Bug ID
int64 961
1.91M
| Comment ID
int64 3.98k
17.1M
| Author
stringlengths 8
48
| Comment Text
stringlengths 1
64.3k
⌀ |
|---|---|---|---|
1,856,261 | 16,600,908 |
User Agent: Mozilla/5.0 (Android 8.0.0; Mobile; rv:109.0) Gecko/120.0 Firefox/120.0
Steps to reproduce:
1. Go to the 3-dots menu on the top right, and select Add-Ons.
2. If you have one selectable in this menu installed (e.g. uBlock Origin), click on it.
3. Exit this menu.
4. The three buttons on the top left are missing, leaving the home button by itself. Doubles are still accessible by the 3-dot menu.
Actual results:
The reload, back, and forward buttons on the top left disappear upon exiting the addon's menu. They seem to reappear upon searching a query in a new tab, but there may be other triggers both to the disappearing and reappearing I have not found.
Expected results:
They should have remained as they were before the menu was opened, and visible. It worked as intended in older versions (the only other version I've tested being sometime around 108-110)
This is not reproducible on smaller mobile devices such as phones (though perhaps in landscape mode?) due to the screen size hiding them anyway.
|
|
1,859,581 | 16,627,200 |
Created attachment 9358901
The new private tab in HCM
**Found in**
* Nightly 120.0a1
**Affected versions**
* Nightly 120.0a1
**Tested platforms**
* Affected platforms: Ubuntu 22.04, macOS 11.7, Windows 10.
**Preconditions**
* In about:config, set "browser.privatebrowsing.felt-privacy-v1" to true
* Enable High Contrast Mode.
**Steps to reproduce**
1. Open a private window.
2. Observe the visibility of text in the banners and the "X" button.
**Expected result**
* The text within the banners and the "X" button should be easily visible in High Contrast Mode.
**Actual result**
* With the new private theme, the text within the banners and the "X" control button is difficult to read due to insufficient contrast.
**Regression range**
* Not a regression.
|
|
1,859,793 | 16,628,712 |
Created attachment 9359109
rec of the issue 2.gif
**[Notes]:**
- We have observed in the [experimenter ticket](https://experimenter.services.mozilla.com/nimbus/set-default-pdf-handler-with-annotations/summary) that the following part of the JSON that is contained in the "Treatment B" branch value is not present in the "Treatment A" branch value for the Set as Default action:
`{
"type": "BLOCK_MESSAGE",
"data": {
"id": "PDF_HANDLER_ANNOTATIONS:treatment_a"
}`
**[Affected versions]:**
- Firefox Release 118.0.2 - Build ID: 20231009140911
- Firefox RC 119.0 - Build ID: 20231013091412
**[Affected Platforms]:**
- Windows 10 x64
**[Prerequisites]:**
- Have a Firefox profile with the following prefs in the "about:config" page:
- `nimbus.debug` set to `true`
- `browser.newtabpage.activity-stream.asrouter.devtoolsEnabled` set to `true`
**[Steps to reproduce]:**
1. Open the browser with the profile from the prerequisites.
2. Navigate to "about:studies?optin_slug=set-default-pdf-handler-with-annotations&optin_branch=treatment-a&optin_collection=nimbus-preview".
3. Navigate to the "about:newtab#devtools" page and search for the "PDF_HANDLER_ANNOTATIONS:treatment_a" message.
4. Click the "Show" button from the message section.
5. Click the "Set as Default" button from the "PDF Handler".
6. Observe the message from the "AS Router Admin"
**[Expected result]:**
- The message section is grayed out and the "Unblock" button is displayed.
**[Actual result]:**
- Nothing happens, the message remains active.
**[Additional Notes]:**
- Due to this issue the message will be redisplayed to users if the default PDF handler is changed to another known browser.
- This issue is not reproducible with the Treatment B branch of the experiment. The "PDF_HANDLER_ANNOTATIONS:treatment_b" is blocked if the "Set as Default" button is clicked.
- Attached a screen recording of the issue.
|
|
1,859,825 | 16,628,913 |
Created attachment 9359145
testcase.html
Found while fuzzing m-c 20231017-71d9492d06c4 (--enable-address-sanitizer --enable-fuzzing)
To reproduce via Grizzly Replay:
```
$ pip install fuzzfetch grizzly-framework
$ python -m fuzzfetch -a --fuzzing -n firefox
$ python -m grizzly.replay ./firefox/firefox testcase.html
```
```
==203086==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000110 (pc 0x7f7c85906453 bp 0x7ffda5004f70 sp 0x7ffda5004ee0 T0)
==203086==The signal is caused by a READ memory access.
==203086==Hint: address points to the zero page.
#0 0x7f7c85906453 in load /builds/worker/fetches/sysroot-x86_64-linux-gnu/usr/lib/gcc/x86_64-linux-gnu/8/../../../../include/c++/8/bits/atomic_base.h:396:9
#1 0x7f7c85906453 in load /builds/worker/workspace/obj-build/dist/include/mozilla/Atomics.h:195:17
#2 0x7f7c85906453 in operator bool /builds/worker/workspace/obj-build/dist/include/mozilla/Atomics.h:496:12
#3 0x7f7c85906453 in IsWritable /builds/worker/checkouts/gecko/xpcom/ds/PLDHashTable.h:111:36
#4 0x7f7c85906453 in StartWriteOp /builds/worker/checkouts/gecko/xpcom/ds/PLDHashTable.h:140:5
#5 0x7f7c85906453 in AutoWriteOp /builds/worker/checkouts/gecko/xpcom/ds/PLDHashTable.cpp:40:59
#6 0x7f7c85906453 in PLDHashTable::Remove(void const*) /builds/worker/checkouts/gecko/xpcom/ds/PLDHashTable.cpp:528:15
#7 0x7f7c8c9c0a50 in RemoveEntry /builds/worker/workspace/obj-build/dist/include/nsTHashtable.h:346:12
#8 0x7f7c8c9c0a50 in RemoveEntry /builds/worker/workspace/obj-build/dist/include/nsTHashtable.h:825:37
#9 0x7f7c8c9c0a50 in Remove /builds/worker/workspace/obj-build/dist/include/nsTHashSet.h:117:41
#10 0x7f7c8c9c0a50 in UntrackBuffer /builds/worker/checkouts/gecko/dom/webgpu/Device.cpp:102:63
#11 0x7f7c8c9c0a50 in mozilla::webgpu::Buffer::Drop() /builds/worker/checkouts/gecko/dom/webgpu/Buffer.cpp:153:15
#12 0x7f7c8c9c1781 in mozilla::webgpu::Buffer::~Buffer() /builds/worker/checkouts/gecko/dom/webgpu/Buffer.cpp:54:3
#13 0x7f7c8c9e1540 in DeleteCycleCollectable /builds/worker/workspace/obj-build/dist/include/mozilla/webgpu/Buffer.h:44:3
#14 0x7f7c8c9e1540 in mozilla::webgpu::Buffer::cycleCollection::DeleteCycleCollectable(void*) /builds/worker/workspace/obj-build/dist/include/mozilla/webgpu/Buffer.h:44:3
#15 0x7f7c858900c6 in SnowWhiteKiller::~SnowWhiteKiller() /builds/worker/checkouts/gecko/xpcom/base/nsCycleCollector.cpp:2472:7
#16 0x7f7c8588ee75 in nsCycleCollector::FreeSnowWhite(bool) /builds/worker/checkouts/gecko/xpcom/base/nsCycleCollector.cpp:2662:3
#17 0x7f7c8589ac17 in nsCycleCollector::BeginCollection(mozilla::CCReason, ccIsManual, nsICycleCollectorListener*) /builds/worker/checkouts/gecko/xpcom/base/nsCycleCollector.cpp:3659:3
#18 0x7f7c85899931 in nsCycleCollector::Collect(mozilla::CCReason, ccIsManual, js::SliceBudget&, nsICycleCollectorListener*, bool) /builds/worker/checkouts/gecko/xpcom/base/nsCycleCollector.cpp:3483:9
#19 0x7f7c8589921a in nsCycleCollector::ShutdownCollect() /builds/worker/checkouts/gecko/xpcom/base/nsCycleCollector.cpp:3417:20
#20 0x7f7c8589c5c6 in nsCycleCollector::Shutdown(bool) /builds/worker/checkouts/gecko/xpcom/base/nsCycleCollector.cpp:3721:5
#21 0x7f7c8589ecc0 in nsCycleCollector_shutdown(bool) /builds/worker/checkouts/gecko/xpcom/base/nsCycleCollector.cpp:4045:18
#22 0x7f7c85b471f2 in mozilla::ShutdownXPCOM(nsIServiceManager*) /builds/worker/checkouts/gecko/xpcom/build/XPCOMInit.cpp:702:3
#23 0x7f7c96640c42 in XRE_InitChildProcess(int, char**, XREChildData const*) /builds/worker/checkouts/gecko/toolkit/xre/nsEmbedFunctions.cpp:660:16
#24 0x562e10d2007c in content_process_main /builds/worker/checkouts/gecko/browser/app/../../ipc/contentproc/plugin-container.cpp:57:28
#25 0x562e10d2007c in main /builds/worker/checkouts/gecko/browser/app/nsBrowserApp.cpp:375:18
#26 0x7f7cad029d8f in __libc_start_call_main csu/../sysdeps/nptl/libc_start_call_main.h:58:16
#27 0x7f7cad029e3f in __libc_start_main csu/../csu/libc-start.c:392:3
#28 0x562e10c44388 in _start (/home/user/workspace/browsers/m-c-20231018094117-fuzzing-asan-opt/firefox+0xdc388) (BuildId: ca2f30af2fc2908e9ae9232cd28460de69599e13)
```
|
|
1,859,913 | 16,629,348 |
We recently made a change to the shopping CFRs that made them harder to dismiss. They can only be dismissed by clicking the close X or by clicking the link that opens the bottom sheet. The issue is that this results in the page and the UI being unresponsive. So we want to revert back to the original behavior where they were dismissible by clicking outside.
I will file an additional bug with a request for new behavior.
|
|
1,859,978 | 16,629,738 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36
Steps to reproduce:
(1) Set up a new mail account
(2) Import OpenPGP S/E subkeys + a matching public key
(3) A password dialog is shown, the entered password should be used by TB to set a new password for the private subkeys and store it in [ProfD]/encrypted-openpgp-passphrase.txt
(4) Compose a new message, enable OpenPGP signing, and click the Send button.
Actual results:
Some password has been stored in [ProfD]/encrypted-openpgp-passphrase.txt in Step (3) above. However, another password prompt is displayed when sending the message, where the user is required to enter the original OpenPGP S subkey password. I.e., the subkey password has not been changed during the import process.
Expected results:
No password prompt should be shown when sending an OpenPGP-signed message. The subkey password should have been selected randomly during the key import process, stored in [ProfD]/encrypted-openpgp-passphrase.txt, and then used again as needed.
Additional information:
* The problem stared occurring after one of the recent updates.
* Possibly relevant: I have imported into TB only my public key and S/E subkeys, but not the private key itself. I know that https://support.mozilla.org/en-US/kb/openpgp-thunderbird-howto-and-faq#w_what-types-of-openpgp-keys-are-supported states that "Certain keys that are incomplete, for example those using an offline primary key" may not be supported, but I had been using the very same setup before (even with earlier versions of Supernova), and then it was working flawlessly.
* Possibly relevant #2: https://bugzilla.mozilla.org/show_bug.cgi?id=1629195
* Setting up a new TB profile does not lead to any improvement.
|
|
1,860,130 | 16,630,549 |
### Steps to reproduce
1. open `https://addons.mozilla.org/en-US/firefox/addon/tree-style-tab/` in a new Fenix tab
2. click the readerview icon
### Expected behavior
The content of the tree-style-tab document to be rendered in readerview mode
### Actual behavior
The tab gets reloaded automatically on Nightly, while on Beta and Release the tab stays empty, in both cases the tab doesn't enter the readerview mode.
### Device information
Reproduced on Fenix 120 on the emulator (and 119 and 118 on a real android device, Pixel 6a with Android 13).
### Any additional information?
While investigating this issue I determined that the "tab reload" part of the actual behavior is triggered by [readerview.js itself from here when it fails to retrive a doc for the page we want to render in readerview mode](https://searchfox.org/mozilla-mobile/rev/d09c6e0a19680781d290b83f2200126598e31f7b/firefox-android/android-components/components/feature/readerview/src/main/assets/extensions/readerview/readerview.js#330).
The issue can be triggered consistently on the addons.mozilla.org detail page because retrieving the page using XMLHttpRequest seems to be failing (because it is hitting a `CORS related error Cross-Origin Request Blocked: The Same Origin Policy disallows reading the remote resource at https://addons.mozilla.org/en-US/firefox/addon/tree-style-tab/. (Reason: CORS header ‘Access-Control-Allow-Origin’ missing). Status code: 200.)` error), and in that case we will be trying to retrieve the doc through the `getSerializedDoc` runtime message, which was also not providing a cached doc.
After looking more closely to why the doc wasn't returned by `getSerializedDoc`, I applied some more fixes to the readerview background script to make it to correctly provide the reply to the getSerializedDoc message, and with just that change applied locally I was able to enter and exit readerview mode on the tree-style-tab AMO detail page as expected.
|
|
1,860,602 | 16,634,970 |
Created attachment 9359849
Sample pdf file.pdf
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.0 Safari/605.1.15
Steps to reproduce:
1. Open Firefox.
2. Open new tab and drop "Sample pdf file.pdf" onto it
3. Go to page 2
4. No box is checked yet except int the first and second paragraphes
5. Click on print button from the pdf viewer toolbar
6. See page 2 preview
Actual results:
Two boxes are checked
Expected results:
No box must be checked
|
|
1,860,898 | 16,636,901 |
Scope of this bug is to debug and fix UI updates seen in Milestone message in Fx119 and Fx120.
|
|
1,861,215 | 16,638,411 |
See https://bugzilla.mozilla.org/show_bug.cgi?id=1858144#c5
|
|
1,861,707 | 16,641,739 |
Steps to reproduce:
I'm using this script:
```
<script>
(async () => {
const imports = {env:{s:"test"}};
const wasmModule = await WebAssembly.instantiateStreaming(fetch('bug.wasm'),imports)
console.log(wasmModule.instance.exports.f());
})()
</script>
```
to load this Wasm code:
```
(module
(import "env" "s" (global $s (ref any)))
(func (export "f") (result (ref any))
(global.get $s))
)
```
Actual results:
I get the following error:
```
LinkError: import object field 's' is not a Object-or-null value required for non-externref reference type
```
Expected results:
I would expect the string to be successfully imported. It works in Chrome, and the spec does seem to allow this.
|
|
1,861,722 | 16,641,777 |
Steps to reproduce:
I'm using this script:
```
<script>
(async () => {
const imports = {env:{s:"test"}};
const wasmModule = await WebAssembly.instantiateStreaming(fetch('bug.wasm'),imports)
console.log(wasmModule.instance.exports.f());
})()
</script>
```
to load this Wasm code:
```
<script>
(async () => {
const imports = {env:{s:"test"}};
const wasmModule = await WebAssembly.instantiateStreaming(fetch('bug.wasm'),imports)
console.log(wasmModule.instance.exports.f());
})()
</script>
```
Actual results:
I get a compilation error:
```
CompileError: wasm validation error: at offset 41: unrecognized opcode: fb 1a
```
Expected results:
This should compile without error, since `extern.internalize` (now called `any.convert_extern`) is a constant instruction according to the spec.
|
|
1,861,849 | 16,642,708 |
This seems like it happens only sometimes, and it is often resolved on a refresh.
I think this is due to how we use region to setup stories. We need region to know if we use stories or not, and that can take a moment on first profile startup. We may know the region before we start setting up newtab, and we may not.
If we know, we can get ahead and start and everything works as expected.
If we don't know region yet, we setup newtab anyway, and listen for when the region is set. In this case, newtab is about to setup stories but is going to fail, then set the `loaded` flag. Once the region is ready, we then do another attempt to setup stories. This attempt however, checks that `loaded` is true. There is a case where region is ready late, and `loaded` is not ready yet, causing newtab to render a handful of empty stories.
To test, load a new profile and check the first newtab that is opened (you don't need to open it manually), do this a few times, I can usually hit it 1 out of 4.
If you refresh, the region should be ready now because there is a second cache update that should be ready for startup, but even that's not a guarantee to be ready before region, I've just never been able to trigger it. Just very unlikely.
Expected: should see the normal set of stories.
Actual: Empty boxes instead.
|
|
1,862,982 | 16,651,624 |
**[Affected versions]:**
- Firefox Beta 120.0b5 - Build ID: 20231101092803
- Firefox Nightly 121.0a1 - Build ID: 20231103093836
**[Affected Platforms]:**
- Windows 10 x64
- macOS 14.0
**[Prerequisites]:**
- Have the `cookiebanners.service.mode.privateBrowsing` pref set to `1` on a new profile.
**[Steps to reproduce]:**
1. Open the browser from the prerequisites.
2. Navigate to cnn.com.
3. Click the shield icon from the address bar.
4. Click the "Cookie Banner Blocker" option from the dropdown.
5. Observe the `Turn Off/On` and `Cancel` buttons.
**[Expected result]:**
- The button labels are centered in the button.
**[Actual result]:**
- Beta: Both button labels are off center and are placed higher in the button when compared to similar buttons.
- Nightly: The `Cancel` button is off center and is placed higher in the button than the `Turn Off/On` one.
**[Notes]:**
- Attached [screenshots of the issue](https://drive.google.com/drive/folders/1OinLgHsbdLG_UiTeR97uVXe0Mi0lywSn?usp=sharing).
- The issue is either not reproducible on Linux or it's not very visible.
|
|
1,863,545 | 16,657,501 |
Created attachment 9362361
1699361916225.JPEG
### Steps to reproduce
1. Open a product detailed page from Amazon, BestBuy, or Walmart, and search for a product without quality reviews
for example: https://www.amazon.com/dp/B0CJFKSWKW/ref=sspa_mw_detail_1?ie=UTF8&psc=1&sp_csd=d2lkZ2V0TmFtZT1zcF9waG9uZV9kZXRhaWwp13NParams
2. Tap the price tag icon in the URL to view the bottom sheet.
3. Expand the bottom sheet by swiping up the bar above the title.
4. Observe if the "Review checker is powered by Fakespot by Mozilla" text is displayed at the bottom.
### Expected behavior
The "Review checker is powered by Fakespot by Mozilla" text should be displayed at all times in the bottom sheet, as shown the Figma design.
### Actual behavior
The "Review checker is powered by Fakespot by Mozilla" text is not displayed for products without analysis, nor during the analyzing process.
### Device information
* Firefox version: Nightly 121.0a1 from 11/07, Beta 120.0b7
* Android device: LG G7 fit (Android 8.1.0), Samsung Galaxy A53 5G (Android 13) (reproducible with any device).
### Any additional information?
After the analysis is complete, the "Review checker is powered by Fakespot by Mozilla" text is displayed on the bottom sheet.
|
|
1,863,549 | 16,657,515 |
Created attachment 9362370
Screenshot_1.jpg
### Prerequisites
Enroll in the "Android Review Checker Beta Smoke Test" Nimbus Experiment.
### Steps to reproduce
1. Go to an amazon.com, bestbuy.com, or walmart.com product detailed page.
2. Tap on the shopping icon.
3. On the bottom sheet, expand the "How we determine review quality" by tapping the down chevron.
4. Tap on the "How Fakespot detemines review quality" hyperlink.
### Expected behavior
Tapping on the "How Fakespot detemines review quality" hyperlink redirects the user to a SUMO article.
### Actual behavior
Tapping the "How Fakespot detemines review quality" hyperlink redirects the user to a "Placeholder 1855991" page.
If the page is still in progress, maybe replace the "Placeholder 1855991" text to "In progress".
### Device information
* Firefox version: Beta 120.0b7, Nightly 121.0a1
* Android devices: all
|
|
1,863,692 | 16,658,407 |
Created attachment 9362531
session restore windows.webm
**Found in**
- Firefox 120.0b7
**Affected versions**
- Firefox 120.0b7
- Firefox 121.0a1
**Tested platforms**
- Affected platforms: Windows 11, macOS 13.6.1, Ubuntu 22.04
- Unaffected platforms:
**Steps to reproduce**
1. Launch Firefox
2. Access a couple websites in new windows
3. Modify the height/width of the windows and then exit Firefox via Menu
4. Reopen Firefox with the same profile
5. Click the Restore Previous Session option from the Hamburger menu
**Expected result**
- The previous window size and location on screen are preserved as before
**Actual result**
- The window location and size are not remembered when using restore session
**Regression range**
- Will see for a regression
**Additional notes**
- The issue can be seen in the attachment
|
|
1,863,783 | 16,659,211 |
Profile is here: https://share.firefox.dev/49vQtJ9
I was switching back and forth between tabs. I'm seeing a bunch of time in chrome://browser/content/firefoxview/opentabs.mjs:render()
It seems like Firefox View is adding about 300ms to every tab switch.
|
|
1,863,910 | 16,660,018 |
Created attachment 9362780
testJavascript.pdf
Steps to reproduce:
PDF Document with embeded JavaScript code opened in Firefox 119, the code isn't work, isn't run. The same Document open in Firefox 118 work fine.
Expected results:
The attached document would be at the first input fields, after focus lost create an alert dialog.
|
|
1,864,136 | 16,661,529 |
Created attachment 9362986
Sample pdf file.pdf
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.0 Safari/605.1.15
Steps to reproduce:
I'm using Firefox nightly (121.0a1).
Open Sample pdf file.pdf file with FF.
Focusing on page 2 where there are some radio and checkbox buttons.
After clicking on any radio and checkbox buttons, go to print view.
After repeating these steps (maybe several times) and changing boxes, you will see the print view doesn't display correctly.
Please watch the attached video.
Actual results:
Print view does not display the actual form.
|
|
1,864,246 | 16,662,269 |
Steps to reproduce:
Build options:
```
/bin/sh ../../gecko-dev/js/src/configure --enable-debug --disable-optimize --disable-shared-js --disable-tests
```
Test case:
```
this.evalStencil(this);
gc();
```
Actual results:
Error message:
```
Hit MOZ_CRASH(Invalid object. Dead wrapper?) at /media/Store/JS-engines/SpiderMonkey/gecko-dev/js/src/vm/JSObject.h:649
```
Stack backtrace
```
JSObject::maybeUnwrapAs<js::StencilObject>(JSObject * this) (gecko-dev/js/src/vm/JSObject.h:649)
EvalStencil(JSContext * cx, uint32_t argc, JS::Value * vp) (gecko-dev/js/src/builtin/TestingFunctions.cpp:7174)
CallJSNative(JSContext * cx, js::Native native, js::CallReason reason, const JS::CallArgs & args) (gecko-dev/js/src/vm/Interpreter.cpp:472)
js::InternalCallOrConstruct(JSContext * cx, const JS::CallArgs & args, js::MaybeConstruct construct, js::CallReason reason) (gecko-dev/js/src/vm/Interpreter.cpp:566)
InternalCall(JSContext * cx, const js::AnyInvokeArgs & args, js::CallReason reason) (gecko-dev/js/src/vm/Interpreter.cpp:633)
js::CallFromStack(JSContext * cx, const JS::CallArgs & args, js::CallReason reason) (gecko-dev/js/src/vm/Interpreter.cpp:638)
js::Interpret(JSContext * cx, js::RunState & state) (gecko-dev/js/src/vm/Interpreter.cpp:3053)
MaybeEnterInterpreterTrampoline(JSContext * cx, js::RunState & state) (gecko-dev/js/src/vm/Interpreter.cpp:386)
js::RunScript(JSContext * cx, js::RunState & state) (gecko-dev/js/src/vm/Interpreter.cpp:444)
js::ExecuteKernel(JSContext * cx, JS::HandleScript script, JS::HandleObject envChainArg, js::AbstractFramePtr evalInFrame, JS::MutableHandleValue result) (gecko-dev/js/src/vm/Interpreter.cpp:831)
js::Execute(JSContext * cx, JS::HandleScript script, JS::HandleObject envChain, JS::MutableHandleValue rval) (gecko-dev/js/src/vm/Interpreter.cpp:863)
ExecuteScript(JSContext * cx, JS::HandleObject envChain, JS::HandleScript script, JS::MutableHandleValue rval) (gecko-dev/js/src/vm/CompilationAndEvaluation.cpp:494)
JS_ExecuteScript(JSContext * cx, JS::HandleScript scriptArg) (gecko-dev/js/src/vm/CompilationAndEvaluation.cpp:518)
RunFile(JSContext * cx, const char * filename, FILE * file, CompileUtf8 compileMethod, bool compileOnly, bool fullParse) (gecko-dev/js/src/shell/js.cpp:1218)
Process(JSContext * cx, const char * filename, bool forceTTY, FileKind kind) (gecko-dev/js/src/shell/js.cpp:1798)
ProcessArgs(JSContext * cx, js::cli::OptionParser * op) (gecko-dev/js/src/shell/js.cpp:10873)
Shell(JSContext * cx, js::cli::OptionParser * op) (gecko-dev/js/src/shell/js.cpp:11135)
main(int argc, char ** argv) (gecko-dev/js/src/shell/js.cpp:11539)
```
|
|
1,864,257 | 16,662,323 |
Steps to reproduce:
version:
commit 456995bd895bc6c5d1dcd582ff72acbf7e7a28e0
Build options:
```
/bin/sh ../../gecko-dev/js/src/configure --enable-debug --disable-optimize --disable-shared-js --disable-tests
```
Test case:
```
function foo() { return 0; }
let x = new foo();
let y = 0;
x ^= y;
Float64Array.columnNumber = --x;
this.evaluate("valueOf", Float64Array);
```
Actual results:
Error message:
```
Assertion failure: value_ != 0, at dist/include/js/ColumnNumber.h:255
```
Stack backtrace
```
JS::detail::MaybeLimitedColumnNumber<1073741823u>::valid(const JS::detail::MaybeLimitedColumnNumber<1073741823u> * this) (gecko-dev/js/public/ColumnNumber.h:255)
JS::detail::MaybeLimitedColumnNumber<1073741823u>::MaybeLimitedColumnNumber(JS::detail::MaybeLimitedColumnNumber<1073741823u> * this, uint32_t value) (gecko-dev/js/public/ColumnNumber.h:161)
JS::LimitedColumnNumberOneOrigin::MaybeLimitedColumnNumber(JS::LimitedColumnNumberOneOrigin * this) (gecko-dev/js/public/ColumnNumber.h:289)
JS::LimitedColumnNumberOneOrigin::fromUnlimited(uint32_t value) (gecko-dev/js/public/ColumnNumber.h:304)
JS::LimitedColumnNumberOneOrigin::fromUnlimited(const JS::detail::MaybeLimitedColumnNumber<0u> & value) (gecko-dev/js/public/ColumnNumber.h:308)
js::ScriptSource::initFromOptions(js::ScriptSource * this, js::FrontendContext * fc, const JS::ReadOnlyCompileOptions & options) (gecko-dev/js/src/vm/JSScript.cpp:1911)
js::frontend::CompilationInput::initScriptSource(js::frontend::CompilationInput * this, js::FrontendContext * fc) (gecko-dev/js/src/frontend/Stencil.cpp:1355)
js::frontend::CompilationInput::initForGlobal(js::frontend::CompilationInput * this, js::FrontendContext * fc) (gecko-dev/js/src/frontend/CompilationStencil.h:701)
CompileGlobalScriptToStencilAndMaybeInstantiate<char16_t>(JSContext * maybeCx, js::FrontendContext * fc, js::LifoAlloc & tempLifoAlloc,
js::frontend::CompilationInput & input, js::frontend::ScopeBindingCache * scopeCache, JS::SourceText<char16_t> & srcBuf, js::ScopeKind scopeKind, js::frontend::ExtraBindingInfoVector * maybeExtraBindings, BytecodeCompilerOutput & output) (gecko-dev/js/src/frontend/BytecodeCompiler.cpp:315)
CompileGlobalScriptToStencilImpl<char16_t>(JSContext * maybeCx, js::FrontendContext * fc, js::LifoAlloc & tempLifoAlloc, js::frontend::CompilationInput & input, js::frontend::ScopeBindingCache * scopeCache, JS::SourceText<char16_t> & srcBuf, js::ScopeKind scopeKind) (gecko-dev/js/src/frontend/BytecodeCompiler.cpp:407)
js::frontend::CompileGlobalScriptToStencil(JSContext * cx, js::FrontendContext * fc, js::LifoAlloc & tempLifoAlloc, js::frontend::CompilationInput & input, js::frontend::ScopeBindingCache * scopeCache, JS::SourceText<char16_t> & srcBuf, js::ScopeKind scopeKind) (gecko-dev/js/src/frontend/BytecodeCompiler.cpp:419)
CompileGlobalScriptToStencilImpl<char16_t>(JSContext * cx, const JS::ReadOnlyCompileOptions & options, JS::SourceText<char16_t> & srcBuf) (gecko-dev/js/src/frontend/Stencil.cpp:5437)
JS::CompileGlobalScriptToStencil(JSContext * cx, const JS::ReadOnlyCompileOptions & options, JS::SourceText<char16_t> & srcBuf) (gecko-dev/js/src/frontend/Stencil.cpp:5457)
Evaluate(JSContext * cx, unsigned int argc, JS::Value * vp) (gecko-dev/js/src/shell/js.cpp:2700)
CallJSNative(JSContext * cx, js::Native native, js::CallReason reason, const JS::CallArgs & args) (gecko-dev/js/src/vm/Interpreter.cpp:472)
js::InternalCallOrConstruct(JSContext * cx, const JS::CallArgs & args, js::MaybeConstruct construct, js::CallReason reason) (gecko-dev/js/src/vm/Interpreter.cpp:566)
InternalCall(JSContext * cx, const js::AnyInvokeArgs & args, js::CallReason reason) (gecko-dev/js/src/vm/Interpreter.cpp:633)
js::CallFromStack(JSContext * cx, const JS::CallArgs & args, js::CallReason reason) (gecko-dev/js/src/vm/Interpreter.cpp:638)
js::Interpret(JSContext * cx, js::RunState & state) (gecko-dev/js/src/vm/Interpreter.cpp:3053)
MaybeEnterInterpreterTrampoline(JSContext * cx, js::RunState & state) (gecko-dev/js/src/vm/Interpreter.cpp:386)
js::RunScript(JSContext * cx, js::RunState & state) (gecko-dev/js/src/vm/Interpreter.cpp:444)
js::ExecuteKernel(JSContext * cx, JS::HandleScript script, JS::HandleObject envChainArg, js::AbstractFramePtr evalInFrame, JS::MutableHandleValue result) (gecko-dev/js/src/vm/Interpreter.cpp:831)
js::Execute(JSContext * cx, JS::HandleScript script, JS::HandleObject envChain, JS::MutableHandleValue rval) (gecko-dev/js/src/vm/Interpreter.cpp:863)
ExecuteScript(JSContext * cx, JS::HandleObject envChain, JS::HandleScript script, JS::MutableHandleValue rval) (gecko-dev/js/src/vm/CompilationAndEvaluation.cpp:494)
JS_ExecuteScript(JSContext * cx, JS::HandleScript scriptArg) (gecko-dev/js/src/vm/CompilationAndEvaluation.cpp:518)
RunFile(JSContext * cx, const char * filename, FILE * file, CompileUtf8 compileMethod, bool compileOnly, bool fullParse) (gecko-dev/js/src/shell/js.cpp:1218)
Process(JSContext * cx, const char * filename, bool forceTTY, FileKind kind) (gecko-dev/js/src/shell/js.cpp:1798)
ProcessArgs(JSContext * cx, js::cli::OptionParser * op) (gecko-dev/js/src/shell/js.cpp:10873)
Shell(JSContext * cx, js::cli::OptionParser * op) (gecko-dev/js/src/shell/js.cpp:11135)
main(int argc, char ** argv) (gecko-dev/js/src/shell/js.cpp:11539)
```
|
|
1,864,390 | 16,664,804 |
Created attachment 9363214
radio button focus.jpg
**[Affected versions]:**
- Firefox Release 119.0.1 - Build ID: 20231106151204
- Firefox Beta 120.0b6 - Build ID: 20231103091743
**[Affected Platforms]:**
- Windows 10 x64
- Windows 11 x64
**[Prerequisites]:**
- Have a new Firefox profile.
- Firefox is not set as the default browser.
- Have the following prefs in the "about:config" page.
`nimbus.debug `pref set to `true`.
`browser.ping-centre.log` pref set to `true`.
- Have the "Browser Console" opened.
**[Steps to reproduce]:**
1. Open the browser with the profile from the prerequisites.
2. Navigate to `about:studies?optin_slug=ask-new-profiles-if-they-are-a-new-or-existing-firefox-user-fx119&optin_branch=treatment-b&optin_collection=nimbus-preview`.
3. Restart the browser.
4. Navigate to “about:welcome”.
5. Press the “Tab” key until any of the radio buttons is focused.
6. Observe the radio button.
**[Expected result]:**
- The focus around the button is completely visible.
**[Actual result]:**
- The focus around the button is cut.
**[Notes]:**
- This issue is reproducible on both "Treatment A" and “Treatment B branches of the experiment.
- Attached is a screenshot of the issue.
|
|
1,865,044 | 16,670,032 |
The following testcase crashes on mozilla-central revision 20231115-571c293cedd2 (opt build, run with --fuzzing-safe --ion-offthread-compile=off --more-compartments):
a = newGlobal();
a.b = this;
a.eval("(" + function () {
Debugger(b).onExceptionUnwind = function () {}
} + ")()");
function c(d) {
binary = wasmTextToBinary(d)
e = new WebAssembly.Module(binary)
return new WebAssembly.Instance(e)
}
f = c(`
(type $g
(func (param i64 i64 funcref) (result i64))
)
(func (export "vis") (param i64 i64 funcref) (result i64)
local.get 0
local.get 1
local.get 2
local.get 2
ref.cast (ref $g)
return_call_ref $g
)
`);
h = f.exports["vis"];
i = new WebAssembly.Function({
parameters: [ "i64", "i64", "funcref" ], results: ["i64"]
}, function () {});
h(3n, 1n, i);
Backtrace:
received signal SIGSEGV, Segmentation fault.
#0 0x0000555556e3f2ce in js::jit::JSJitFrameIter::machineState() const ()
#1 0x0000555556f38bd9 in js::jit::InlineFrameIterator::resetOn(js::jit::JSJitFrameIter const*) ()
#2 0x0000555556f37469 in js::FrameIter::settleOnActivation() ()
#3 0x00005555575a1639 in js::Debugger::fireExceptionUnwind(JSContext*, JS::Handle<JS::Value>, js::ResumeMode&, JS::MutableHandle<JS::Value>) ()
#4 0x00005555575a13cb in js::DebugAPI::slowPathOnExceptionUnwind(JSContext*, js::AbstractFramePtr) ()
#5 0x000055555777f09b in js::wasm::HandleThrow(JSContext*, js::wasm::WasmFrameIter&, js::jit::ResumeFromException*) ()
#6 0x000055555777fb23 in WasmHandleThrow(js::jit::ResumeFromException*) ()
#7 0x000026372362a49e in ?? ()
[...]
#11 0x0000000000000000 in ?? ()
rax 0x7fffffffc080 140737488339072
rbx 0x7fffffffb6b0 140737488336560
rcx 0x0 0
rdx 0x2637235fa289 42018258526857
rsi 0xccccccff6b1efbe9 -3689348597337359383
rdi 0x7fffffffb6b0 140737488336560
rbp 0x7fffffffb6a0 140737488336544
rsp 0x7fffffffb680 140737488336512
r8 0x2637235fa289 42018258526857
r9 0x0 0
r10 0x1 1
r11 0x55555669b720 93825010349856
r12 0x0 0
r13 0x10b 267
r14 0x7fffffffb920 140737488337184
r15 0x7fffffffb768 140737488336744
rip 0x555556e3f2ce <js::jit::JSJitFrameIter::machineState() const+94>
=> 0x555556e3f2ce <_ZNK2js3jit14JSJitFrameIter12machineStateEv+94>: mov 0x30(%rsi),%rdi
0x555556e3f2d2 <_ZNK2js3jit14JSJitFrameIter12machineStateEv+98>: mov (%rdi),%rsi
Marking s-s due to crash address but likely sec-moderate at most due to debugger involvement.
|
|
1,849,723 | 16,547,594 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:109.0) Gecko/20100101 Firefox/117.0
Steps to reproduce:
- Apply the following theme with a sidebar_text color defined: https://addons.mozilla.org/en-CA/firefox/addon/dim-a-simple-dark-theme/
- Show the Bookmarks sidebar (Ctrl/Cmd + B)
Bookmarks should appear with light grey text, as specified in the theme’s sidebar_text value.
- Click the dropdown “Bookmarks” header
- Select “Synced tabs”
Synced tab titles appear black.
Actual results:
The Synced Tabs are extremely difficult to read, as they appear as black text on a near black background.
Expected results:
The text should adhere to the sidebar_text color shown in the Bookmarks and History tabs (#c8c2c5). As noted in https://bugzilla.mozilla.org/show_bug.cgi?id=1418602#c55, the sidebar* attributes were meant to apply to Synced Tabs as well.
|
|
1,849,876 | 16,548,654 |
Created attachment 9350008
plop.pdf
Steps to reproduce the problem:
1. Add some text in one of the fields
What is the expected behavior? (add screenshot)
Only the filed field must have a value
What went wrong? (add screenshot)
The other field is field with the same value.
It's a regression from bug 1839774 and especially from bug 1838855.
|
|
1,850,127 | 16,550,422 |
**[Affected versions]:**
- Firefox Release 116
- Firefox Beta 117
**[Affected Platforms]:**
- Windows 10x64
- macOS Ventura 13.5
- Ubuntu 22.04 x64
**[Prerequisites]:**
- Have an experiment that sets the feature “off” on the control branch and “on” on the treatment branch, and a rollout that sets the feature on for the opaqueResponseBlocking feature config.
- Have a user.js file for the experiment to be able to enroll in the control branch.
**[Steps to reproduce]:**
1. Use the [user.js](https://drive.google.com/drive/folders/1MW_44FaYGB-6i53NMe5yJyUAdSA7U5Z-) file to enroll in the Treatment A branch of the experiment and restart the browser.
2. Navigate to “about:telemetry#events-tab” and observe the events.
3. Navigate to “about:studies” and remove the experiment.
4. Navigate to “about:telemetry#events-tab” and observe the events.
**[Expected result]:**
- Step 2:
- An enroll event is registered for the experiment.
- An enroll event is registered for the rollout.
- Step 3: The experiment is removed and the user is still enrolled in the rollout.
**[Actual result]:**
- Step 2:
- An enroll event is registered for the experiment.
- An enroll event is registered for the rollout.
- Step 3: The user is unenrolled from both the experiment and the rollout.
**[Notes]:**
- The issue might be reproducible on Firefox Nightly also, if the rollout sets a value different from the default value for the "browser.opaqueResponseBlocking" pref.
|
|
1,850,954 | 16,557,514 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/117.0
Steps to reproduce:
//Javascript to change the protocol of an URL instance
const url = new URL("https://www.anypage.com?query=something")
url.protocol = "nlp:"
console.log(url.href)
Actual results:
//protocol of the URL object stays on "http" or "https" and does not change.
url.href equals "https://www.anypage.com?query=something"
Expected results:
//this worked before upgrading to Firefox version 117
//protocol of the URL object changes to "nlp:"
url.href equals "nlp://www.anypage.com?query=something"
|
|
1,850,993 | 16,557,804 |
When I watch YouTube videos in full screen mode, I often notice that the link status bar with a URL is shown in the bottom left corner, overlapping the video. It's annoyingly persistent; moving the mouse around over the video in full screen mode doesn't make it go away.
I've finally found out how this happens: It's because, after I click the video on the home page, I don't hover any other link before I enter full screen mode. And clicking the video link makes the link disappear without a navigation and in such a way that we don't notice that the mouse has left the link.
Steps to reproduce:
1. Go to https://www.youtube.com/
2. Optionally: Visit a random video, change the video mode to "Theater mode" (button next to the fullscreen button), and go back to the home page.
3. On the home page, move your mouse over a random video (link status bar appears) and click it.
4. On the video page, don't move your mouse over any other links.
5. Optionally: Double-click the video that's now playing to make it fullscreen.
Expected results:
After clicking the video on the home page, the link URL status bar should disappear.
Actual results:
The status bar remains visible. It is stuck until you hover another link, for example a suggested video or a user name in a comment. Moving your mouse over the video controls does not cause the status bar to hide.
It's easier to get into this state with theater mode because the right column of recommended videos is less likely to be under your mouse immediately after navigating to a video.
|
|
1,851,135 | 16,558,627 |
Steps to reproduce:
SpiderMonkey version:
commit: 8f09f6a6ff0d8b6ea75a1e1279a06ba02479578c
Test case
```js
function foo(a1, a2, a3) {
let x = a1.hasOwnProperty(foo);
function bar(a15) {
const ret = `
x = foo;
`;
return ret;
}
a3.toString = bar;
for (let i = 0; i < 100; i++) {
for (let j = 0; j < 100; j++) {
this.isValidJSON(String.fromCharCode());
}
}
class Baz {}
const s = a3.toString(a2);
const p = new Baz();
function Bling() {}
const b = new Bling();
b.sameZoneAs = b;
const t = this.newGlobal(b).Debugger;
t(p).getNewestFrame().eval(s);
return bar;
}
foo(this, foo, foo);
```
Exec:
```
./dist/bin/js --baseline-warmup-threshold=50 --ion-warmup-threshold=200 --ion-check-range-analysis --ion-extra-checks ./test.js
```
Actual results:
Error:
```
Assertion failure: whyMagic() == why, at /media/Store/Project/js-engines/SpiderMonkey/gecko-dev-build/debug_8f09f6_20230901/dist/include/js/Value.h:868
```
|
|
1,851,517 | 16,563,030 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/116.0
Steps to reproduce:
I opened a PDF in Firefox, edited a large text box, switched to another Firefox window, and switched back to the PDF window.
Actual results:
The edit I had made to the large text box disappeared, reverting the text box to its state before I edited it.
Expected results:
The large text box should have preserved my edit and remained the active text box when I switched back to the PDF window.
|
|
1,851,678 | 16,564,347 |
This crash looks like it might be a regression starting in 116.
Crash report: https://crash-stats.mozilla.org/report/index/f94340dc-fbad-44d7-b340-a08590230902
Exception: `mozilla.appservices.autofill.AutofillApiException$SqlException`
Message: reason=Error executing SQL: database or disk is full
```
Frame Module Function Source In app?
0 mozilla.appservices.autofill.FfiConverterTypeAutofillApiError read autofill.kt:16
1 mozilla.appservices.autofill.FfiConverterTypeAutofillApiError read autofill.kt:1
2 mozilla.appservices.autofill.FfiConverter$DefaultImpls liftFromRustBuffer autofill.kt:13
3 mozilla.appservices.autofill.FfiConverterRustBuffer$DefaultImpls liftFromRustBuffer autofill.kt:6
4 mozilla.appservices.autofill.FfiConverterTypeAutofillApiError liftFromRustBuffer autofill.kt:2
5 mozilla.appservices.autofill.FfiConverterTypeAutofillApiError liftFromRustBuffer autofill.kt:1
6 mozilla.appservices.autofill.FfiConverterRustBuffer$DefaultImpls lift autofill.kt:7
7 mozilla.appservices.autofill.FfiConverterTypeAutofillApiError lift autofill.kt:3
8 mozilla.appservices.autofill.FfiConverterTypeAutofillApiError lift autofill.kt:2
9 mozilla.appservices.autofill.AutofillApiException$ErrorHandler lift autofill.kt:2
10 mozilla.appservices.autofill.AutofillApiException$ErrorHandler lift autofill.kt:1
11 mozilla.appservices.autofill.AutofillKt checkCallStatus autofill.kt:75
12 mozilla.appservices.autofill.AutofillKt access$checkCallStatus autofill.kt:1
13 mozilla.appservices.autofill.Store <init> autofill.kt:5
14 mozilla.components.service.sync.autofill.AutofillCreditCardsAddressesStorage$conn$2 invoke AutofillCreditCardsAddressesStorage.kt:30
15 kotlin.SynchronizedLazyImpl getValue LazyJVM.kt:21
16 mozilla.components.service.sync.autofill.AutofillCreditCardsAddressesStorage getConn$service_sync_autofill_release AutofillCreditCardsAddressesStorage.kt:3
17 mozilla.components.service.sync.autofill.AutofillCreditCardsAddressesStorage$getAllAddresses$2 invokeSuspend AutofillCreditCardsAddressesStorage.kt:6
18 mozilla.components.service.sync.autofill.AutofillCreditCardsAddressesStorage$getAllAddresses$2 invoke AutofillCreditCardsAddressesStorage.kt:13
19 kotlinx.coroutines.intrinsics.UndispatchedKt startUndispatchedOrReturn Undispatched.kt:5
20 kotlinx.coroutines.BuildersKt withContext :76
21 mozilla.components.service.sync.autofill.AutofillCreditCardsAddressesStorage getAllAddresses AutofillCreditCardsAddressesStorage.kt:11
22 org.mozilla.fenix.FenixApplication$setStartupMetrics$4 invokeSuspend FenixApplication.kt:89
23 kotlin.coroutines.jvm.internal.BaseContinuationImpl resumeWith ContinuationImpl.kt:9
24 kotlinx.coroutines.DispatchedTask run DispatchedTask.kt:112
25 kotlinx.coroutines.internal.LimitedDispatcher$Worker run LimitedDispatcher.kt:4
26 kotlinx.coroutines.scheduling.TaskImpl run Tasks.kt:3
27 kotlinx.coroutines.scheduling.CoroutineScheduler$Worker run CoroutineScheduler.kt:96
```
|
|
1,865,154 | 16,670,672 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:121.0) Gecko/20100101 Firefox/121.0
Steps to reproduce:
1. Make sure "Open previous windows and tabs" is disabled in settings, so Firefox shouldn't automatically restore previous sessions.
2. Open a few normal (website) tabs. Additionally, open the Customize Firefox page.
3. Without closing tabs separately, close Firefox.
4. Reopen Firefox.
Actual results:
Despite that "Open previous windows and tabs" is disabled in settings, Firefox still reopens all tabs (including the Customize Firefox page) from the previous session.
In the browser toolbox, an error can be seen:
```
SessionStore: The session file is invalid: TypeError: can't access property "title", tabState.entries[activeIndex] is undefined SessionStore.sys.mjs:1246:19
initSession resource:///modules/sessionstore/SessionStore.sys.mjs:1246
onBeforeBrowserWindowShown resource:///modules/sessionstore/SessionStore.sys.mjs:2204
```
The error happens because the session store [tries to access](https://searchfox.org/mozilla-central/source/browser/components/sessionstore/SessionStore.sys.mjs#6522-6529) the title/URL of the latest entry for that tab. However, a tab with the Customize Firefox page open does not have any `entries` (as can be seen in the below snippet from the session file), so the code fails.
```json
[
{
"entries": [
{
"url": "https://example.com/",
"title": "Example Domain",
"cacheKey": 0,
"ID": 4,
"docshellUUID": "{41171a80-cc58-4f63-9953-e6de426b45f1}",
"resultPrincipalURI": null,
"hasUserInteraction": false,
"triggeringPrincipal_base64": "{\"3\":{}}",
"docIdentifier": 5,
"persist": true
}
],
"lastAccessed": 1699821941611,
"hidden": false,
"searchMode": null,
"userContextId": 0,
"attributes": {},
"image": "...",
"index": 1,
"requestedIndex": 0
},
{
"entries": [],
"lastAccessed": 1699821947072,
"hidden": false,
"searchMode": null,
"userContextId": 0,
"attributes": {
"customizemode": "true"
},
"image": "chrome://browser/skin/customize.svg",
"index": null
}
]
```
Expected results:
Firefox should not restore the previous session if that is not enabled, and the error shouldn't happen.
Specifically, I think this could be fixed by checking if `tabState.entries` has any elements before accessing them. This is also done [elsewhere](https://searchfox.org/mozilla-central/source/browser/components/sessionstore/SessionStore.sys.mjs#1282-1292) in the same file, but this check wasn't applied to everywhere it's needed for some reason.
|
|
1,865,341 | 16,671,618 |
Created attachment 9364149
PDF_characters_lose.png
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/119.0
Steps to reproduce:
1. Open PDF file that was saved to disk: https://brzozow.sr.gov.pl/container/srbrzozow/druki-i-wzory/zal-nr-03-wniosek-o-zezwolenie-na-samochod.pdf
2. Add some Texts with Polish characters.
3. Save edited PDF file.
4. Edit the edited PDF file again and add some texts with Polish characters.
5. Save the file again.
Actual results:
PDF file edited twice lost Polish characters.
Polish characters were replaced with different characters.
Expected results:
Edited PDF file contents should look exactly as they looked before saving to file.
Polish characters shouldn't be changed.
|
|
1,865,421 | 16,672,085 |
CPU time measurements are instrumented inside FOG IPC and are GIFFT-ed to Legacy Telemetry probes. One of those is `power.cpu_time_per_process_type_ms` which is a keyed scalar mirrored to by a `labeled_counter` of the same name.
The Glean metric is fine, but it seems as though something happened in Firefox 114 that made the parent-process-specific scalar go empty.
Now, there are some differences between Legacy Telemetry and FOG that could account for some differences:
1) Glean's data is process-agnostic. The Glean `power.cpu_time_per_process_type_ms` will be all processes' data all in one. The unexpectedly-empty legacy telemetry collection is the `payload.processes.parent` variety.
- Really, though, in most cases of humans using Firefox there should be something in there.
2) Glean can record later in shutdown than Legacy, and for cpu timings tied to IPC which can be tied to process lifecycles ending during shutdown... there could be an effect.
- See above refutation. Also: why did this work until Fx114?
This bug is in need of a regression range. But in taking a look for likely FOG changes in that timescale, bug 1672273 stands out as it both
a) Changes the CPU instrumentation code
- (though only the per-thread instrumentation, not the per-process-type ones)
b) Changes how labeled metrics are done
- (though the GIFFT mirroring was kept exactly the same)
Also, `test_GIFFT.js` and `test_GIFFTIPC.js` both have labeled counters that work. Is this C++-specific? How?
This bug needs to be investigated to determine the mechanism of reproduction and how it slipped the net of our test coverage. Then it needs to be fixed and validated. And then, if test coverage confidence isn't restored, we may wish to introduce monitoring.
|
|
1,866,014 | 16,678,267 |
7 crashes on macOS from 2+ installations of Firefox 122.0a1 20231121183118
Crash report: https://crash-stats.mozilla.org/report/index/2f0f39ec-eb1f-452e-ab5c-34fee0231122
MOZ_CRASH Reason: ```assertion failed: input_channel_count >= input_channels_to_ignore + output_channel_count```
Top 10 frames of crashing thread:
```
0 XUL MOZ_Crash mfbt/Assertions.h:281
0 XUL RustMozCrash mozglue/static/rust/wrappers.cpp:18
1 XUL mozglue_static::panic_hook mozglue/static/rust/lib.rs:96
2 XUL core::ops::function::Fn::call library/core/src/ops/function.rs:79
3 XUL <alloc::boxed::Box<F, A> as core::ops::function::Fn<Args>>::call library/alloc/src/boxed.rs:2021
3 XUL std::panicking::rust_panic_with_hook library/std/src/panicking.rs:735
4 XUL std::panicking::begin_panic_handler::{{closure}} library/std/src/panicking.rs:601
5 XUL std::sys_common::backtrace::__rust_end_short_backtrace library/std/src/sys_common/backtrace.rs:170
6 XUL rust_begin_unwind library/std/src/panicking.rs:597
7 XUL core::panicking::panic_fmt library/core/src/panicking.rs:72
```
|
|
1,866,035 | 16,678,463 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:120.0) Gecko/20100101 Firefox/120.0
Steps to reproduce:
- have a client/personal certificate in the Windows certificate store
- create new profile
- open it
- open a private browsing window
- open a web site, for example www.ijs.si
- open a new tab
- in the new tab type into the URL bar an address of a web site that uses client certificates, for example https://netstik.sparkasse.si/
- select a certifiicate in the dialog (maybe not shown if the user only has one certificate to choose from)
Actual results:
From now on, whenever there is network activity in the tab that uses a client certificate, firefox becomes very slow. Both the private window and the regular window. For example switching tabs with ctrl+tab takes from one to five seconds.
Whenever that happens, about:processes shows big CPU value for the item "Firefox (19784)", something between 60 and 95%. Windows Task manager does not show a lot, but the Task manager in Windows 11 is very weird, so that does not mean much.
Expected results:
Not slowdown.
I think this started happening today, after updating to version 120.
|
|
1,866,069 | 16,678,980 |
**[Affected versions]:**
- Firefox Beta 121.0b1 - Build ID: 20231120154937
- Firefox Nightly 122.0a1 - Build ID: 20231122095244
**[Affected Platforms]:**
- macOS 13.1
- Ubuntu 22.04 x64
**[Prerequisites]:**
- Have a Screen Reader software installed and enabled (I have used VoiceOver on macOS and Orca on Ubuntu 22.04).
- Have a Firefox profile with the following prefs in the "about:config" page:
- `browser.urlbar.quicksuggest.contextualOptIn` set to `true`
- `browser.search.region` set to `US`
**[Steps to reproduce]:**
1. Open the browser using the profile from the prerequisites.
2. Open a New Tab and click the "Address Bar".
3. Navigate through the "Address Bar" dropdown and observe the behavior.
**[Expected result]:**
- All the elements from the "Opt-In" section are successfully recognized by the screen reader software.
**[Actual result]:**
- Only the buttons and the "Learn more" hyperlink from the "Opt-In" section are recognized by the screen reader software.
**[Notes]:**
- This issue is reproducible regardless if the "Opt-In" section is displayed at the top or at the bottom of the "Address Bar" doorhanger.
- This issue is not reproducible on Windows 10 x64 with the built-in screen reader (Narrator). All the elements are correctly recognized by the screen reader if the "Opt-In" section is displayed at the top of the “Address Bar” doorhanger. However, the "Opt-In" section cannot be accessed through the screen reader software if the section is displayed in the bottom part of the doorhanger.
|
|
1,866,076 | 16,679,076 |
Created attachment 9364989
optin section is not dismissed.mp4
**[Affected versions]:**
- Firefox Beta 121.0b1 - Build ID: 20231120154937
- Firefox Nightly 122.0a1 - Build ID: 20231121183118
**[Affected Platforms]:**
- Windows 10 x64
- Ubuntu 22.04 x64
- macOS 13.6
**[Prerequisites]:**
- Have a Firefox 121 or 122 en* build installed.
- Have the following prefs in the "about:config" page:
- browser.search.region set to US;
**[Steps to reproduce]:**
1. Open the browser with the profile from the prerequisites.
2. Navigate to about:config page.
3. Set the "browser.urlbar.quicksuggest.contextualOptIn" pref to true and "browser.urlbar.quicksuggest.contextualOptIn.topPosition" pref to false.
4. Restart the browser.
5. Click on the address bar.
6. Click on the "Allow suggestions" or "Not Now" button.
7. Observe the behavior.
**[Expected result]:**
- The Opt-In section is dismissed no errors are displayed in the browser console.
**[Actual result]:**
- The Opt-In section is NOT dismissed and an error is displayed in the browser console.
**[Notes]:**
- The Opt-In section disappears if the zero-prefix Address Bar is dismissed and re-displayed. Also, the buttons seem to work as expected even though the message remains displayed.
- The issue is only reproducible for the first impression of the Opt-In section.
- This issue is reproducible only when the Opt-In section is displayed in the bottom part. Is not reproducible with the Opt-In section when is displayed at the top.
- The following error is displayed in the browser console: "TypeError: queryContext is undefined UrlbarProviderQuickSuggestContextualOptIn.sys.mjs:94:7"
- Attached is a screen recording of the issue.
|
|
1,866,292 | 16,680,614 |
**[Notes]:**
- This issue is reproducible with all types of Glean events: impression, engagement, and abandonment.
- This issue is reproducible when clicking any of the “Learn more” link and the “Allow suggestions” or “Not now” buttons from the Opt-In section.
**[Affected versions]:**
- Firefox Beta 121.0b2 - Build ID: 20231122091904
- Firefox Nightly 122.0a1 - Build ID: 20231123095534
**[Affected Platforms]:**
- Windows 10 x64
- Ubuntu 22.04 x64
- macOS 13.1
**[Prerequisites]:**
- Have an en* Firefox build installed.
- Have the following prefs in the "about:config" page:
- `browser.search.region` set to `US`;
- `browser.urlbar.quicksuggest.contextualOptIn` set to `true`;
- `browser.urlbar.quicksuggest.contextualOptIn.topPosition` set to `true`;
- `devtools.chrome.enabled` set to `true`;
**[Steps to reproduce]:**
1. Open the browser with the profile from the prerequisites.
2. Navigate to the “about:glean” page and add a tag name in the first field from "About Testing".
3. Select “events” from the dropdown and click the “Apply settings and submit ping” button.
4. Run the following snippet in the Browser Console:
`Services.fog.setMetricsFeatureConfig(JSON.stringify({
"urlbar.abandonment": true,
"urlbar.engagement": true,
"urlbar.impression": true,
})
);`
5. Open a new tab and click inside the Address Bar.
6. Click the “Allow suggestions” button from the Opt-In section.
7. Go back to the “about:glean” page and click again the “Apply settings and submit ping” button.
8. Open the last event from https://debug-ping-preview.firebaseapp.com/pings/<tag from step 2>.
9. Observe the engagement event.
**[Expected result]:**
- The event contains specific values for each parameter.
**[Actual result]:**
- The “results” and “selected_result” parameters have “unknown” values.
**[Additional notes]:**
- This issue is reproducible only when the Opt-In section is on Top.
- This issue is not reproducible when the Opt-In section is on the Bottom since the section is outside the results area and the event doesn't register the same way:
- the engagement event is not registered at all and the impression and abandonment ones don't contain anything related to the Opt-In section nor any "unknown" values (event [example](https://debug-ping-preview.firebaseapp.com/pings/CF-1-DESKTOP/b197c8f5-1dd2-4f8e-882f-16d9c5801700)).
- Attached is a [screen recording](https://drive.google.com/file/d/1GUZxiSGyBvH0iZfT1Wnf-89-xVMJIQq6/view?usp=sharing) of the issue.
|
|
1,866,479 | 16,681,646 |
Perfherder has detected a build_metrics performance regression from push [238d07f2df955a10d01dd4e6197356c768bf6ba0](https://hg.mozilla.org/integration/autoland/pushloghtml?changeset=238d07f2df955a10d01dd4e6197356c768bf6ba0). As author of one of the patches included in that push, we need your help to address this regression.
### Regressions:
| **Ratio** | **Test** | **Platform** | **Options** | **Absolute values (old vs new)** |
|--|--|--|--|--|
| [0.38%](https://treeherder.mozilla.org/perfherder/graphs?timerange=1209600&series=autoland,2614662,1,2) | installer size | osx-cross | | 88,738,826.67 -> 89,073,890.08 |
Details of the alert can be found in the [alert summary](https://treeherder.mozilla.org/perfherder/alerts?id=40324), including links to graphs and comparisons for each of the affected tests. Please follow our [guide to handling regression bugs](https://wiki.mozilla.org/TestEngineering/Performance/Handling_regression_bugs) and **let us know your plans within 3 business days, or the patch(es) may be backed out** in accordance with our [regression policy](https://www.mozilla.org/en-US/about/governance/policies/regressions/).
For more information on performance sheriffing please see our [FAQ](https://wiki.mozilla.org/TestEngineering/Performance/FAQ).
|
|
1,866,501 | 16,681,788 |
Created attachment 9365366
2023-11-24_16h33_49.png
**Found in**
* Nightly 122.0a1 (2023-11-24)
**Affected versions**
* Nightly 122.0a1 (2023-11-24)
* Beta 121.0b2
**Affected platforms**
* ALL
**Preconditions:**
browser.shopping.experience2023.enabled - true
browser.shopping.experience2023.optedIn = 0
Reach enough product details pages to have the Survey card displayed when reaching a product details page.
**Steps to reproduce**
1. Reach https://www.amazon.com/dp/B08GF9K41Z/ref=sbl_dpx_office-desks_B0BP743MBC_0
2. Resize the Review checker sidebar to a smaller size.
**Expected result**
* The X Close button and text from the Survey card should be properly displayed.
**Actual result**
* The X Close button from the Survey card will disappear and some of the text is truncated.
**Regression range**
Not Applicable
|
|
1,866,620 | 16,682,410 |
Currently the narrow layout for `browser.urlbar.quicksuggest.contextualOptIn.topPosition=false` looks way worse than for `browser.urlbar.quicksuggest.contextualOptIn.topPosition=true`. We should ideally fix this for the experiment to prevent potentially skewing results by making `topPosition=false` less usable.
|
|
1,866,772 | 16,685,437 |
Created attachment 9365598
problematic example History <>.db
Steps to reproduce:
Tried to import Safari browsing history using the built-in import tool.
Actual results:
Importing Safari history on macos which is located at ~/Library/Safari/History.db silently hangs forever in case some problematic entries are present inside Safari's "history_items" table. Specifically, those items are slightly unusual, and below is a list of example entries that cause a problem.
http://-z((((((((((())))})sszz/
http://<country-code>.archive.ubuntu.com/ubuntu/
http://The Summarit performance is of a very high level and certainly equal to that of the Summicron.
http://q3_us_knob_v1.5.bin.zip/
http://www.195.19.96.179/
Removing those lines from Safari database manually allows the import to finish successfully. I could only find problems with http:// entries and not https://
A sample problematic History.db file is attached. It has just one problematic line from the list above.
Expected results:
Firefox definitely mustn't just stall on import. Problematic lines should either be quietly discarded or a comprehensive error report outlining the necessary steps to remedy the problem should be presented to the user.
|
|
1,866,799 | 16,685,556 |
Split off from https://bugzilla.mozilla.org/show_bug.cgi?id=1865792.
`NoExistingAuthFlow` is also a crasher: https://crash-stats.mozilla.org/report/index/9926eab7-85d2-462f-a279-46fcd0231121
|
|
1,866,811 | 16,685,648 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:120.0) Gecko/20100101 Firefox/120.0
Steps to reproduce:
run Date.parse("0000-00-00") in the Javascript-Console
Actual results:
943916400000
Expected results:
NaN
|
|
1,866,839 | 16,685,927 |
The following testcase crashes on mozilla-central revision 20231127-edf9292c0bc9 (debug build, run with --fuzzing-safe --ion-offthread-compile=off --more-compartments):
b = {}
c = d = new Proxy(b, {
"deleteProperty"() {
e = newGlobal()
e.parent = c
e.eval("var dbg = new Debugger(parent); dbg.onEnterFrame = function() {}")}
})
delete d.footrue;
function h(i) {
binary = wasmTextToBinary(i)
try {
j = new WebAssembly.Module(binary)
} catch {}
return new WebAssembly.Instance(j)
}
ins = h(`
(func $f (param i64 i64)
)
(func $g (param i64)
i64.const 6
i64.const 3
return_call $f
)
(func (export "run") (param i64)
local.get 0
call $g
)`)
ins.exports.run(100n)
Backtrace:
received signal SIGSEGV, Segmentation fault.
0x592d458e in js::wasm::GetNearestEffectiveInstance(js::wasm::Frame const*) ()
#0 0x592d458e in js::wasm::GetNearestEffectiveInstance(js::wasm::Frame const*) ()
#1 0x592c37bf in js::wasm::DebugFrame::hasGlobal(js::GlobalObject const*) const ()
#2 0x58649ba9 in js::Debugger::removeDebuggeeGlobal(JS::GCContext*, js::GlobalObject*, mozilla::detail::HashTable<js::WeakHeapPtr<js::GlobalObject*> const, mozilla::HashSet<js::WeakHeapPtr<js::GlobalObject*>, js::StableCellHasher<js::WeakHeapPtr<js::GlobalObject*> >, js::TrackedAllocPolicy<(js::TrackingKind)1> >::SetHashPolicy, js::TrackedAllocPolicy<(js::TrackingKind)1> >::Enum*, js::Debugger::FromSweep) ()
#3 0x58648d47 in js::DebugAPI::sweepAll(JS::GCContext*) ()
#4 0x589b6dc6 in js::gc::GCRuntime::sweepDebuggerOnMainThread(JS::GCContext*) ()
#5 0x589b83f3 in js::gc::GCRuntime::beginSweepingSweepGroup(JS::GCContext*, js::SliceBudget&) ()
#6 0x589dec16 in sweepaction::SweepActionCall::run(js::gc::SweepAction::Args&) ()
#7 0x589e6868 in sweepaction::SweepActionSequence::run(js::gc::SweepAction::Args&) ()
#8 0x589de773 in sweepaction::SweepActionForEach<js::gc::SweepGroupsIter, JSRuntime*>::run(js::gc::SweepAction::Args&) ()
#9 0x589bd737 in js::gc::GCRuntime::performSweepActions(js::SliceBudget&) ()
#10 0x58900940 in js::gc::GCRuntime::incrementalSlice(js::SliceBudget&, JS::GCReason, bool) ()
#11 0x58903e1b in js::gc::GCRuntime::gcCycle(bool, js::SliceBudget const&, JS::GCReason) ()
#12 0x589054da in js::gc::GCRuntime::collect(bool, js::SliceBudget const&, JS::GCReason) ()
#13 0x588ce530 in js::gc::GCRuntime::gc(JS::GCOptions, JS::GCReason) ()
#14 0x5832385c in JSRuntime::destroyRuntime() ()
#15 0x5819fe8b in js::DestroyContext(JSContext*) ()
#16 0x57db9365 in main ()
eax 0x56702301 1450189569
ebx 0x598cd2b8 1502401208
ecx 0x598ceedc 1502408412
edx 0x0 0
esi 0xffffbf68 -16536
edi 0x593118a7 1496389799
ebp 0xffffc948 4294953288
esp 0xffffc920 4294953248
eip 0x592d458e <js::wasm::GetNearestEffectiveInstance(js::wasm::Frame const*)+350>
=> 0x592d458e <_ZN2js4wasm27GetNearestEffectiveInstanceEPKNS0_5FrameE+350>: movl $0x8b,0x0
0x592d4598 <_ZN2js4wasm27GetNearestEffectiveInstanceEPKNS0_5FrameE+360>: call 0x57e51260 <abort>
Likely not more than sec-moderate at most due to Debugger involved. This test only reproduces on 32-bit but I have seen similar signatures on 64-bit.
|
|
1,909,525 | 17,038,842 |
### Steps to reproduce
1. Open a web page.
2. Tools menu > Turn on Reader View
3. Open a different web page in a second tab.
### Expected behavior
The new web page should not open in Reader View.
### Actual behavior
The new web page opens in Reader View.
### Device information
* Firefox version: 130.0a1
* Android device model: Moto G5 (Android 8.1) and Samsung Galaxy A32 (Android 13)
|
|
1,908,596 | 17,030,526 |
Created attachment 9413443
Snackbars.png
### Steps to reproduce
1. Perform the following actions that trigger a snackbar.
2. Observe the snackbar.
a. add a bookmark -> Open a webpage, open the three-dot menu, tap the bookmark's "Add" button.
b. delete bookmark from edit view -> Bookmark a webpage, then tap the 3-dot menu button from the Navigation toolbar, and tap the "Edit" button next to the Bookmarks option. > Tap the trashcan icon in the top-right corner. > Tap on the DELETE option from the dialogue.
c. close all tabs from tabs tray -> Open several tabs and open the Tabs Tray. > Tap the ⋮ button. Select "Close all tabs" from the overlay. - only when address bar is positioned to top
d. save to collection from 3 dot menu -> Open a webpage, open the three-dot menu and tap the "Save to collection" option. > Fill in the name. > Save.
e. rename collection -> Create a new collection (or use an existing one). > Expand the collection and tap the collection menu, select Rename collection. > Rename the collection. > Tap the Ok (enter) button from the keyboard.
f. delete collection -> Create a new collection (or use an existing one). > Expand the collection and open the overflow menu, tap Delete collection.
g. add a website to shortcuts -> Open a webpage, open the three-dot menu and tap the "Add to shortcuts" option.
h. remove a website from shortcuts -> Long tap the pinned shortcut from homepage, then select "Remove". or open the three-dot menu and tap the "Remove from shortcuts" option.
i. copy URL -> Open a webpage > Long tap the URL > Tap "Copy"
j. context menu options -> Open a webpage > Long tap a link > Tap "Open link in new tab"
k. context menu options -> Open a webpage > Long tap a link > Tap "Open link in private tab"
l. context menu options -> Open a webpage > Long tap a link > Tap "Copy link"
m. context menu options -> Open a webpage > Long tap a linked image > Tap "Open image in new tab"
n. context menu options -> Open a webpage > Long tap a linked image > Tap "Open image location"
### Expected behavior
The snackbars do not overlap the navigation bar/the address bar.
### Actual behavior
Several snackbars are displayed on top of the navigation bar/the address bar.
### Device information
* Firefox version: Nightly 130.0a1 from 07/18
* Android device model: Google Pixel 8 Pro
* Android OS version: Android 14
|
|
1,910,410 | 17,046,803 |
STR
Create a rollout on the prefFlips experiment with:
```
{
"prefs": {
"extensions.logging.enabled": {
"branch": "default",
"value": 123
}
}
}
```
ER: The rollout should immediately unenroll after enrolling
AR: It stays enrolled but the pref does not change.
|
|
1,909,793 | 17,040,721 |
Steps to reproduce:
Build js
Actual results:
Build failure due to MacroAssembler::move8ZeroExtend unimplemented
Expected results:
Build successful.
|
|
1,909,206 | 17,036,671 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:128.0) Gecko/20100101 Firefox/128.0
Steps to reproduce:
This got lost when fdlibm was re-vendored, because the fix wasn't created as a patch in `modules/fdlibm/patches`
If compiling to WASI using a mozconfig like so:
```
ac_add_options --enable-project=js
ac_add_options --enable-application=js
ac_add_options --target=wasm32-unknown-wasi
ac_add_options --without-system-zlib
ac_add_options --disable-jit
ac_add_options --disable-shared-js
ac_add_options --disable-shared-memory
ac_add_options --disable-tests
ac_add_options --disable-clang-plugin
ac_add_options --enable-jitspew
ac_add_options --enable-optimize=-O3
ac_add_options --enable-portable-baseline-interp
mk_add_options AUTOCLOBBER=1
ac_add_options --host=aarch64-apple-darwin
ac_add_options --enable-debug
```
Actual results:
```
47 In file included from /Users/jakechampion/mozilla-unified/obj-wasm32-unknown-wasi/dist/system_wrappers/signal.h:3:
0:03.48 /Users/jakechampion/.mozbuild/sysroot-wasm32-wasi/include/signal.h:2:2: error: "wasm lacks signal support; to enable minimal signal emulation, compile with -D_WASI_EMULATED_SIGNAL and link with -lwasi-emulated-signal"
0:03.48 2 | #error "wasm lacks signal support; to enable minimal signal emulation, \
```
Expected results:
A successful build should have occurred
|
|
1,909,472 | 17,038,466 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:128.0) Gecko/20100101 Firefox/128.0
Steps to reproduce:
Attempt to build using wasi with intl/icu in debug mode
Actual results:
The build fails with
>undefined symbol: char const* mozilla::detail::TypeToString<unsigned long>()
Expected results:
The build should succeed
|
|
1,909,478 | 17,038,517 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:128.0) Gecko/20100101 Firefox/128.0
Steps to reproduce:
Attempt to build using wasi with intl/icu in debug mode
Actual results:
The builds fails with:
>.mozbuild/sysroot-wasm32-wasi/include/signal.h:2:2: error: "wasm lacks signal support; to enable minimal signal emulation, compile with -D_WASI_EMULATED_SIGNAL and link with -lwasi-emulated-signal"
Expected results:
The build should succeed
|
|
1,909,587 | 17,039,272 |
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:128.0) Gecko/20100101 Firefox/128.0
Steps to reproduce:
Build Firefox on MIPS64 system
Actual results:
Error:
```
In file included from Unified_cpp_js_src_jit15.cpp:29:
/var/cache/acbs/build/acbs.qgwua8gz/firefox-128.0/js/src/jit/mips64/MoveEmitter-mips64.cpp: In member function ‘virtual void js::jit::MoveEmitterMIPS64::breakCycle(const js::jit::MoveOperand&, const js::jit::MoveOperand&, js::jit::MoveOp::Type, uint32_t)’:
/var/cache/acbs/build/acbs.qgwua8gz/firefox-128.0/js/src/jit/mips64/MoveEmitter-mips64.cpp:33:19: error: could not convert ‘js::jit::ABIType::Float64’ from ‘js::jit::ABIType’ to ‘int’
33 | case ABIType::Float64:
| ~~~~~~~~~^~~~~~~
| |
| js::jit::ABIType
/var/cache/acbs/build/acbs.qgwua8gz/firefox-128.0/js/src/jit/mips64/MoveEmitter-mips64.cpp: In member function ‘virtual void js::jit::MoveEmitterMIPS64::completeCycle(const js::jit::MoveOperand&, const js::jit::MoveOperand&, js::jit::MoveOp::Type, uint32_t)’:
/var/cache/acbs/build/acbs.qgwua8gz/firefox-128.0/js/src/jit/mips64/MoveEmitter-mips64.cpp:88:19: error: could not convert ‘js::jit::ABIType::Float64’ from ‘js::jit::ABIType’ to ‘int’
88 | case ABIType::Float64:
| ~~~~~~~~~^~~~~~~
| |
| js::jit::ABIType
```
Expected results:
Build works.
|
|
1,899,516 | 16,958,130 |
Starting with Ubuntu 24.04 userns are not available by default to non packaged versions (they ship an AppArmor profile for Snap and packaged versions), and prior to bug 1884347 would (purposedly) crash processes. We might want to inform our population of users running tarball that they need to create an AppArmor profile
|
|
1,899,561 | 16,958,584 |
I narrowed this regression is caused by bug 1898705.
## Environment
- macOS 14.5 (Apple Silicon)
- Firefox Nightly https://hg.mozilla.org/mozilla-central/rev/e1da1893a0c56d2752be4e6eb1efa16c58723a55
## Steps to Reproduce
1. Install web extension that provides a sidebar.
- https://addons.mozilla.org/ja/firefox/addon/tree-style-tab/ is probably nice to reproduce.
2. Open the sidebar for the extension installed in _step 1_.
3. Narrow the sidebar width to the minimum width (e.g. 200px).
## Actual Result
- I seem webextension's sidebar has viewport width that is widely than the sidebar width.
- For example, with https://addons.mozilla.org/ja/firefox/addon/tree-style-tab/, it's closed button will not be visible in the sidebar if the sidebar is narrowed
## Expected Result
- webextension's sidebar should have the same viewport width that is same with the containing sidebar width.
|
|
1,899,804 | 16,960,333 |
User Agent: Mozilla/5.0 (Android 14; Mobile; rv:128.0) Gecko/128.0 Firefox/128.0
Steps to reproduce:
Open this pdf:
https://www.vodafone.de/media/downloads/pdf/2060-sepa-basis-lastschrift-mandat-festnetz.pdf
Make a edit in a field of the pdf
Leave the field
The barcode on the left turns into rubbish
Actual results:
The barcode on the left turns into rubbish
Expected results:
Only the edited field should have changed
|
|
1,899,953 | 16,961,242 |
Steps to reproduce:
1. Download a file
2. Cancel download via system notification
Actual results:
File will not deleted when download is cancelled.
Expected results:
Fenix should auto delete it.
|
|
1,899,997 | 16,961,493 |
Created attachment 9404976
az_recorder_20240531_130931.mp4
### Steps to reproduce
1. Launch Fenix.
2. While on homescreen page, tap the Recents button to open the list of recently used apps.
3. Tap on Fenix icon, and then select "Open in pop-up view" option.
### Expected behavior
There are no layout issues.
### Actual behavior
There are a "+" button and a search bar in the first phase, and after taping on the "+" button there is expanded a second search bar. Only the expended search bar si active.
### Device information
* Firefox version: Firefox Nightly 128 (2024-05-31)
* Android device model: Samsung Galaxy Tab S8
* Android OS version: Android 12
### Any additional information?
This issue is a regression: Last good build [2024-05-13] - First bad build [2024-05-14]
|
|
1,900,029 | 16,961,759 |
Created attachment 9405000
the option does not block the message.mp4
**[Affected versions]:**
- Firefox Nightly 128.0a1 - Build ID: 20240529214854
- Firefox Beta 127.0b8 - Build ID: 20240529091551
- Firefox Release 126.0.1 - Build ID: 20240526221752
**[Affected Platforms]:**
- Windows 10
- macOS 13.4
- Ubuntu 22.04
**[Prerequisites]:**
- Have following prefs set:
- browser.search.region = US
- nimbus.debug = true
- browser.newtabpage.activity-stream.asrouter.devtoolsEnabled = true
**[Steps to reproduce]:**
1. Open the browser using the profile from the prerequisites.
2. Enroll in any of the treatment branches of the experiment (eg: [treatment-a](about:studies?optin_slug=fx-view-discoverability-early-day-users&optin_branch=treatment-a&optin_collection=nimbus-preview)).
3. Trigger the Firefox View callout message using asrouter or naturally.
4. Click on the additional dropdown button and select the "Don't show this recommendation again" option.
5. Navigate to about:asrouter#devtools and search for the "FX_VIEW_DISCOVERABILITY_EARLY_DAY_USERS" message.
6. Observe the message.
**[Expected result]:**
- The message is blocked.
**[Actual result]:**
- The message is NOT blocked.
**[ Notes]:**
- Attached is a screen recording of the issue.
|
|
1,900,526 | 16,966,899 |
Created attachment 9405438
new_pages_le_max_memory.js
Steps to reproduce:
Steps to reproduce:
Checkout commit d9496bfef09039b2642da45585ca821c36917c6d and invoke the js shell as follows:
```
./js-spidermonkey-shell --fast-warmup --fuzzing-safe new_pages_le_max_memory.js
```
Actual results:
```
Assertion failure: newPages <= wasm::MaxMemoryPages(t) && newPages.byteLength() < ArrayBufferObject::ByteLengthLimit, at vm/ArrayBufferObject.cpp:1678
```
|
|
1,900,740 | 16,968,338 |
Created attachment 9405634
type2state.js
Steps to reproduce:
Checkout commit d9496bfef09039b2642da45585ca821c36917c6d and invoke the js shell as follows:
```
./js-spidermonkey-shell --fast-warmup --fuzzing-safe frameptr_has_cached_saved_frame.js
```
Actual results:
```
Assertion failure: state == Type2State<T>::result, at dist/include/mozilla/MaybeOneOf.h:60
```
|
|
1,900,947 | 16,969,845 |
Created attachment 9405822
screenshot
STR:
1. Make sure that reader.improved_text_menu.enabled is set to true
2. Open a page where you can use the reader mode
3. Enable the reader mode
4. Press the button for text and layout settings
Actual:
The order of buttons for the font size is wrong - the button for increasing the font size is left, the button for decreasing the font size is right. This is unusual and different to all other controls of the reader mode settings.
Expected:
The button for decreasing the font size is left, the button for increasing the font size is right.
Additional question: Is there a reason why the font size controls are buttons, while all other controls that control numbers in the new settings are sliders? Shouldn't that be consistent?
needinfo? Irene Ni as you implemented the font size controls in the new settings.
|
|
1,901,298 | 16,972,150 |
With a LWT installed there is no focus ring shown when a tab is focused.
STR
1. Install a LWT (Light/Dark/Alpenglow all reproduce)
2. Press F6, Shift+Tab to the New Tab button
3. Press Shift+Tab again
4. Press Left/Right to change tabs
Expected results: The tab has a focus ring
Actual results: No focus ring is shown
|
|
1,901,410 | 16,972,739 |
Created attachment 9406243
bug.js
Steps to reproduce:
This bug is flaky and must be executed multiple times in order to trigger the assert.
Checkout commit 15778b8c32f8535624fff2af36fc669e65a9af3 and invoke the js shell as follows:
```
/root/js-spidermonkey-shell --fast-warmup --fuzzing-safe --gc-zeal=22,98 <testcase>
```
Actual results:
```
Assertion failure: prop.isNativeProperty() && prop.propertyInfo().isDataProperty() && prop.propertyInfo().writable(), at /root/src/js/src/jit/VMFunctions.cpp:2058
```
|
|
1,901,413 | 16,972,742 |
Created attachment 9406246
bug.js
Steps to reproduce:
Checkout commit 15778b8c32f8535624fff2af36fc669e65a9af3 and invoke the js shell as follows:
```
/root/js-spidermonkey-shell --fuzzing-safe <testcase>
```
Actual results:
```
Hit MOZ_CRASH(invalid UTF-8 string: ReportInvalidCharacter) at /root/src/js/src/vm/CharacterEncoding.cpp:322
```
|
|
1,901,499 | 16,974,893 |
Created attachment 9406338
Extensions success link not worning.mp4
**[Affected versions]:**
- Firefox Nightly 128.0a1 (Build ID: )
**[Affected Platforms]:**
- Windows 10
- Windows 11
**[Prerequisites]:**
- Have a non Firefox browser installed and make sure that you have at least one compatible add-on installed.
- Have the following prefs in the "about:config" page:
- nimbus.debug set to true
- browser.newtabpage.activity-stream.asrouter.devtoolsEnabled set to true
**[Steps to reproduce]:**
1. Open the browser using the profile from the prerequisites.
2. Navigate to "about:asrouter".
3. In the "Messages" section copy any of the messages from the treatment branches of [this experiment](https://experimenter.services.mozilla.com/nimbus/embedded-import-spotlight-existing-users/summary).
4. Click the "Modify" button.
5. Click the "Import to Firefox" button.
6. Click the "x of x extensions" link.
7. Observe the behavior.
**[Expected result]:**
- The "about:addons" page is opened in a new tab in the background.
**[Actual result]:**
- Nothing happens.
**[ Notes]:**
- The issue is NOT reproducible if the Import modal is triggered from "about:preferences#general" or "about:logins" page.
- Here is the error displayed in the browser console when the link is clicked:
- `TypeError: can't access property "ownerGlobal", browser is null MigrationWizardParent.sys.mjs:845:18`
- Attached is a screen recording of the issue.
|
|
1,901,664 | 16,976,055 |
Created attachment 9406543
testcase
Testcase is attached.
```
2058 MOZ_ASSERT(v.isSymbol() || v.isBigInt());
(gdb) bt
#0 js::ToNumberSlow (cx=0x7ffff6635100, v_=v_@entry=..., out=out@entry=0x7fffffffb0c0) at /home/ubumain/trees/mozilla-central/js/src/jsnum.cpp:2058
#1 0x000055555795b4cd in js::ToNumber (cx=0x7ffff7a1ca60 <_IO_stdfile_2_lock>, vp=...) at /home/ubumain/trees/mozilla-central/js/src/jsnum.h:244
#2 js::ToNumericSlow (cx=0x7ffff7a1ca60 <_IO_stdfile_2_lock>, vp=vp@entry=...) at /home/ubumain/trees/mozilla-central/js/src/jsnum.cpp:2084
#3 0x00005555572b5458 in js::ToNumeric (cx=0x7ffff6635100, vp=...) at /home/ubumain/trees/mozilla-central/js/src/jsnum.h:260
#4 js::PowOperation (cx=cx@entry=0x7ffff6635100, lhs=..., rhs=..., res=...) at /home/ubumain/trees/mozilla-central/js/src/vm/Interpreter-inl.h:801
#5 0x00005555572b8eaa in js::PowValues (cx=0x7ffff7a1ca60 <_IO_stdfile_2_lock>, cx@entry=0x7ffff6635100, lhs=..., rhs=..., res=..., res@entry=...) at /home/ubumain/trees/mozilla-central/js/src/vm/Interpreter.cpp:4807
#6 0x0000555557e92b8c in js::jit::DoBinaryArithFallback (cx=0x7ffff6635100, frame=0x7fffffffb458, stub=0x7ffff6532ee8, lhs=..., rhs=..., ret=...) at /home/ubumain/trees/mozilla-central/js/src/jit/BaselineIC.cpp:2295
#7 0x00000a1c8a24072f in ?? ()
#8 0xfff9800000000000 in ?? ()
#9 0x0000000000000065 in ?? ()
#10 0x00007fffffffb4a0 in ?? ()
#11 0x00000a1c8a268f3f in ?? ()
#12 0x0000000000000001 in ?? ()
#13 0x00007fffffffb458 in ?? ()
#14 0x00007ffff6532ee8 in ?? ()
#15 0xfff8800000000000 in ?? ()
#16 0xfffa800000000009 in ?? ()
#17 0xfffa800000000009 in ?? ()
#18 0xfff8800000000000 in ?? ()
#19 0xfff9800000000000 in ?? ()
#20 0xfff9800000000000 in ?? ()
#21 0x00000bddd8765510 in ?? ()
#22 0x00007ffff6644975 in ?? ()
#23 0x00007ffff6532ea8 in ?? ()
#24 0x00000bddd873f088 in ?? ()
#25 0x00007ffff6532de0 in ?? ()
#26 0x0000000000000000 in ?? ()
(gdb)
```
```
The first bad revision is:
changeset: https://hg.mozilla.org/mozilla-central/rev/7dc6122ebd68
user: Denis Palmeiro
date: Wed Nov 22 01:53:40 2023 +0000
summary: Bug 1847258: Use the warmup counter when the last IC is attached as the Ion hint threshold, and adjust inlining heuristics when the hint is used. r=iain
```
Run with `--fuzzing-safe --no-threads --baseline-eager`, compile with `AR=ar sh ../configure --enable-debug --enable-debug-symbols --with-ccache --enable-nspr-build --enable-ctypes --enable-gczeal --enable-rust-simd --disable-tests`, tested on m-c rev 9e49a1b86e40.
I did a quick search of previous similar assertion failures and most (if not all) were s-s, so setting this one as well.
Jan, do you mind taking a look as a start? Or perhaps you might wish to forward to Denis/Iain as necessary.
|
|
1,901,954 | 16,977,951 |
Created attachment 9406919
maintenanceservice-2.log
**Found in**
* when testing updates for Firefox 128.0b1
**Affected versions**
* beta versions prior to Firefox 127.0b8
**Tested platforms**
* Affected platforms: Windows 10, Windows 11.
* Unaffected platforms: Ubuntu 22.04, macOS 13
**Steps to reproduce**
1. Install a beta version that is older than Firefox 127.0b8 (for eg Firefox 127.0b7)
2. Trigger an update on the default beta channel.
**Expected result**
* Firefox should be updated to Firefox 127 build 2 and then to Firefox 128.0b1.
**Actual result**
* After triggering the update the 'Firefox Update' dialog with the message 'Firefox is installing your updates and will start in a few moments' is displayed and stalls.
* After approx 20 minutes, the 'Firefox update' dialog is dismissed and Firefox is restarted to Firefox 127 build 2, but the update to 128.0b1 fails. - please see the attached maintenanceservice log.
* Some Firefox process remain open in the background, if I manually close those processes then the update to Firefox 128 beta is done.
**Regression range**
* All the beta versions prior to Firefox 127.0b8 seems to be affected
**Additional notes**
* The issue is reproducible only with the MSI and EXE installers.
* Update are properly done on ZIP versions.
* Updates from versions older than Firefox 127.0b7 to Firefox 128 beta fail on the second attempt. The first attempt is successful.
|
|
1,888,744 | 16,867,810 |
Steps to reproduce:
On git commit 28cc363411d2029aed04c969c8f98785cae110db the attached sample asserts in the js-shell when invoked as `obj-x86_64-pc-linux-gnu/dist/bin/js --fuzzing-safe crash.js`
```
x = [0,0,0,0,0,0,0,0];
function f1() {
return enableGeckoProfiling();
}
x.sort(f1);
```
```
#0 GetTopProfilingJitFrame (act=0x7fffffffc9d0) at js/src/vm/GeckoProfiler.cpp:62
#1 js::GeckoProfilerRuntime::enable (this=0x7ffff602f378, enabled=true) at js/src/vm/GeckoProfiler.cpp:111
#2 0x000055555718e2fb in EnableGeckoProfiling (cx=cx@entry=0x7ffff6039100, argc=<optimized out>, vp=<optimized out>)
at js/src/shell/js.cpp:7366
#3 0x00005555572693b7 in CallJSNative (cx=cx@entry=0x7ffff6039100,
native=native@entry=0x55555718e240 <EnableGeckoProfiling(JSContext*, unsigned int, JS::Value*)>,
reason=reason@entry=js::CallReason::Call, args=...) at js/src/vm/Interpreter.cpp:479
#4 0x00005555572685d2 in js::InternalCallOrConstruct (cx=0x7ffff6039100, args=..., construct=construct@entry=js::NO_CONSTRUCT,
reason=js::CallReason::Call) at js/src/vm/Interpreter.cpp:573
#5 0x000055555726a326 in InternalCall (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, args=..., reason=1505770368)
at js/src/vm/Interpreter.cpp:640
#6 0x000055555727e5b1 in js::CallFromStack (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, args=..., reason=<optimized out>)
at js/src/vm/Interpreter.cpp:645
#7 js::Interpret (cx=0x7ffff6039100, state=...) at js/src/vm/Interpreter.cpp:3060
#8 0x0000555557267b27 in MaybeEnterInterpreterTrampoline (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, cx@entry=0x7ffff6039100, state=...)
at js/src/vm/Interpreter.cpp:393
#9 0x000055555726781a in js::RunScript (cx=cx@entry=0x7ffff6039100, state=...) at js/src/vm/Interpreter.cpp:451
#10 0x00005555572684ff in js::InternalCallOrConstruct (cx=0x7ffff6039100, args=..., construct=construct@entry=js::NO_CONSTRUCT,
reason=<optimized out>) at js/src/vm/Interpreter.cpp:605
#11 0x000055555726a326 in InternalCall (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, cx@entry=0x7ffff6039100, args=..., reason=1505770368,
reason@entry=js::CallReason::Call) at js/src/vm/Interpreter.cpp:640
#12 0x000055555726a50e in js::Call (cx=cx@entry=0x7ffff6039100, fval=fval@entry=..., thisv=thisv@entry=..., args=..., rval=...,
reason=reason@entry=js::CallReason::Call) at js/src/vm/Interpreter.cpp:672
#13 0x0000555558225118 in js::jit::InvokeFunction (cx=cx@entry=0x7ffff6039100, obj=obj@entry=..., constructing=<optimized out>,
ignoresReturnValue=false, argc=argc@entry=2, argv=argv@entry=0x7fffffffc830, rval=...)
at js/src/jit/VMFunctions.cpp:545
#14 0x0000555558225c99 in js::jit::InvokeFromInterpreterStub (cx=0x7ffff6039100, frame=<optimized out>)
at js/src/jit/VMFunctions.cpp:569
#15 0x000037f7787c1f20 in ?? ()
#16 0x00007fffffffc900 in ?? ()
```
|
|
1,888,902 | 16,870,672 |
Steps to reproduce:
On git commit 28cc363411d2029aed04c969c8f98785cae110db the attached sample asserts in the js-shell when invoked as `obj-x86_64-pc-linux-gnu/dist/bin/js --fuzzing-safe crash.js`
Bisecting did not identify a recent regressor, commits from end of 2022 are affected already.
This particular crash is release-asserted but flagging as s-s just in case the root cause triggers other problems.
```
const v0 = `
function F1() {
const v11 = registerModule("module1", parseModule(\`import {} from "module2";
import {} from "module3";\`));
const v13 = "await 1;";
drainJobQueue();
registerModule("module2", parseModule(v13));
registerModule("module3", parseModule(v0));
moduleLink(v11);
moduleEvaluate(v11);
}
F1();
`;
eval(v0);
```
```
#0 js::ModuleObject::getCycleRoot (this=0x15d82ec3e138) at js/src/builtin/ModuleObject.cpp:1267
#1 0x00005555576ad7f9 in GatherAvailableModuleAncestors (cx=cx@entry=0x7ffff7439100, module=module@entry=..., execList=execList@entry=...)
at js/src/vm/Modules.cpp:1776
#2 0x00005555576acb40 in js::AsyncModuleExecutionFulfilled (cx=cx@entry=0x7ffff7439100, module=module@entry=...)
at js/src/vm/Modules.cpp:1867
#3 0x000055555741f8be in AsyncModuleExecutionFulfilledHandler (cx=cx@entry=0x7ffff7439100, argc=<optimised out>, vp=<optimised out>)
at js/src/vm/AsyncFunction.cpp:301
#4 0x0000555557264fc7 in CallJSNative (cx=cx@entry=0x7ffff7439100,
native=native@entry=0x55555741f760 <AsyncModuleExecutionFulfilledHandler(JSContext*, unsigned int, JS::Value*)>,
reason=reason@entry=js::CallReason::Call, args=...) at js/src/vm/Interpreter.cpp:479
#5 0x00005555572641e2 in js::InternalCallOrConstruct (cx=0x7ffff7439100, args=..., construct=construct@entry=js::NO_CONSTRUCT,
reason=js::CallReason::Call) at js/src/vm/Interpreter.cpp:573
#6 0x0000555557265f36 in InternalCall (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, cx@entry=0x7ffff7439100, args=..., reason=1505742272,
reason@entry=js::CallReason::Call) at js/src/vm/Interpreter.cpp:640
#7 0x000055555726611e in js::Call (cx=cx@entry=0x7ffff7439100, fval=fval@entry=..., thisv=thisv@entry=..., args=..., rval=rval@entry=...,
reason=reason@entry=js::CallReason::Call) at js/src/vm/Interpreter.cpp:672
#8 0x0000555557340b25 in js::Call (cx=0x7ffff7439100, fval=..., thisv=..., arg0=..., rval=...) at js/src/vm/Interpreter.h:116
#9 0x000055555772f0cf in PromiseReactionJob (cx=cx@entry=0x7ffff7439100, argc=<optimised out>, vp=<optimised out>)
at js/src/builtin/Promise.cpp:2256
#10 0x0000555557264fc7 in CallJSNative (cx=cx@entry=0x7ffff7439100,
native=native@entry=0x55555772df40 <PromiseReactionJob(JSContext*, unsigned int, JS::Value*)>, reason=reason@entry=js::CallReason::Call, args=...)
at js/src/vm/Interpreter.cpp:479
#11 0x00005555572641e2 in js::InternalCallOrConstruct (cx=0x7ffff7439100, args=..., construct=construct@entry=js::NO_CONSTRUCT,
reason=js::CallReason::Call) at js/src/vm/Interpreter.cpp:573
#12 0x0000555557265f36 in InternalCall (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, cx@entry=0x7ffff7439100, args=..., reason=1505742272,
reason@entry=js::CallReason::Call) at js/src/vm/Interpreter.cpp:640
#13 0x000055555726611e in js::Call (cx=cx@entry=0x7ffff7439100, fval=..., thisv=..., args=..., rval=rval@entry=...,
reason=reason@entry=js::CallReason::Call) at js/src/vm/Interpreter.cpp:672
#14 0x000055555746e8cf in JS::Call (cx=cx@entry=0x7ffff7439100, thisv=..., fval=..., args=..., rval=rval@entry=...)
at js/src/vm/CallAndConstruct.cpp:119
#15 0x00005555575b87ef in JS::Call (cx=0x7ffff7439100, thisv=..., funObj=..., args=..., rval=...)
at obj-x86_64-pc-linux-gnu/dist/include/js/CallAndConstruct.h:110
#16 js::InternalJobQueue::runJobs (this=0x7ffff7423c80, cx=0x7ffff7439100) at js/src/vm/JSContext.cpp:872
#17 0x00005555575b7eb0 in js::RunJobs (cx=cx@entry=0x7ffff7439100) at js/src/vm/JSContext.cpp:809
#18 0x000055555716d7f8 in RunShellJobs (cx=cx@entry=0x7ffff7439100) at js/src/shell/js.cpp:1294
#19 0x000055555718bd6d in DrainJobQueue (cx=cx@entry=0x7ffff7439100, argc=<optimised out>, vp=<optimised out>)
at js/src/shell/js.cpp:1316
#20 0x0000555557264fc7 in CallJSNative (cx=cx@entry=0x7ffff7439100,
native=native@entry=0x55555718bce0 <DrainJobQueue(JSContext*, unsigned int, JS::Value*)>, reason=reason@entry=js::CallReason::Call, args=...)
at js/src/vm/Interpreter.cpp:479
#21 0x00005555572641e2 in js::InternalCallOrConstruct (cx=0x7ffff7439100, args=..., construct=construct@entry=js::NO_CONSTRUCT,
reason=js::CallReason::Call) at js/src/vm/Interpreter.cpp:573
#22 0x0000555557265f36 in InternalCall (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, args=..., reason=1505742272)
at js/src/vm/Interpreter.cpp:640
#23 0x000055555727a1c1 in js::CallFromStack (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, args=..., reason=<optimised out>)
at js/src/vm/Interpreter.cpp:645
#24 js::Interpret (cx=0x7ffff7439100, state=...) at js/src/vm/Interpreter.cpp:3060
#25 0x0000555557263737 in MaybeEnterInterpreterTrampoline (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, cx@entry=0x7ffff7439100, state=...)
at js/src/vm/Interpreter.cpp:393
#26 0x000055555726342a in js::RunScript (cx=cx@entry=0x7ffff7439100, state=...) at js/src/vm/Interpreter.cpp:451
#27 0x0000555557268402 in js::ExecuteKernel (cx=cx@entry=0x7ffff7439100, script=script@entry=..., envChainArg=envChainArg@entry=...,
evalInFrame=evalInFrame@entry=..., result=result@entry=...) at js/src/vm/Interpreter.cpp:838
#28 0x0000555557268c0d in js::Execute (cx=cx@entry=0x7ffff7439100, script=script@entry=..., envChain=envChain@entry=..., rval=rval@entry=...)
at js/src/vm/Interpreter.cpp:870
#29 0x0000555557372dfe in js::ModuleObject::execute (cx=0x7ffff7439100, self=...) at js/src/builtin/ModuleObject.cpp:1412
#30 0x00005555576ac120 in InnerModuleEvaluation (cx=0x7ffff7439100, module=..., stack=stack@entry=..., index=3, indexOut=indexOut@entry=0x7fffffffc858)
at js/src/vm/Modules.cpp:1694
#31 0x00005555576abd5a in InnerModuleEvaluation (cx=0x7ffff7439100, module=..., stack=..., index=2, index@entry=0,
indexOut=indexOut@entry=0x7fffffffc8d8) at js/src/vm/Modules.cpp:1611
#32 0x00005555576a272a in js::ModuleEvaluate (cx=cx@entry=0x7ffff7439100, moduleArg=moduleArg@entry=..., result=result@entry=...)
at js/src/vm/Modules.cpp:1465
#33 0x000055555717b463 in ModuleEvaluate (cx=cx@entry=0x7ffff7439100, argc=<optimised out>, vp=<optimised out>)
at js/src/shell/js.cpp:5488
#34 0x0000555557264fc7 in CallJSNative (cx=cx@entry=0x7ffff7439100,
native=native@entry=0x55555717b240 <ModuleEvaluate(JSContext*, unsigned int, JS::Value*)>, reason=reason@entry=js::CallReason::Call, args=...)
at js/src/vm/Interpreter.cpp:479
#35 0x00005555572641e2 in js::InternalCallOrConstruct (cx=0x7ffff7439100, args=..., construct=construct@entry=js::NO_CONSTRUCT,
reason=js::CallReason::Call) at js/src/vm/Interpreter.cpp:573
#36 0x0000555557265f36 in InternalCall (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, args=..., reason=1505742272)
at js/src/vm/Interpreter.cpp:640
#37 0x000055555727a1c1 in js::CallFromStack (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, args=..., reason=<optimised out>)
at js/src/vm/Interpreter.cpp:645
#38 js::Interpret (cx=0x7ffff7439100, state=...) at js/src/vm/Interpreter.cpp:3060
#39 0x0000555557263737 in MaybeEnterInterpreterTrampoline (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, cx@entry=0x7ffff7439100, state=...)
at js/src/vm/Interpreter.cpp:393
#40 0x000055555726342a in js::RunScript (cx=cx@entry=0x7ffff7439100, state=...) at js/src/vm/Interpreter.cpp:451
#41 0x0000555557268402 in js::ExecuteKernel (cx=cx@entry=0x7ffff7439100, script=script@entry=..., envChainArg=envChainArg@entry=...,
evalInFrame=evalInFrame@entry=..., result=result@entry=...) at js/src/vm/Interpreter.cpp:838
#42 0x00005555572fceff in EvalKernel (cx=0x7ffff7439100, v=v@entry=..., evalType=evalType@entry=DIRECT_EVAL, caller=..., env=env@entry=...,
pc=0x7ffff744561c "y\001", vp=...) at js/src/builtin/Eval.cpp:352
#43 0x00005555572feeed in js::DirectEval (cx=0x7ffff7439100, v=..., vp=...) at js/src/builtin/Eval.cpp:379
#44 0x0000555557279715 in js::Interpret (cx=0x7ffff7439100, state=...) at js/src/vm/Interpreter.cpp:2945
#45 0x0000555557263737 in MaybeEnterInterpreterTrampoline (cx=0x7ffff7a008e0 <_IO_stdfile_2_lock>, cx@entry=0x7ffff7439100, state=...)
at js/src/vm/Interpreter.cpp:393
#46 0x000055555726342a in js::RunScript (cx=cx@entry=0x7ffff7439100, state=...) at js/src/vm/Interpreter.cpp:451
#47 0x0000555557268402 in js::ExecuteKernel (cx=cx@entry=0x7ffff7439100, script=script@entry=..., envChainArg=envChainArg@entry=...,
evalInFrame=evalInFrame@entry=..., result=result@entry=...) at js/src/vm/Interpreter.cpp:838
#48 0x0000555557268c0d in js::Execute (cx=cx@entry=0x7ffff7439100, script=script@entry=..., envChain=..., rval=rval@entry=...)
at js/src/vm/Interpreter.cpp:870
#49 0x00005555574b3daa in ExecuteScript (cx=cx@entry=0x7ffff7439100, envChain=..., script=..., rval=rval@entry=...)
at js/src/vm/CompilationAndEvaluation.cpp:494
#50 0x00005555574b4028 in JS_ExecuteScript (cx=cx@entry=0x7ffff7439100, scriptArg=scriptArg@entry=...)
at js/src/vm/CompilationAndEvaluation.cpp:518
#51 0x00005555571a0b78 in RunFile (cx=0x7ffff7439100, filename=<optimised out>, file=<optimised out>, compileMethod=CompileUtf8::DontInflate,
compileOnly=false, fullParse=<optimised out>) at js/src/shell/js.cpp:1196
#52 0x00005555571a003e in Process (cx=cx@entry=0x7ffff7439100, filename=0x0, forceTTY=<optimised out>, kind=kind@entry=FileScript)
at js/src/shell/js.cpp:1775
#53 0x000055555715c073 in ProcessArgs (cx=0x7ffff7439100, op=0x7fffffffe638) at js/src/shell/js.cpp:11124
#54 Shell (cx=0x7ffff7439100, op=op@entry=0x7fffffffe638) at js/src/shell/js.cpp:11383
#55 0x0000555557154729 in main (argc=<optimised out>, argv=0x7fffffffe8c8) at js/src/shell/js.cpp:11891
```
|
|
1,889,122 | 16,872,503 |
Created attachment 9394499
before moving to next field.png
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:124.0) Gecko/20100101 Firefox/124.0
Steps to reproduce:
Visit "https://nj.gov/mvc/pdf/license/BA-208.pdf" and start filling out form. In the zip code field enter a 5 digit number starting with zero. Hit tab or select next field and the leading zero gets truncated.
Actual results:
When you move to the next field after entering the zip code with a leading zero, the leading zero is truncated. For example, entering 01234 results in 1234 when the cursor is moved to next field.
Expected results:
The number should present as entered since zip codes can begin with zeros.
|
|
1,889,473 | 16,874,869 |
It is difficult to reproduce this but sometimes when dragging or resizing the region near the edge of the window, the window will scroll when we go to set focus to either copy or download even though the buttons are still visible.
|
|
1,889,616 | 16,875,834 |
Created attachment 9394993
MicroSurvey - Card 1.png
**[Affected Versions]:**
* Firefox Release 121.0, BUILD ID: 20231211174248
* Firefox Release 123.0.1, BUILD ID: 20240304104836
* Firefox Release 124.0.2, BUILD ID: 20240401114208
* Firefox Beta 125.0b5, BUILD ID: 20240327092250
**[Affected Platforms]:**
* macOS 14.3
**[Prerequisites]:**
* Have a Firefox “de” locale installed.
* Have the `nimbus.debug` pref set to `true` in the “about:config” page.
**[Steps to reproduce]:**
1. Open the browser using the profile from the prerequisites.
2. Navigate to the following link: about:studies?optin_slug=new-tab-microsurvey-relevance-satisfaction-existing-users-us-ca-de-gb&optin_branch=treatment-a&optin_collection=nimbus-preview
3. Open a new tab.
4. Observe the title from the Microsurvey card displayed in the bottom-right part of the screen.
**[Expected result]:**
* The title should be in line with the Firefox logo.
**[Actual result]:**
* The title is displayed on the next row, under the Firefox logo.
**[Notes]:**
* This issue is reproducible only on macOS. On Windows and Linux there is a different behavior tracked in bug 1889341
* This issue is reproducible only on the DE locale.
* Attaching a screenshot of the issue:
|
|
1,889,646 | 16,875,940 |
Created attachment 9395014
Gif showing the issue
**Found in**
* Latest Nightly 126.0a1
**Affected versions**
* Latest Nightly 126.0a1
**Tested platforms**
* Affected platforms: Windows 10, Windows 11, macOS 13 and Ubuntu 22.04
* Unaffected platforms: none
**Preconditions**
* Have `browser.tabs.cardPreview.enabled` set to `true`
**Steps to reproduce**
1. Open a few tabs
2. Click on menu, for example Hamburger Menu or Download panel.
3. Hover over a tab
4. Without leaving the tab strip click on all the tabs
**Expected result**
* The opened menu is closed when clicking a tab OR tab preview is not shown when the menu is opened, just as it does when a context menu is opened (only on Mac).
**Actual result**
* Menus, panels will not close if preview is visible and clicking on different tabs without leaving the tabstrip.
**Regression range**
* Not a regression since this is also reproducible on old Nightly build from 2024-01-14 where this feature officially landed in Nightly.
**Additional notes**
* On Chrome if I have their hamburger menu open or context menu I can't use the tab preview, but on some menus like Bookmark or Users I can though it will close immediately when clicking on a tab.
|
|
1,889,668 | 16,876,083 |
Created attachment 9395043
dark previews.gif
**Found in**
* 126.0a1
**Affected versions**
* 126.0a1
**Tested platforms**
* Affected platforms: Ubuntu 22, Ubuntu 23
* Unaffected platforms: Windows, macOS
**Preconditions**
* browser.tabs.cardPreview.enabled set to true
**Steps to reproduce**
1. Open several tabs and hover over them to trigger the preview.
2. Open a few more new tabs.
3. Hover again all the open tabs and observe the preview.
**Expected result**
The tab hover preview is visible for all the opened tabs.
**Actual result**
The tab hover preview appears dark for more then half of the tabs.
**Regression range**
* Not a regression since this is a new feature.
**Notes**
* The dark hover preview is frequently visible on the about:newtab pages.
* Video attached showing the issue.
|
|
1,889,703 | 16,876,309 |
Currently about:welcome import screen no longer has the default "Import Browser Data" title. This should be introduced as the default title.
Route to `about:welcome` select import checkbox, route to import screen, notice `Import Browser Data` title missing.
|
|
1,889,885 | 16,877,220 |
**Found in**
* version: 125.0b8
**Affected versions**
* version:125.0b8
**Tested platforms**
* Affected platforms: Windows, Ubuntu
* Unaffected platforms: Mac
**Preconditions**
* enable following prefs:
* browser.shopping.experience2023.enabled = true
* browser.shopping.experience2023.optedIn =1
**Steps to reproduce**
1. Open FF with above prefs enabled.
2. Go to link https://www.amazon.com/gp/video/detail/B0C6XBFVVW/ref=atv_hm_hom_c_lZOsi7_2_1?jic=8%7CEgNhbGw%3D
3. From Review Checker->Settings->turn off review checker.
4. From address bar open review checker.
5. Check review checker display.
**Expected result**
* Checker review dialog should be correctly displayed.
**Actual result**
* Checker review dialog is cut.
** Regression range:
2024-04-05T11:43:09.660000: INFO : Narrowed integration regression window from [935d4767, a3f5efd3] (3 builds) to [935d4767, def04dad] (2 builds) (~1 steps left)
2024-04-05T11:43:11.310000: DEBUG : Found commit message:
Bug 1876173 - Add attribute to allow RDM to be disabled for aboutwelcome spotlight r=omc-reviewers,aminomancer
app_name: firefox
build_date: 2024-03-25 23:16:15.703000
build_file: C:\Users\svuser_1\.mozilla\mozregression\persist\935d47673d41--autoland--target.zip
build_type: integration
build_url: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/ZUt0iigQR0GrsT-CL9HCnQ/runs/0/artifacts/public%2Fbuild%2Ftarget.zip
changeset: 935d47673d4131314e486197594ae3495e9a30d1
pushlog_url: https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=935d47673d4131314e486197594ae3495e9a30d1&tochange=f0c093accca57074b185d7b3a5a76937ae8c57ed
repo_name: autoland
repo_url: https://hg.mozilla.org/integration/autoland
task_id: ZUt0iigQR0GrsT-CL9HCnQ
|
|
1,890,186 | 16,879,075 |
Created attachment 9395423
firefox_bug_report_screenshot.jpg
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:125.0) Gecko/20100101 Firefox/125.0
Steps to reproduce:
1. Go to any video conferencing site
2. Join a meeting
3. Try to select which audio input (and/or output) device should be used
Actual results:
The list with available devices is not complete. Also testing this causes Firefox to stop responding, but I do not know how related is that.
See attached screenshot for discrepancy between options in Firefox vs. Sound settings in macos.
Expected results:
The list with available devices should be complete and let me choose audio output device.
|
|
1,890,775 | 16,885,164 |
**Steps to reproduce**
1. Navigate to `data:text/html,<meta charset=utf8><style>%0Ahtml{}body{color:red;background:gold;}</style>Hello`
1. Open the inspector and select the `body` node
1. In the `body` rule, unselect the `color` property
**Expected results**
The `Hello` text gets black, the background color stays gold
**Actual results**
The `Hello` text gets black, the background color gets white, and if you navigate to the style editor, you can see that the stylesheet is empty
---
The stylesheet has the following content:
```css
html{}body{color:red;background:gold;}
```
When `InspectorUtils.replaceBlockRuleBodyTextInStylesheet` is called to replace the content of the second rule, `get_byte_index_from_line_and_column` ([glue.rs](https://searchfox.org/mozilla-central/rev/1f27a4022f9f1269d897526c1c892a57743e650c/servo/ports/geckolib/glue.rs#9230-9268)) returns an erroneous index as it doesn't account for the line break character.
This means that `Servo_ReplaceBlockRuleBodyTextInStylesheetText` ([glue.rs](https://searchfox.org/mozilla-central/rev/1f27a4022f9f1269d897526c1c892a57743e650c/servo/ports/geckolib/glue.rs#9173-9228)) includes the closing curly bracket of the previous rule when building the string it should parse (i.e. `}body{color:red;background:gold;}`).
This is a parsing error as the parser didn't encountered any opening bracket before, and the function will return null.
So we should fix the index computation, but also maybe avoid emptying the stylesheet when the function returns null.
|
|
1,890,802 | 16,885,293 |
Created attachment 9396035
Screenshot with the issue
**Found in**
* Nightly 126.0a1
**Affected versions**
* Nightly 126.0a1
**Tested platforms**
* Affected platforms: macOS 13, Ubuntu 22.04.
* Unaffected platforms: Windows 10 x64
**Preconditions**
* Enable High Contrast Mode.
**Steps to reproduce**
1. Enable Screenshots by pressing the Cmd+Shift+S keys.
2. Observe the Cancel button and mouse hover over it.
**Expected result**
* The Cancel button should be visible, and the mouse hover effect should be noticeable.
**Actual result**
* The Cancel button and the mouse hover effect are barely visible. Please see the screenshot attached.
**Regression range**
* Yes, this is not reproducible on Firefox 125.0b5 or earlier versions when Screenshots component is enabled (screenshots.browser.component.enabled is set to true).
* Found commit message:
Bug 1881405: Use the `forced-colors` media query instead of `prefers-contrast` when styles depend on system colors in privacy.css r=ayeddi,settings-reviewers,desktop-theme-reviewers,emilio
* Pushlog:
https://hg.mozilla.org/integration/autoland/json-pushes?changeset=033b661b349ec9f221fe7453bef9e57d74fcade2&full=1
**Additional notes**
* This issue is not reproducible on Windows 10.
|
|
1,890,861 | 16,885,651 |
Created attachment 9396098
Screenshot_20240409-124252.png
|
|
1,891,130 | 16,887,443 |
### Steps to reproduce
1. Use a mozconfig WITHOUT artifact mode.
2. Run `./mach build`.
3. Run `./mach gradle assembleWithGeckoBinariesFocusRelease`.
4. Observe.
5. Run `./mach gradle assembleWithGeckoBinariesFenixRelease`.
6. Observe.
### Expected behaviour
* At step 4, Focus builds and installs on the connected device.
* At step 6, **Fenix** builds and installs on the connected device.
### Actual behaviour
* At step 4, Focus builds and installs on the connected device.
* At step 6, **Fenix** fails the build.
### Device information
* Firefox version: n/a
* Android device model: n/a
* Android OS version: n/a
### Any additional information?
Final output from `mach gradle assembleWithGeckoBinariesFenixRelease`:
```
FAILURE: Build failed with an exception.
* What went wrong:
Could not determine the dependencies of task ':fenix:lintVitalReportWithGeckoBinariesFenixRelease'.
> Could not resolve all task dependencies for configuration ':fenix:lintChecks'.
> Could not resolve project :mozilla-lint-rules.
Required by:
project :fenix
> A dependency was declared on configuration 'default' which is not declared in the descriptor for project :mozilla-lint-rules.
```
|
|
1,891,468 | 16,892,825 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:124.0) Gecko/20100101 Firefox/124.0
Steps to reproduce:
- Visit a site that sets an invalid HSTS header ("HSTS: max-age=20; includeSubDomains=")
- Visit the site via HTTP and observe whether it is upgraded: http://sub.headers.websec.saarland/_hp/tests/upgrade-hsts.sub.html?resp_type=parsing&browser_id=1&label=HSTS&first_id=32689&last_id=32689&scheme=http&t_resp_id=32689&t_element_relation=direct_direct&t_resp_origin=https://sub.headers.websec.saarland
- Visit sub.site via HTTP and observe whether it is upgraded: http://sub.headers.websec.saarland/_hp/tests/upgrade-hsts.sub.html?resp_type=parsing&browser_id=1&label=HSTS&first_id=32689&last_id=32689&scheme=http&t_resp_id=32689&t_element_relation=subdomain_subdomain&t_resp_origin=https://sub.headers.websec.saarland
Actual results:
- Firefox upgrades both requests, the HSTS header seems to be interpreted as `max-age=60; includeSubDomains`.
- Chromium says the header is invalid, both are not updated
- Safari only upgrades the first request, seems to be interpreted as `max-age=60;`
Expected results:
Chromium's behavior seems to be correct. An empty directive value is not allowed according to the HSTS syntax and the specification states that: ""UAs MUST ignore any STS header field containing directives, or other header field value data, that does not conform to the syntax defined in this specification." https://www.rfc-editor.org/rfc/rfc6797#section-6.1
|
|
1,891,729 | 16,894,428 |
Created attachment 9396894
mainmenu.mp4
### Steps to reproduce
1. Launch Fenix and navigate to any site that with a text box (ie wikipedia.org)
2. Tap on the search field.
3. Tap on 3 dots menu.
### Expected behavior
The main menu is displayed from the bottom of the screen.
### Actual behavior
The keyboard is hiding and the menu remains displayed in the middle of the screen.
### Device information
* Firefox version: Firefox Nightly 127 (2024-04-16), Firefox 126 Beta 1
* Android device model: OnePLus 6T (Android 11), Xiaomi 12T (Android 12)
### Any additional information?
* This issue is a regression: Last Good build (2024-03-18) - First Bad build (2024-03-20) - there is no Nightly build from 2024-03-19
|
|
1,892,125 | 16,897,036 |
STR:
- Use ctrl+shift+s or the screenshots toolbar button or the context menu to open the Screenshots UI on any page
- Click either the Save Visible or Save Full Page
- In the resulting preview dialog, click on the "copy" or "dialog" text rather than the icon in either of the buttons
ER:
- The screenshot is copied or downloaded
AR:
- Nothing happens. Its necessary to click on the icon to actually copy/download.
Ultimately the best fix here is to use `moz-button` and `moz-button-group` for these buttons. We have all this custom css in/around [`.preview-button`](https://searchfox.org/mozilla-central/rev/c7df16ffad1f12a19c81c16bce0b65e4a15304d0/browser/components/screenshots/content/screenshots.css#18-47) which isn't great.
But, specifically, the issue here is that we shouldn't be overriding the `pointer` property, and we should remove the `pointer-events: none` in favor of examining the `event.currentTarget` rather than `event.target` in the [click event listener.](https://searchfox.org/mozilla-central/rev/c7df16ffad1f12a19c81c16bce0b65e4a15304d0/browser/components/screenshots/content/screenshots.js#103)
|
|
1,895,909 | 16,927,950 |
While resizing an image to a smaller size, the image starts sliding down and moves outside the resize box.
|
|
1,896,002 | 16,928,555 |
STR:
1. Make sure "always ask where to save files" is checked in about:preferences
2. Open screenshots
3. Take a full page or visible screenshot
4. Click the download button
5. File picker should open, close file picker without saving screenshot
AR:
The screenshots preview buttons are disabled
ER:
The screenshots preview buttons should be enabled
|
|
1,896,398 | 16,932,957 |
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:125.0) Gecko/20100101 Firefox/125.0
Steps to reproduce:
OS: Arch Linux
Graphics: Intel Iris Xe (12th Gen)
Desktop Environment: KDE Wayland
1. Open https://www.flightradar24.com
2. Zoom in/out and drag around QUICKLY
Actual results:
The tab crashed. On some circumstances, the whole browser freezes.
Firefox outputs this:
```
WARNING: failed to create memfd: Too many open files: file /usr/src/debug/firefox/firefox-125.0.3/ipc/chromium/src/base/shared_memory_posix.cc:312
```
Expected results:
The tab won't crash.
I raised the rlim NOFILE to the hard limit (524288 on my machine) and rerun that tab. It runs without any problem.
`strace -e prlimit64` shows that firefox itself raises rlim NOFILE to 4096 by default, which turns out to be not enough.
Relates to Bug 1401776
The limit was hard-coded at https://searchfox.org/mozilla-central/source/toolkit/xre/nsAppRunner.cpp#3762.
I think we should raise that limit to a much higher one to accommodate some quirky websites.
|
|
1,896,871 | 16,936,212 |
### Steps to reproduce
1. install temporarily `dnr-dynamic-with-options` extension example from the webextensions-examples github repo (https://github.com/mdn/webextensions-examples/tree/main/dnr-dynamic-with-options), e.g. using web-ext cli tool:
```
web-ext run -t firefox-android -s /path/to/dnr-dynamic-with-options --android-device=emulator-5554 --firefox-apk org.mozilla.fenix.debug --firefox-apk-component org.mozilla.fenix.debug.App
```
2. open the extension settings page from the Fenix Addon Manager UI
3. Request and Grant a couple more host permissions by pasting the following snipped in the "Allowed host permissions" text area and then clicking the "Grant host permissions" button (and then grant the permission in the permissions dialog opened right after):
```
["*://example.com/*", "https://developer.mozilla.org/*"]
```
4. Open the Permissions UI
### Expected behavior
Expect the optional permissions UI to list 3 host permissions:
- `Allow for all sites`
- `*://example.com/*`
- `https://developer.mozilla.org/*`
(like on Firefox Desktop when using the same STR described here)
### Actual behavior
Only `Allow for all sites` is listed and can be revoked, the other two granted host permissions are not listed and so can't be seen nor revoked by the user from the Firefox for Android AddonManager UI.
### Device information
* Firefox version: Nightly 128 with Bug 1811563 patch applied
* Android device model: any
* Android OS version: any
### Any additional information?
An optional host permission match pattern allows extensions to request more specific host permissions when they are actually allowed subsets of the more broad host permission explicitl listed in the manifest.
As an example an "all urls" kind of host permission (e.g. "<all_urls>", "*://*/*" etc.) allows the extension to request to the user to grant more specific host permissions (e.g. the extension may request "*://example.com/*", and the user may then decide to revoke the permission on "<all_urls>" but leave the extension to be granted access to "*://example.com/*" if that optional host permission isn't also revoked).
This behavios currently works as described above on the permissions view available in the Firefox Desktop about:addons page, while the permissions view available in the Firefox for Android AddonManager UI as implemented in Bug 1811563 does not yet list granted host permissions that are not also explicitly listed in the manifest file.
One possible approach to ensure we also list granted host permissions not explicitly listed in the manifest file may be to tweak the `newFromWebExtension` method defined in `mobile/android/android-components/components/feature/addons/src/main/java/mozilla/components/feature/addons/Addon.kt` by composing the optionalOrigin and grantedOptionalOrigins arrays:
```
diff --git a/mobile/android/android-components/components/feature/addons/src/main/java/mozilla/components/feature/addons/Addon.kt b/mobile/android/android-components/components/feature/addons/src/main/java/mozilla/components/feature/addons/Addon.kt
--- a/mobile/android/android-components/components/feature/addons/src/main/java/mozilla/components/feature/addons/Addon.kt
+++ b/mobile/android/android-components/components/feature/addons/src/main/java/mozilla/components/feature/addons/Addon.kt
@@ -436,12 +436,14 @@ data class Addon(
)
} ?: emptyList()
- val optionalOrigins = metadata?.optionalOrigins?.map { origin ->
+ val optionalOriginsOrEmptyList = metadata?.optionalOrigins ?: emptyList()
+ val allOrigins = (optionalOriginsOrEmptyList + grantedOptionalOrigins).toSet().toTypedArray()
+ val optionalOrigins = allOrigins.map { origin ->
Permission(
name = origin,
granted = grantedOptionalOrigins.contains(origin),
)
- } ?: emptyList()
+ }
return Addon(
id = extension.id,
```
One caveat of only applying the change from this diff is that the granted host permissions listed in the permissions screen are going to disappear as soon as the granted host permission not explicitly listed in the manifest is revoked, instead of being still listed (unless the user explicitly navigate away from the permissions view) to allow the user to change idea and grant it back.
It may be worth considering looking more closely into this corner case and if we could keep those host permissions in the UI to allow the user to still grant it back if they didn't navigate away from the permissions view (but we may consider leaving it to a separate followup if it requires more investigations and/or refactoring).
|
|
1,897,354 | 16,939,350 |
User Agent: Mozilla/5.0 (Android 14; Mobile; rv:128.0) Gecko/128.0 Firefox/128.0
Steps to reproduce:
Post latest update, I see the address bar is now longer touching the edge, even there is more space on the left side compared to right side
|
|
1,898,437 | 16,949,026 |
**Found in**
* Fx 128.0a1
**Affected versions**
* Fx 128.0a1
**Affected platforms**
* all
**Steps to reproduce**
1. Launch Firefox.
2. Perform a search using the address bar and wait for the SERP to load.
3. Click the address bar again and re-submit the search link.
3.1.(if needed) Click the address bar, hit escape to have the link displayed instead of the persistent search term.
**Expected result**
* SERP is reached again without any displayed errors in the console.
**Actual result**
* The following error is displayed in the browser console.
```
Could not record event: TypeError: can't access property "providerName", details.result is undefined
```
**Regression range**
* Looking for a regression ASAP.
**Additional notes**
* This occurs for all the search engines.
```
Could not record event: TypeError: can't access property "providerName", details.result is undefined
#notifyEngagement resource:///modules/UrlbarProvidersManager.sys.mjs:454
notifyEngagementChange resource:///modules/UrlbarProvidersManager.sys.mjs:420
#internalRecord resource:///modules/UrlbarController.sys.mjs:983
record resource:///modules/UrlbarController.sys.mjs:843
handleNavigation resource:///modules/UrlbarInput.sys.mjs:773
handleCommand resource:///modules/UrlbarInput.sys.mjs:624
handleKeyNavigation resource:///modules/UrlbarController.sys.mjs:345
_on_keydown resource:///modules/UrlbarInput.sys.mjs:4091
maybeDeferEvent resource:///modules/UrlbarEventBufferer.sys.mjs:159
_on_keydown resource:///modules/UrlbarInput.sys.mjs:4090
handleEvent resource:///modules/UrlbarInput.sys.mjs:588
```
|
|
1,898,802 | 16,951,928 |
Created attachment 9403874
jobRequests-5-7VZUM4YBFUSB.pdf
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:126.0) Gecko/20100101 Firefox/126.0
Steps to reproduce:
I generated a PDF using a Java application I have created (https://github.com/Aliuken/JobVacanciesApp_Java11)
Actual results:
The file is not properly rendered in Firefox
Expected results:
Using Chrome or Adobe Reader the file is rendered correctly. I attach the PDF file.
|
|
1,905,623 | 17,007,663 |
**Environment:**
Operating system: All Desktop
Firefox version: Firefox 127.0/Firefox 129.0a1
**Steps to reproduce:**
1. Open a downloaded PDF file from the site (provided in attachment)
2. Observe the Japanese punctuation
**Expected Behavior:**
Japanese punctuation is rendered as expected
**Actual Behavior:**
Items are misaligned
**Notes:**
- Reproduces regardless of the status of ETP
- Reproduces in Firefox Nightly, Firefox Release
- Does not reproduce in Chrome
Created from https://github.com/webcompat/web-bugs/issues/138659
|
|
1,906,001 | 17,010,183 |
Created attachment 9410926
image.png
The Normal/Active tab page button is not being properly colored by the `TabRow` + `Tab` component.
The [`tabCounterAlpha`](https://searchfox.org/mozilla-central/source/mobile/android/fenix/app/src/main/java/org/mozilla/fenix/tabstray/TabsTrayBanner.kt#199) `val` just needs to get moved to within the `Tab` block at [line 236](https://searchfox.org/mozilla-central/source/mobile/android/fenix/app/src/main/java/org/mozilla/fenix/tabstray/TabsTrayBanner.kt#236) for it to get the correct `LocalContentColor` within the localized context of the selected/unselected state of the `Tab`.
This can be verified using the Compose preview inside the file.
|
|
1,906,013 | 17,010,223 |
Created attachment 9410932
debug shell stack
```
(function (stdlib, foreign, heap) {
"use asm";
var Infinity = stdlib.Infinity;
var v = new stdlib.Int8Array(heap);
function f() {
return 0;
return v[0] | ~~Infinity;
}
return f;
});
```
```
406 MOZ_ASSERT(func_.callSiteLineNums.length() == lastReadCallSite_);
(gdb) bt
#0 (anonymous namespace)::FunctionCompiler::finish (this=0x7fffffff6fc0) at /home/skymainubu/trees/mozilla-central/js/src/wasm/WasmIonCompile.cpp:406
#1 IonBuildMIR (d=..., codeMeta=..., func=..., locals=..., mir=..., tryNotes=..., observedFeatures=0x7fffffff79b4, error=<optimized out>) at /home/skymainubu/trees/mozilla-central/js/src/wasm/WasmIonCompile.cpp:9362
#2 0x00005555586ce205 in js::wasm::IonCompileFunctions (codeMeta=..., compilerEnv=..., lifo=..., inputs=..., code=0x7ffff6bc73b0, error=error@entry=0x0) at /home/skymainubu/trees/mozilla-central/js/src/wasm/WasmIonCompile.cpp:9429
#3 0x00005555586ae66b in ExecuteCompileTask (task=0x7ffff6bc7000, error=0x0) at /home/skymainubu/trees/mozilla-central/js/src/wasm/WasmGenerator.cpp:517
#4 0x00005555586af4b7 in js::wasm::ModuleGenerator::locallyCompileCurrentTask (this=0x7fffffff9690) at /home/skymainubu/trees/mozilla-central/js/src/wasm/WasmGenerator.cpp:603
#5 js::wasm::ModuleGenerator::finishFuncDefs (this=0x7fffffff9690) at /home/skymainubu/trees/mozilla-central/js/src/wasm/WasmGenerator.cpp:734
/snip
```
```
The first bad revision is:
changeset: https://hg.mozilla.org/mozilla-central/rev/5698ddb336ec
user: Ryan Hunt
date: Tue Jun 25 15:18:39 2024 +0000
summary: Bug 1885771 - wasm: Use call site line number mechanism for asm.js truncate. r=yury
```
Run with `--fuzzing-safe --no-threads --no-baseline --no-ion`, compile with `AR=ar sh ../configure --enable-debug --enable-debug-symbols --with-ccache --enable-nspr-build --enable-ctypes --enable-gczeal --enable-rust-simd --disable-tests`, tested on m-c rev d25c9bdacb64.
Setting s-s to be safe. Ryan, is bug 1885771 a likely regressor?
|
|
1,906,022 | 17,010,264 |
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:129.0) Gecko/20100101 Firefox/129.0
Steps to reproduce:
I ran Firefox 129.0a1 20240701221157 to 20240702212506 on Wayland in Plasma 6.1.1 in a Fedora 40 KDE installation. I had previously enabled the menu bar. I selected Help in the menu bar and then About Nightly.
Actual results:
A box with a white outline but without an image was shown to the left of Update to 129.0a1 (2024-07-02) in About Firefox Nightly with 129.0a1 20240701221157 to 20240702212506. The middle of the box was black like the background of the About Firefox Nightly window which might've meant the box was transparent. This box wasn't shown with 129.0a1 20240701094548 or earlier. I bisected between 129.0a1 20240701094548 and 20240701221157 with mozregression. The first bad change was for Bug 1905704 - Use single image element for about dialog updater icons.
17:24.41 INFO: Narrowed integration regression window from [d88355e1, 0558f0d3] (3 builds) to [d88355e1, 942908c2] (2 builds) (~1 steps left)
17:24.41 INFO: No more integration revisions, bisection finished.
17:24.41 INFO: Last good revision: d88355e1c5cdeac1e4b56016c3193ef019861a0e
17:24.41 INFO: First bad revision: 942908c28bf32f5dd392fc4418eda3d9477ccea9
17:24.41 INFO: Pushlog:
https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=d88355e1c5cdeac1e4b56016c3193ef019861a0e&tochange=942908c28bf32f5dd392fc4418eda3d9477ccea9
The box with the white outline was also shown to the left of the Updates disabled by system administrator when using mozregression.
Expected results:
An image should have been shown where the box with the white outline but no image was shown to the left of Update to 129.0a1 (2024-07-02) in About Firefox Nightly. Such images as spinners or a checkmark are shown where the box with the white outline was at other steps of checking for updates or updating.
|
|
1,906,342 | 17,012,509 |
For the similar reproduce step mentioned in [bug 1904937 comment0](https://bugzilla.mozilla.org/show_bug.cgi?id=1904937#c0).
I can see sometime the infinite buffering appears again, when auto eviction kicks in, because it incorrectly evict the data which we're going to play soon. We should remove data furthest from the current playback position.
|
|
1,906,682 | 17,016,068 |
Created attachment 9411551
Screenshot_20240708-112256.png
### Steps to reproduce
1. Open firefox nightly app with new nav toolbar enabled.
2. The homepage button looks misaligned.
### Expected behavior
- The homepage button needs to be center aligned.
- The touch target size of the tab tray icon should match other icons in the nav toolbar.
### Actual behavior
- The homepage button is off center.
- The touch target size of the tab tray icon does not match other icons in the nav toolbar.
### Device information
* Firefox version: Nightly
* Android device model: Pixel 6
* Android OS version: Latest
### Any additional information?
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.