine has no file!
Stack Trace:
0:ENSURE_WARN(false,_installLocation: engine has no file!,2147500037)
1:()
2:()
3:()
4:isDefault([object Object],8,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object])
5:SRCH_SVC_getDefault([object Object])
6:getDefaultEngines([object Object])
7:EngineStore()
8:engineManager_init()
9:onload([object Event]
The search engines window which contain no search engine files will appear,
then it does not response when you clink on any button in that window, it will only close when the X button is pressed on.
Reproducible: Always
Steps to Reproduce:
1.Open Firefox 3.1b2
2.Manage Search engines in the search toolbar
Actual Results:
Firefox gives an error.
ASSERT: *** Search: _installLocation: engine has no file!
Stack Trace:
0:ENSURE_WARN(false,_installLocation: engine has no file!,2147500037)
1:()
2:()
3:()
4:isDefault([object Object],8,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object])
5:SRCH_SVC_getDefault([object Object])
6:getDefaultEngines([object Object])
7:EngineStore()
8:engineManager_init()
9:onload([object Event]
The search engines window which contain no search engines files will appear,
then it does not response when you clink on any button in that window, it will only close when the X button is pressed on.
Expected Results:
A normal window containing your search engines should appear.
|
|
478,445 | 3,972,669 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1b2) Gecko/20081201 Firefox/3.1b2
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1b2) Gecko/20081201 Firefox/3.1b2
The arc function produces a different graphical result when the strokeText function is called afterward.
Reproducible: Always
Steps to Reproduce:
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<style type="text/css">
canvas { border: 2px solid red; }
</style>
<script type="text/javascript" language="JavaScript">
function main() {
var canvas = document.getElementById('test');
var ctx = canvas.getContext('2d');
drawarc(ctx,400,400);
drawarc(ctx,450,400);
ctx.fillText("23",60,50);
drawarc(ctx,500,400);
// strokeText seems to affect previous arc draw somehow in Firefox 3.1b2
// The arc appears longer and fatter
ctx.strokeText("93",60,80);
}
function drawarc(ctx,x,y) {
var radcvt=Math.PI/180;
var ang=30;
var rad=31;
ctx.beginPath();
ctx.arc(x,y,rad,0,ang*radcvt,false);
ctx.closePath();
ctx.fill();
}
</script>
</head>
<body>
<div style="position:absolute;left:50px;top:50px">
<canvas id="test" width="1000" height="1000"></canvas>
<script type="text/javascript" language="JavaScript">
main();
</script>
</div>
</body>
</html>
Actual Results:
Three arcs are produced with the last one different than the first two.
Expected Results:
Three arcs should be produced that are identical.
|
|
479,461 | 3,982,934 |
User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1b3pre) Gecko/20090220 Shiretoko/3.1b3pre
Build Identifier: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1b3pre) Gecko/20090220 Shiretoko/3.1b3pre
A bug regarding private browsing have returned. I'm using the nightly build as of 2009-20-2.
Reproducible: Always
Steps to Reproduce:
1. Have some documents open
2. Enter private browsing mode
3. Close the window
4. Exit private browsing mode
Actual Results:
A window with an empty document is opened
Expected Results:
Should get original session restored
Also, in this state, it thinks you're no longer in private browsing mode.
Work-around: Re-enter private browsing, leave the document open, exit private browsing.
There really should not be any dependency on having a document open in order to properly exit private browsing. I wonder what happens if you have no documents open and you try to ENTER private browsing. Maybe I'll try that.
|
|
466,527 | 3,868,818 |
Steps to reproduce:
1) Set your start page to about:crashes, about:config (or about: anything), or http://ejohn.org/files/border-image/iui/samples/music.html#_home
2) Change the mailnews.start_page_override.mstone to something other than your current setting.
3) (re)Start Thunderbird
Expected Results:
What's new page in a new tab with CSS loaded and looking pretty.
Actual Results:
What's new page in a new tab with no css.
I've taken a brief look, but I'm stumped at the moment.
|
|
467,123 | 3,874,126 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1b3pre) Gecko/20081128 Minefield/3.1b3pre (.NET CLR 3.5.30729)
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1b3pre) Gecko/20081128 Minefield/3.1b3pre
// In a JS component, I load a DOM (from chrome://)
var vXhrComp = Components.classes["@mozilla.org/xmlextras/
xmlhttprequest;1"];
var vReq = vXhrComp.createInstance
(Components.interfaces.nsIXMLHttpRequest);
vReq.open("GET", "chrome://mychrome/content/myFile.xml", false);
vReq.send(null);
var vXml = vReq.responseXML;
//I want to "backup" my dom
var vXmlBackup = vXml.cloneNode(true);
The cloneNode failed : Security Error: Content at moz-nullprincipal:
{c0db3ee1-d7c3-47e4-b6d9-0a08830f15c9} may not load or link to //
chrome://mychrome/content/myFile.xml.
I think it's a problem with transfering nsIPrincipal to the new
Document.
In a xul page in chrome, this code works fine.
Reproducible: Always
Steps to Reproduce:
1. Install the xpi (see attached testcase)
2. In JSconsole write : window.openDialog("chrome://bugclonenode/content/", "bugclonenode", "chrome, resizable");
3. Click on the button : XHR load document : it's ok:
4. Click a second time : call the cloneNode method on the already loaded Document : nok
It works in firefox 3.0.4, but the security warning is published (in the JsConsole).
In firefox 3.1 (and xulrunner 1.9.1), the cloneNode() throw an exception.
|
|
468,991 | 3,889,429 |
User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1b3pre) Gecko/20081210 Shiretoko/3.1b3pre
Build Identifier: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1b3pre) Gecko/20081210 Shiretoko/3.1b3pre
Hello.
I found following problem with firefox 3.1beta2 + 3.1beta3pre on mac.
I start Firefox - first Window opens.
After that I open a second window and try to customize the toolbar (in the new second window!!!).
But, there opens only a white window without content.
Only in the first opened window the customize-window opens normally.
Thanks & regards
Mehmet
Reproducible: Always
Steps to Reproduce:
1. Open Firefox 3.1b2 or 3.1b3pre
2. Open a new "second" window
3. Open in the "second" window "customize..." with rightclick on the toolbar.
Actual Results:
The customize-window is white without content to cusomize the toolbar.
Expected Results:
customize-window should open in the second with content to customize the toolbar.
|
|
1,757,604 | 15,796,530 |
+++ This bug was initially created as a clone of Bug #1460538 +++
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.13; rv:60.0) Gecko/20100101 Firefox/60.0
Build ID: 20180503143129
Steps to reproduce:
By using the reflected URL in some special resource URIs it is possible to inject CSS and bypass CSP: http://demo.vwzq.net/php/ff_chrome.php
Stylesheet injection is not XSS, but still harmful.
Content:
<link rel="stylesheet" href="chrome://browser/content/places/?'*{}*{color:red}">
Actual results:
The reflected stylesheet is interpreted bypassing the strict CSP policy.
Expected results:
I guess the mitigation here should be to avoid reflecting any user content, or at least avoid loading directories.
---
This appears to still be broken.
|
|
1,510,494 | 13,783,391 |
To lessen the attack surface when writing files in unrestricted locations such as the log file and the update.status file the updater should write these files to a restricted location and then copy them to the final location.
*edit* s/then copy them to the final location/the unelevated updater should copy them to the update directory/
|
|
1,510,553 | 13,783,919 |
Created attachment 9028239
rec of the issue.gif
[Affected versions]:
- Firefox 64 and above
[Affected Platforms]:
- All Windows
- All Linux
- All Mac
[Prerequisites]:
- Have an RTL Firefox locale build (eg. Firefox AR) installed.
- Have a Firefox profile with the following prefs in the "about:config" page:
- "browser.newtabpage.activity-stream.asrouterExperimentEnabled" with "true" value;
[Steps to reproduce]:
1. Open the browser with the profile from prerequisites.
2. Navigate to "about:config#asrouter" page.
3. Scroll down to the "SNIPPETS_SEND_TO_DEVICE_TEST" section and click the "Show" button.
4. Click the "Install now" button.
5. Observe the hyperlink from ".The intended recipient of the email must have consented. Learn more" string.
[Expected result]:
- Only the "Learn more" string is hyperlinked.
[Actual result]:
- The "." character is also hyperlinked.
[Additional Notes]:
- This issue is also reproducible on the other Firefox locales but is more visible on the RTL ones.
- This issue is not reproducible with the "Newsletter" or with the "Simple" snippets.
- Attached a screen recording of the issue.
|
|
1,510,565 | 13,784,042 |
Created attachment 9028255
Screen Shot 2018-11-28 at 12.11.34.png
|
|
1,510,934 | 13,786,362 |
Filed by: dluca [at] mozilla.com
https://treeherder.mozilla.org/logviewer.html#?job_id=214547954&repo=autoland
https://queue.taskcluster.net/v1/task/FZ76eTYeQw6ToqAOo7WDLQ/runs/0/artifacts/public/logs/live_backing.log
[task 2018-11-29T06:21:52.615Z] 06:21:52 INFO - GECKO(7014) | MEMORY STAT vsizeMaxContiguous not supported in this build configuration.
[task 2018-11-29T06:21:52.616Z] 06:21:52 INFO - GECKO(7014) | MEMORY STAT | vsize 621MB | residentFast 293MB | heapAllocated 105MB
[task 2018-11-29T06:21:52.617Z] 06:21:52 INFO - TEST-OK | dom/tests/mochitest/ajax/offline/browser_disableAppcache.js | took 4055ms
[task 2018-11-29T06:21:52.653Z] 06:21:52 INFO - checking window state
[task 2018-11-29T06:21:54.667Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7378
[task 2018-11-29T06:21:54.724Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7344
[task 2018-11-29T06:21:54.725Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7304
[task 2018-11-29T06:21:54.733Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7132
[task 2018-11-29T06:21:54.735Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7184
[task 2018-11-29T06:21:54.736Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7272
[task 2018-11-29T06:21:54.744Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7225
[task 2018-11-29T06:21:54.744Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7330
[task 2018-11-29T06:21:54.781Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7215
[task 2018-11-29T06:21:54.782Z] 06:21:54 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7109
[task 2018-11-29T06:21:55.127Z] 06:21:55 INFO - GECKO(7014) | Completed ShutdownLeaks collections in process 7014
[task 2018-11-29T06:21:55.128Z] 06:21:55 INFO - TEST-START | Shutdown
[task 2018-11-29T06:21:55.129Z] 06:21:55 INFO - Browser Chrome Test Summary
[task 2018-11-29T06:21:55.131Z] 06:21:55 INFO - Passed: 6
[task 2018-11-29T06:21:55.132Z] 06:21:55 INFO - Failed: 2
[task 2018-11-29T06:21:55.133Z] 06:21:55 INFO - Todo: 0
[task 2018-11-29T06:21:55.140Z] 06:21:55 INFO - Mode: e10s
[task 2018-11-29T06:21:55.144Z] 06:21:55 INFO - *** End BrowserChrome Test Results ***
[task 2018-11-29T06:21:55.627Z] 06:21:55 INFO - GECKO(7014) | 1543472515621 Marionette DEBUG Received observer notification xpcom-will-shutdown
[task 2018-11-29T06:21:55.627Z] 06:21:55 INFO - GECKO(7014) | 1543472515621 Marionette INFO Stopped listening on port 2828
[task 2018-11-29T06:21:55.627Z] 06:21:55 INFO - GECKO(7014) | 1543472515621 Marionette DEBUG Remote service is inactive
[task 2018-11-29T06:21:55.958Z] 06:21:55 INFO - TEST-INFO | Main app process: exit 0
|
|
1,511,303 | 13,788,986 |
When visiting sites with iframes containing bare-bone sites that have clearly no intention of using any kind of storage, cookies, service workers, appcache, or parent-frame interaction, we still do a storage access check for the iframe origin, resulting in a STATE_COOKIES_LOADED entry in the content blocking log. This is an example stack I recorded in lldb:
* frame #0: 0x0000000101a16df0 XUL`nsGlobalWindowOuter::NotifyContentBlockingState(this=0x000000011fd2e400, aState=32768, aChannel=0x000000011ffa0000, aBlocked=false, aURIHint=0x000000011ff78a00) at nsGlobalWindowOuter.cpp:5404 [opt]
frame #1: 0x00000001053bf2b2 XUL`mozilla::AntiTrackingCommon::NotifyBlockingDecision(aChannel=0x000000011ffa0000, aDecision=eAllow, aRejectedReason=0) at AntiTrackingCommon.cpp:1459 [opt]
frame #2: 0x00000001019a5ca5 XUL`nsContentUtils::StorageDisabledByAntiTracking(aWindow=<unavailable>, aChannel=0x000000011ffa0000, aPrincipal=<unavailable>, aURI=<unavailable>) at nsContentUtils.cpp:9120 [opt]
frame #3: 0x00000001019a4dd6 XUL`nsContentUtils::InternalStorageAllowedForPrincipal(aPrincipal=0x0000000146c89dc0, aWindow=<unavailable>, aURI=0x0000000000000000, aChannel=0x000000011ffa0000) at nsContentUtils.cpp:9214 [opt]
frame #4: 0x00000001019a5294 XUL`nsContentUtils::StorageAllowedForChannel(aChannel=0x000000011ffa0000) at nsContentUtils.cpp:8901 [opt]
frame #5: 0x0000000103bfa39d XUL`mozilla::dom::ServiceWorkerInterceptController::ShouldPrepareForIntercept(this=<unavailable>, aURI=0x000000011ff78a00, aChannel=0x000000011ffa0000, aShouldIntercept=0x00007ffeefbfb2bf) at ServiceWorkerInterceptController.cpp:41 [opt]
frame #6: 0x0000000100a1d986 XUL`mozilla::net::HttpBaseChannel::ShouldIntercept(this=<unavailable>, aURI=<unavailable>) at HttpBaseChannel.cpp:3291 [opt]
frame #7: 0x0000000100a3c34b XUL`mozilla::net::HttpChannelChild::ShouldInterceptURI(this=<unavailable>, aURI=<unavailable>, aShouldUpgrade=<unavailable>) at HttpChannelChild.cpp:4018 [opt]
frame #8: 0x0000000100a3b2d5 XUL`mozilla::net::HttpChannelChild::AsyncOpen(this=0x000000011ffa0000, listener=0x000000011ffa0000, aContext=<unavailable>) at HttpChannelChild.cpp:2677 [opt]
frame #9: 0x0000000100a3ae35 XUL`mozilla::net::HttpChannelChild::AsyncOpen2(this=0x000000011ffa0000, aListener=<unavailable>) at HttpChannelChild.cpp:2732 [opt]
I'm quite sure there are more of these instances, maybe AppCache or Quota Manager.
The problem with that is of course that developers will not understand the criteria by which we flag an iframe as having cookies and users might perceive a higher "threat" from third party cookies than was actually present.
We should try to reduce the number of these false positives as much as possible and document the remaining edge cases.
|
|
1,511,358 | 13,789,364 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3625.0 Safari/537.36
Steps to reproduce:
1. Place an element on the lower corner of a container element.
2. Give a z-index of -1
Actual results:
Resizing container element doesn't work
Example:
https://codepen.io/esanzgar/pen/vQvWpg
Other browsers work as expected.
Expected results:
The container element should be resizable.
Also notice that the color is not correct. It shouldn't be blue but purple-ish
|
|
1,511,375 | 13,789,495 |
User Agent: Mozilla/5.0 (Android 7.1.1; Tablet; rv:65.0) Gecko/65.0 Firefox/65.0
Steps to reproduce:
1. Open https://github.com/mozilla-mobile
2. Request desktop site
3. Zoom in
4. Scroll down
Actual results:
The scroll bar appears higher than it is supposed to.
Expected results:
The scroll bar should be closer to the edge.
|
|
1,511,477 | 13,790,234 |
Created attachment 9029043
trackers.txt
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:63.0) Gecko/20100101 Firefox/63.0
Steps to reproduce:
Visit huffingtonpost.com
Actual results:
only 18 trackers blocked
Expected results:
WV reported 40+ trackers blocked. Something closer to that number
|
|
1,511,751 | 13,793,150 |
[Tracking Requested - why for this release]:
E.g. here on Bugzilla, the main view tells me "Tracking Cookies Blocked", and when I open the subview I only see "From This Site: https://bugzilla.mozilla.org", which I assume wasn't actually blocked. Or was it? If it wasn't, what other cookies were blocked as "Tracking Cookies Blocked" suggests? I really can't tell from the UI.
|
|
1,511,878 | 13,795,495 |
Created attachment 9029330
graytext_2cases.html
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.14; rv:63.0) Gecko/20100101 Firefox/63.0
Steps to reproduce:
Open the attached file graytext_2cases.html. The expanded select list has the first item disabled, but it is still black, not gray.
Actual results:
The first option, which is disabled, is black, not gray in the expanded select list
Expected results:
The disabled item should have been gray in the expanded list.
Chrome has the same problem, but Safari does not. It only appears to impact OS X, and maybe only 10.14. Windows browsers are not impacted.
|
|
1,512,046 | 13,796,659 |
**Steps to reproduce**
1. Open the console
2. Evaluate the following:
```
x = new Error("ooops")
x.stack = "a@http://exampl.com:1:1\n" +
"grouped@http://react.js:1:1\n" +
"grouped2@http://react.js:1:1";
console.error(x)
```
3. You should see a callstack with a React group, click on it
**Expected results**
The group is expanded
**Actual results**
The group is expanded, but so is the message (the arrow icon on the left is rotated, and the console.error callstack is displayed under the react callstack)
---
This is because the group click propagates, which shouldn't.
|
|
1,512,047 | 13,796,663 |
Steps to Reproduce:
1. Open Nightly
2. Go to about:preferences#search
3. Click at "Find more search engines"
4. Select search engine and click at "Add to Firefox" button.
Expected Result:
The add-on is successfully installed and added to the list of available engines and is properly set as a default search engine.
Actual Result:
No response.
|
|
1,512,166 | 13,797,536 |
[Affected versions]
Firefox 65.0a1 (2018-12-04)
[Affected platforms]
Ubuntu 16.04, Windows 7/10, Mac OS 10.13
[Steps to reproduce]
1. Open Firefox with a new profile
2. Go to dailymotion.com
3. Set Content Blocking to Strict in about:preferences#privacy
4. Check the Control Panel section: Cookies and Trackers are blocked
5. Turn off blocking for this site
[Expected result]
The Cookies and Trackers should be displayed with a different label since tracking protection is disabled for that site. "Detected" would be a suggestion.
[Actual result]
Cookies and Trackers are still labeled as "blocked".
[Note]
If the lists for both of them are reached, the listed cookies and tracker are not blocked even if the label from the control center suggested it.
|
|
1,512,180 | 13,797,645 |
Beginning with either this morning's (20181205102000) or last evening's updates, opening Bookmark folders using Open All in Tabs no longer works on Linux x86_64.
I couldn't find where this was already filed. I'll try to find the regressor using mozregression.
|
|
1,512,187 | 13,797,682 |
Created attachment 9029618
Screen Shot 2018-11-30 at 3.00.44 PM.png
I've attached a screenshot of the new proposed spotlight visuals.
the dotted outline is photon blue-60.
the background is photon blue-40 at 1% alpha.
it fades after 2 seconds with a 500ms animation.
|
|
1,512,311 | 13,798,346 |
Clicking the "What's new" link in Nightly Updates opens a new blank tab. The address bar is empty but it seems to be about:blank. This seems to have started with bug 1503681 changing the behavior of target=_blank.
Regression range:
https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=b526d511169a3ba8ea647d6de9183a62df0da1c8&tochange=e6a4330eb15f1536f505d6794c4efac3ca9fcf7e
|
|
1,512,625 | 13,800,399 |
[Affected versions]
Firefox 65.0a1
[Affected platforms]
Ubuntu 16.04, Windows 7/10, Mac OS 10.13
[Steps to reproduce]
1. Open the Firefox browser and Set the following preferences in about:config
* Make sure dom.storage_access.enabled is TRUE (default in Nightly)
* Make sure urlclassifier.trackingAnnotationTable.testEntries is set to "storage-access.englehardt-tracker.com,imminent-archeology.glitch.me".
* Make sure network.cookie.cookieBehavior is 4 (default in Nightly)
* Make sure dom.storage_access.auto_grants is FALSE
2. Reach https://storage-access.englehardt-tracker.com/index.html and Click anywhere on the page.
3. Reach https://imminent-archeology.glitch.me/ and Click anywhere on the page.
4. Reach https://senglehardt.com/test/storage_access/ and click the "call document.requestStorageAccess".
5. Click the "Allow on Any site" button from the Permissions Prompt.
6. Reach https://ehsanakhgari.org/test/storage_access/ and click the "Set and Get cookie via document.cookie"
[Expected result]
Since its the same Tracker and it has been allowed on any site, the user should be able to Set and get Cookies on both https://ehsanakhgari.org/test/storage_access/ , https://senglehardt.com/test/storage_access/
[Actual result]
The same tracker and cookie are not allowed on other websites after the user selects the "Allow on any site" from the Permissions Doorhanger panel.
|
|
1,512,645 | 13,800,498 |
Created attachment 9029973
Screen Shot 2018-12-07 at 09.56.08.png
|
|
1,512,665 | 13,800,565 |
Created attachment 9029981
DevTOOls_zoom_Changes.gif
[Affected versions]:
- Win10x64, Ubuntu 16.04 x64, macOS10.12.6
[Affected platforms]:
- Firefox: 65.0a1 (2018-12-07)
[Steps to reproduce]:
1. Launch Firefox;
2. Access https://www.w3schools.com/tags/tryit.asp?filename=tryhtml_iframe
3. Right click on any element;
4. Click on the Inspect Element option;
5. In the inspector add any property (ex: add to element {color: red;});
6. Click on the Changes tab;
7. Click on the page so that the inspector is no longer tagged as active;
8. Zoom in/out a couple of times;
[Expected result]:
- the tracked changes in Changes tab are still displayed;
[Actual result]:
- the tab (apparently) resets to the default state;
[Regression range]:
- bug 1506170 appears to be at fault
[Additional notes]:
- attached recording with the issue;
- Preference: devtools.inspector.changes.enabled - true | if the Changes tab not displayed in the inspector;
- managed to reproduce in other pages with iframes in them, but on a locally saved file the issue did not manifest.
|
|
1,512,731 | 13,800,968 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:63.0) Gecko/20100101 Firefox/63.0
Steps to reproduce:
1. Install some Noto fonts for scripts that otherwise cannot render.
2. Install the latest Windows update (KB4469342).
3. Open firefox and check for the Wikipedia page, relevant to the block of characters you just installed the font for.
(I am not sure if the order of steps 1 and 2 can be switched) (The update is important, since it was not happening before the update)
Actual results:
The font fails to load and all characters display as the fallback apperance.
Expected results:
The font should have loaded and rendered as intended.
|
|
1,512,857 | 13,801,884 |
See the spec: https://design.firefox.com/photon/components/buttons.html#default-3
To the 8px padding 10px are added from https://searchfox.org/mozilla-central/source/toolkit/themes/shared/in-content/common.inc.css#249
This leads to invisible font size label.
PS: Is it okay that the menulist popups are center aligned? This looks a bit weird. Worth for a new bug?
|
|
1,512,867 | 13,801,987 |
https://searchfox.org/mozilla-central/rev/9abd7f512c9b29071776e7ba3f991d5b5fd59ff6/toolkit/components/places/SyncedBookmarksMirror.jsm#3128 is subtly wrong. For remote merge states where the item doesn't exist locally, we should always return true, even if the remote node indicates the item doesn't need to be merged. In theory, we could return true and always take the remote state, but that'll trigger a lot of unneeded writes for keywords and tags if they didn't actually change.
For symmetry, we should also do the same for local items: if we're taking the local state, and the item doesn't exist remotely, always upload it.
Both catch the case where the mirror's `needsMerge` flags are out of sync with Places; for example, the mirror says an item doesn't need to be merged, but isn't known locally, or Places has an item with `syncStatus = NORMAL` that doesn't exist at all in the mirror.
|
|
1,512,883 | 13,802,086 |
Created attachment 9030166
clip-path crash.html
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:64.0) Gecko/20100101 Firefox/64.0
Steps to reproduce:
I took an element equipped with a clip-path with the value of `circle(<any value>)` and a hover effect which makes the circle value bigger or smaller. Then I went crazy and put a transition with a big negative value on its easing (something like `cubic-bezier(0,0,1,-20)`, how big it must be depends on the circle's size).
Actual results:
Triggering the hover effect, the circle() value transitioned to a negative value, an unexpected behavior, making the tab crash.
See the attachment for a demonstration
Expected results:
Gecko should not have tried to re-render the clip-path with a negative value, not making the tab crash.
I am pretty certain that this bug can also be applied to other clip-path values (other than circle) and maybe other CSS properties as well
|
|
1,512,958 | 13,803,587 |
Created attachment 9030212
testAudio.zip
User Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:65.0) Gecko/20100101 Firefox/65.0
Steps to reproduce:
- Take a look at the attached sample and run it, in Nightly 65.
- Click to play the sound.
- When you switch between tabs, the audio should pause or resume, as per the visibilitychange listener: when you go back to the tab after 2 or 3 seconds, the sound resumes as expected. However, if you stay outside from the tab just some seconds more (let's say 10 seconds), the sound will resume when you'll be back to the tab but will be stopped quite immediately after that.
Actual results:
The sound has stopped playing, instead of resuming.
Expected results:
The sound should have resumed. This issue seems to only apply to Audio with WebAudio/createMediaElementSource (there is another test case in the html file, which works fine for "basic" Audio).
|
|
1,513,007 | 13,804,098 | null |
|
1,513,035 | 13,804,302 |
The styles that were changed to update the button spec also affected the menu inputs.
STR:
- go to about:preferences
- under "fonts and colors" click to open the menu
AR:
- each menu item is centered
ER:
- The menu items are left-aligned in LTR and right-aligned in RTL
I believe this is due to this rule https://searchfox.org/mozilla-central/source/toolkit/themes/shared/in-content/common.inc.css#177
|
|
1,513,076 | 13,804,616 |
In the section for Do Not Track in about:preferences#privacy, it says:
Send websites a “Do Not Track” signal that you don’t want to be tracked.
* Always
* Only when Nightly is set to block Detected Trackers
The second option no longer makes sense, because we don't use the terminology "Detected Trackers" anymore. We need to change this string.
Betsy has proposed:
Only when Nightly is set to block known trackers
|
|
1,513,135 | 13,804,953 |
Created attachment 9030367
video of repro (sorry it's swf)
User Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:63.0) Gecko/20100101 Firefox/63.0
Steps to reproduce:
See attached video. User indicates that they want to download a file, and my application sends that file over HTTP.
Actual results:
Firefox's download UI indicates that no data has arrived and a 0 byte ".part" file is created in the download folder. A Wireshark trace (attached) shows that all of the data has arrived, and it looks ok to me. The problem occurs about half the time with Firefox 63, and it seems to require the file to be above a certain size (a few MB). It seems like a recent regression, I wasn't able to reproduce in FF 60. Chrome downloads fine.
Expected results:
File should be saved to disk.
|
|
1,520,791 | 13,868,032 |
Since the patch on bug 1506608 landed the value of `docShellIsActive` is no longer `true` for the initial tab.
Steps:
1) Start Firefox
2) Open the Scratch pad
3) Switch to Browser environment
4) Run: `alert(window.gBrowser.selectedTab.linkedBrowser.docShellIsActive)`
Expected result:
The flag should report that the docshell is active
Current result:
The flag tells that the docshell is not active
Once a new tab gets opened the value is correctly initialized.
I don't know which implications that generally has and how important it is to get fixed. As such I will mark it as tracking firefox 65.
|
|
1,520,914 | 13,869,321 |
Created attachment 9037368
Screen Shot 2019-01-17 at 11.09.39 AM.png
`about:newtab#devtools-ds` goes blank after turning ON the toggle "enabled" switch.
This happens only for the very first time on the system (Mac/Windows) , and the next time when toggled ON/OFF, page is not blank. (cached)
*Steps to Reproduce:*
1. Set browser.newtabpage.activity-stream.asrouter.devtoolsEnabled to true in about:config
2. Visit about:newtab#devtools-ds. Ensure you see devtools
3. Try turning ON the new tab experiment off/on via the "enabled" checkbox
*Actual Result:*
`about:newtab#devtools-ds` goes blank
*Expected Result:*
Blank page not expected.
cc: @Kate Hudson :k88hudson
|
|
1,520,949 | 13,869,714 |
+++ This bug was initially created as a clone of Bug #1510964 +++
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:63.0) Gecko/20100101 Firefox/63.0
Characters that are not in the a-z range aren't imported properly from Chrome (on Windows), probably due to a character encoding problem somewhere along the migration process.
See attachment 9032392 and bug 1510964 comment 8.
|
|
1,521,214 | 13,871,935 |
Created attachment 9037702
update-ptr.diff
I found that the [cursor of `BufferReader`][cursor] may be set to a wrong position(maybe out of bound) if `BufferReader::Seek()` is called just after `BufferReader::Read()`.
#### Example
Assume a buffer has `10` bytes from address `0` to `10`, and we read `4` bytes (by `ReadU32()`) every time.
1. At the first, the cursor to read the buffer is set to `0`, and the remaining length is `10`.
2. After 2 rounds, the cursor will point to `8`.
3. At the 3 round, the call `ReadU32()` will fail since there are only `2` bytes left and the remaining length will be [set to `0`][remaining-zero].
- The `Offset()` now will be `10` since it's [calculated by *length* - *remaining*][offset], where *length* and *remaining* are `10` and `0` respectively.
- However, the cursor isn't updated! It's still `8`.
4. When `Seek(0)` is called, the new cursor will be set to `-2`
- the new cursor is [calculated by `new cursor = current cursor - current offset + new offset`][seek-cursor], where the *current cursor* is `8`, *current offset* is `10`, and the *new offset* is `0`.
5. If `Read()` is called again, we will read something from `-2`
[cursor]: https://searchfox.org/mozilla-central/rev/dac799c9f4e9f5f05c1071cba94f2522aa31f7eb/dom/media/BufferReader.h#293
[read]: https://searchfox.org/mozilla-central/rev/dac799c9f4e9f5f05c1071cba94f2522aa31f7eb/dom/media/BufferReader.h#149
[seek]: https://searchfox.org/mozilla-central/rev/dac799c9f4e9f5f05c1071cba94f2522aa31f7eb/dom/media/BufferReader.h#224
[remaining-zero]: https://searchfox.org/mozilla-central/rev/dac799c9f4e9f5f05c1071cba94f2522aa31f7eb/dom/media/BufferReader.h#151
[offset]: https://searchfox.org/mozilla-central/rev/dac799c9f4e9f5f05c1071cba94f2522aa31f7eb/dom/media/BufferReader.h#47
[seek-cursor]: https://searchfox.org/mozilla-central/rev/dac799c9f4e9f5f05c1071cba94f2522aa31f7eb/dom/media/BufferReader.h#231
|
|
1,521,233 | 13,872,048 |
+++ This bug was initially created as a clone of Bug #1521231 +++
See bug 1520522 comment 3
|
|
1,521,304 | 13,872,471 |
1. https://www.infowars.com/breaking-news/
Linux x86_64 Nightly. wait a few seconds.
Linux x86_64 Beta. wait a minute or two.
May require the live show to be playing in the page. It just stopped being reproducible when the "Watch Live now" started throwing HLS.js error networkError. Appears to be using https://videojs.com/. I wasn't able to reproduce with any examples at videojs.com. I'm going to file any way in the event the error is temporary and we will be able to reproduce later.
Opt doesn't appear to crash.
2. Assertion failure: IsAcquired() && mOwningThread == PR_GetCurrentThread(), at xpcom/threads/BlockingResourceBase.cpp:369
#01: mozilla::DataChannelConnection::ReceiveCallback(socket*, void*, unsigned long, sctp_rcvinfo, int) (netwerk/sctp/datachannel/DataChannel.cpp:2317)
#02: mozilla::receive_cb(socket*, sctp_sockstore, void*, unsigned long, sctp_rcvinfo, int, void*) (netwerk/sctp/datachannel/DataChannel.cpp:241)
#03: sctp_invoke_recv_callback (netwerk/sctp/src/netinet/sctputil.c:4859)
#04: sctp_add_to_readq (netwerk/sctp/src/netinet/sctputil.c:4968)
#05: sctp_notify_send_failed (netwerk/sctp/src/netinet/sctputil.c:3150)
#06: sctp_ulp_notify (:?)
#07: sctp_release_pr_sctp_chunk (netwerk/sctp/src/netinet/sctputil.c:5119)
#08: sctp_t3rxt_timer (:?)
#09: sctp_timeout_handler (netwerk/sctp/src/netinet/sctputil.c:1763)
#10: sctp_handle_tick (netwerk/sctp/src/netinet/sctp_callout.c:167)
#11: start_thread (pthread_create.c:?)
#12: __GI___clone (:?)
#13: ??? (???:???)
jesup: I made this security senstive because of bug 1493347. Could you take a look and see if this is really security sensitive?
Previously this url had thrown:
Assertion failure: false (MOZ_ASSERT_UNREACHABLE: Should always be able to find the appropriate sample description! Malformed mp4?), at /builds/worker/workspace/build/src/dom/media/mp4/Index.cpp:260
Assertion failure: false (MOZ_ASSERT_UNREACHABLE: Should always be able to find the appropriate sample description! Malformed mp4?), at z:/build/build/src/dom/media/mp4/Index.cpp:260
#01: mozilla::MP4TrackDemuxer::GetNextSample() [dom/media/mp4/MP4Demuxer.cpp:419]
#02: mozilla::MP4TrackDemuxer::GetSamples(int) [dom/media/mp4/MP4Demuxer.cpp:480]
#03: mozilla::TrackBuffersManager::DoDemuxVideo() [dom/media/mediasource/TrackBuffersManager.cpp:1396]
#04: mozilla::TrackBuffersManager::CodedFrameProcessing() [dom/media/mediasource/TrackBuffersManager.cpp:1366]
#05: mozilla::TrackBuffersManager::SegmentParserLoop() [dom/media/mediasource/TrackBuffersManager.cpp:809]
#06: nsresult mozilla::detail::RunnableMethodImpl<mozilla::TrackBuffersManager *,void (mozilla::TrackBuffersManager::*)(),1,mozilla::RunnableKind::Standard>::Run() [xpcom/threads/nsThreadUtils.h:1161]
#07: mozilla::TaskQueue::Runner::Run() [xpcom/threads/TaskQueue.cpp:200]
#08: nsThreadPool::Run() [xpcom/threads/nsThreadPool.cpp:243]
#09: nsThread::ProcessNextEvent(bool,bool *) [xpcom/threads/nsThread.cpp:1154]
#10: NS_ProcessNextEvent(nsIThread *,bool) [xpcom/threads/nsThreadUtils.cpp:468]
#11: mozilla::ipc::MessagePumpForNonMainThreads::Run(base::MessagePump::Delegate *) [ipc/glue/MessagePump.cpp:304]
#12: MessageLoop::RunHandler() [ipc/chromium/src/base/message_loop.cc:309]
#13: MessageLoop::Run() [ipc/chromium/src/base/message_loop.cc:291]
#14: nsThread::ThreadFunc(void *) [xpcom/threads/nsThread.cpp:452]
#15: PR_NativeRunThread [nsprpub/pr/src/threads/combined/pruthr.c:406]
#16: static unsigned int pr_root(void *) [nsprpub/pr/src/md/windows/w95thred.c:138]
#17: ucrtbase.dll + 0x1c4ce
#18: KERNEL32.DLL + 0x13dc4
#19: static void patched_BaseThreadInitThunk(int, void *, void *) [mozglue/build/WindowsDllBlocklist.cpp:724]
#20: ntdll.dll + 0x73691
On Windows and Linux.
|
|
1,521,446 | 13,874,636 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:66.0) Gecko/20100101 Firefox/66.0
Steps to reproduce:
* Switch ON the option: Options > General > Browsing > Search for text when you start typing
* Open Firefox Home page (by opening a new tab, or going to about:home)
* Click in the "Search the web" box
* Start typing
Actual results:
The text appears in the search panel at the bottom (search in page)
Expected results:
Text should appear in the address bar, to allow searching the web.
|
|
1,521,484 | 13,874,950 |
Calling 'get next: 10 / 20 / 50 pushes' loads them but doesn't add "fromchange <revision>" to 'Active Filters' bar anymore.
Steps to reproduce:
1. Load a tree, e.g. https://treeherder.mozilla.org/#/jobs?repo=autoland&resultStatus=testfailed%2Cbusted%2Cexception%2Cusercancel%2Cretry&group_state=expanded
2. Go to the bottom and let it load 10 more pushes.
Actual result:
10 more pushes loaded but bar "Active Filters:" has no rule "fromchange". URL contains fromchange.
3. Reload the page.
Actual result:
"Active Filters:" contains "fromchange".
|
|
1,521,499 | 13,875,110 |
Created attachment 9037944
unknown event.png
**[Affected versions]:**
- Firefox Nightly 66.0a1, Build ID 20190120213632
**[Affected Platforms]:**
- All Windows
- All Mac
- All Linux
**[Steps to reproduce]:**
1. Start the browser and open a new tab.
2. Open the Browser Console (Ctrl/Cmd+Shift+J) and write something in the in-content search bar.
3. Observe the Browser Console.
**[Expected results]:**
- No errors are displayed.
**[Actual results]:**
- "Unknown event: ["activity_stream", "event", "SEARCH_HANDOFF"]" error is displayed.
**[Notes]:**
- Attached a screenshot of the issue.
|
|
1,521,577 | 13,875,694 |
During 1-1 in appear.in, the far end reported that audio was behind about 1.5 secs comparing to video. I run mozregression with Dan as the remote party and the response was:
daaecb62f373 Andreas Pehrson — Bug 1509548 - Make MediaStreamGraph pull data per track instead of per stream. r=padenot
311cee86bc66 Andreas Pehrson — Bug 1509548 - Remove early PullNewData return. r=padenot
dfde7b2c53a6 Andreas Pehrson — Bug 1509548 - Clean up what appears to be wip-leftover gunk in DecodedStream. r=jya
8a064f0dbb2f Andreas Pehrson — Bug 1509548 - Rename track-end-time methods in StreamTracks. r=padenot
1d1b47e23537 Andreas Pehrson — Bug 1509548 - Remove the concept of a known tracks time from MediaStreamGraph. r=padenot
In order to reproduce it you need to enter in an appear.in room and observe the time lag of audio compering to video (built in devices). It's 100% reproducible on my Linux box. I haven't observed it on OSX.
Feel free to ping me if you need a remote party to reproduce it or even further debugging.
|
|
1,521,643 | 13,876,087 |
*Tested on :*
*FF Nightly version :* 66.0a1 (2019-01-20)
*OS :* Mac and Windows 10 Pro
Depends on the bug https://bugzilla.mozilla.org/show_bug.cgi?id=1514018
*Steps to Reproduce:*
1. Go to Config and Set `browser.newtabpage.activity-stream.discoverystream.config` to `{"enabled":true,"layout_endpoint":"https://getpocket.com/v3/newtab/layout?version=1&consumer_key=40249-e88c401e1b1f2242d9e441c4&layout_variant=dev-test-1"}`
2. Open new tab
3. Observe Hover interactions
*Bug :*
* Hover works as expected only when placed on the item title.
* Hover anywhere within the item section. Hover on the `excerpt` or on `image` or on the `publisher/sponsor'. Wiredup link is not highlighted.
*Expected Result:*
Hover anywhere within the item section. Hover on the `image title` `excerpt` or on `image` or on the `publisher/sponsor' should Highlight the image title
|
|
1,521,677 | 13,876,338 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.14; rv:66.0) Gecko/20100101 Firefox/66.0
Steps to reproduce:
Login to Basecamp3 dashboard
Actual results:
- Many JS features not working.
- Like (cmd + j ) for jump menu not working
- Support window (click question mark at bottom right) not working
Console errors:
TypeError: can't define non-configurable property on WindowProxy
libraries.js:17
<anonymous> libraries.js:17 n3Webpack
ReferenceError: CalendarDate is not defined[Learn More]
desktop-3aba787f2e87930c151dfd33da306bac9b8cda3961016c34001d75fda612b889.js:2:25033
<anonymous> https://3.basecamp-static.com/assets/desktop-3aba787f2e87930c151dfd33da306bac9b8cda3961016c34001d75fda612b889.js:2 <anonymous> https://3.basecamp-static.com/assets/desktop-3aba787f2e87930c151dfd33da306bac9b8cda3961016c34001d75fda612b889.js:2
TypeError: BC.current is undefined[Learn More]
Expected results:
Should work dashboard without any issues.
|
|
1,522,092 | 13,879,554 |
[Affected versions]:
Nightly 66.0a1 - Build ID: 20190123070847
[Affected platforms]:
Windows 7/10 x64
Mac OS 10.14
Ubuntu 18.04
[Steps to reproduce]:
1. Go to cnn.com/videos
2. Open the Site Information panel
3. Select the Allow/Block dropdown using keyboard only navigation
[Expected]:
The dropdown should be selectable with keyboard only navigation
[Actual Results]:
The dropdown box can't be selected, thus the allow/block state can't be changed using keyboard only navigation.
|
|
1,522,381 | 13,881,542 |
I'm seeing an issue that when you are logged in to sync on nightly on windows, the sync panel in preferences doesn't load. The error trace is something like:
```
TypeError: date is undefined[Learn More] TelemetryUtils.jsm:121:5
truncateToHours resource://gre/modules/TelemetryUtils.jsm:121
SyncTelemetryImpl resource://services-sync/telemetry.js:482
<anonymous> resource://services-sync/telemetry.js:799
<anonymous> resource://services-sync/service.js:31
getter resource://services-sync/main.js:21
ensureLoaded jar:file:///C:/Program Files/Nightly/omni.ja!/components/Weave.js:75
init chrome://browser/content/preferences/in-content/sync.js:85
init chrome://browser/content/preferences/in-content/preferences.js:43
init_category_if_required chrome://browser/content/preferences/in-content/preferences.js:36
gotoPref chrome://browser/content/preferences/in-content/preferences.js:166
init_all chrome://browser/content/preferences/in-content/preferences.js:70
_fireOnSelect chrome://global/content/elements/richlistbox.js:305
selectItem chrome://global/content/elements/richlistbox.js:431
onxblmousedown chrome://global/content/bindings/richlistbox.xml:125
```
From debugging this, it looks like the offending line is[telemetry.js#480](https://searchfox.org/mozilla-central/source/services/sync/modules/telemetry.js#480)
```javascript
let sessionStartDate = Services.startup.getStartupInfo().main;
```
For me, let sessionStartDate = Services.startup.getStartupInfo() returns an object that does NOT have a main property:
```javascript
{process:"2107-04-18T20:00:31.345Z",start:"2019-01-24T04:32:27.973Z"}
```
This then gets passed to [TelemetryUtils.jsm#120](https://searchfox.org/mozilla-central/source/toolkit/components/telemetry/app/TelemetryUtils.jsm#120)
So either that changed, or my system is borked. But Ive confirmed this on two different windows machines with different profiles and sync accounts. (havent checked extensions yet).
|
|
1,522,773 | 13,883,953 |
Filed by: ncsoregi [at] mozilla.com
https://treeherder.mozilla.org/logviewer.html#?job_id=223972870&repo=mozilla-central
https://queue.taskcluster.net/v1/task/H4vQcnknRjeIX5EBIb5nPQ/runs/0/artifacts/public/logs/live_backing.log
Created based on https://bugzilla.mozilla.org/show_bug.cgi?id=1346506#c9
[task 2019-01-25T07:48:53.624Z] 07:48:53 INFO - TEST-START | /mediacapture-streams/MediaStreamTrack-MediaElement-disabled-audio-is-silence.https.html
[task 2019-01-25T07:48:53.645Z] 07:48:53 INFO - PID 5440 | ###!!! [Parent][RunMessage] Error: Channel closing: too late to send/recv, messages will be lost
[task 2019-01-25T07:48:54.631Z] 07:48:54 INFO - PID 5440 | -----------------------------------------------------
[task 2019-01-25T07:48:54.632Z] 07:48:54 INFO - PID 5440 | Suppressions used:
[task 2019-01-25T07:48:54.632Z] 07:48:54 INFO - PID 5440 | count bytes template
[task 2019-01-25T07:48:54.634Z] 07:48:54 INFO - PID 5440 | 500 15968 nsComponentManagerImpl
[task 2019-01-25T07:48:54.634Z] 07:48:54 INFO - PID 5440 | 2 416 mozJSComponentLoader::LoadModule
[task 2019-01-25T07:48:54.634Z] 07:48:54 INFO - PID 5440 | 611 17713 libfontconfig.so
[task 2019-01-25T07:48:54.634Z] 07:48:54 INFO - PID 5440 | 1 29 libglib-2.0.so
[task 2019-01-25T07:48:54.634Z] 07:48:54 INFO - PID 5440 | -----------------------------------------------------
[task 2019-01-25T07:48:54.742Z] 07:48:54 INFO - PID 5440 | ###!!! [Child][MessageChannel] Error: (msgtype=0x350100,name=PContent::Msg_AccumulateChildHistograms) Closed channel: cannot send/recv
[task 2019-01-25T07:48:54.743Z] 07:48:54 INFO - PID 5440 | ###!!! [Child][MessageChannel] Error: (msgtype=0x350101,name=PContent::Msg_AccumulateChildKeyedHistograms) Closed channel: cannot send/recv
[task 2019-01-25T07:48:54.743Z] 07:48:54 INFO - PID 5440 | ###!!! [Child][MessageChannel] Error: (msgtype=0x350105,name=PContent::Msg_RecordDiscardedData) Closed channel: cannot send/recv
[task 2019-01-25T07:48:54.880Z] 07:48:54 INFO - PID 5440 | Assertion failure: Request::mDisconnected, at /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/MozPromise.h:429
[task 2019-01-25T07:48:54.881Z] 07:48:54 INFO - PID 5440 | AddressSanitizer:DEADLYSIGNAL
[task 2019-01-25T07:48:54.881Z] 07:48:54 INFO - PID 5440 | =================================================================
[task 2019-01-25T07:48:54.881Z] 07:48:54 ERROR - PID 5440 | ==6464==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000001 (pc 0x7fc386da11e8 bp 0x7ffc436ef450 sp 0x7ffc436ef440 T0)
[task 2019-01-25T07:48:54.882Z] 07:48:54 INFO - PID 5440 | ==6464==The signal is caused by a WRITE memory access.
[task 2019-01-25T07:48:54.882Z] 07:48:54 INFO - PID 5440 | ==6464==Hint: address points to the zero page.
[task 2019-01-25T07:48:55.033Z] 07:48:55 INFO - Closing window 55834574849
[task 2019-01-25T07:48:55.098Z] 07:48:55 INFO - PID 5440 | JavaScript error: , line 0: InvalidStateError: An attempt was made to use an object that is not, or is no longer, usable
[task 2019-01-25T07:48:55.129Z] 07:48:55 INFO - .
[task 2019-01-25T07:48:55.130Z] 07:48:55 INFO - TEST-OK | /mediacapture-streams/MediaStreamTrack-MediaElement-disabled-audio-is-silence.https.html | took 1505ms
[task 2019-01-25T07:48:55.131Z] 07:48:55 INFO - TEST-START | /mediacapture-streams/MediaStreamTrack-MediaElement-disabled-video-is-black.https.html
[task 2019-01-25T07:48:55.153Z] 07:48:55 INFO - PID 5440 | SharedThreadPool in xpcom-shutdown-threads. Waiting for pools "CubebOperation"
[task 2019-01-25T07:48:55.706Z] 07:48:55 INFO - PID 5440 | #0 0x7fc386da11e7 in mozilla::MozPromise<bool, nsresult, true>::ThenValueBase::AssertIsDead() /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/MozPromise.h:429:9
[task 2019-01-25T07:48:55.708Z] 07:48:55 INFO - PID 5440 | #1 0x7fc386da1870 in mozilla::MozPromise<bool, nsresult, true>::AssertIsDead() /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/MozPromise.h:945:13
[task 2019-01-25T07:48:55.710Z] 07:48:55 INFO - PID 5440 | #2 0x7fc386da1b88 in mozilla::MozPromise<bool, nsresult, true>::~MozPromise() /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/MozPromise.h:986:5
[task 2019-01-25T07:48:55.711Z] 07:48:55 INFO - PID 5440 | #3 0x7fc386da16fd in mozilla::MozPromise<bool, nsresult, true>::Private::~Private() /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/MozPromise.h:242:9
[task 2019-01-25T07:48:55.733Z] 07:48:55 INFO - PID 5440 | #4 0x7fc38bc9fcc0 in Release /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/MozPromise.h:137:3
[task 2019-01-25T07:48:55.733Z] 07:48:55 INFO - PID 5440 | #5 0x7fc38bc9fcc0 in Release /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/RefPtr.h:45
[task 2019-01-25T07:48:55.733Z] 07:48:55 INFO - PID 5440 | #6 0x7fc38bc9fcc0 in Release /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/RefPtr.h:362
[task 2019-01-25T07:48:55.733Z] 07:48:55 INFO - PID 5440 | #7 0x7fc38bc9fcc0 in ~RefPtr /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/RefPtr.h:76
[task 2019-01-25T07:48:55.733Z] 07:48:55 INFO - PID 5440 | #8 0x7fc38bc9fcc0 in ~MozPromiseHolder /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/MozPromise.h:1116
[task 2019-01-25T07:48:55.734Z] 07:48:55 INFO - PID 5440 | #9 0x7fc38bc9fcc0 in ~ /builds/worker/workspace/build/src/dom/media/MediaStreamGraph.cpp:3042
[task 2019-01-25T07:48:55.734Z] 07:48:55 INFO - PID 5440 | #10 0x7fc38bc9fcc0 in ~LambdaRunnable /builds/worker/workspace/build/src/dom/media/systemservices/MediaUtils.h:66
[task 2019-01-25T07:48:55.734Z] 07:48:55 INFO - PID 5440 | #11 0x7fc38bc9fcc0 in mozilla::media::LambdaRunnable<mozilla::MediaInputPort::BlockSourceTrackId(int, mozilla::BlockingMode)::$_0>::~LambdaRunnable() /builds/worker/workspace/build/src/dom/media/systemservices/MediaUtils.h:66
[task 2019-01-25T07:48:55.749Z] 07:48:55 INFO - PID 5440 | #12 0x7fc3851b378c in mozilla::Runnable::Release() /builds/worker/workspace/build/src/xpcom/threads/nsThreadUtils.cpp:49:1
[task 2019-01-25T07:48:55.751Z] 07:48:55 INFO - PID 5440 | #13 0x7fc38518163d in Release /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/RefPtr.h:45:40
[task 2019-01-25T07:48:55.752Z] 07:48:55 INFO - PID 5440 | #14 0x7fc38518163d in Release /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/RefPtr.h:362
[task 2019-01-25T07:48:55.754Z] 07:48:55 INFO - PID 5440 | #15 0x7fc38518163d in ~RefPtr /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/RefPtr.h:76
[task 2019-01-25T07:48:55.755Z] 07:48:55 INFO - PID 5440 | #16 0x7fc38518163d in ~Runner /builds/worker/workspace/build/src/xpcom/threads/AbstractThread.cpp:109
[task 2019-01-25T07:48:55.755Z] 07:48:55 INFO - PID 5440 | #17 0x7fc38518163d in mozilla::EventTargetWrapper::Runner::~Runner() /builds/worker/workspace/build/src/xpcom/threads/AbstractThread.cpp:109
[task 2019-01-25T07:48:55.757Z] 07:48:55 INFO - PID 5440 | #18 0x7fc3851b3b4c in Release /builds/worker/workspace/build/src/xpcom/threads/nsThreadUtils.cpp:49:1
[task 2019-01-25T07:48:55.758Z] 07:48:55 INFO - PID 5440 | #19 0x7fc3851b3b4c in mozilla::CancelableRunnable::Release() /builds/worker/workspace/build/src/xpcom/threads/nsThreadUtils.cpp:71
[task 2019-01-25T07:48:55.760Z] 07:48:55 INFO - PID 5440 | #20 0x7fc38bc9598c in ~nsCOMPtr_base /builds/worker/workspace/build/src/obj-firefox/dist/include/nsCOMPtr.h:331:7
[task 2019-01-25T07:48:55.761Z] 07:48:55 INFO - PID 5440 | #21 0x7fc38bc9598c in Destruct /builds/worker/workspace/build/src/obj-firefox/dist/include/nsTArray.h:522
[task 2019-01-25T07:48:55.761Z] 07:48:55 INFO - PID 5440 | #22 0x7fc38bc9598c in DestructRange /builds/worker/workspace/build/src/obj-firefox/dist/include/nsTArray.h:2178
[task 2019-01-25T07:48:55.762Z] 07:48:55 INFO - PID 5440 | #23 0x7fc38bc9598c in ClearAndRetainStorage /builds/worker/workspace/build/src/obj-firefox/dist/include/nsTArray.h:1296
[task 2019-01-25T07:48:55.763Z] 07:48:55 INFO - PID 5440 | #24 0x7fc38bc9598c in Clear /builds/worker/workspace/build/src/obj-firefox/dist/include/nsTArray.h:1759
[task 2019-01-25T07:48:55.766Z] 07:48:55 INFO - PID 5440 | #25 0x7fc38bc9598c in mozilla::(anonymous namespace)::MediaStreamGraphShutDownRunnable::Run() /builds/worker/workspace/build/src/dom/media/MediaStreamGraph.cpp:1546
[task 2019-01-25T07:48:55.767Z] 07:48:55 INFO - PID 5440 | #26 0x7fc3851848f6 in mozilla::AutoTaskDispatcher::TaskGroupRunnable::Run() /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/TaskDispatcher.h:197:35
[task 2019-01-25T07:48:55.768Z] 07:48:55 INFO - PID 5440 | #27 0x7fc385180e1b in mozilla::EventTargetWrapper::Runner::Run() /builds/worker/workspace/build/src/xpcom/threads/AbstractThread.cpp:139:32
[task 2019-01-25T07:48:55.769Z] 07:48:55 INFO - PID 5440 | #28 0x7fc385176901 in mozilla::SchedulerGroup::Runnable::Run() /builds/worker/workspace/build/src/xpcom/threads/SchedulerGroup.cpp:299:32
[task 2019-01-25T07:48:55.769Z] 07:48:55 INFO - PID 5440 | #29 0x7fc3851a69d6 in nsThread::ProcessNextEvent(bool, bool*) /builds/worker/workspace/build/src/xpcom/threads/nsThread.cpp:1160:14
[task 2019-01-25T07:48:55.770Z] 07:48:55 INFO - PID 5440 | #30 0x7fc3851d0af1 in NS_InvokeByIndex /builds/worker/workspace/build/src/xpcom/reflect/xptcall/md/unix/xptcinvoke_asm_x86_64_unix.S:106
[task 2019-01-25T07:48:55.780Z] 07:48:55 INFO - PID 5440 | #31 0x7fc386c98113 in Invoke /builds/worker/workspace/build/src/js/xpconnect/src/XPCWrappedNative.cpp:1628:10
[task 2019-01-25T07:48:55.781Z] 07:48:55 INFO - PID 5440 | #32 0x7fc386c98113 in Call /builds/worker/workspace/build/src/js/xpconnect/src/XPCWrappedNative.cpp:1186
[task 2019-01-25T07:48:55.781Z] 07:48:55 INFO - PID 5440 | #33 0x7fc386c98113 in XPCWrappedNative::CallMethod(XPCCallContext&, XPCWrappedNative::CallMode) /builds/worker/workspace/build/src/js/xpconnect/src/XPCWrappedNative.cpp:1152
[task 2019-01-25T07:48:55.784Z] 07:48:55 INFO - PID 5440 | #34 0x7fc386c9e166 in XPC_WN_CallMethod(JSContext*, unsigned int, JS::Value*) /builds/worker/workspace/build/src/js/xpconnect/src/XPCWrappedNativeJSOps.cpp:946:10
[task 2019-01-25T07:48:55.802Z] 07:48:55 INFO - PID 5440 | #35 0x194c174b2c0f (<unknown module>)
[task 2019-01-25T07:48:55.802Z] 07:48:55 INFO - PID 5440 | AddressSanitizer can not provide additional info.
[task 2019-01-25T07:48:55.802Z] 07:48:55 INFO - PID 5440 | SUMMARY: AddressSanitizer: SEGV /builds/worker/workspace/build/src/obj-firefox/dist/include/mozilla/MozPromise.h:429:9 in mozilla::MozPromise<bool, nsresult, true>::ThenValueBase::AssertIsDead()
[task 2019-01-25T07:48:55.802Z] 07:48:55 INFO - PID 5440 | ==6464==ABORTING
|
|
1,522,779 | 13,883,990 |
Created attachment 9039040
<input type="date">
See screenshots.
This is a regression from bug 1512489.
|
|
1,522,810 | 13,884,229 |
There seem to be some issues affecting the recommendations shown in about:addons, while in Private Browsing Mode we do not get the default extensions listed and also the 'Allow Nightly to make personalized extension recommendations' does not seem to affect the recommendations in any way.
In order to reproduce the issue:
1. Open about:config using any profile and set the 'extensions.webservice.discoverURL' pref to point to the dev environment: https://discovery.addons-dev.allizom.org/%LOCALE%/firefox/discovery/pane/%VERSION%/%OS%/%COMPATIBILITY_MODE%
2. Go to about:addons and select the Get Add-ons tab, notice the recommended extensions listed here.
3. Open a new Private Browsing window and go to the about:addons again
4. Go to options and uncheck the 'Allow Nightly to make personalized extension recommendations' checkbox located in Privacy and Security, then go back to about:addons and see the recommendations.
Expected results:
Step 3, 4 - The default recommended addons should be shown instead of personalized ones
Actual results:
We are shown the same list of recommendations regardless of the checkbox or private browsing mode. In some remote cases while in private browsing mode, I've also had recommendations change upon refreshing the page.
|
|
1,522,823 | 13,884,289 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.0.3578.98 Safari/537.36
Steps to reproduce:
- Install the Firefox ADMX templated
- Browse to the extensions options for Firefox
- Configured "Extensions to uninstall" with a * for the value
Actual results:
Nothing (kind of expected)
Expected results:
- We use this currently with Google Chrome's GPO settings.
- We need to block/blacklist all extensions except for what we want to allow and/or force install.
- Using a * is a easy way to do this if it works, but it does not work, nor is the naming or GPO options really geared towards that.
- There is no "whitelist" settings for permitted extensions
|
|
1,522,860 | 13,884,583 |
Created attachment 9039116
keyboard navigation broken on Top Sites.gif
**[Affected versions]:**
- Firefox Nightly 66.0a1, Build ID 20190125095457
- Firefox Dev Edition 66.0b1, Build ID 20190121133415
**[Affected Platforms]:**
- All Windows
- All Mac
- All Linux
**[Steps to reproduce]:**
1. Start the browser and open a new tab.
2. Using the keyboard, navigate to the 2nd Top Site's context menu.
3. Press Space and tab through to the last option.
4. Press Tab once more and observe the behavior.
**[Expected results]:**
- The focus is moved to the 3rd Top Site.
**[Actual results]:**
- The focus is moved to the 1st Top Site's context menu.
**[Regression range]:**
21:29.75 INFO: Narrowed inbound regression window from [2809c756, 6479cf44] (3 builds) to [2809c756, 226b130b] (2 builds) (~1 steps left)
21:29.75 INFO: No more inbound revisions, bisection finished.
21:29.75 INFO: Last good revision: 2809c756574d28c25d031eecd41779d3392c44d4
21:29.75 INFO: First bad revision: 226b130b6360f05f7329b3a836177af5bb32fd1d
21:29.75 INFO: Pushlog:
https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=2809c756574d28c25d031eecd41779d3392c44d4&tochange=226b130b6360f05f7329b3a836177af5bb32fd1d
Looks like this was introduced by something in Bug 1519303.
**[Notes]:**
- The issue is also reproducible on the Recommended by Pocket and Highlight sections.
- Attached a screen recording of the issue.
|
|
1,522,901 | 13,884,900 |
**Steps to reproduce**
1. Open the console
1. Evaluate the following:
```
{
var a = () => trace();
var b = () => trace();
var trace = () => console.trace();
a();
b();
}
```
**Expected results**
I see 2 different console.trace message:
```
console.trace() debugger eval code:4:21
trace debugger eval code:4
a debugger eval code:2
<anonymous> debugger eval code:5
```
and
```
console.trace() debugger eval code:4:21
trace debugger eval code:4
b debugger eval code:3
<anonymous> debugger eval code:6
```
**Actual results**
In only get the first trace, with a repeat badge of `2`.
---
|
|
1,523,064 | 13,886,023 |
Created attachment 9039313
IMG_9D2F16152400-1.jpeg
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.14; rv:66.0) Gecko/20100101 Firefox/66.0
Steps to reproduce:
For some webpages Reader View on iOS cannot pull main content, where the desktop version of Firefox Reader View works just fine. Example of such a page: https://www.thedailystar.net/environment/news/coca-cola-pepsi-tout-plastic-recycling-rare-joint-appearance-1693171
Actual results:
Steps:
1. Go to https://www.thedailystar.net/environment/news/coca-cola-pepsi-tout-plastic-recycling-rare-joint-appearance-1693171 from Firefox on iOS.
2. Click the Reader View icon.
Reader view was not able to pull main content instead it was just showing the heading and HTML source of featured image. Screenshot attached.
Expected results:
Reader view should be able to pull main content and display without any issue.
|
|
1,523,160 | 13,886,645 |
Hi,
you forgot to add descriptions for the SSLVersionMin and SSLVersionMax policies in bug 1522182.
|
|
1,523,245 | 13,888,525 |
**Steps to reproduce**
1. Go to https://nchevobbe.github.io/demo/console-test-app.html
1. Open the console
1. Scroll to the bottom, and click on the `trace` button in the `React/Redix` section.
1. In the console, you should see 2 console.trace messages, on the last one, click on the `React` group to expand it
1. Right click and select the `Copy message` action
1. Paste the result in the console input (or any other input)
**Expected results**
I get the following:
```
console.trace() App.js:17
render App.js:17
React 12
finishClassComponent
updateClassComponent
beginWork
performUnitOfWork
workLoop
renderRoot
performWorkOnRoot
performWork
performSyncWork
interactiveUpdates$1
interactiveUpdates
dispatchInteractiveEvent
```
**Actual results**
I get the following (note the extra new lines):
```
console.trace() App.js:17
render App.js:17
React 12
finishClassComponent
updateClassComponent
beginWork
performUnitOfWork
workLoop
renderRoot
performWorkOnRoot
performWork
performSyncWork
interactiveUpdates$1
interactiveUpdates
dispatchInteractiveEvent
```
---
|
|
1,523,381 | 13,889,586 | null |
|
1,523,408 | 13,889,836 |
I'm guessing this affects both activity stream and discovery stream:
> Warning: attempting to write 5837 bytes to preference browser.newtabpage.activity-stream.feeds.section.topstories.spoc.impressions. This is bad for general performance and memory usage. Such an amount of data should rather be written to an external file. This preference will not be sent to any content processes.
|
|
1,504,657 | 13,735,207 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:64.0) Gecko/20100101 Firefox/64.0
Steps to reproduce:
1) Go to www.jp.dk (A website for a major Danish newspaper)
2) Select an article and wait for the page to finish loading
3) Try to browse back
Actual results:
When trying to browse back, the page loads forever. On Chrome I have no issue.
|
|
1,504,686 | 13,736,475 |
DuckDuck go does a redirect on the search so we are double counting.
Need to either get DuckDuckGo to fix or update our engine to include the param they are adding.
|
|
1,504,725 | 13,736,840 |
Nothing happens when clicking the "Apply and restart" button after switching browser languages on Beta.
There is an error in the console: ReferenceError: event is not defined
|
|
1,504,766 | 13,737,176 |
Created attachment 9022687
Correct color for background tabs on Beta
I had held off updating for a few days because of the logout issue, so I'm not sure when this issue might have begun.
I first saw it when updating the morning for build 65.0a1 (2018-11-04).
The focused color of background tabs is light rather than dark. The colors look correct when the window loses focus though.
I tested in a fresh profile in beta and nightly. I've attached screenshots of both.
|
|
1,504,816 | 13,737,574 |
The changes in bug 1475228 may have introduced a use-after-free in nsJSUtils::CompileFunction. The new code looks like this:
JS::Rooted<JSFunction*> fun(cx);
JS::SourceBufferHolder source(PromiseFlatString(aBody).get(), aBody.Length(),
JS::SourceBufferHolder::NoOwnership);
if (!JS::CompileFunction(cx, aScopeChain, aOptions,
PromiseFlatCString(aName).get(),
aArgCount, aArgArray,
source, &fun))
if aBody is not flat, PromiseFlatString(aBody) makes a copy and constructs a temporary, .get() returns a pointer into that temporary, and the whole thing goes out of scope before JS::CompileFunction is called so at that point "source" is pointing into freed memory.
Now it's possible that all callsites that reach this point are passing flat strings; I haven't really dug through so far. But it would be a lot safer to not rely on that.
Bug 1503086 is going to fix this up on nightly, but we don't really want to backport it, so may want a more-targeted fix for backport.
Since SourceBufferHolder doesn't require null-termination, we presumably don't even need PromiseFlatString here. We can just BeginReading().
|
|
1,504,854 | 13,737,833 |
This is something Verdi and I noticed today. If you type @g so that @google is autofilled, the popup shows results that match @g. It should show only the single "Search with Google" heuristic result, the way it does if you actually typed out @google yourself.
|
|
1,505,024 | 13,739,045 |
Created attachment 9022921
image.png
**Steps to reproduce**
1. Open the console
2. Hit Cmd+F (or Ctrl+F if not on OSX)
3. The filter input should be focused
4. Hit Cmd+F (or Ctrl+F if not on OSX) again
**Expected results**
The focus stays on the input
**Actual results**
The filter input is blurred, and a weird search UI is displayed in the JsTerm
---
What gets triggered is the codeMirror "built-in" search UI.
It is enabled via the "search" addon.
|
|
1,505,063 | 13,739,385 |
Created attachment 9022973
devTools_emptyRDM.gif
[Affected versions]:
- 65.0a1 (2018-11-06)
[Affected platforms]:
- Win 10x64, Ubuntu 16.04LTS, macOS 10.13
[Steps to reproduce]:
1. Launch Firefox;
2. Open any local file;
3. Press F12 on the keyboard;
4. Click on the RDM button;
[Expected result]:
- the file/page is displayed;
[Actual result]:
- page content is empty;
- address is empty;
- previous-next (navigation)buttons are not enabled;
[Regression range]:
- bug 1472212 appears to be at fault;
[Additional notes]:
- attached recording with the issue;
- pressing home then back, does display the content but the devTools inspector sections are emptied out;
|
|
1,505,069 | 13,739,441 |
Not sure why --lwt-text-color was picked here. For starters, this is only set in the case of a webextension theme, end even then this won't necessarily do the right thing.
|
|
1,505,310 | 13,741,190 |
Created attachment 9023199
request.txt
This request is related to the shield study we want to run in 64 that looks at new certificate error page impact on retention and engagement (see bug 1501004 and bug 1503572. We need additional telemetry enabled on release for this study and the request is to have the probes detailed on bug 1503572 enabled on release.
|
|
1,505,515 | 13,742,595 |
We dereference nodeInfo here:
https://searchfox.org/mozilla-central/rev/6e0e603f4852b8e571e5b8ae133e772b18b6016e/mobile/android/geckoview/src/main/java/org/mozilla/geckoview/SessionAccessibility.java#153
and here:
https://searchfox.org/mozilla-central/rev/6e0e603f4852b8e571e5b8ae133e772b18b6016e/mobile/android/geckoview/src/main/java/org/mozilla/geckoview/SessionAccessibility.java#144
|
|
1,505,551 | 13,742,848 |
In order to allow independent roll-outs for CFR/Snippets/Onboarding features, we need to split configuration into different preferences.
We should also ensure this works with the requirements from shield's roll-out mechanism.
|
|
1,505,801 | 13,744,588 |
Created attachment 9023644
firefox-find-ui.png
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:65.0) Gecko/20100101 Firefox/65.0
Steps to reproduce:
Bring up "find in page" UI (Ctrl+F), begin typing.
65.0a1 (2018-11-07)
Actual results:
The entry box and next/previous arrow buttons begin correct, but the remaining UI (Highlight All, Match Case, etc.) is white and effectively invisible as a result.
After entering text, the next/previous arrows also change to white, effectively disappearing into the background color as well.
Expected results:
UI elements should use a sufficiently contrasting foreground color (the arrows are correct, prior to text entry) to be visible at all times.
|
|
1,505,897 | 13,745,284 |
No-Jun reported this bug on GitHub:
https://github.com/mozilla-mobile/focus-android/issues/3859
STR:
1. Load https://www.booking.com/
RESULT:
Fatal signal 11 (SIGSEGV), code 1, fault addr 0x0 in tid 12585 (Web Content)
DEBUG : Cause: null pointer dereference
Looks like some linker issue:
#00 pc 0005df7c /system/bin/linker (__dl_syscall+32)
#01 pc 000190bf /system/bin/linker (__dl__ZL13resend_signalP7siginfob+54)
|
|
1,505,991 | 13,745,968 |
Filed by: ncsoregi [at] mozilla.com
https://treeherder.mozilla.org/logviewer.html#?job_id=210697483&repo=autoland
https://queue.taskcluster.net/v1/task/aAGkQNo2Tjuvv-MEvW72Jw/runs/0/artifacts/public/logs/live_backing.log
00:11:22 INFO - 1541722282614 Marionette TRACE 1 <- [1,32,null,{"value":"iVBORw0KGgoAAAANSUhEUgAABAAAAAMHCAYAAACnvyW1AAAgAElEQVR4nO3dT2yb953ncR8HO4fd02BuW6DXrZFTDrnMXooAnZPdADmk ... IAAAAABBAAAAAAIIAAAAAAAAEEAAAAAAggAAAAAEAAAQAAAAACCAAAAAAQQAAAAACAAAIAAAAABBAAAAAAIIAAAAAAAAH+AyQLt0odbR2mAAAAAElFTkSuQmCC"}]
00:11:22 INFO - 1541722282617 Marionette TRACE 1 -> [0,33,"Marionette:SetContext",{"value":"chrome"}]
00:11:22 INFO - 1541722282618 Marionette TRACE 1 <- [1,33,null,{"value":null}]
00:11:22 INFO - 1541722282620 Marionette TRACE 1 -> [0,34,"Marionette:GetContext",{}]
00:11:22 INFO - 1541722282621 Marionette TRACE 1 <- [1,34,null,{"value":"chrome"}]
00:11:22 INFO - 1541722282627 Marionette TRACE 1 -> [0,35,"Marionette:SetContext",{"value":"content"}]
00:11:22 INFO - 1541722282629 Marionette TRACE 1 <- [1,35,null,{"value":null}]
00:11:22 INFO - 1541722282631 Marionette TRACE 1 -> [0,36,"WebDriver:GetPageSource",{}]
00:11:22 INFO - 1541722282639 Marionette TRACE 1 <- [1,36,null,{"value":"<html><head></head><body></body></html>"}]
00:11:22 INFO - 1541722282641 Marionette TRACE 1 -> [0,37,"Marionette:SetContext",{"value":"chrome"}]
00:11:22 INFO - 1541722282642 Marionette TRACE 1 <- [1,37,null,{"value":null}]
00:11:22 INFO - TEST-UNEXPECTED-FAIL | telemetry\marionette\tests\client\test_main_tab_scalars.py TestMainTabScalars.test_main_tab_scalars | AssertionError: Error waiting for ping: Timed out after 60.0 seconds
00:11:22 INFO - Traceback (most recent call last):
00:11:22 INFO - File "Z:\task_1541721296\build\venv\lib\site-packages\marionette_harness\marionette_test\testcases.py", line 159, in run
00:11:22 INFO - testMethod()
00:11:22 INFO - File "Z:\task_1541721296\build\tests\telemetry\marionette\tests\client\test_main_tab_scalars.py", line 40, in test_main_tab_scalars
00:11:22 INFO - [ping] = self.wait_for_pings(self.restart_browser, check_is_shutdown_ping, 1)
00:11:22 INFO - File "Z:\task_1541721296\build\venv\lib\site-packages\telemetry_harness\testcase.py", line 117, in wait_for_pings
00:11:22 INFO - self.fail("Error waiting for ping: {}".format(e.message))
00:11:22 INFO - TEST-INFO took 84333ms
|
|
1,506,102 | 13,746,792 |
Simple scenario: take an Italian build and enable intl.multilingual.enabled
At this point English ("Inglese (Stati Uniti)") will be available in the dropdown, but it can only translate Fluent parts without a language pack.
I'm not exactly sure what a solution could be for this, besides hiding en-US until we fully migrate to Fluent, in lack of a language pack?
|
|
1,506,114 | 13,746,891 |
Created attachment 9023950
library.gif
[Affected versions]:
Firefox 65.0a1 (BuildId:20181108100100)
[Unffected versions]:
Firefox 64.0b7 (BuildId:20181105164654)
Firefox 63.0.1 (BuildId:20181030165643)
Firefox 60.3.0esr (BuildId:20181017185317)
[Affected platforms]:
Windows 10 64bit
[Steps to reproduce]:
1. Launch Firefox.
2. Bookmark any webpage and observe the library button.
[Expected result]:
The library button is not displaying a fading animation.
[Actual result]:
The library button has a fading animation.
[Regression range]:
This seems to be a recent regression:
Last good revision: b9a0d600e885107a741ad7e98a9be6aeb0c6d2b4
First bad revision: a672ddc3be9ff55553aa5ffeead42a505e38bb94
Pushlog:
https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=b9a0d600e885107a741ad7e98a9be6aeb0c6d2b4&tochange=a672ddc3be9ff55553aa5ffeead42a505e38bb94
[Notes]:
For further information regarding this issue, please observe the attached screencast.
|
|
1,506,128 | 13,746,987 |
Created attachment 9023961
share.gif
[Affected versions]:
Firefox 65.0a1 (BuildId:20181108100100)
Firefox 64.0b7 (BuildId:20181105164654)
[Unaffected versions]:
Firefox 63.0.1 (BuildId:20181030165643)
Firefox 60.3.0esr (BuildId:20181017185317)
[Affected platforms]:
Windows 10 64bit
[Steps to reproduce]:
1. Launch Firefox.
2. Click the Page Actions button.
3. Right click on the "share" button.
4. Click the "Add to address bar" option.
[Expected result]:
The share button icon is displayed inside the address bar.
[Actual result]:
The share button icon is not visible.
[Regression range]:
This behavior was introduced with Bug 1363169
[Notes]:
For further information regarding this issue, please observe the attached screencast.
|
|
1,506,225 | 13,747,804 |
This is required in order to support versioning of messages
|
|
1,506,230 | 13,747,848 |
Created attachment 9024070
Short video of the issue
[Precondition]
Navigate to about:config and enable the Changes feat. ( devtools.inspector.changes.enabled - True)
[STR]
1. Launch Firefox and enable the Inspector.
2. Make sure Changes tab is not focused when the Inspector is enabled. (If it is, just change the focus to a different 3rd pane tab, close the inspector end re-enable).
3. Perform any change. (edit property names / values / add rules / delete rules / disable rules / Font Changes ).
4. Navigate to Changes tab.
[AR]
None of the changes performed are recorded in the Changes tab.
I've added a short video for better understanding.
[ER]
Changes are tracked even if the Changes tab is not in focus.
|
|
1,506,454 | 13,749,706 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.13; rv:65.0) Gecko/20100101 Firefox/65.0
Steps to reproduce:
When I worked on `unhandledrejection` event for Servo, I found this internal slot was missing in SM. So, I've sent a PR to mozjs in https://github.com/servo/mozjs/pull/150 .
I'll send a patch for this issue.
Actual results:
The `GetPromiseIsHandled` internal slot is missing
Expected results:
We should have `GetPromiseIsHandled` internal slot in Promise
|
|
1,506,457 | 13,749,724 |
Created attachment 9024292
Screenshot 2018-11-11 at 16.07.55.png
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.13; rv:60.0) Gecko/20100101 Firefox/60.0
Steps to reproduce:
Just open a new private windows in full screen mode on macOS (the bug also occurs if the private window is not open in full screen and put after full screen).
Actual results:
The top right "private windows" icon is duplicated (see screenshot attached)!
If we put the windows back in normal mode (not full screen) the duplicate icon disappear.
Expected results:
The top right icon should not be duplicated in full screen mode.
|
|
1,506,495 | 13,749,972 |
Created attachment 9024331
color.jpg
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.14; rv:65.0) Gecko/20100101 Firefox/65.0
Steps to reproduce:
gfx.color_management.enablev4 = true
gfx.color_management.mode = 1
gfx.color_management.display_profile = /Library/ColorSync/Profiles/SyncMaster.icc
Actual results:
(attached image)
Firefox 64.0 = Safari 12 ≠ Firefox 65.0
Expected results:
Color management seems to not work.
|
|
1,506,531 | 13,751,405 |
Created attachment 9024366
Order Summary page.png
Created attachment 9023247 [details]
Screenshot of issue
[Affected versions]:
Nightly 65.0a1
[Affected platforms]:
Windows 10 x 64, Windows 7 x32, Mac OS X 10.13 and on Ubuntu 16.04 x64.
[Prerequisites]:
- set the pref dom.payments.request.enabled to "true"
- set the pref browser.search.region to "US"
- make sure to have at least one Shipping Address and saved CC
[Steps to reproduce]:
1. Go to "https://rsolomakhin.github.io/pr/single/" and click on "Buy"
2. Select any address and a Payment Method
3. Observe shipping address, shipping options and payment method fields.
[Expected Result]:
The field should be properly filled in with the selected option from the drop-down menu.
[Actual Result]:
The fields are populated twice with the selected option from the drop-down.
[Note]:
This bug is similar to what is happening on Windows with High-Contrast
https://bugzilla.mozilla.org/show_bug.cgi?id=1505359
|
|
1,506,640 | 13,752,116 |
The basic/bug908915.js test fails when I run it locally with zeal mode 10 and the JITs disabled.
$ JS_GC_ZEAL=10 ./jit-test/jit_test.py optdebug-build/shell --args='--no-baseline --no-ion' bug908915
Assertion failure: found(), at /home/jon/clone/inbound/js/src/optdebug-build/dist/include/mozilla/HashTable.h:1273
Exit code: -11
FAIL - basic/bug908915.js
|
|
1,506,709 | 13,752,893 |
...disable full tree a11y by default.
|
|
1,506,734 | 13,753,061 |
For Linux, we need to allow support for full paths and distribution directory.
|
|
1,506,751 | 13,753,159 |
User Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.77 Safari/537.36
Steps to reproduce:
Load page
Select element with media query max-width rule
Change screen width to trigger css border color changes on element
Refreshed using Ctrl + F5 (and also ctrl + R)
Video of problem here: https://i.redd.it/usia2q4c6mx11.gif
Tested also with other rules in media query. Same bug.
Tested on clean install of Firefox, Developer edition and Portable edition, and Nightly v65.x
Windows 7, 64bit, FF v63.0.1
Actual results:
Before refresh style updated correctly in devtools rules pane
After refresh style doesn't change to reflect border color change
Note: the style on the page is changing correctly, but the rule in the devtools pane is not after a page refresh. Closing and reopening tab/browser restores correct behavior. It's the refresh that causes the bug.
Expected results:
After page refresh rules should update in rules pane to reflect css style changes based on media query
|
|
1,506,792 | 13,753,547 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:65.0) Gecko/20100101 Firefox/65.0
Steps to reproduce:
1. Load this minimal HTML: <style>body::after{content:""}</style><hr>. Alternatively, https://www.topicbox.com will do.
2. Open the Inspector tab of the dev tools, and look at its representation of the DOM tree.
Actual results:
The <body> node in the tree is shown as expanded, but contains no children.
Expected results:
The <body> should display all its children (an <hr> in my minimal example, or a lot more on https://www.topicbox.com).
Note that you can work around this bug by closing the dev tools and reopening them. But you’ll need to do this on every page load, which is tedious.
This has been annoying me for many, many months (it’s been a bug from the very start of the new dev tools, I think), but for some reason I never got round to minimising it and filing this bug report until today.
|
|
1,506,825 | 13,753,770 |
Environment:
Operating system: Windows 10 Pro
Firefox version: Firefox Nightly 65.0a1 (2018-11-12) (64-bit)
Prerequisites:
Enable Basic Tracking Protection
Steps to reproduce:
1. Navigate to: https://na85.lightning.force.com/lightning/setup/SetupOneHome/home
2. Login withvalid credentials.
Expected Behavior:
The site is loaded.
Actual Behavior:
The site remains in a loading state.
Notes:
1. Screenshot attached: http://prntscr.com/lhss4z
|
|
1,506,838 | 13,753,873 |
Created attachment 9024721
scene1.png
Send to device in 64.0b8 has some major issues with missing default values, wrong parsing of input and display of snippet
- scene1_button_label no default value. Must default to Learn More.
- scene2_dismiss_button_text no default value - Must default to Dismiss
- scene2_button_label has default value - Must default to Send
- scene2_input_placeholder has default value - Must default to YOUR EMAIL HERE
- scene2_title is not displayed when set
- scene2_text is not rich text
- locale - must default to en-us
- country - must default o us
- newsletters must default to '' (empty string, not undefined)
- include_sms values are True and False as strings. must default to False
- Buttons appear larger and text smaller in the new template compared to the old one. (see screenshot. Current template above, new ASR template bellow)
- Space of scene2_icon is wrong (see screenshot. Current template above, new ASR template bellow)
- Snippet container width is smaller in the new template (see screenshot. Current template above, new ASR template bellow)
- Overall the old templates look nicer compared to the old ones (see screenshot. Current template above, new ASR template bellow)
Templates must match what we have for the current AS system so we can automatically migrate the hundreds of active Snippets.
Ensuring that fields work (i.e. can be populated) and default values are set will shield a proper migration to ASR.
I haven't completed testing all implemented templates. Same issues may be applicable to other templates as well. Can you please spend some time to QA currently implemented templates?
I'm available to help interpret current templates and answer any questions if needed.
See also https://github.com/mozmeao/snippets-service/issues/831
|
|
1,506,867 | 13,754,040 |
**Steps to reproduce**
1. Open the console
2. Evaluate `x = {y: {z: 2}}`
3. Type x["y"].
**Expected results**
The autocomplete popup is displayed, and contains the `z` item
**Actual results**
The autocomplete popup is not displayed
---
This can be confusing because for example `x.y[` will show the popup.
---
This is because in JsPropertyProvider we split on `.` to get all the properties [1].
This is a very naive solution that fails in a lot of cases (for example `(x).`).
Instead of doing this, we could re-use the AST we already create a few lines up [2], and traverse it in order to retrieve the property chain.
If we can't, we can always fallback to split on ".".
[1] https://searchfox.org/mozilla-central/rev/7f7c353e969e61a6a85201cc8ad3c3de12ac30d8/devtools/shared/webconsole/js-property-provider.js#338
[2] https://searchfox.org/mozilla-central/rev/7f7c353e969e61a6a85201cc8ad3c3de12ac30d8/devtools/shared/webconsole/js-property-provider.js#290-294
|
|
1,506,869 | 13,754,061 |
From bug 1505847 comment 4:
> This change removed the hover/checked/active states from the findbar buttons
> on macOS.
>
> There's this rule that needed to be taken care of:
> https://searchfox.org/mozilla-central/source/browser/themes/osx/browser.
> css#211-216
|
|
1,506,976 | 13,755,018 |
We have various pieces of instrumentation for measuring page load until a certain paint. At the moment, all these paint measurements take their "end" timestamp at the time the display list for that paint is built.
However, this skips a large part of the painting pipeline: tile allocation, rasterization (which can happen on a different thread), time until the transaction is processed in the compositor thread's event queue, compositing.
We should build upon the framework from bug 1501442 to obtain the correct timestamp here. This will have two advantages:
- The measured value reflects reality more closely.
- It's easier to associate these timings / markers with screenshots taken in the compositor. This makes it easier to assess their validity.
Especially the latter is really hard to do correctly with the current setup.
|
|
1,507,026 | 13,755,395 |
Looks like I didn't use Java's synchronized blocks correctly. I was using the method modifier, and it does not work with a mix of inner classes.
|
|
1,507,055 | 13,755,566 |
This requires a Google Cloud account (you can get one for free).
STR:
1. Go to Google Cloud Console: https://console.cloud.google.com
2. From the menu on the left hand side, under Stackdriver, click on Monitoring.
3. You'll get redirected to your Google Account login page, select an account there.
4. You'll be redirected to <https://app.google.stackdriver.com/account/login2/?next=https%3A%2F%2Fapp.google.stackdriver.com%2Fsocial%2Fcomplete%2Fgoogle-oauth2%2F>.
5. Press Log in with Google.
6. Back to step 4.
Now the user enters a never ending loop.
Andrea, any chance you could take a look please? Thanks!
|
|
1,507,138 | 13,756,380 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:63.0) Gecko/20100101 Firefox/63.0
Steps to reproduce:
::: script :::
let proxySetting = {
proxyType: "manual",
http: "proxy.org:80",
ssl: "proxy.org:80"
};
browser.proxy.settings.set({value:proxySetting})
.then(()=>{
browser.proxy.settings.get({})
.then((got) => {
console.log(got.value);
});
});
Actual results:
If a port 80 is passed by a script, it will be deleted accoording to the log.
In the GUI (General / Network Proxy / Settings...),
the port is then displayed with the value zero
and Firefox will also use the value zero in the following for a proxy.
::: log :::
autoConfigUrl: ""
autoLogin: false
ftp: ""
http: "proxy.org"
httpProxyAll: false
passthrough: "localhost, 127.0.0.1"
proxyDNS: false
proxyType: "manual"
socks: ""
socksVersion: 5
ssl: "proxy.org"
::: about:config :::
network.proxy.http_port = 0
network.proxy.ssl_port = 0
Expected results:
The specified port 80 must be retained.
It seems that special error handling for port 80 is also present in other protocols, e.q. ssl, ftp.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.