status
stringclasses 1
value | repo_name
stringlengths 9
24
| repo_url
stringlengths 28
43
| issue_id
int64 1
104k
| updated_files
stringlengths 8
1.76k
| title
stringlengths 4
369
| body
stringlengths 0
254k
⌀ | issue_url
stringlengths 37
56
| pull_url
stringlengths 37
54
| before_fix_sha
stringlengths 40
40
| after_fix_sha
stringlengths 40
40
| report_datetime
timestamp[ns, tz=UTC] | language
stringclasses 5
values | commit_datetime
timestamp[us, tz=UTC] |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
closed | electron/electron | https://github.com/electron/electron | 37,772 | ["patches/node/.patches", "patches/node/fix_increase_concurrency_in_v8platform_postjob.patch", "patches/v8/.patches", "patches/v8/revert_wasm_simplify_compilejstowasmwrapperjob.patch"] | chore: remove Simplify CompileJSToWasmWrapperJob revert patch | - 4347597: [wasm] Simplify CompileJSToWasmWrapperJob | https://chromium-review.googlesource.com/c/v8/v8/+/4347597 was reverted in #37701 because it caused many failures to the node tests, eg: https://app.circleci.com/pipelines/github/electron/electron/67031/workflows/eaedbe8d-2f77-47f0-a729-840ceed6c411/jobs/1475853. We should remove the patch as a followup to #37701. | https://github.com/electron/electron/issues/37772 | https://github.com/electron/electron/pull/37782 | 601217218759d088f12680a428fead8c8b26a5e0 | ef657bdc9d28a38321509e636f2f928cb6c2d981 | 2023-03-31T00:08:41Z | c++ | 2023-04-01T12:47:54Z |
closed | electron/electron | https://github.com/electron/electron | 37,761 | ["shell/browser/api/electron_api_web_contents_mac.mm"] | [Bug]: After entering and exiting full screen, app can't quit | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.2.0
### What operating system are you using?
macOS
### Operating System Version
macOS Ventura 13.2 (22D49)
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
1. Clone the quick start project
2. Start the project and hit `Fn+F` twice (enter and exit full screen)
The app should be able to quit.
### Actual Behavior
The app can't quit.
### Testcase Gist URL
https://github.com/electron/electron-quick-start
### Additional Information
The following is logged in the Terminal:
```
2023-03-30 09:06:20.327 Electron[66185:406517] not in fullscreen state
2023-03-30 09:06:22.768 Electron[66185:406517] not in fullscreen state
```
This feels high priority. | https://github.com/electron/electron/issues/37761 | https://github.com/electron/electron/pull/37803 | e9d5c3517cf3a6090826d7e8ca91c2d059a7caca | 3f42040a1b46f3c3bb338730f30f793ceadee386 | 2023-03-30T06:04:04Z | c++ | 2023-04-04T13:16:59Z |
closed | electron/electron | https://github.com/electron/electron | 37,737 | ["lib/browser/api/web-contents.ts", "shell/browser/web_contents_preferences.cc", "spec/api-browser-window-spec.ts", "spec/guest-window-manager-spec.ts", "spec/lib/screen-helpers.ts"] | [Bug]: setWindowOpenHandler changes default stylesheet for plain text files in dark theme | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
>=22.0.0
### What operating system are you using?
Windows
### Operating System Version
Windows 10 Enterprise 22H2
### What arch are you using?
x64
### Last Known Working Electron version
21
### Expected Behavior
Click on the link to open the plain text file in a new window.
Text should be readable, e.g. white text on dark background.
### Actual Behavior
Text is not readable. It's white text on white background.
This seems to only happen if there is a `setWindowOpenHandler`. If you comment out that line, a dark background is applied.
### Testcase Gist URL
https://gist.github.com/ajafff/8d9e8eb8cc0a7ec2297f2445c54f5526
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37737 | https://github.com/electron/electron/pull/36914 | d95f9d2c636775946717a95fefee1ada51524a79 | a26343f3e084028b920a7ea44e6b221350e3546c | 2023-03-28T10:17:49Z | c++ | 2023-05-02T21:44:34Z |
closed | electron/electron | https://github.com/electron/electron | 37,642 | ["shell/browser/api/electron_api_browser_view.cc", "shell/browser/api/electron_api_browser_view.h", "spec/api-browser-view-spec.ts"] | [Bug]: BrowserWindow.removeBrowserView will crash the app if the browserView is destroyed | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.4
### What operating system are you using?
macOS
### Operating System Version
13.0
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
BrowserWindow.removeBrowserView should remove the browserView regardless of whether browserView is destroyed or not.
### Actual Behavior
The app will exit with code SIGSEGV.
### Testcase Gist URL
```Javascript
const {app, BrowserWindow, BrowserView} = require('electron')
const path = require('path')
async function createWindow () {
// Create the browser window.
const mainWindow = new BrowserWindow({
width: 800,
height: 600,
webPreferences: {
preload: path.join(__dirname, 'preload.js')
}
})
const view = new BrowserView({
preload: path.join(__dirname, 'preload.js')
});
view.webContents.on('destroyed', () => {
mainWindow.removeBrowserView(view)
});
mainWindow.addBrowserView(view);
view.setBounds({x:0, y:0, width:300, height:300});
// and load the index.html of the app.
await mainWindow.loadFile('index.html')
await view.webContents.loadFile('index.html')
view.webContents.close();
// Open the DevTools.
// mainWindow.webContents.openDevTools()
}
// This method will be called when Electron has finished
// initialization and is ready to create browser windows.
// Some APIs can only be used after this event occurs.
app.whenReady().then(() => {
createWindow()
app.on('activate', function () {
// On macOS it's common to re-create a window in the app when the
// dock icon is clicked and there are no other windows open.
if (BrowserWindow.getAllWindows().length === 0) createWindow()
})
})
// Quit when all windows are closed, except on macOS. There, it's common
// for applications and their menu bar to stay active until the user quits
// explicitly with Cmd + Q.
app.on('window-all-closed', function () {
if (process.platform !== 'darwin') app.quit()
})
```
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37642 | https://github.com/electron/electron/pull/38842 | ce6fe040feefe326a557132fb81137097d4a39c1 | a00a25376d339ba78c4a1efbae2de05fd534332e | 2023-03-22T03:57:18Z | c++ | 2023-06-21T19:20:54Z |
closed | electron/electron | https://github.com/electron/electron | 37,628 | ["shell/browser/ui/cocoa/electron_ns_window.mm"] | [Bug]: BrowserWindows.on('swipe') doesn't trigger when swiping on BrowserViews | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.4
### What operating system are you using?
macOS
### Operating System Version
macOS Ventura 13.2.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
22.3.3
### Expected Behavior
```js
const win = new BrowserWindow({ width: 800, height: 600 })
const view = new BrowserView();
win.setBrowserView(view);
view.setBounds({ x: 0, y: 0, width: 300, height: 300 });
view.webContents.loadURL('https://electronjs.org');
win.on('swipe', (e, direction) => {
console.log('swipe', direction);
});
```
`BrowserWindow.on('swipe')` should be called whenever users swipe anywhere in the window (inside or outside of the `BrowserView` WebContents).
### Actual Behavior
`BrowserWindow.on('swipe')` is only called when swiping inside the `BrowserWindow` WebContents (outside of the `BrowserView`, not when swiping inside the `BrowserView`.
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37628 | https://github.com/electron/electron/pull/37946 | f03315ad4d16d0d066f7c3f8a35ab06d297fb440 | fceeb64e7c835868bec48442cfb4c7450bddf959 | 2023-03-21T00:56:00Z | c++ | 2023-04-13T09:28:22Z |
closed | electron/electron | https://github.com/electron/electron | 37,568 | ["shell/browser/api/electron_api_cookies.cc", "spec/api-net-spec.ts", "spec/api-session-spec.ts"] | [Bug]: Cookie is not set from parition when domain is present | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.0.0
### What operating system are you using?
macOS
### Operating System Version
macOS Venture 13.0.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
The following code when a valid session cookie is set in the rederer throws the following error. I am able to set a cookie when no domain is present.
```
(node:52646) UnhandledPromiseRejectionWarning: Error: Failed to parse cookie
```
```typescript
const createWindow = async (): Promise<void> => {
console.log();
const partitionSession = session.fromPartition("persist:federated");
const sessionCookie = store.get("__session");
if (sessionCookie) {
await partitionSession.cookies.set(sessionCookie);
}
// Create the browser window.
mainWindow = new BrowserWindow({
height: 600,
width: 800,
webPreferences: {
preload: MAIN_WINDOW_PRELOAD_WEBPACK_ENTRY,
/**
* The next two items are not normally set and required
* because of our nodejs usage
*/
nodeIntegration: true,
contextIsolation: false,
partition: "persist:federated",
session: partitionSession,
},
});
// and load the index.html of the app.
mainWindow.loadURL(MAIN_WINDOW_WEBPACK_ENTRY);
// Open the DevTools.
mainWindow.webContents.openDevTools();
const cookies = mainWindow.webContents.session.cookies;
cookies.on("changed", (event, cookie, cause, removed) => {
if (cookie.name === "__session")
store.set("__session", {
url: "http://localhost:3000",
name: cookie.name,
value: cookie.value,
domain: cookie.domain,
path: cookie.path,
secure: cookie.secure,
httpOnly: cookie.httpOnly,
sameSite: cookie.sameSite,
});
});
};
```
### Actual Behavior
A cookie should be set when the domain option is present.
### Testcase Gist URL
https://gist.github.com/8ee681e77bbcf7603642532fc27b0334
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37568 | https://github.com/electron/electron/pull/37586 | 48d0b09ad932aabc91ef072c862f3489202e40fd | b8f970c1c710c7e43cff6770fa845b96445cdaf8 | 2023-03-13T17:16:17Z | c++ | 2023-03-16T12:48:14Z |
closed | electron/electron | https://github.com/electron/electron | 37,565 | ["shell/browser/api/message_port.cc", "spec/api-ipc-spec.ts"] | [Bug]: Crash when attempting to transfer binary data objects through `MessagePortMain` | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.3
### What operating system are you using?
macOS
### Operating System Version
Ventura 13.2.1
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
Currently, `MessagePortMain` only supports transfer of other `MessagePortMain` instances. Attempting to transfer an object through `MessagePortMain` which the API doesn't support should safely error.
### Actual Behavior
The API correctly errors when attempting to transfer a variety of unsupported objects. However, when attempting to transfer an `ArrayBuffer` or a `TypedArray`, the main process immediately segfaults.
If the main process owns the channel's other port (i.e. it owns both ports) and the `app` `ready` event has been fired, then the main process logs the following message before it crashes:
```
[3841:0313/201342.125729:ERROR:child_thread_impl.cc(234)] Invalid PlatformChannel receive right
```
If a renderer owns the other port, or the `ready` event has not yet been fired, the segfault still occurs, but no message is logged.
It seems reasonable to assume that it's breaking specifically on binary data, but it's possible that it also breaks on other kinds of objects I have not tested.
Note that if the other port is owned by a renderer, the segfault doesn't occur if the renderer attempts to transfer an `ArrayBuffer` or `TypedArray` back to the main process. It only occurs if `MessagePortMain` instigates the transfer.
In case it's of interest, I do not appear to get any crash report dumped when trying to use `crashReporter`.
### Testcase Gist URL
https://gist.github.com/ed433ab70e6d3862675af9b1c2731f10
### Additional Information
Note that this issue is **not** about whether the API *should* support transfer of `ArrayBuffer`. This issue is just about the fact that given it currently doesn't, it should error instead of crashing.
Ping @nornagon | https://github.com/electron/electron/issues/37565 | https://github.com/electron/electron/pull/37585 | 1e106c8aa41b18b02ade44a54e6bd558d725fc37 | b27e4cae21547e7ddf3a6199d5bed0e621d4eee6 | 2023-03-13T11:23:34Z | c++ | 2023-03-27T17:56:55Z |
closed | electron/electron | https://github.com/electron/electron | 37,560 | ["docs/api/app.md", "shell/browser/api/electron_api_app.cc", "shell/browser/browser.h", "shell/browser/browser_mac.mm", "shell/common/platform_util.h", "shell/common/platform_util_mac.mm", "spec/api-app-spec.ts"] | [Bug]: Open at login with setLoginItemSettings in MAS build is not working | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
16.0.6
### What operating system are you using?
macOS
### Operating System Version
Version 13.2 (22D49)
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
Setting app to open at login with setLoginItemSettings api should work both locally and with MAS build.
Locally the app is added to login items but when i build it for MAS this doesn't work.
### Actual Behavior
Setting app to open at login with setLoginItemSettings api works locally but doesn't work in MAS build.
In MAS build the app is not added to login items.
<img width="477" alt="image" src="https://user-images.githubusercontent.com/17458685/224576455-a9145dfc-7121-4097-9fe4-a472f0c2ead4.png">
### Testcase Gist URL
_No response_
### Additional Information
According to docs https://www.electronjs.org/docs/latest/api/app#appsetloginitemsettingssettings-macos-windows setting login item in MAS build should work by using the following API
```js
app.setLoginItemSettings({
openAtLogin: openAtLogin,
})
```
In the docs on limitations of MAS build there is no mention of any limitation of using above API https://www.electronjs.org/docs/latest/tutorial/mac-app-store-submission-guide#limitations-of-mas-build
Here is possible related issue and PRs:
- https://github.com/electron/electron/issues/7312
- https://github.com/electron/electron/pull/11144
- https://github.com/electron/electron/pull/10856 | https://github.com/electron/electron/issues/37560 | https://github.com/electron/electron/pull/37244 | 6d0d350e138494cd68dd3d4fa73b2719da4575bf | f7b1c75c72a8dcab4157408f92bd3771606d8029 | 2023-03-12T22:11:32Z | c++ | 2023-10-16T16:25:11Z |
closed | electron/electron | https://github.com/electron/electron | 37,515 | ["shell/browser/api/electron_api_web_contents.cc"] | [Bug]: app drag regions capture clicks on windows with frame | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.1
### What operating system are you using?
Windows
### Operating System Version
Windows 10 22H2
### What arch are you using?
x64
### Last Known Working Electron version
22.2.1
### Expected Behavior
WebContents containing `-webkit-app-region: drag` should not be draggable or capture clicks when a window includes a frame.
### Actual Behavior
WebContents containing `-webkit-app-region: drag` captures clicks and is not draggable when window includes a frame.
### Testcase Gist URL
https://gist.github.com/samuelmaddock/ce68e2160f80c576bbf0bc3a8809363d
### Additional Information
This likely isn't a bug, but rather a change in Chromium that might need to be documented in our release notes.
@MarshallOfSound mentioned that `-webkit-app-region` was recently implemented for non-frameless windows in Chrome, likely leading to this change of behavior. | https://github.com/electron/electron/issues/37515 | https://github.com/electron/electron/pull/37594 | b72f81ab5b79bd86856bd46ad99a540fa082a4de | 4c6092e151b070db2b87358761db83b99be511a5 | 2023-03-06T23:01:15Z | c++ | 2023-03-28T14:52:28Z |
closed | electron/electron | https://github.com/electron/electron | 37,504 | ["shell/renderer/api/electron_api_context_bridge.cc"] | [Bug]: native iterators don't survive the contextbridge, but Symbol.iterator does | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.0
### What operating system are you using?
Windows
### Operating System Version
Windows 10 Pro Version 10.0.19044 Build 19044
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
We expected native iterables to survive the contextbridge, because custom iterable implementations do.
preload
```js
contextBridge.exposeInMainWorld("test_api", {
iterate(iterable) {
const values = [];
for (const item of iterable) {
values.push(item);
}
return values;
}
});
```
renderer
```js
function* generate() {
for (const i of [1, 2, 3]) {
yield i;
}
}
const iterable = generate();
const result = window.test_api.iterate(iterable);
```
### Actual Behavior
## Full example
See the full working example including test cases [here](https://github.com/MareikeTaeubner/mcve-contextbridge-symbols)
## What we try to do
We try to invoke a function `iterate` defined in the preload script with a parameter that is an iterator from the renderer process:
```js
contextBridge.exposeInMainWorld("test_api", {
iterate(iterable) {
const values = [];
for (const item of iterable) {
values.push(item);
}
return values;
}
});
```
## What we see
If we call this function with a native iterable from the renderer process, it will fail with `Error: iterable is not iterable`:
```js
function* generate() {
for (const i of [1, 2, 3]) {
yield i;
}
}
const iterable = generate();
const result = window.test_api.iterate(iterable);
```
However, if we implement the iterable ourselves, it will work. We see that, contrary to what the [documentation](https://www.electronjs.org/docs/latest/api/context-bridge#parameter--error--return-type-support) says, the `Symbol.iterator` survives.
```js
const values = [1, 2, 3];
let index = 0;
const iterable = {
[Symbol.iterator]: function () {
return {
next() {
if (index >= values.length) {
return { done: true };
} else {
return { value: values[index++] };
}
},
};
},
};
const result = window.test_api.iterate(iterable);
```
### Testcase Gist URL
[test sample as electron fiddle gist](https://gist.github.com/7f45335a82bdd797c439b9067ca723b4)
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37504 | https://github.com/electron/electron/pull/37593 | b27e4cae21547e7ddf3a6199d5bed0e621d4eee6 | 97b19a794676217de1bf1756f003fe7515b8c040 | 2023-03-06T12:55:35Z | c++ | 2023-03-28T00:36:55Z |
closed | electron/electron | https://github.com/electron/electron | 37,495 | ["docs/api/app.md"] | [Bug]: Window showing on 'open-url' with preventDefault() | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
21.3.3
### What operating system are you using?
macOS
### Operating System Version
macOS Monterey
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
The window should not show up.
### Actual Behavior
The window shows up.
### Testcase Gist URL
https://gist.github.com/vothvovo/00b6932c52d49c2ecf42853bd5db778a
### Additional Information
You have to build the gist to test it because otherwise the custom uri schema won't register. | https://github.com/electron/electron/issues/37495 | https://github.com/electron/electron/pull/37564 | 3b69a542fbb1a84a3d2b531a152f18e9e88b1666 | e480cb7103df6702d0364cf76d1f20724efc49dc | 2023-03-04T13:08:55Z | c++ | 2023-03-14T13:17:28Z |
closed | electron/electron | https://github.com/electron/electron | 37,487 | ["shell/browser/native_window_mac.mm"] | [Bug]: Restoring dock icon after creating multiple overlay windows creates duplicates | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [x] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.1, 24.0.0-alpha.7, 25.0.0-nightly.20230302
### What operating system are you using?
macOS
### Operating System Version
macOS Ventura 13.2
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
Creating several windows with `visibleOnFullScreen: true` and calling `app.dock.show()` should only show one icon in the dock.
### Actual Behavior
Creating several windows with `visibleOnFullScreen: true` and calling `app.dock.show()` has a timing issue: it allows multiple dock icons to appear and detach, surviving even after you close the application.
<img width="552" alt="screenshot: duplicate dock icons" src="https://user-images.githubusercontent.com/10053423/222592340-f68f4ecd-1ccd-4553-9540-245e2ac6de4a.png">
_These don't go away when you quit the app._
### Testcase Gist URL
https://gist.github.com/PsychoLlama/fcb2790372b6e4957beac9be7e543d3d
### Additional Information
I'm not very good with native code, but I think I've tracked it down. It appears to be similar to the bug described in `DockHide`: https://github.com/electron/electron/blob/ed7b5c44a2c7bfd57a44a04891452e967f07dd8e/shell/browser/browser_mac.mm#L426-L434
The mechanism for hiding the dock is here: https://github.com/electron/electron/blob/ed7b5c44a2c7bfd57a44a04891452e967f07dd8e/shell/browser/browser_mac.mm#L440-L444
It's pretty much the same implementation in `win.setVisibleOnAllWorkspaces`: https://github.com/electron/electron/blob/ed7b5c44a2c7bfd57a44a04891452e967f07dd8e/shell/browser/native_window_mac.mm#L1291-L1292
`dock.hide()` avoids the bug by aborting if you call it within 1 second of calling `dock.show()`. Since `SetVisibleOnAllWorkspaces` doesn't have a guard, calling `dock.show()` can result in the same bug, creating an orphan dock icon for almost every window.
---
Use case: We need to draw over windows we don't own, so we create a transparent overlay for each display. It's applied to every macOS space and has to render on top of fullscreen windows too:
```typescript
win.setVisibleOnAllWorkspaces(true, {
visibleOnFullScreen: true,
})
```
Calling that method implicitly hides the dock icon (I understand that's an unavoidable behavior), but we can restore it with `app.dock.show()`. I called that method after creating an overlay.
That's where I ran into the bug. If you have more than one display, we'll create multiple windows and call `dock.show()` multiple times, creating a race condition on the macOS APIs and generating duplicate dock icons. | https://github.com/electron/electron/issues/37487 | https://github.com/electron/electron/pull/37599 | 7ed3c7a359c78c725ba02eaf5edd28968e356a32 | eb613ef3d474f374874e5494a57bb06639811bdd | 2023-03-03T00:54:47Z | c++ | 2023-03-20T14:30:49Z |
closed | electron/electron | https://github.com/electron/electron | 37,476 | ["shell/browser/ui/cocoa/electron_ns_window.mm", "spec/api-browser-window-spec.ts"] | [Bug]: BrowserWindow.previewFile causes window to not update (appearing frozen) | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.1, 24.0.0-alpha.6, 25.0.0-nightly-20230301
### What operating system are you using?
macOS
### Operating System Version
macOS 13.2.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
n/a
### Expected Behavior
When using `BrowserWindow.previewFile` and then choosing to open the file from QuickLook, the `BrowserWindow` content should update/respond normally.
### Actual Behavior
When using `BrowserWindow.previewFile` and then choosing to open the file from QuickLook, the `BrowserWindow` content stops updating visually. The window isn't completely frozen though, as javascript is still running and clicking on the page initiates event handlers. Timers seem to fire more slowly, however.
### Testcase Gist URL
https://gist.github.com/a15492e41ae5c4dacb1152c35bea5cc1
### Additional Information
The [testcase fiddle](https://gist.github.com/a15492e41ae5c4dacb1152c35bea5cc1) updates a button's text to a random string every 10 milliseconds, queries the button's text, and prints it to the console.
This video shows that after opening a .log file in Console.app via `BrowserWindow.previewFile`, the button text stops visually updating in the window, but it does update as shown in the devtools console. And clicking on the button causes a console message to be logged as well.
https://www.loom.com/share/4fbd0f43c61b47f3ab00acf64d2a196f
This test case uses Console.app, but we have users also reporting this issue for opening .pdf files in Acrobat Pro. It doesn't reproduce for me with .txt files opening TextEdit.
Reproducing the bug also seems to require that the window not be in the default position. It might also require moving the other application window (Console.app) after it's opened. | https://github.com/electron/electron/issues/37476 | https://github.com/electron/electron/pull/37530 | e480cb7103df6702d0364cf76d1f20724efc49dc | bf1cc1aeb2a55a07d8ed794df1e6a630221bd69a | 2023-03-02T23:41:22Z | c++ | 2023-03-14T13:41:34Z |
closed | electron/electron | https://github.com/electron/electron | 37,465 | ["docs/tutorial/accessibility.md", "shell/browser/mac/electron_application.mm"] | [Bug]: `AXManualAccessibility` attribute can't be set (`kAXErrorAttributeUnsupported`) | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.1 (Electron Fiddle), 22.0.2 (demo app launched from Fiddle), 23.1.0 (Element), 19.1.9 (VS Code)
### What operating system are you using?
macOS
### Operating System Version
macOS Catalina 10.15.7
### What arch are you using?
x64
### Last Known Working Electron version
unknown
### Expected Behavior
As described [in the Electron docs](https://www.electronjs.org/docs/latest/tutorial/accessibility/#macos), it should be possible to set the `AXManualAccessibility` attribute[^axma] of an Electron-based application through the Accessibility API, to force Electron/Chromium to build the full tree of accessibility info.
[^axma]: Added in #10305. The private `AXEnhancedUserInterface` attribute (used by VoiceOver) also causes Electron/Chromium to build the full a11y tree, but has some unwanted side effects.
### Actual Behavior
Attempting to set `AXManualAccessibility` always results in an error (`kAXErrorAttributeUnsupported` or its equivalent), as if the attribute doesn't exist. I can replicate this in Swift, AppleScript (GUI scripting through the System Events app), and [Hammerspoon](https://github.com/Hammerspoon/hammerspoon) (through [the `hs.axuielement` module](https://www.hammerspoon.org/docs/hs.axuielement.html)). UIElementInspector also doesn't list it as an existing attribute.
### Testcase Gist URL
_No response_
### Additional Information
I believe this is a continuation of #30644, which was auto-closed as stale.
I haven't included a test case gist because this happens with _any_ Electron app I try - from the default code that Fiddle opens with, all the way up to VS Code.
I can see in electron_application.mm [where it checks if that attribute is what's being set](https://github.com/electron/electron/blob/9b20b3a722a7b807c161813cf2b616e74aee660d/shell/browser/mac/electron_application.mm#L182), but I don't see any code that tells the Accessibility API that that attribute actually _exists_ to be set. It looks to me like fixing that would involve overriding [the `accessibilityAttributeNames` method](https://developer.apple.com/documentation/objectivec/nsobject/1525181-accessibilityattributenames?language=objc) ([example here](https://developer.apple.com/library/archive/documentation/Cocoa/Conceptual/Accessibility/cocoaAXSupportingAttributes/cocoaAXSupportAttributes.html)). That method is from a deprecated informal protocol that suggests the `NSAccessibility` protocol instead, but:
a) `NSAccessibility` doesn't seem to have any replacement for `accessibilityAttributeNames` - the new protocol doesn't seem to support custom attributes (or custom anything, really).
b) `accessibilitySetValue:forAttribute:` - which Electron is already overriding - is also part of that deprecated informal protocol, and also doesn't seem to have a replacement.
#### Examples of various ways I've tried setting the attribute:
<details>
<summary>Swift example</summary>
```swift
#!/usr/bin/swift
import Cocoa
let name = CommandLine.arguments.count >= 2 ? CommandLine.arguments[1] : "Electron"
let pid = NSWorkspace.shared.runningApplications.first(where: {$0.localizedName == name})!.processIdentifier
let axApp = AXUIElementCreateApplication(pid)
let error = AXUIElementSetAttributeValue(axApp, "AXManualAccessibility" as CFString, true as CFTypeRef)
dump(error.rawValue)
// (couldn't get Swift to return the error value's name outside the REPL, so I'm using the rawValue)
// should be 0
// actually returns -25205 (AXError.attributeUnsupported)
```
</details>
<details>
<summary>AppleScript example</summary>
```applescript
#!/usr/bin/osascript
-- Note: you'll probably have to run this from Script Editor.
-- Running in the terminal will give the error "Not authorized to send Apple events to System Events."
on run argv
if (count argv) ≥ 1 then
set appName to first item of argv
else
set appName to "Electron"
end if
tell application "System Events"
tell application process appName
set value of attribute "AXManualAccessibility" to true
end tell
end tell
end run
```
</details>
<details>
<summary>Hammerspoon (Lua) example</summary>
```lua
-- paste this into Hammerspoon console
-- (all at once - if you run the lines separately, the locals will disappear between lines)
local appName = 'Electron' -- or whatever
local app = hs.application.get(appName)
local axApp = hs.axuielement.applicationElement(app)
return axApp:setAttributeValue('AXManualAccessibility', true)
-- returns:
-- nil Attribute is not supported by target
-- you can also do:
-- axApp.AXManualAccessibility = true
-- but that just fails silently
```
</details> | https://github.com/electron/electron/issues/37465 | https://github.com/electron/electron/pull/38102 | a0a44f07dda3562bcd7039e0270eae32be7549d6 | f35b9b3f06663e657a2f3b36cc021cf4b109ab26 | 2023-03-01T21:51:06Z | c++ | 2023-04-26T17:41:56Z |
closed | electron/electron | https://github.com/electron/electron | 37,463 | ["patches/chromium/desktop_media_list.patch", "shell/browser/api/electron_api_desktop_capturer.cc", "shell/browser/api/electron_api_desktop_capturer.h"] | [Bug]: DesktopCapturer doesn't work in Ubuntu 22.04.2 LTS | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.1
### What operating system are you using?
Ubuntu
### Operating System Version
22.04.2
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
Get access to screens for capture recording using desktopCapturer in electron Api
### Actual Behavior
There are difficulties to report the problem because no specific error is received in the terminal, only the message below
```bash
tiago@tghpereira:~/Área de Trabalho/electron/test$ yarn start
yarn run v1.22.19
$ electron .
/home/tiago/Área de Trabalho/electron/test/node_modules/electron/dist/electron exited with signal SIGSEGV
error Command failed with exit code 1.
```
I followed the official Electronjs documentation step by step and tried several approaches mentioned in forums to find ways to solve the problem, but without success.
Please help us with this issue which is greatly affecting many packages.
### Testcase Gist URL
https://gist.github.com/68498fa7437a78b9daf435f3413c9a43
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37463 | https://github.com/electron/electron/pull/38833 | 3e3152008ff569632e0b36d53ac9a24fae515709 | 905e41bbddd72f4d19c294b430d1a7c282e22a5d | 2023-03-01T16:20:13Z | c++ | 2023-07-11T08:21:11Z |
closed | electron/electron | https://github.com/electron/electron | 37,436 | ["docs/api/web-contents.md", "docs/fiddles/features/web-bluetooth/index.html", "docs/fiddles/features/web-bluetooth/main.js", "docs/fiddles/features/web-bluetooth/preload.js", "docs/fiddles/features/web-bluetooth/renderer.js", "shell/browser/lib/bluetooth_chooser.cc", "shell/browser/lib/bluetooth_chooser.h"] | [Bug]: How can I cancel a bluetooth request? | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23
### What operating system are you using?
Windows
### Operating System Version
win10
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
navigator.bluetooth.requestDevice return null after a while, if there is no device found.
### Actual Behavior
navigator.bluetooth.requestDevice always bolck if there is no device
### Testcase Gist URL
https://gist.github.com/a5d16943212179f7c033d5bd70531404
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37436 | https://github.com/electron/electron/pull/37601 | 42e7cd9b3f3e16d07f162716b4f1346e32004aec | 6a6908c4c887c5f685f5e172ff01afc02ebcbc65 | 2023-02-28T12:13:57Z | c++ | 2023-03-27T13:31:15Z |
closed | electron/electron | https://github.com/electron/electron | 37,424 | ["shell/browser/api/electron_api_web_contents.cc", "spec/api-web-contents-spec.ts"] | [Bug]: takeHeapSnapshot is always throwing an error | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.2
### What operating system are you using?
macOS
### Operating System Version
macOS 12.5
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
`process.takeHeapSnapshot(path)` and `window.webContents.takeHeapSnapshot(path)` should generate a heap snapshot at the given path.
### Actual Behavior
`Error: takeHeapSnapshot failed` is returned.
### Testcase Gist URL
https://gist.github.com/0302cbfbd787648265b60efe85b7c3d7
### Additional Information
This provides a minimum reproducible gist, same issue as https://github.com/electron/electron/issues/28758 | https://github.com/electron/electron/issues/37424 | https://github.com/electron/electron/pull/37434 | 8f2917db0169c81d293db7e32902252612ac68f5 | 9b20b3a722a7b807c161813cf2b616e74aee660d | 2023-02-27T14:55:15Z | c++ | 2023-03-01T15:50:36Z |
closed | electron/electron | https://github.com/electron/electron | 37,419 | ["shell/browser/api/electron_api_browser_view.cc", "shell/browser/api/electron_api_browser_window.cc", "shell/browser/api/electron_api_web_contents.cc", "spec/api-browser-view-spec.ts"] | [Bug]: Electron will crash when I call BrowserView.destroy(). | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.0
### What operating system are you using?
Windows
### Operating System Version
Windows 11 22H2 22621.1265
### What arch are you using?
x64
### Last Known Working Electron version
21.4.2
### Expected Behavior
Electron will not exit and browserView will close.
### Actual Behavior
Electron exited with code 3221225477.
### Testcase Gist URL
_No response_
### Additional Information
I publish fiddle to github failed. /(ㄒoㄒ)/~~
main.js
```
// Modules to control application life and create native browser window
const { app, BrowserWindow, BrowserView, ipcMain, dialog } = require('electron')
const path = require('path')
function createWindow() {
// Create the browser window.
const mainWindow = new BrowserWindow({
width: 800,
height: 600,
webPreferences: {
nodeIntegration: false,
contextIsolation: true,
preload: path.join(__dirname, 'preload.js')
}
})
// and load the index.html of the app.
mainWindow.loadFile('index.html')
const mainView = new BrowserView({
webPreferences: {
nodeIntegration: false,
contextIsolation: true,
}
})
mainView.webContents.loadURL('https://boardmix.cn/app/')
mainWindow.setBrowserView(mainView);
const windowBounds = mainWindow.getContentBounds();
const viewWidth = windowBounds.width - 2;
const viewHeight = windowBounds.height - 200 - 1;
mainView.setBounds({ x: 1, y: 200, width: viewWidth, height: viewHeight });
ipcMain.on('close', () => {
if (!mainView.webContents || mainView.webContents.isDestroyed()) {
dialog.showMessageBox({message: 'webContents is destroyed'})
} else {
mainView.webContents.destroy();
}
})
// Open the DevTools.
// mainWindow.webContents.openDevTools()
}
// This method will be called when Electron has finished
// initialization and is ready to create browser windows.
// Some APIs can only be used after this event occurs.
app.whenReady().then(() => {
createWindow()
app.on('activate', function () {
// On macOS it's common to re-create a window in the app when the
// dock icon is clicked and there are no other windows open.
if (BrowserWindow.getAllWindows().length === 0) createWindow()
})
})
// Quit when all windows are closed, except on macOS. There, it's common
// for applications and their menu bar to stay active until the user quits
// explicitly with Cmd + Q.
app.on('window-all-closed', function () {
if (process.platform !== 'darwin') app.quit()
})
// In this file you can include the rest of your app's specific main process
// code. You can also put them in separate files and require them here.
```
renderer.js
```
document.querySelector('h1').addEventListener('click', () => {
desktopManagerApi.close();
})
```
preload.js
```
const { contextBridge, ipcRenderer } = require('electron')
contextBridge.exposeInMainWorld('desktopManagerApi', {
close() {
ipcRenderer.send('close')
}
});
```
But, if i call `webContents.close()`, the `destroyed` event not be emitted. `webContents.isDestroyed()` is false. | https://github.com/electron/electron/issues/37419 | https://github.com/electron/electron/pull/37420 | 8fb0f430301578ba94b175f0b097fb2752f5ddf9 | 5e25d23794a4b3b1bda35c79fcc1b2fdd787ad69 | 2023-02-27T08:07:17Z | c++ | 2023-03-01T10:35:06Z |
closed | electron/electron | https://github.com/electron/electron | 37,417 | ["patches/node/.patches", "patches/node/lib_fix_broadcastchannel_initialization_location.patch"] | [Bug]: BroadcastChannel support broken in version 23 | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.1
### What operating system are you using?
macOS
### Operating System Version
Montery 12.6.3
### What arch are you using?
x64
### Last Known Working Electron version
22.3.1. Stops working in 23.0.0.
### Expected Behavior
Setting up a new BroadcastChannel() in two BrowserWindows with `{ contextIsolation: false }` shall work and communicated events to each other.
### Actual Behavior
BrowserChannels do not communicate between BrowserWindows when contextIsolation is false.
### Testcase Gist URL
[Repro](https://gist.github.com/dfahlander/cb9dfccd6e32101843800feade72049c)
Notes to repro:
1. Run it with Electron 22
2. Move the browser window so you see the other browser window behind
3. Click Post message button and see how messages arrive on the other window
4. Run it with Electron 23
5. Repeat step 2 and 3 but now messages aren't propagated to the other window.
### Additional Information
It seems as the BroadcastChannel we see in a window, is the NodeJS version of BroadcastChannel and not the DOM version because it has the `unref()` method that is unique to Node's version. I don't know if this could be the reason why it does not work as in Electron version 22, where BroadcastChannel is absent in the node thread but present and working fine within BrowserWindows.
| https://github.com/electron/electron/issues/37417 | https://github.com/electron/electron/pull/37421 | 2e03bdb9b624c24646b5104ce972e0800f622683 | 87f2a1d572a78c41da293eeb467966f76da7ba9d | 2023-02-26T23:39:10Z | c++ | 2023-02-28T22:26:37Z |
closed | electron/electron | https://github.com/electron/electron | 37,414 | ["shell/browser/ui/cocoa/electron_ns_window.mm"] | [Bug]: Frameless window is not draggable when running MAS build | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.1
### What operating system are you using?
macOS
### Operating System Version
macOS Big Sur
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
22.3.1
### Expected Behavior
Window should be draggable.
### Actual Behavior
Window is not draggable.
### Testcase Gist URL
https://gist.github.com/1387d3e19cbf2717cd21bbf3befaf159
### Additional Information
To be noted, dragging works as expected on standard Electron builds but consistently fails when using Mac Apple Store builds (>= v23).
@nornagon I understand you've been working on some related issues recently, please let me know if this is a known/fixed issue and I'll close this one, thanks! | https://github.com/electron/electron/issues/37414 | https://github.com/electron/electron/pull/37466 | 692876c73740d039ac840c91c719f7e9cf3b7e76 | 76c825d6193e6da540b64190ff2d2298d53da5ea | 2023-02-26T00:16:38Z | c++ | 2023-03-02T19:21:51Z |
closed | electron/electron | https://github.com/electron/electron | 37,404 | ["patches/chromium/fix_harden_blink_scriptstate_maybefrom.patch", "shell/common/node_bindings.cc", "shell/common/node_bindings.h", "shell/renderer/electron_renderer_client.cc", "shell/renderer/web_worker_observer.cc", "spec/chromium-spec.ts"] | [Bug]: proxyquire can't load modules after window.open method is called | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.0.0
### What operating system are you using?
Windows
### Operating System Version
Windows 11, Version22H2, OS Build 22621.1265
### What arch are you using?
x64
### Last Known Working Electron version
22.3.1
### Expected Behavior
proxyquire should load modules as it does with electron version 22.3.1.
### Actual Behavior
Attempt to use proxyquire in tests with electron of version 23 an higher leads to the following error message:
```
TypeError: Script methods can only be called on script instances.
at Script.runInThisContext (node:vm:129:12)
at Object.runInThisContext (node:vm:313:38)
at wrapSafe (node:internal/modules/cjs/loader:1083:15)
at Module._compile (node:internal/modules/cjs/loader:1130:27)
at Module._extensions..js (node:internal/modules/cjs/loader:1229:10)
at require.extensions.<computed> (C:\work\proxyquire-test\node_modules\proxyquire\lib\proxyquire.js:311:43)
at Module.load (node:internal/modules/cjs/loader:1044:32)
at Module._load (node:internal/modules/cjs/loader:885:12)
at f._load (node:electron/js2c/asar_bundle:2:13330)
at o._load (node:electron/js2c/renderer_init:2:3109)
at Module.require (node:internal/modules/cjs/loader:1068:19)
at Proxyquire._withoutCache (C:\work\proxyquire-test\node_modules\proxyquire\lib\proxyquire.js:222:12)
at Proxyquire.load (C:\work\proxyquire-test\node_modules\proxyquire\lib\proxyquire.js:129:15)
at Context.<anonymous> (C:\work\proxyquire-test\test.spec.js:7:5)
at callFn (C:\work\proxyquire-test\node_modules\mocha\lib\runnable.js:366:21)
at Runnable.run (C:\work\proxyquire-test\node_modules\mocha\lib\runnable.js:354:5)
at Runner.runTest (C:\work\proxyquire-test\node_modules\mocha\lib\runner.js:678:10)
at C:\work\proxyquire-test\node_modules\mocha\lib\runner.js:801:12
at next (C:\work\proxyquire-test\node_modules\mocha\lib\runner.js:593:14)
at C:\work\proxyquire-test\node_modules\mocha\lib\runner.js:603:7
at next (C:\work\proxyquire-test\node_modules\mocha\lib\runner.js:486:14)
at Immediate._onImmediate (C:\work\proxyquire-test\node_modules\mocha\lib\runner.js:571:5)
at process.processImmediate (node:internal/timers:471:21)
```
### Testcase Gist URL
https://gist.github.com/kyrylo-hrechykhin/f0afe7c469145c51d4df39fbea97ec95
### Additional Information
#### Workarounds:
- avoid calling window.open method completely
- separate electron-mocha runs to the one where window.open method is called and the one where proxyquire is used. If it happens in two different files, it is easy to do.
#### Details:
Issue is probably caused by node version upgrade in electron 23. But as it worked in electron 22, I consider it as a regression. I also could not reproduce this issue in non-electron environment.
#### Run testcase gist locally:
Run the following commands in the folder where all the files specified in the attached gist.
```
yarn
yarn start
```
Please upgrade/downgrade electron versions to see actual/expected results. | https://github.com/electron/electron/issues/37404 | https://github.com/electron/electron/pull/38754 | 4ab0a5ade47534db85d333e17f40530cc9726475 | 8874306dc021f2beb4c2b7cd49b8f3bec614e373 | 2023-02-24T16:47:53Z | c++ | 2023-07-18T08:41:50Z |
closed | electron/electron | https://github.com/electron/electron | 37,378 | ["shell/browser/api/electron_api_browser_view.cc", "shell/browser/api/electron_api_browser_window.cc", "shell/browser/api/electron_api_web_contents.cc", "spec/api-browser-view-spec.ts"] | [Bug]: No way to close a browserView | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.0
### What operating system are you using?
Windows
### Operating System Version
10.0.19042
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
There should be a way to close a browserView.
### Actual Behavior
Neither window.close() nor webContents.close() has any effect.
You can play a video on youtube and hear the sound continue after .removeBrowserView is called. So that does not change the behavior either. Also calling window.close() in the devtools has no effect.
```
const electron = require( "electron" )
electron.app.whenReady().then(()=>{
let main = new electron.BrowserWindow({
"height" : 700,
"width" : 800,
show: true,
});
let view = new electron.BrowserView({});
view.webContents.loadURL("https://www.youtube.com");
main.addBrowserView(view);
view.setBounds({x: 0, y: 0, width: 800, height: 700})
view.webContents.openDevTools({ mode: "detach" });
setTimeout(() => {
console.log("calling webContents.close()")
//main.removeBrowserView(view);
//view.webContents.close();
view.webContents.executeJavaScript("window.close()");
}, 5000);
})
```
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37378 | https://github.com/electron/electron/pull/37420 | 8fb0f430301578ba94b175f0b097fb2752f5ddf9 | 5e25d23794a4b3b1bda35c79fcc1b2fdd787ad69 | 2023-02-22T08:43:45Z | c++ | 2023-03-01T10:35:06Z |
closed | electron/electron | https://github.com/electron/electron | 37,376 | ["shell/browser/notifications/mac/cocoa_notification.mm"] | [Bug]: Notification using `hasReply` removes the first action | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.0
### What operating system are you using?
macOS
### Operating System Version
13.2.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
When showing a main process [Notification](https://www.electronjs.org/docs/latest/api/notification) (i.e. not the browser [Notification](https://developer.mozilla.org/en-US/docs/Web/API/Notification)) and using `hasReply: true` and `actions: [...]` then the first action is missing from the list.
Example:
```javascript
const notification = new Notification({
title: 'Hello 2',
subtitle: '#general',
body: 'blabla',
hasReply: true,
replyPlaceholder: 'Type your reply...',
actions: [
{ type: 'button', text: 'Action 1' },
{ type: 'button', text: 'Action 2' },
],
});
notification.show();
```
I expect to get this notification with `Reply`, `Action 1`, and `Action 2`:
### Actual Behavior
I get this notification:
### Testcase Gist URL
https://gist.github.com/stefansundin/bb92e0d0cf83439e176b3d970e533104
### Additional Information
I tested my fiddle with electron v20.0.0 and it had the same bug. | https://github.com/electron/electron/issues/37376 | https://github.com/electron/electron/pull/37381 | 1f390119fe82249a856965fadbcf4fab89766e67 | 8fb0f430301578ba94b175f0b097fb2752f5ddf9 | 2023-02-22T01:02:21Z | c++ | 2023-03-01T08:46:56Z |
closed | electron/electron | https://github.com/electron/electron | 37,356 | ["shell/browser/api/electron_api_browser_view.cc", "shell/browser/api/electron_api_browser_window.cc", "shell/browser/api/electron_api_web_contents.cc", "spec/api-browser-view-spec.ts"] | [Bug]: Destroyed event not emitted on close for custom BrowserView.webContents | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.3.0
### What operating system are you using?
Windows
### Operating System Version
Windows 10 version 21H2
### What arch are you using?
x64
### Last Known Working Electron version
22.2.1
### Expected Behavior
I've created a custom BrowserView, added it to my main BrowserWindow via `addBrowserView()`, and navigated its webContents to a file. Upon receiving a message from the renderer process, I attach a handler to the webContents' 'destroyed' event, then close the contents via `webContents.close()`. Since I'm omitting the 'waitForBeforeUnload' option in my close() call, I expect the 'destroyed' event to be emitted, and my attached handler to run.
### Actual Behavior
The 'destroyed' event is not emitted, and my attached handler never run, until my main BrowserWindow is closed.
### Testcase Gist URL
https://gist.github.com/mgalla10/cdff38e750ae49d01500445b019493ca
### Additional Information
To repro, run the attached gist with v22.2.1 and click the "Close" button. After dismissing a message box saying "Closing page 2", that page will disappear and another "Page 2 closed" message box will appear. Run the gist again with v22.3.0 and notice that after dismissing the first message box, the page doesn't seem to disappear, and the second message box never appears. | https://github.com/electron/electron/issues/37356 | https://github.com/electron/electron/pull/37420 | 8fb0f430301578ba94b175f0b097fb2752f5ddf9 | 5e25d23794a4b3b1bda35c79fcc1b2fdd787ad69 | 2023-02-20T21:20:50Z | c++ | 2023-03-01T10:35:06Z |
closed | electron/electron | https://github.com/electron/electron | 37,352 | ["docs/api/web-contents.md", "shell/browser/api/electron_api_web_contents.cc", "spec/api-web-contents-spec.ts"] | [Feature Request]: Add event to WebContents to react to changes of `isCurrentlyAudible` | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a feature request that matches the one I want to file, without success.
### Problem Description
Chromium only shows the button to mute/unmute a tab when it actually has media playing. Once the media has paused, the button will eventually fade away. Electron does not offer a way to react to this change.
### Proposed Solution
Either a single change event like `audible-change` or two symmetrical events along the lines of `did-become-audible` and `stopped-being-audible`.
### Alternatives Considered
Poll `isCurrentlyAudible`
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37352 | https://github.com/electron/electron/pull/37366 | c8f715f9a1c116ccb3796e3290fea89989cc0f6e | 512e56baf75bce86aef11184af881cff67d8e6a2 | 2023-02-20T12:29:43Z | c++ | 2023-03-06T16:00:24Z |
closed | electron/electron | https://github.com/electron/electron | 37,340 | ["docs/api/native-image.md", "docs/breaking-changes.md", "shell/common/api/electron_api_native_image_win.cc", "spec/api-native-image-spec.ts", "spec/fixtures/assets/capybara.png"] | [Bug]: getSize() is incorrect if the nativeImage is created by createThumbnailFromPath() | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
21.0.0
### What operating system are you using?
macOS
### Operating System Version
macOS 13.2.1 (22D68)
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
```js
const image = await nativeImage.createThumbnailFromPath(filePath, {width: 2000, height:2000})
const size = image.getSize() //<- expect the real thumbnail size, but it's always {2000, 2000}
```
getSize() is incorrect if the nativeImage is created by createThumbnailFromPath()
### Actual Behavior
size = the maxSize passed in
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37340 | https://github.com/electron/electron/pull/37362 | f33bf2a27113bd9cd92bdd35233107779fc79f47 | 8ee58e18fd31a7d33ccab78820d61bab9ec1ec69 | 2023-02-19T03:16:15Z | c++ | 2023-03-09T02:48:29Z |
closed | electron/electron | https://github.com/electron/electron | 37,311 | ["shell/browser/api/electron_api_web_contents.cc", "shell/browser/native_window_mac.mm", "spec/api-browser-window-spec.ts"] | [Bug]: html fullscreen not working post v18 | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.0
### What operating system are you using?
macOS
### Operating System Version
macOS Monterey 12.6.3
### What arch are you using?
x64
### Last Known Working Electron version
18.3.15
### Expected Behavior
With the `BroswerWindow` set to `fullscreenable: false`, when hitting the fullscreen button on a video (on YouTube, Tubi, etc...) the video should 'fullscreen' to the window. When clicking it again, it should restore.
### Actual Behavior
In some cases (e.g on YouTube) the fullscreen works and `enter-html-full-screen` is emitted. When clicking it again the video does not restore though `leave-html-full-screen` will be emitted.
Any subsequent press of the video's fullscreen button will do nothing and emit nothing.
In other cases (e.g. sometimes on Tubi) neither fullscreen or the restore works at all.
### Testcase Gist URL
https://gist.github.com/jtvberg/8673c71b5dc489a285ac405d6bec8540
### Additional Information
This stopped working as expected somewhere between v18 and v19.
I have also tried v20-v23.
I am referring to the button in actual video host not the fullscreen button on the window. | https://github.com/electron/electron/issues/37311 | https://github.com/electron/electron/pull/37348 | 868676aa5ccf68793333d40124e49f0d0b175246 | 32c60b29bbd9abcaedbe9d65b2a4eae78e1c2fca | 2023-02-16T21:04:43Z | c++ | 2023-02-21T11:11:34Z |
closed | electron/electron | https://github.com/electron/electron | 37,305 | ["shell/browser/ui/cocoa/electron_ns_window_delegate.mm"] | [Bug]: Aspect Ratio locking and MinHeight/Width appear incompatible | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.0
### What operating system are you using?
macOS
### Operating System Version
13.2
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
IF a BW has minHeight of 300
AND an aspect ratio of X.
The BW should never get smaller than 300px tall.
### Actual Behavior
It's tough to describe but it appears as though minHeight is either ignored, or being scaled in some way.
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37305 | https://github.com/electron/electron/pull/37306 | 49df19214ea3abaf0ad91adf3d374cba32abd521 | 8f2917db0169c81d293db7e32902252612ac68f5 | 2023-02-16T13:45:48Z | c++ | 2023-03-01T15:50:14Z |
closed | electron/electron | https://github.com/electron/electron | 37,301 | ["filenames.gni", "shell/browser/api/electron_api_web_contents.h", "shell/browser/ui/cocoa/delayed_native_view_host.cc", "shell/browser/ui/cocoa/delayed_native_view_host.h", "shell/browser/ui/cocoa/delayed_native_view_host.mm", "shell/browser/ui/cocoa/electron_inspectable_web_contents_view.h", "shell/browser/ui/cocoa/electron_inspectable_web_contents_view.mm"] | [Bug]: context menus cannot be displayed when "-webkit-app-region: drag" is enabled | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.1.0
### What operating system are you using?
macOS
### Operating System Version
macOS Ventura 13.2.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
22.x
### Expected Behavior
I want to show the context menu in "-webkit-app-region: drag" enabled area,
### Actual Behavior
No response and no reaction.
### Testcase Gist URL
https://gist.github.com/3fe6a69c1eba1d7f85caf7572ba77fd2
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37301 | https://github.com/electron/electron/pull/37386 | a3e3efe4c4d6ed2b40f89eafbd034d8a1744a3a7 | e27905c7654e119464ba149f1643cd2770575159 | 2023-02-16T08:25:26Z | c++ | 2023-02-24T00:05:30Z |
closed | electron/electron | https://github.com/electron/electron | 37,293 | ["docs/api/session.md", "shell/browser/electron_browser_context.cc"] | [Feature Request]: Expose setDisplayMediaRequestHandler disable_local_echo flag to enable audio output during recording | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a feature request that matches the one I want to file, without success.
### Problem Description
- We want to record the electron browser window for both audio and video.
- The `BrowserWindow` instance loads an external URL, which calls `window.navigator.mediaDevices.getDisplayMedia()` at some point.
- We've attached a handler to `setDisplayMediaRequestHandler()` in our preload, which returns the following: `callback({ video: source, audio: mainBrowser.webContents.mainFrame });`
* The video source is obtained through `desktopCapturer`, but I don't think that has anything to do with this bug.
- When we assemble the stream to the `MediaRecorder` and perform recording, the browser window loses audio output when the recording starts, and resumes output when the recording stops.
- The recording itself properly records video and any audio made from the window during that time.
After a lot of investigation, I noticed a flag here in the PR that implements `setDisplayMediaRequestHandler()` (#30702):
[shell/browser/electron_browser_context.cc](https://github.com/electron/electron/commit/221bb513265128e004a0469f3b2b1c89387ba56f#diff-fa823350727ad86e0ac08c18d6f17e24a62d653b43a08242e4608f5a8ee8e5b9R501)
```
} else if (result_dict.Get("audio", &rfh)) {
devices.audio_device = blink::MediaStreamDevice(
request.audio_type,
content::WebContentsMediaCaptureId(rfh->GetProcess()->GetID(),
rfh->GetRoutingID(),
/* disable_local_echo= */ true)
```
And thought `disable_local_echo` might control audio loopback behavior.
I compiled a custom version of electron with that flag set to `false` and repeated the above steps, which correctly allowed audio to come through the speakers while recording was happening. The recording itself retained video and audio.
I am on an M2 Mac running Ventura 13.2.1.
(_I have zero C++ experience and just used intuition to find this while looking at the PR for `setDisplayMediaRequestHandler` and followed the electron build guides to test my hypothesis._)
### Proposed Solution
May I ask that an option be exposed for `setDisplayMediaRequestHandler` that controls the value of that particular flag? Or, have the option be part of the callback for the handler.
### Alternatives Considered
Have not discovered any alternatives. Literature around `setDisplayMediaRequestHandler` is limited as it's a recently-released feature.
### Additional Information
This bug ticket describes the behavior we experienced, although the context is a bit different (It references `getUserMedia` and `chrome.tabCapture`.
https://bugs.chromium.org/p/chromium/issues/detail?id=1403733 | https://github.com/electron/electron/issues/37293 | https://github.com/electron/electron/pull/37315 | 5e25d23794a4b3b1bda35c79fcc1b2fdd787ad69 | 49df19214ea3abaf0ad91adf3d374cba32abd521 | 2023-02-16T03:19:19Z | c++ | 2023-03-01T10:37:19Z |
closed | electron/electron | https://github.com/electron/electron | 37,292 | ["patches/chromium/mas_no_private_api.patch"] | [MAS Rejection]: Symbols: _sandbox_compile_string, _sandbox_free_profile, _sandbox_free_params, _sandbox_create_params, _sandbox_set_param | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
### Electron Version
23.0.0
### Rejection Email
Guideline 2.5.1 - Performance - Software Requirements
Your app uses or references the following non-public or deprecated APIs:
Symbols: _sandbox_compile_string, _sandbox_free_profile, _sandbox_free_params, _sandbox_create_params, _sandbox_set_param
The use of non-public or deprecated APIs is not permitted on the App Store, as they can lead to a poor user experience should these APIs change and are otherwise not supported on Apple platforms.
Next Steps
It would be appropriate to revise your binary and remove any references to the non-public or deprecated APIs identified above.
If you are using third-party libraries, update to the most recent version of those libraries. If you do not have access to the libraries' source, the following command line tools can help you identify the location of problematic code:
- The "strings" tool can output a list of the methods the library calls.
- The "otool -ov" tool will output the Objective-C class structures and their defined methods.
Resources
- We are constantly reevaluating and identifying non-public APIs that you may have been using for an extended period of time. Always use [public APIs and frameworks](https://developer.apple.com/documentation/) and ensure they are up to date.
- If there are no alternatives for providing the functionality your app requires, you can use [Feedback Assistant](https://feedbackassistant.apple.com/welcome) to submit an enhancement request.
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37292 | https://github.com/electron/electron/pull/37309 | a92fd2aa05c0cfcb5809268eb8a4f1d4d3f91074 | 85cf56d80b9073091303f807e584362501bb8fda | 2023-02-16T02:14:18Z | c++ | 2023-02-21T10:44:18Z |
closed | electron/electron | https://github.com/electron/electron | 37,162 | ["shell/browser/api/electron_api_browser_window.cc", "shell/browser/api/electron_api_web_contents.cc"] | [Bug]: window.close() destroys all child BrowserView-s even if window's 'beforeunload' handler has 'e.returnValue = false' | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.2.0
### What operating system are you using?
Windows
### Operating System Version
Windows 10
### What arch are you using?
x64
### Last Known Working Electron version
21.4.1
### Expected Behavior
If 'beforeunload' event listener for a window has
`event.returnValue = false`
non of it's child Browser Views are destroyed/closed
### Actual Behavior
If 'beforeunload' event listener for a window has
`event.returnValue = false`
all of it's child Browser Views will be detroyed
### Testcase Gist URL
https://gist.github.com/YauheniBH-EF/190ea8ff42c09a15763705ed4647b02e
### Additional Information
I have a feeling that it's related to https://github.com/electron/electron/pull/35509 this PR changes, but i'm not proficient enough in Electron/C++ to check/fix it by myself. | https://github.com/electron/electron/issues/37162 | https://github.com/electron/electron/pull/37205 | 4d6f230d2108c19ac862577e83443a2eec53a183 | 8eee4f2df1982ae52f1ed1a7ca3e2183ba701c4f | 2023-02-07T11:35:20Z | c++ | 2023-02-14T17:40:37Z |
closed | electron/electron | https://github.com/electron/electron | 37,158 | ["docs/api/session.md", "spec/chromium-spec.ts"] | [Bug]: There is no `clipboard-sanitized-write` permission listed in docs | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
17.0.0
### What operating system are you using?
macOS
### Operating System Version
Does not matter
### What arch are you using?
x64
### Last Known Working Electron version
16.0.0
### Expected Behavior
This is not a bug.
I noticed my webContents requests permission named `clipboard-sanitized-write` when attempting to write data to clipboard.
I did a search in docs https://www.electronjs.org/docs/latest/api/session#sessetpermissionrequesthandlerhandler and found only `clipboard-read` permission.
It would be great to update the documentation :)
Expected: there is `clipboard-sanitized-write` in docsthe
### Actual Behavior
There is no `clipboard-sanitized-write` permission in docs.
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37158 | https://github.com/electron/electron/pull/37173 | 478ce969143cd8d069ab4def0992e8d5719445cf | e5e9186d613277c855d913fa21d0f03496c2ff87 | 2023-02-07T09:06:56Z | c++ | 2023-02-09T10:38:39Z |
closed | electron/electron | https://github.com/electron/electron | 37,127 | ["docs/api/web-contents.md", "lib/browser/api/web-contents.ts"] | [Bug]: A6 not accepted as pageSize in settings of webContents.print | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.2.0
### What operating system are you using?
macOS
### Operating System Version
Ventura 13.0.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
web content.print to accept A6 as pageSize in settings
### Actual Behavior
throwing an error : Error: Unsupported pageSize: A6
at _.print (node:electron/js2c/browser_init:2:76160)
### Testcase Gist URL
_No response_
### Additional Information
You need to add in this file electron/lib/browser/api/web-contents.ts, to the constant PDFPageSizes the A6 configuration | https://github.com/electron/electron/issues/37127 | https://github.com/electron/electron/pull/37159 | cb03c6516b2729ddf915314d5a72fbacb772a665 | 889859df5bb74e6c18b1c2f21c28ade80a366611 | 2023-02-03T10:09:02Z | c++ | 2023-02-14T10:44:34Z |
closed | electron/electron | https://github.com/electron/electron | 37,090 | ["docs/api/web-contents.md", "docs/fiddles/features/web-bluetooth/index.html", "docs/fiddles/features/web-bluetooth/main.js", "docs/fiddles/features/web-bluetooth/preload.js", "docs/fiddles/features/web-bluetooth/renderer.js", "shell/browser/lib/bluetooth_chooser.cc", "shell/browser/lib/bluetooth_chooser.h"] | [Bug]: Web Bluetooth: Always cancelled request after first dismiss | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.1.0
### What operating system are you using?
Windows
### Operating System Version
Windows 10 Pro 22H2
### What arch are you using?
x64
### Last Known Working Electron version
None
### Expected Behavior
Even if it didn't found any devices, after i dismiss the request, it should perform another scan next time i call requestDevice.
### Actual Behavior
After first dismiss the requestDevice, all later navigator.bluetooth.requestDevice return User cancelled immediatly.
### Testcase Gist URL
https://gist.github.com/111927e2d69b53c6c151eb6aa5070948
### Additional Information
I have create a simple tst code on Fiddle, see above.
I have alreadly check #10763, who seem like having the same problem but somehow he never repoduce it anymore. But i think i found the issue.
It's really simple to reproduce infact, just get "select-bluetooth-device" event, and run navigator.bluetooth.requestDevice({ filters: [{ services: [0xffff] }] }) in console, and just run it again, it will return user cancelled immediatly.
In fact, there's kinda a way to prevent that happend, is to call the callback in "select-bluetooth-device" event with an empty string when you need to cancel the discovery.
But, that's the main issue here, if there's no filtered device found at all, then you never get the callback function to call with an empty string. if then you dismiss the request, it will keep returning "user cancelled" immediatly everytime you call requesrDevice later.
the Fiddle example is demonstrating this. click scan, if there's no device found, click cancel, console will prompt "User cancelled the requestDevice()". if you uncomment the "acceptAllDevices" in renderer.js, which will found all the device, will not have this problem.
Is there a way to get the callback funtion? or a better way to cancel the discovery?
Thank you. | https://github.com/electron/electron/issues/37090 | https://github.com/electron/electron/pull/37601 | 42e7cd9b3f3e16d07f162716b4f1346e32004aec | 6a6908c4c887c5f685f5e172ff01afc02ebcbc65 | 2023-02-01T06:01:03Z | c++ | 2023-03-27T13:31:15Z |
closed | electron/electron | https://github.com/electron/electron | 37,049 | ["patches/chromium/printing.patch"] | [Bug]: Main process crash when trying to open print dialog | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23
### What operating system are you using?
Windows
### Operating System Version
Windows 11 22H2
### What arch are you using?
x64
### Last Known Working Electron version
22.1.0
### Expected Behavior
Opens print dialog.
### Actual Behavior
Main process crash, tested on macos and windows 11.
### Testcase Gist URL
https://gist.github.com/thomasLechaptois/7169a5188de0aa448b3e18ab813a41f8
### Additional Information
All Electron 23 alpha and beta up to beta 5 are affected by this crash.
| https://github.com/electron/electron/issues/37049 | https://github.com/electron/electron/pull/37052 | fcc7a869f218059d0de22db1ea3d63c2e9c192ce | ce35bda80580cd355f5cdf31d39f9b4da82822e4 | 2023-01-27T14:25:35Z | c++ | 2023-01-31T11:06:11Z |
closed | electron/electron | https://github.com/electron/electron | 37,040 | ["docs/api/browser-window.md"] | [Feature Request]: Allow clearing of aspect ratio | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a feature request that matches the one I want to file, without success.
### Problem Description
`window.setAspectRatio` allows us to specify the resizing aspect ratio for a window, but how can we reset this value to allow any aspect ratio? Perhaps `window.clearAspectRatio` (which might also warrant the addtion of `window.getAspectRatio`).
### Proposed Solution
I want to go between enforcing and not enforcing aspect ratio, e.g.:
```
window.setAspectRatio(bounds);
// later...
if (window.getAspectRatio()) {
window.clearAspectRatio();
}
```
### Alternatives Considered
Current workaround is to destroy the window once it has aspect ratio set and re-create a new one without aspect ratio.
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37040 | https://github.com/electron/electron/pull/37074 | 01b4e3b521a46df1c9e810da2e6244767fd6288e | 730a07ad623cfb72b7028b368a399a6fe593f740 | 2023-01-26T23:10:27Z | c++ | 2023-01-31T20:36:09Z |
closed | electron/electron | https://github.com/electron/electron | 37,038 | ["shell/renderer/electron_renderer_client.cc", "shell/renderer/web_worker_observer.cc", "shell/renderer/web_worker_observer.h", "spec/fixtures/crash-cases/worker-multiple-destroy/index.html", "spec/fixtures/crash-cases/worker-multiple-destroy/index.js", "spec/fixtures/crash-cases/worker-multiple-destroy/worklet.js"] | [Bug]: nodeIntegrationInWorker: true makes the app crash using AudioWorklet | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.1.0
### What operating system are you using?
MacOS
### Operating System Version
Ventura 13.2
### What arch are you using?
arm64 (and x64)
### Last Known Working Electron version
9
### Expected Behavior
The app shouldn't crash.
### Actual Behavior
Setting `nodeIntegrationInWorker: true` will make the app crash when using an AudioWorklet.
I forked the `electron-quickstart` project in order to be able to easily reproduce the issue.
The changes can be found here: https://github.com/electron/electron-quick-start/compare/master...rmnbrd:electron-quick-start:issue-node-integration-in-worker-and-audio-worklet
**Reproduction steps**
- Clone the example I've linked right above
- Run `yarn && yarn start`
- Click the `Start AudioWorklet` button **twice**
- Click the `Reload` button
- **App crashes**
### Testcase Gist URL
https://github.com/electron/electron-quick-start/compare/master...rmnbrd:electron-quick-start:issue-node-integration-in-worker-and-audio-worklet
### Additional Information
We have tested on several versions of Electron (22, 21, 20, 19).
It is worth noting that:
- We could not experience that issue on v9 and below.
- The bug is not reproducible on v22 if `nodeIntegrationInWorker` is set to `false`.
- The bug is not reproducible on Chrome 109.
---
Here you will find the complete crash report:
https://pastebin.com/wrKshgK3 | https://github.com/electron/electron/issues/37038 | https://github.com/electron/electron/pull/37050 | ce35bda80580cd355f5cdf31d39f9b4da82822e4 | 23739c644ba8dfa6cec5417cdf4e2550ca4b94b8 | 2023-01-26T15:43:30Z | c++ | 2023-01-31T11:29:29Z |
closed | electron/electron | https://github.com/electron/electron | 37,028 | [".github/workflows/release_dependency_versions.yml"] | GH Workflow Fails to Trigger `mksnapshot` Release for `23-x-y` Prereleases | Example failure: https://github.com/electron/electron/actions/runs/3991558400
I think there are two issues here:
* First looks like a permission error, possibly because [the permissions for the workflow](https://github.com/electron/electron/blob/a30a9c7c4faa5aca476dd593dd02bbfac43432df/.github/workflows/release_dependency_versions.yml#L10-L11) are too restrictive? Not sure what the correct permission is to trigger a cross-repo workflow.
* It looks like [`update-version.js` in `mksnapshot`](https://github.com/electron/mksnapshot/blob/ca16d69d8fd1257caceac224aa5397d12d95ae64/script/update-version.js#L4) would reject prerelease versions | https://github.com/electron/electron/issues/37028 | https://github.com/electron/electron/pull/37036 | fd761ec8f74d8a4c77864ab86ef6994a5fb72e1d | 2dc76d0d8087bdd882cf3e0aa8dc4e4cd643e9b9 | 2023-01-26T00:52:01Z | c++ | 2023-02-01T18:12:19Z |
closed | electron/electron | https://github.com/electron/electron | 37,008 | ["shell/browser/native_window.cc", "shell/browser/native_window_views.cc"] | [Bug]: Unable to resize, with cursor drag, from edge of window that is set to -webkit-app-region:drag | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.0.0-beta.5
### What operating system are you using?
Windows
### Operating System Version
Windows 11 Pro (10.0.22621 Build 22621)
### What arch are you using?
x64
### Last Known Working Electron version
22.0.3
### Expected Behavior
User is able to resize window from the edge of window that is set to `webkit-app-region:drag` , by dragging from this edge. (reproducible up till version 22.0.3)
### Actual Behavior
User is unable to resize window from the edge of window that is set to `webkit-app-region:drag` , by dragging from this edge.
https://user-images.githubusercontent.com/15127636/214312123-018f7ea2-2e9c-4e6c-8a8f-1489943007c3.mp4
### Testcase Gist URL
https://gist.github.com/takumus/9054bde77c8206143232fed2659bf108
### Additional Information
_No response_ | https://github.com/electron/electron/issues/37008 | https://github.com/electron/electron/pull/37016 | 1486cbdf6410e569f619d1ca7ceea0dd3380bf02 | 0026fdb78a68c6a4067191826bd1bfc1d9700998 | 2023-01-24T14:17:18Z | c++ | 2023-01-26T13:04:19Z |
closed | electron/electron | https://github.com/electron/electron | 36,970 | ["patches/node/fix_expose_the_built-in_electron_module_via_the_esm_loader.patch"] | [Bug]: Regression: #35930 broke CJS via ESM loading when exporting null/undefined with ELECTRON_RUN_AS_NODE | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.3
### What operating system are you using?
Ubuntu
### Operating System Version
Ubuntu 22.10
### What arch are you using?
x64
### Last Known Working Electron version
20.3.1 / 21.1.0
### Expected Behavior
1. `npm i [email protected] ava`
2. `ELECTRON_RUN_AS_NODE=true ./node_modules/.bin/electron ./node_modules/.bin/ava`
3. Works (can't find any tests obviously)
I was successfully running my esm tests via `ELECTRON_RUN_AS_NODE` until now
### Actual Behavior
1. `npm i [email protected] ava`
2. `ELECTRON_RUN_AS_NODE=true ./node_modules/.bin/electron ./node_modules/.bin/ava`
3. Crashes
```
node:internal/modules/esm/translators:178
if (!ObjectPrototypeHasOwnProperty(exports, exportName) ||
^
TypeError: Cannot convert undefined or null to object
at hasOwnProperty (<anonymous>)
at ModuleWrap.<anonymous> (node:internal/modules/esm/translators:178:12)
at ModuleJob.run (node:internal/modules/esm/module_job:193:25)
at async Promise.all (index 0)
at async ESMLoader.import (node:internal/modules/esm/loader:533:24)
at async loadESM (node:internal/process/esm_loader:91:5)
at async handleMainPromise (node:internal/modules/run_main:65:12)
```
### Testcase Gist URL
_No response_
### Additional Information
I was able to track this down and repro without ava as well (https://github.com/avajs/ava/discussions/3155). It appears you can no longer export `null` because it crashes the esm loader patch.
index.mjs
```js
import m from './module.cjs';
```
module.cjs
```js
module.exports = null;
```
`ELECTRON_RUN_AS_NODE=true ./node_modules/.bin/electron index.mjs`
It broke between `20.3.1` and `20.3.2` because of #35930 . If you're curious, in ava this comes from this dep https://github.com/jamiebuilds/ci-parallel-vars/blob/994940e4882bdddf046a5c076b49d700449e4b38/index.js#L41-L56
> Weirdly, this is like the 4th time I've tried this.
@MarshallOfSound fifth time is the charm | https://github.com/electron/electron/issues/36970 | https://github.com/electron/electron/pull/37009 | 4bc6b15f53a3e4f961c2ade3c96fa9229ec00367 | 58beec1da2cacd59a91a556b86f47ec0147b9c36 | 2023-01-20T09:12:15Z | c++ | 2023-01-25T21:03:47Z |
closed | electron/electron | https://github.com/electron/electron | 36,906 | ["shell/browser/ui/autofill_popup.cc"] | [Bug]: Datalist positioning is broken in Electron v22.x.y | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.x.y
### What operating system are you using?
Windows
### Operating System Version
Windows 10 build 19045.2486
### What arch are you using?
x64
### Last Known Working Electron version
21.3.4
### Expected Behavior
I expect that a datalist when displayed, should be positioned horizontally in line with it's bound input box; either above or below depending on space available.
See Image below from Electron fiddle using Electron v21.3.4
__
### Actual Behavior
In Electron v22.0..0+ The datalist is positioned offset to the left of its bound input box by quite a distance. In some instance it can appear as if the datalist is clamped to the left margin.
See Image below from Electron fiddle using Electron v22.0.0 although the same can be reproduced using any of the currently available v22.0.y -> 22.0.1, 22.0.2 And a quick test with v23.0.0-beta2 also exhibity the same behaviour.
### Testcase Gist URL
https://gist.github.com/e50cb256d567ed25b7bbf12b9cbe5417
### Additional Information
_No response_ | https://github.com/electron/electron/issues/36906 | https://github.com/electron/electron/pull/36934 | 9630e26e6dec503bc82ca4b9f87ebf0344311ca5 | 55c818d0a84400621c0817c8a2e4cc17d9eb24c8 | 2023-01-13T11:34:12Z | c++ | 2023-01-19T18:44:23Z |
closed | electron/electron | https://github.com/electron/electron | 36,858 | ["patches/chromium/fix_harden_blink_scriptstate_maybefrom.patch", "shell/common/node_bindings.cc", "shell/common/node_bindings.h", "shell/renderer/electron_renderer_client.cc", "shell/renderer/web_worker_observer.cc", "spec/chromium-spec.ts"] | [Bug]: Uncaught illegal state exception and renderer crash with node immediates and closing child window | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
13.x, 17.x, 22.0.0
### What operating system are you using?
Windows
### Operating System Version
Windows 11 (22621.963)
### What arch are you using?
x64
### Last Known Working Electron version
Unknown
### Expected Behavior
Closing a child window should not produce an 'Uncaught illegal state' exception. Furthermore, it shouldn't cause the renderer process to exit with code 7 (which can specifically occur if there is a node native immediate being used, see additional information below).
### Actual Behavior
A race condition can occur when using `setImmediate` (or using a node module that sets up native immediates) in a node environment and closing a child window that has a separate node environment (but shares the underlying isolate and event loop).
Note: This seems to occur when the BrowserWindow webPreferences sandbox is set to false.
The behavior here is that the console will report `Uncaught illegal state` and in certain cases the renderer process will simply exit with code 7. The following was also reported to stderr:
```
illegal access
(Use `electron --trace-uncaught ...` to show where the exception was thrown)
```
### Testcase Gist URL
https://gist.github.com/markh-discord/1bb303c57d1a20aeee95aeedbc403ab3
### Additional Information
Note: This seems to occur only when the main BrowserWindow webPreferences sandbox is set to false.
The reproduce project in the gist may require a few attempts, but when clicking the button to open the window it will automatically close. Around 1-10 attempts may be needed to display the error in the console. To improve the simplicity of this I used Javascript `setImmediate` which seems to be more recoverable as it doesn't exit the process. In the steps below there is a node native immediate being setup by the node https module that does cause the renderer process to eventually exit.
Reproduce steps aren't precise since this is a race condition, however the following path seems to hit this:
1. node `Environment A` is created and `Environment::InitializeLibuv` is run and sets up its `Environment::CheckImmediate` method in `uv_check_start`.
a. This will be interesting later, but it's called via the event loop `uv_run`: https://github.com/libuv/libuv/blob/e972c6705f197e12d211b598fc514bbea051425d/src/win/core.c#L645
1. Window is opened, a new node Environment (I'll call `Environment B`) is created: https://github.com/electron/electron/blob/dfe501941cd9e7092829c3ef8e27e322047cf736/shell/renderer/electron_renderer_client.cc#L91-L92
a. This environment uses the existing `NodeBindings::isolate_data_`, which also shares the isolate and uv_loop_ from `Environment A`: https://github.com/electron/electron/blob/dfe501941cd9e7092829c3ef8e27e322047cf736/shell/common/node_bindings.cc#L493
2. node `Environment A` has a native immediate pushed from http post work being done by the node http module.
a. This can occur in a few places via `env()->SetImmediate(...)`, here's one as an example: https://github.com/nodejs/node/blob/e8db11c5b2169931238db6c02f6d7f80b3c6e759/src/crypto/crypto_tls.cc#L593
3. Window is closed, causing `Environment B` to begin cleanup:
a. `node::FreeEnvironment` is called, and sets up JS to be disallowed: https://github.com/nodejs/node/blob/2af83ea0f8a1594c0d0bf4bc4b3035234bcd895e/src/api/environment.cc#L420-L421
b. `node::CleanupHandles` is eventually called, this starts up the event loop via `uv_run`: https://github.com/nodejs/node/blob/2af83ea0f8a1594c0d0bf4bc4b3035234bcd895e/src/env.cc#LL966C21-L966C21
c. Per 1a above, the `uv_check_invoke` is called, and the `Environment A`'s `CheckImmediate` is run
4. At this point we're now running things from `Environment A`, in particular it will begin executing the native immediates that it has. Note that we're still in the same call stack and the isolate's state to disallow JS is still flagged.
5. Since a native immediate was added in step 2 above, this begins running, eventually for the http stream work it tries to callback into Javascript via `node::ReportWritesToJSStreamListener::OnStreamAfterReqFinished`:
a. https://github.com/nodejs/node/blob/2af83ea0f8a1594c0d0bf4bc4b3035234bcd895e/src/stream_base.cc#L632
6. This then fails the check on if JS is allowed to be executed (setup on shared isolate from `Environment B` in step 3a):
a. https://chromium.googlesource.com/v8/v8/+/roll/src/execution.cc
```
if (!ThrowOnJavascriptExecution::IsAllowed(isolate)) {
isolate->ThrowIllegalOperation();
```
7. Call stack begins unwinding and eventually hits: `errors::TriggerUncaughtException`: https://github.com/nodejs/node/blob/2af83ea0f8a1594c0d0bf4bc4b3035234bcd895e/src/env.cc#L1097
8. This in turn attempts to execute a `fatal_exception_function.As<Function>()->Call`: https://github.com/nodejs/node/blob/2af83ea0f8a1594c0d0bf4bc4b3035234bcd895e/src/node_errors.cc#L1126-L1127
a. Since JS is still disallowed we get _another_ `v8::internal::Isolate::ThrowIllegalOperation` triggered
9. This unwinds and executes the `TryCatchScope` dtor setup in the above scope: https://github.com/nodejs/node/blob/2af83ea0f8a1594c0d0bf4bc4b3035234bcd895e/src/node_errors.cc#L1116-L1117
10. The `TryCatchScope::~TryCatchScope` then builds the message to dumps and eventually causes the `env_->Exit` to be called: https://github.com/nodejs/node/blob/2af83ea0f8a1594c0d0bf4bc4b3035234bcd895e/src/node_errors.cc#L572
11. This then runs the `Environment::Exit` and eventually will exit the process: https://github.com/nodejs/node/blob/2af83ea0f8a1594c0d0bf4bc4b3035234bcd895e/src/env.cc#L1548
| https://github.com/electron/electron/issues/36858 | https://github.com/electron/electron/pull/38754 | 4ab0a5ade47534db85d333e17f40530cc9726475 | 8874306dc021f2beb4c2b7cd49b8f3bec614e373 | 2023-01-10T20:50:12Z | c++ | 2023-07-18T08:41:50Z |
closed | electron/electron | https://github.com/electron/electron | 36,842 | [".github/workflows/stale.yml"] | [Bug]: Close Stale Issues has not been working since December 5th 2022 | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
17.0.0
### What operating system are you using?
macOS
### Operating System Version
macOS 13.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
Close Stale issues runs and checks out the issues and the number of days before the stale issue is closed.
### Actual Behavior
The Close Stale Issues workflow is failing every day with the error: "[Invalid workflow file: .github/workflows/stale.yml#L27](https://github.com/electron/electron/actions/runs/3870067541/workflow)
The workflow is not valid. .github/workflows/stale.yml (Line: 27, Col: 5): Required property is missing: runs-on"
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/36842 | https://github.com/electron/electron/pull/36843 | 1a9c338c92a50ad2891953a7fd2cecc2e7414bc0 | f56a26c4f70dca6e9468ba15a82007d4a9ad53d6 | 2023-01-09T14:28:22Z | c++ | 2023-01-09T18:16:58Z |
closed | electron/electron | https://github.com/electron/electron | 36,829 | ["shell/browser/electron_browser_main_parts.cc", "shell/renderer/electron_renderer_client.cc", "spec/fixtures/api/unhandled-rejection-handled.js", "spec/fixtures/api/unhandled-rejection.js", "spec/node-spec.ts"] | [Bug]: When logging unhandled rejection stack, stack is logged twice | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.0
### What operating system are you using?
Ubuntu
### Operating System Version
Ubuntu 22.10
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
When adding an unhandledrejection listener that prints error.stack, the error stack should only be printed once.
```js
const handleUnhandledRejection = (reason) => {
console.error(`Unhandled Rejection: ${reason.stack}`);
};
const main = async () => {
process.on("unhandledRejection", handleUnhandledRejection);
throw new Error("oops");
};
main();
```
This is the output in NodeJS:
```
Unhandled Rejection: Error: oops
at main (/tmp/electron-unhandled-rejection-bug/index.js:7:9)
at Object.<anonymous> (/tmp/electron-unhandled-rejection-bug/index.js:10:1)
at Module._compile (node:internal/modules/cjs/loader:1218:14)
at Module._extensions..js (node:internal/modules/cjs/loader:1272:10)
at Module.load (node:internal/modules/cjs/loader:1081:32)
at Module._load (node:internal/modules/cjs/loader:922:12)
at Function.executeUserEntryPoint [as runMain] (node:internal/modules/run_main:81:12)
at node:internal/main/run_main_module:23:47
```
### Actual Behavior
In Electron, the unhandled rejection stack is printed twice:
```
Unhandled Rejection: Error: oops
at main (/tmp/electron-unhandled-rejection-bug/index.js:7:9)
at Object.<anonymous> (/tmp/electron-unhandled-rejection-bug/index.js:10:1)
at Module._compile (node:internal/modules/cjs/loader:1141:14)
at Module._extensions..js (node:internal/modules/cjs/loader:1196:10)
at Module.load (node:internal/modules/cjs/loader:1011:32)
at Module._load (node:internal/modules/cjs/loader:846:12)
at f._load (node:electron/js2c/asar_bundle:2:13328)
at loadApplicationPackage (/tmp/electron-unhandled-rejection-bug/node_modules/electron/dist/resources/default_app.asar/main.js:121:16)
at Object.<anonymous> (/tmp/electron-unhandled-rejection-bug/node_modules/electron/dist/resources/default_app.asar/main.js:233:9)
at Module._compile (node:internal/modules/cjs/loader:1141:14)
(node:14838) UnhandledPromiseRejectionWarning: Error: oops
at main (/tmp/electron-unhandled-rejection-bug/index.js:7:9)
at Object.<anonymous> (/tmp/electron-unhandled-rejection-bug/index.js:10:1)
at Module._compile (node:internal/modules/cjs/loader:1141:14)
at Module._extensions..js (node:internal/modules/cjs/loader:1196:10)
at Module.load (node:internal/modules/cjs/loader:1011:32)
at Module._load (node:internal/modules/cjs/loader:846:12)
at f._load (node:electron/js2c/asar_bundle:2:13328)
at loadApplicationPackage (/tmp/electron-unhandled-rejection-bug/node_modules/electron/dist/resources/default_app.asar/main.js:121:16)
at Object.<anonymous> (/tmp/electron-unhandled-rejection-bug/node_modules/electron/dist/resources/default_app.asar/main.js:233:9)
at Module._compile (node:internal/modules/cjs/loader:1141:14)
(Use `electron --trace-warnings ...` to show where the warning was created)
(node:14838) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). To terminate the node process on unhandled promise rejection, use the CLI flag `--unhandled-rejections=strict` (see https://nodejs.org/api/cli.html#cli_unhandled_rejections_mode). (rejection id: 1)
```
Overall, the NodeJS output is cleaner / easier to understand.
### Testcase Gist URL
https://github.com/SimonSiefke/electron-unhandled-rejection-bug
### Additional Information
## Possible solution
Replacing https://github.com/electron/electron/blob/90af7d7fe2fc70984404474feec4ac2c827794fe/shell/browser/electron_browser_main_parts.cc#L281-L282
with an unhandled rejection handler next to this function https://github.com/electron/electron/blob/90af7d7fe2fc70984404474feec4ac2c827794fe/lib/browser/init.ts#L22
```js
process.on('unhandledRejection', function (error) {
// Do nothing if the user has a custom unhandled rejection handler.
if (process.listenerCount('unhandledRejection') > 1) {
return;
}
import('electron')
.then(({ dialog }) => {
const stack = error.stack ? error.stack : `${error.name}: ${error.message}`;
const message = 'Unhandled Rejection\n' + stack;
dialog.showErrorBox('A JavaScript error occurred in the main process', message);
});
});
```
would keep the current behaviour of not crashing the main process on unhandled rejections
## Related issues
- https://github.com/electron/electron/issues/36528 | https://github.com/electron/electron/issues/36829 | https://github.com/electron/electron/pull/37464 | 17ccb3c6ecca1b16fb3e10504ffe0135b4e1794c | 829fb4f586ee4f0c5b6bbbc6ae9c9853a710eba6 | 2023-01-08T18:29:36Z | c++ | 2023-03-06T10:04:43Z |
closed | electron/electron | https://github.com/electron/electron | 36,826 | ["docs/api/structures/display.md", "shell/common/gin_converters/gfx_converter.cc", "spec/api-screen-spec.ts"] | [Feature Request]: Expose Display label property | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a feature request that matches the one I want to file, without success.
### Problem Description
In line with #6899 and #25252 I would like request support for getting the display label. It looks like chromium provides a "User-friendly label, determined by the platform." that isn't currently used by electron.
https://chromium.googlesource.com/chromium/src/+/master/ui/display/display.h
### Proposed Solution
Based off of similar previous accepted changes I think these are the edits necessary:
In gfx_converter.cc line 147:
`dict.Set("label", val.label());`
In api-screen-spec.ts line 37:
`expect(display).to.have.property('label').that.is.a('string');`
And finally in the docs display.md line 4:
`* `label` string - User-friendly label, determined by the platform.`
### Alternatives Considered
No alternatives necessary.
### Additional Information
I would be happy to contribute the changes myself, but I am a little confused by the contribution workflow and unfortunately I couldn't get a successful build using the build-tools. | https://github.com/electron/electron/issues/36826 | https://github.com/electron/electron/pull/36855 | a7bc57922010062784b2be877ba778323499e0f9 | 2c56a06ad36c624d680c499725d5d9ec3e4d88b0 | 2023-01-07T20:57:54Z | c++ | 2023-01-18T06:44:40Z |
closed | electron/electron | https://github.com/electron/electron | 36,721 | ["shell/browser/ui/message_box_mac.mm"] | [Bug]: macOS confirmation dialog does not show blue border around focused button when it is default | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22
### What operating system are you using?
macOS
### Operating System Version
12.6.2
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
A focused button shows a blue border even when default button:
### Actual Behavior
A focused button does not show the blue border:
### Testcase Gist URL
https://gist.github.com/d9bd650ffc1efcf61016cbc81ea35652
### Additional Information
This only seems to happen when using non-standard labels for the buttons. It requires this macOS setting:
| https://github.com/electron/electron/issues/36721 | https://github.com/electron/electron/pull/36772 | 42cda4a893e5c502ffe4bd1eb0f086709b586bd3 | 32288ac9c588fe8c3b814338aaf5df7fddbce276 | 2022-12-21T06:31:39Z | c++ | 2023-01-05T08:56:38Z |
closed | electron/electron | https://github.com/electron/electron | 36,715 | ["docs/api/web-contents.md", "docs/api/webview-tag.md", "lib/browser/web-view-events.ts", "shell/browser/api/electron_api_web_contents.cc", "shell/browser/api/electron_api_web_contents.h", "shell/browser/ui/inspectable_web_contents.cc", "shell/browser/ui/inspectable_web_contents_delegate.h"] | [Bug]: Developer tools click link does not create a new page | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.17
### What operating system are you using?
Windows
### Operating System Version
windows 11
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
click the link should create a new page
### Actual Behavior
but nothing happen
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/36715 | https://github.com/electron/electron/pull/36774 | 8d008c977df465d3ad96dd6c8c3daa7afd8bea5f | 7d46d3ec9d5f30138777d1bd12890ebe28ba6c31 | 2022-12-21T01:42:46Z | c++ | 2023-01-26T08:54:26Z |
closed | electron/electron | https://github.com/electron/electron | 36,629 | ["patches/chromium/short-circuit_permissions_checks_in_mediastreamdevicescontroller.patch", "shell/browser/web_contents_permission_helper.cc", "spec/api-session-spec.ts"] | [Bug]: `getUserMedia` triggers extra call to `setPermissionRequestHandler` handler with missing metadata | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.0
### What operating system are you using?
macOS
### Operating System Version
12.6
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
20.3.8
### Expected Behavior
In Electron 20.3.8, registering a handler with `setPermissionRequestHandler` and calling `navigator.getUserMedia({ audio: true }, ...)` in the browser produces one call to the handler, which includes a `securityOrigin` property in its `details`.
### Actual Behavior
Starting in Electron 21.0.0, the handler is also called a second time, with a `details` param that doesn't include a `securityOrigin` or `mediaTypes` property. Returning `false` for this call will cause the `getUserMedia` call to fail. Example log output from Fiddle:
<img width="832" alt="image" src="https://user-images.githubusercontent.com/114442851/206822049-094ec08e-3ed4-4968-b491-419a4790329b.png">
### Testcase Gist URL
https://gist.github.com/rf-figma/35de0678214fe97a8e736b64e1a26ead
| https://github.com/electron/electron/issues/36629 | https://github.com/electron/electron/pull/36787 | 1d9a4ab02cb198de27de8636d0893bcc27c671cf | f31826f4a0c44756305087f11cf0de794e68c447 | 2022-12-10T01:39:19Z | c++ | 2023-01-11T10:55:31Z |
closed | electron/electron | https://github.com/electron/electron | 36,615 | ["patches/chromium/.patches", "patches/chromium/chore_defer_usb_service_getdevices_request_until_usb_service_is.patch", "shell/browser/feature_list.cc"] | [Bug]: WebUSB API device list is always empty (attached devices are ignored). | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
23.0.0-alpha.1
### What operating system are you using?
macOS
### Operating System Version
Ventura 13.0.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
Electron should detect my Novation Launchpad X (a class-compliant, USB MIDI device).
### Actual Behavior
The device shows up in `chrome://usb-internals` (in stock Chrome), but when I call `navigator.usb.getDevices` in Electron, the device does not appear in the device list.
The device list is always empty.
### Testcase Gist URL
_No response_
### Additional Information
Codewise, I basically have this in my `index.js` file:
``` js
const { app, BrowserWindow } = require("electron");
app.whenReady().then(function() {
const selectDevice = function(event, details, callback) {
console.log("device list:", details.deviceList); // this array is always empty
event.preventDefault();
};
const renderer = new BrowserWindow(options); // note: `options` is defined IRL
renderer.webContents.session.setPermissionCheckHandler(() => true);
renderer.webContents.session.on("select-usb-device", selectDevice);
});
```
In the renderer process, I just have a call to `navigator.usb.requestDevice` (inside a `keydown` event listener that only exists to ensure that the call is in response to a user gesture).
Note: The filters I'm passing to the WebUSB API are empty (though I tried Novation's ID (`0x1235`) too). | https://github.com/electron/electron/issues/36615 | https://github.com/electron/electron/pull/37441 | 4e85bb921bef97a5bedb4188d0e360ec4a0b70c0 | efde7a140bc48e0e3e08404f41f6179c26858ce4 | 2022-12-08T19:23:44Z | c++ | 2023-03-07T17:40:40Z |
closed | electron/electron | https://github.com/electron/electron | 36,602 | ["BUILD.gn", "docs/api/tray.md", "filenames.gni", "shell/browser/ui/gtk/menu_gtk.cc", "shell/browser/ui/gtk/menu_gtk.h", "shell/browser/ui/status_icon_gtk.cc", "shell/browser/ui/status_icon_gtk.h", "shell/browser/ui/tray_icon_gtk.cc", "shell/browser/ui/tray_icon_linux.cc", "shell/browser/ui/tray_icon_linux.h"] | [Bug]: No Tray icon for Arch Linux | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.0
### What operating system are you using?
Other Linux
### Operating System Version
Arch Linux, Manjaro
### What arch are you using?
x64
### Last Known Working Electron version
22.0.0
### Expected Behavior
Be able to use Electron [tray API](https://www.electronjs.org/docs/latest/api/tray)
### Actual Behavior
I cannot have any Tray icon showing up since the last release of Electron 22.0.0. I believe we change the underlying implementation for Tray API with [this PR](https://github.com/electron/electron/pull/36333).
With Electron <= 21, it worked fine
Here's the info of OS we're using
```
$ uname -a
Linux tower 6.0.11-arch1-1 #1 SMP PREEMPT_DYNAMIC Fri, 02 Dec 2022 17:25:31 +0000 x86_64 GNU/Linux
```
or
```
$ uname -a
Linux manjaro 5.15.60-1-MANJARO #1 SMP PREEMPT Thu Aug 11 13:14:05 UTC 2022 x86_64 GNU/Linux
```
I attached the minimum reproduction example here. Steps:
- `yarn install`
- then `yarn start`
- cannot see the Tray icon
- `yarn add --dev [email protected]`
- `yarn start` again
- then we see the Tray icon again (should be a T-shirt png)
[electron-tray-22.zip](https://github.com/electron/electron/files/10175635/electron-tray-22.zip)
### Testcase Gist URL
_No response_
### Additional Information
We don't experience this issue for Mac, Windows and Ubuntu | https://github.com/electron/electron/issues/36602 | https://github.com/electron/electron/pull/36815 | 7d46d3ec9d5f30138777d1bd12890ebe28ba6c31 | c303135b02fa5829459c816cd5bb75dd0e6e4aeb | 2022-12-07T11:48:18Z | c++ | 2023-01-26T10:15:55Z |
closed | electron/electron | https://github.com/electron/electron | 36,597 | ["shell/common/api/electron_api_asar.cc"] | [Bug]: memory leak when a worker accesses a file in the asar archive package for the first time | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
17.0.0
### What operating system are you using?
Windows
### Operating System Version
Windows 10 version 21H2
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
After a worker ends, the memory usage falls back to before the worker starts.
### Actual Behavior
Each time a worker is created, the memory usage will be increased a little. Even if the worker is finished, the memory usage will not be reduced.
### Testcase Gist URL
https://gist.github.com/xuwanghu/d61b511abf80aa691375cdfd30a8474e
### Additional Information
The purpose of adding so many dependencies is to make the ASAR file large enough so that the memory leaks are more noticeable.
**How to Reproduce:**
it's on windows:
1. npm install
2. yarn dist
3. run the /dist/win-unpacked/test_leak_for_worker.exe
4. use the Performance Monitor to monitor its working set.
| https://github.com/electron/electron/issues/36597 | https://github.com/electron/electron/pull/36600 | ab890fb8c3883ca60af514546245bf6a000686f1 | f72e6551f093e7f9f02531c6a69f390521463bf8 | 2022-12-07T05:42:31Z | c++ | 2022-12-14T17:37:28Z |
closed | electron/electron | https://github.com/electron/electron | 36,548 | ["docs/api/net.md"] | [Bug]: Documentation for net.request doesn't clearly show constructor options | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.0
### What operating system are you using?
macOS
### Operating System Version
12.6.1
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
The documentation for net.request should include documentation of the options that the function takes: https://www.electronjs.org/docs/latest/api/net#netrequestoptions
### Actual Behavior
Instead, that documentation exists at https://www.electronjs.org/docs/latest/api/client-request, which is not obvious from the net.request docs
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/36548 | https://github.com/electron/electron/pull/36556 | 8acf6039e78725918af5797fad6fc4f7cf1e177c | e1e66fc8ac52af3f951404a647cb82a8389528c3 | 2022-12-04T21:18:48Z | c++ | 2022-12-05T23:17:37Z |
closed | electron/electron | https://github.com/electron/electron | 36,539 | ["shell/browser/browser.h", "shell/browser/browser_mac.mm", "shell/browser/electron_browser_main_parts.cc"] | [Bug]: settrafficLightPosition on macOS calculates incorrectly for RTL system languages | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.0
### What operating system are you using?
macOS
### Operating System Version
macOS ventura 13.0.1
### What arch are you using?
x64
### Last Known Working Electron version
N/A
### Expected Behavior
<img width="807" alt="لقطة الشاشة ٢٠٢٢-١٢-٠٢ في ٧ ٢٨ ٣٣ ص" src="https://user-images.githubusercontent.com/964386/205172660-1e95a7ac-cb60-4873-8a27-e93586547cb0.png">
### Actual Behavior
<img width="804" alt="لقطة الشاشة ٢٠٢٢-١٢-٠٢ في ٧ ٢٨ ٥٧ ص" src="https://user-images.githubusercontent.com/964386/205172682-87a89c5c-a178-4e25-b2a8-923d50f079c5.png">
### Testcase Gist URL
https://gist.github.com/deepak1556/3025154b48cb156126cfdcc74338ec8e
### Additional Information
Set the system preferred language to any RTL language value and restart the OS. When system preferred language is in RTL and application language is set to `en`, the traffic light positions are calculated from outside the window frame positions. | https://github.com/electron/electron/issues/36539 | https://github.com/electron/electron/pull/36839 | 168726a0521dd53f160b216f7b84832b1368fff5 | 414791232a79c783d3d758fe11a864344d513660 | 2022-12-01T22:33:43Z | c++ | 2023-01-10T11:19:00Z |
closed | electron/electron | https://github.com/electron/electron | 36,538 | ["lib/browser/api/web-contents.ts", "shell/browser/web_contents_preferences.cc", "spec/api-browser-window-spec.ts", "spec/guest-window-manager-spec.ts", "spec/lib/screen-helpers.ts"] | [Bug]: setWindowOpenHandler results in incorrect dark mode background color for text/plain content | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
Electron 21.3.1, 22.0.0, and the latest 24.0.0-nightly.20221201
### What operating system are you using?
macOS 13 and Windows 11
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
n/a
### Expected Behavior
When the system is in dark mode, and a popup is opened via `window.open` to a URL returning content of type `text/plain`, the resulting popup window should have a black background with white text. This happens correctly unless `mainWindow.webContents.setWindowOpenHandler` is called.
See https://www.loom.com/share/4af414156762410e994b5e2a56d997c7 for the correct behavior when `mainWindow.webContents.setWindowOpenHandler` is commented out.
### Actual Behavior
With the call to `mainWindow.webContents.setWindowOpenHandler` uncommented, the popup window incorrectly has white text with a white background. See https://www.loom.com/share/15aadf785f3740c5b8f07239ea1e46ed for the actual behavior.
### Testcase Gist URL
https://gist.github.com/36925a2377d0b8c7c67abd9816c2f411 | https://github.com/electron/electron/issues/36538 | https://github.com/electron/electron/pull/36914 | d95f9d2c636775946717a95fefee1ada51524a79 | a26343f3e084028b920a7ea44e6b221350e3546c | 2022-12-01T21:08:20Z | c++ | 2023-05-02T21:44:34Z |
closed | electron/electron | https://github.com/electron/electron | 36,417 | ["shell/browser/api/electron_api_web_contents.cc"] | [Bug]: webview tag background color do not follow browser window when reload webview | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
21.2.0
### What operating system are you using?
macOS
### Operating System Version
macOS Big Sur 11.4
### What arch are you using?
x64
### Last Known Working Electron version
11.4.5
### Expected Behavior
Recently, we upgraded our electron version from 11.4.5 to 21.2.0 and met a problem.
The final problem I located is that webview's background color do not follow browser window when reload webview. So it caused from white -> red instead of browser windows' bg color -> red in my demo.
### Actual Behavior
### Testcase Gist URL
https://gist.github.com/PerfectPan/100900b7758293b8f72b23d0babdb1cc
### Additional Information
In our situation, we use webview to load chrome devtools frontend. This app styles background color asynchronous, so we met this problem. A hack way maybe we modify the inspect.html to hardcode our background color in it. But we want to know if there are any other way to solve this. | https://github.com/electron/electron/issues/36417 | https://github.com/electron/electron/pull/36920 | 1d98b27a6643d6b6cbc4d9d426d6d5acc4b45e2a | b1548c2dbefdcd8f823c18175715fc5d34b09781 | 2022-11-21T13:15:40Z | c++ | 2023-01-18T13:46:47Z |
closed | electron/electron | https://github.com/electron/electron | 36,331 | ["shell/browser/electron_browser_context.cc", "spec/api-media-handler-spec.ts"] | [Bug]: No BrowserCaptureMediaStreamTrack being returned when using preferCurrentTab | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
22.0.0-beta.4
### What operating system are you using?
Windows
### Operating System Version
Windows 11
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
Looking at the tests written here https://github.com/electron/electron/pull/30702/files#diff-cc0dd3c6f0140615357588553aaf9e9f95e6132498626a5d788d8e297d98eee9 I have decided to try to use the Chromium 104+ feature for region capture.
While the api appears to work, I cannot get the api call to return the expected BrowserCaptureMediaStreamTrack, instead the usual MediaStreamTrack is returned.
This causes the ability to use cropTo from the track to fail since it doesnt exist in MediaStreamTrack.
Given that this is still a beta, is this a known issue or do I need to do something different?
### Actual Behavior
I would expect a BrowserCaptureMediaStreamTrack to be returned and track.cropTo(cropTarget) to work as expected, allowing me to share a region of the window/screen selected
### Testcase Gist URL
https://gist.github.com/e277f244fcf0402152466a0fddc9720a
### Additional Information
This code works fine when using it against Chrome | https://github.com/electron/electron/issues/36331 | https://github.com/electron/electron/pull/39074 | e1d63794e5ce74006188ed2b7f1c35b5793caf36 | 2481f94b4efe262c9ae2f5f16cd0845eeb6c4efd | 2022-11-14T11:05:51Z | c++ | 2023-08-23T08:49:24Z |
closed | electron/electron | https://github.com/electron/electron | 36,309 | [".circleci/config/base.yml", "appveyor.yml"] | [Bug]: mksnapshot args shipped with v21 doen't work properly | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
21.2.2
### What operating system are you using?
macOS
### Operating System Version
13
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
20.3.4
### Expected Behavior
Running mksnapshot works normally.
### Actual Behavior
It fails.
It succeeds if I remove
```
--turbo-profiling-input
../../v8/tools/builtins-pgo/arm64.profile
```
from the `mksnapshot_args` file.
I'm not able to find this `arm64.profile` file anywhere to try and build with it being present. (the path mentioned is two levels up, so I'm guessing it is somehow missed while creating the zip and relative path ends up here)
### Testcase Gist URL
_No response_
### Additional Information
Contents of the mksnapshot_args file across recent versions
Note: I'm running mksnapshot via electron-mksnapshot | https://github.com/electron/electron/issues/36309 | https://github.com/electron/electron/pull/36378 | 4f1f263a9a98abb1c4e265c1d19e281312adf5b2 | 7529ebfe0e20ff0456aaab29c22346e35cf074ce | 2022-11-10T15:38:05Z | c++ | 2022-11-17T22:49:12Z |
closed | electron/electron | https://github.com/electron/electron | 36,282 | ["script/node-disabled-tests.json", "shell/app/node_main.cc", "shell/browser/javascript_environment.cc", "shell/browser/javascript_environment.h"] | Enable `WebAssembly.{compileStreaming|instantiateStreaming}` in Node.js | Refs https://github.com/electron/electron/pull/35999.
Refs https://github.com/nodejs/node/pull/42701.
Node.js added support for `WebAssembly.compileStreaming` and `WebAssembly.instantiateStreaming` but they don't work within Electron at the moment.
This is happening because Node.js' logic for enabling them [here](https://github.com/nodejs/node/pull/42701/files#diff-5f2fc380364ef43210af4a28c08f9a0c7ec5f85e3ec1f5c1df5e0204eb9dc721R261) is called via `node::SetIsolateUpForNode`, which by necessity can only be called on an existing isolate. However, the switch to enable it [here](https://source.chromium.org/chromium/chromium/src/+/main:v8/src/wasm/wasm-js.cc;l=2983-2988?q=wasm_streaming_callback&ss=chromium%2Fchromium%2Fsrc) is called by gin during [its own isolate initialization](https://github.com/electron/electron/blob/f2c341b655406434b5a44bc8b1e2418d664c319b/shell/app/node_main.cc#L162), which means that it ends up being undefined. Look into a good way to fix this. | https://github.com/electron/electron/issues/36282 | https://github.com/electron/electron/pull/36420 | 909ee0ed6bbdf57ebaeda027b67a3a44185f6f7d | b90a5baa6dc7f745bf8779dd79d51229a3f8593d | 2022-11-08T10:43:02Z | c++ | 2022-12-05T17:07:49Z |
closed | electron/electron | https://github.com/electron/electron | 36,030 | ["filenames.gni", "patches/node/.patches", "patches/node/enable_crashpad_linux_node_processes.patch", "shell/app/node_main.cc", "shell/common/api/crashpad_support.cc", "shell/common/node_bindings.cc", "spec/api-crash-reporter-spec.ts", "spec/fixtures/apps/crash/fork.js", "spec/fixtures/apps/crash/main.js", "spec/fixtures/apps/crash/node-extra-args.js"] | Enable crashpad support on linux for ELECTRON_RUN_AS_NODE processes | Refs https://github.com/electron/electron/pull/35900/commits/b10f24386cb37d736b10f4831235eedcf1f0db4b
Node.js forked child processes have been relying on breakpad for crash report generation, with https://chromium-review.googlesource.com/c/chromium/src/+/3764621 Linux has moved onto crashpad completely. For these processes to act as crashpad clients, they need the following
* Inherit FD for the handler process communication, which is usually inherited by the chromium spawned child processes via `ElectronBrowserClient::GetAdditionalMappedFilesForChildProcess`. For Node.js the fork is performed by Libuv, we should pass it via stdio option of child_process.fork. The descriptor will eventually be looked up in [PlatformCrashpadInitialization](https://source.chromium.org/chromium/chromium/src/+/main:components/crash/core/app/crashpad_linux.cc;l=197). Also we will need to setup `base::GlobalDescriptors` instance in the child process.
* Pass the PID of the handler process to the child process via the command line flag `--crashpad-handler-pid/crash_reporter::switches::kCrashpadHandlerPid` whose value can be extracted with the same API to obtain the FD [crash_reporter::GetHandlerSocket](https://source.chromium.org/chromium/chromium/src/+/main:components/crash/core/app/crashpad.h;l=236). | https://github.com/electron/electron/issues/36030 | https://github.com/electron/electron/pull/36460 | 16a7bd71024789fcabb9362888ee4638852b1eb1 | 2c723d7e84dfab5ad97fc0927426a0b2cd7a15b5 | 2022-10-14T11:57:59Z | c++ | 2022-11-29T15:33:54Z |
closed | electron/electron | https://github.com/electron/electron | 35,996 | ["patches/chromium/printing.patch"] | [Bug]: Errors about silent printing | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.17
### What operating system are you using?
Windows
### Operating System Version
Windows 10 Pro 21H2
### What arch are you using?
x64
### Last Known Working Electron version
18.3.15
### Expected Behavior
I want to use contents.print ({silent: true}) to execute the print task only once
### Actual Behavior
Use contents.Print({silent: true}) will be executed twice.
Use the following code to monitor the actual printing times.
```js
window. onbeforeprint = (event) => {
console. log('before print');
};
window. onafterprint = (event) => {
console. log('After print');
};
```
This phenomenon will appear in 19 and later versions.
### Testcase Gist URL
https://gist.github.com/feddc9ddf37da270efc034d6ff6431b9
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35996 | https://github.com/electron/electron/pull/38741 | d78f37ec8faae9a3677037e1b40baea81fc93b6c | 46fb0d8f5f3a33c5b9bba2329ac5de946cd66a3c | 2022-10-12T04:43:34Z | c++ | 2023-06-15T14:46:38Z |
closed | electron/electron | https://github.com/electron/electron | 35,994 | ["shell/browser/native_browser_view_mac.mm", "spec/api-browser-view-spec.ts"] | [Bug]: BrowserView setBounds behavior inconsistent between Mac and Windows | ```js
bv.setBounds({x: 0, y: 0, width: 500, height: 40})
```
gives different results on different platforms:
mac:
windows:
Regressed in #34713 I believe. cc @codebytere | https://github.com/electron/electron/issues/35994 | https://github.com/electron/electron/pull/38981 | ec4c9024b9795d360fbd96b09684fdb228966781 | 5a77c75753f560bbc4fe6560441ba88433a561f8 | 2022-10-11T23:15:30Z | c++ | 2023-07-06T07:50:08Z |
closed | electron/electron | https://github.com/electron/electron | 35,971 | ["lib/browser/api/web-contents.ts", "shell/browser/api/electron_api_web_contents.cc"] | [Bug]: Some options of printToPDF don't work | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
21.1.0
### What operating system are you using?
macOS
### Operating System Version
macOS Catalina 10.15.7
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
Some options of `webContents.printToPDF` don't be ignored.
The options are:
- `margins` (`top`, `bottom`, `left`, `right` and `marginType`)
- `pageSize`
- `scale`
For example: When I choose a margin top, like 2" and call `printToPDF`, the PDF should come with margin top with 2"
### Actual Behavior
I defined options, but they don't work.
Am I doing something wrong, maybe using incorrect?
### Testcase Gist URL
https://gist.github.com/lesleyandreza/7699603b20f2459d1fcf3b9480fb22e1
### Additional Information
The options `headerTemplate` and `footerTemplate` works, but in a strange way, because the text comes really small. Maybe this could be a Chromium bug.
<img width="1404" alt="very-small-header" src="https://user-images.githubusercontent.com/14945792/194948569-26c8639a-abe8-4a82-b8ee-1e00e1d38192.png">
But there is a workaround with CSS:
<img width="1400" alt="normal-header" src="https://user-images.githubusercontent.com/14945792/194948843-1ad0fd55-d881-4037-abe0-3801fe44fc29.png">
| https://github.com/electron/electron/issues/35971 | https://github.com/electron/electron/pull/35975 | 0759f3320e905dd7e19c05b038204211dee26e39 | ee7cf5a6d40096a887166cfe714cd6de79052b64 | 2022-10-10T21:33:40Z | c++ | 2022-10-11T23:06:34Z |
closed | electron/electron | https://github.com/electron/electron | 35,967 | ["default_app/main.ts"] | [Bug]: V8 flags and desktop name in package.json does not hornored when running local app using electron cli | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.3
### What operating system are you using?
Other Linux
### Operating System Version
Linux pc 5.19.13-arch1-1.1 #1 SMP PREEMPT_DYNAMIC Wed, 05 Oct 2022 07:37:04 +0000 x86_64 GNU/Linux
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
Launch local app by `electron /path/to/app/dir` , then `/path/to/app/dir/package.json` data (`productName`, `version`, `desktopName`, etc.) is loaded.
### Actual Behavior
~~Electron's default values are used.~~
**Edit**: Name and version are correctly updated. https://github.com/electron/electron/blob/c2cb97ea298412143e34027fac2c28fd031e169a/default_app/main.ts#L99-L106
It looks like `desktopName` and `v8Flags` should be considered too.
ref:https://github.com/electron/electron/blob/c2cb97ea298412143e34027fac2c28fd031e169a/lib/browser/init.ts#L107-L130
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35967 | https://github.com/electron/electron/pull/35997 | 76880be6d2cc466a44cf437bf34bf32f94a9f6ad | 295c5331eeb2cce40419b6a54875fba88d61422d | 2022-10-10T17:10:11Z | c++ | 2022-10-17T08:34:24Z |
closed | electron/electron | https://github.com/electron/electron | 35,916 | ["shell/app/electron_main_delegate.cc", "shell/app/electron_main_delegate.h"] | [Bug]: The global content client will init twice on the startup | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19
### What operating system are you using?
Other (specify below)
### Operating System Version
macOS
### What arch are you using?
Other (specify below)
### Last Known Working Electron version
_No response_
### Expected Behavior
The global content client should init once on startup.
### Actual Behavior
The global content client will init twice on the startup and it cause a memory leak.causes.
### Testcase Gist URL
_No response_
### Additional Information
These days I try to debug the electron source and found that the global content client will init twice on startup because the ElectronMainDelegate didn't override the CreateContentClient.
The first time, it call the `ContentMainDelegate::CreateContentClient` to create an ContentClient instance.
```
void SetContentClient(ContentClient* client) {
/*
if(g_client != nullptr)
{
DCHECK(false);
}
*/
g_client = client;
}
//ContentClientCreator::Create
static void Create(ContentMainDelegate* delegate) {
ContentClient* client = delegate->CreateContentClient();
DCHECK(client);
SetContentClient(client);
}
```
Then the global ContentClient will be set as an ElectronContentClient instance in `ElectronMainDelegate::BasicStartupComplete`
So the first one is never to be used and never to be free until the process exit.
| https://github.com/electron/electron/issues/35916 | https://github.com/electron/electron/pull/35932 | ef00a2a1dade1e76be39d77e236d2926ddcc9a80 | ebb866e63d90fc44168f2b9c97ea2378d4c17831 | 2022-10-05T08:26:58Z | c++ | 2022-10-10T14:48:44Z |
closed | electron/electron | https://github.com/electron/electron | 35,877 | ["script/sysroots.json"] | fix arm linux glibc dependency | https://github.com/electron/electron/pull/34491 will reoccur in 22 unless we fix it, see https://bugs.chromium.org/p/chromium/issues/detail?id=1309965#c5 | https://github.com/electron/electron/issues/35877 | https://github.com/electron/electron/pull/36247 | 0ba0df45232d0c1c4542d6647a862305aff28bb4 | 2008c9a5d0be0c169e8016dac8a90744c504c3f7 | 2022-10-03T19:34:44Z | c++ | 2022-11-07T14:38:08Z |
closed | electron/electron | https://github.com/electron/electron | 35,807 | ["shell/browser/native_window_views.cc", "shell/browser/native_window_views.h", "spec/api-browser-window-spec.ts"] | [Bug]: Menu Bar Visibility | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.0
### What operating system are you using?
Other Linux
### Operating System Version
Fedora 36
### What arch are you using?
x64
### Last Known Working Electron version
19.0.0
### Expected Behavior
If I specify:
```js
win.setMenuBarVisibility(false);
```
then menu bar will be hidden which is great. But if you spam F11 key, menu bar will become visible again.
### Actual Behavior
If I specify:
```js
win.setMenuBarVisibility(false);
```
Menu bar shouldn't be visible even while spamming F11.
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35807 | https://github.com/electron/electron/pull/38599 | 5ee890fb6f7c6acbcfd8e6e765334e6f9aa61850 | c8bdd014c87b5c33b77c845f853b7e52bb31de8f | 2022-09-25T15:23:42Z | c++ | 2023-06-08T10:19:34Z |
closed | electron/electron | https://github.com/electron/electron | 35,724 | ["patches/chromium/.patches", "patches/chromium/revert_x11_keep_windowcache_alive_for_a_time_interval.patch", "spec/api-menu-spec.ts"] | [Bug]: After resize the window manually, clicking the menu does not show submenus | ### Preflight Checklist
- [x] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.1.4
### What operating system are you using?
Other Linux
### Operating System Version
Manjaro Linux 22.0.0
### What arch are you using?
x64
### Last Known Working Electron version
`v19.0.0-alpha.3`
### Expected Behavior
After resize the window manually, clicking the menu should show submenus
### Actual Behavior
After resize the window manually, clicking the menu does not show submenus, both gnome and kde, manjaro and ubuntu
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35724 | https://github.com/electron/electron/pull/37798 | 69586684484c05a0078e3b916239186a5c3d749a | 3c0c4d5c278e575691f9376483dc4f60a287c467 | 2022-09-19T09:46:03Z | c++ | 2023-04-11T06:17:45Z |
closed | electron/electron | https://github.com/electron/electron | 35,715 | ["shell/browser/api/electron_api_web_contents.cc"] | [Feature Request]: Allow for opening docked DevTools, even if WCO is enabled | ### Preflight Checklist
- [x] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a feature request that matches the one I want to file, without success.
### Problem Description
#35209 has introduced a change that forces a `'detach'` state for DevTools, if they are opened on a WC in a window with the Window Controls Overlay enabled.
However, this is not always a desired behavior, especially if DevTools are an integral part of your application.
I must emphasize, the detached state is not a default, but *forced*, which means you cannot override this even with
```js
webContents.openDevTools({ mode: 'bottom' })
// ignores the mode
```
### Proposed Solution
While the `'detach'` by default is good for solving the problem of WCO covering the devtools, there should be an option to override this.
For example, `wc.openDevTools()` will open them in the detached state, but `wc.openDevTools({ mode: 'bottom' })` should respect the `mode` property.
Perhaps, this should work just like it works with the `<webview>` tags, when if you explicitly pass an empty mode, it will open the DevTools in the last docked state.
### Alternatives Considered
None. Disabling WCO isn't a good enough option.
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35715 | https://github.com/electron/electron/pull/35754 | 4ffdd284c398d3e18a71f636422eaa0cf28406da | eb3262cd87f1602cea651f89166b0da0f2ee6e14 | 2022-09-18T00:23:12Z | c++ | 2022-09-22T08:44:27Z |
closed | electron/electron | https://github.com/electron/electron | 35,709 | [".github/workflows/scorecards.yml"] | Enable the OpenSSF Scorecard Github Action and Badge | Hi, I am Joyce and I'm working on behalf of Google and the [OpenSSF][ossf] to help essential open-source projects improve their supply-chain security. The OpenSSF is a non-profit foundation backed by the Linux Foundation, dedicated to improving the security of the open-source community. It counts GitHub as a founding [member][ossf-membership].
The [Scorecard][sc] system combines dozens of automated checks to let maintainers better understand their project's supply-chain security posture. It is developed by the OpenSSF, with [direct support from GitHub][sc-blog].
Considering that the Electron Framework is widely used on cross-platform applications using HTML, CSS and JavaScript, it's been included in the OpenSSF's list of the 100 most critical open-source projects. Thus, the Scorecard Github Action could help you to track and solve security risks involving your repository/project and this way garantee that your repository and the contribution process is safe from malicious sabotage.
However, the OpenSSF has also developed the [Scorecard GitHub Action][sc-gha], which adds the results of its checks to the project's [security dashboard](https://github.com/electron/electron/security), as well as suggestions on how to solve any issues (see examples below). This Action has been adopted by 1600+ projects already.
Would you be interested in a PR which adds this Action? Optionally, it can also publish your results to the OpenSSF REST API, which allows a [badge][badge] with the project's score to be added to its README.
Please, feel free to reach me out in case of any doubt or concern and thanks for the attention on this issue.
![Code scanning dashboard with multiple alerts, including Code-Review and Token-Permissions][img-security]
![Detail of a Token-Permissions alert, indicating the specific file and remediation steps][img-detail]
[badge]: https://openssf.org/blog/2022/09/08/show-off-your-security-score-announcing-scorecards-badges/
[ossf]: https://openssf.org/
[ossf-membership]: https://openssf.org/about/members/
[sc]: https://github.com/ossf/scorecard
[sc-blog]: https://github.blog/2022-01-19-reducing-security-risk-oss-actions-opensff-scorecards-v4/
[sc-gha]: https://github.com/ossf/scorecard-action
[img-security]: https://user-images.githubusercontent.com/15221358/190184391-84ca1844-259a-4b3b-9c86-74adadbea7f1.png
[img-detail]: https://user-images.githubusercontent.com/15221358/190184600-ee8d3b39-077e-416a-8711-1b5fb01cf0b3.png | https://github.com/electron/electron/issues/35709 | https://github.com/electron/electron/pull/35741 | a9ef68f12676fd76998ac9e15c0b10a3b6cd4daf | 05577d0903e18df1d431625376ac8b99aa28f209 | 2022-09-16T16:42:08Z | c++ | 2022-11-15T00:22:10Z |
closed | electron/electron | https://github.com/electron/electron | 35,594 | ["shell/browser/electron_download_manager_delegate.cc"] | [Bug]: File extension appended multiple times to file name when saving | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.1.1
All latest 18.x.y, 19.x.y and 20.x.y versions are affected.
### What operating system are you using?
Windows
### Operating System Version
Windows 10 Version 20H2
### What arch are you using?
x64
### Last Known Working Electron version
19.0.7
### Expected Behavior
When saving/downloading a file, file name should be saved with the given file extension.
### Actual Behavior
File extension is appended three times to file name with double dots, like:
<file_name>.<file_extension>..<file_extension>..<file_extension>
Issue happens for file extensions which are not in registry.
For file types which are in registry, file name is fine.
### Testcase Gist URL
https://gist.github.com/dfa6a6e75b43e475f285df9e0f743650
### Additional Information
Most likely related to the change https://github.com/electron/electron/pull/34723 | https://github.com/electron/electron/issues/35594 | https://github.com/electron/electron/pull/35618 | 9f97c3e50a11b364f02f57209854bb4000ac1136 | 12a7d7eea52409ce02af6d5f51f8208f483f24d2 | 2022-09-07T14:47:00Z | c++ | 2022-09-13T18:47:04Z |
closed | electron/electron | https://github.com/electron/electron | 35,512 | ["docs/tutorial/performance.md"] | [Bug]: electron should not set a default menu | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.x
### What operating system are you using?
Windows
### Operating System Version
Windows 11
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
An application should be in full control of the menu and thus the following code is taking startup CPU cycles when not needed:
https://github.com/electron/electron/blob/bfced8cbfe6bad6d119ea9276d13aa67accc9c9a/lib/browser/init.ts#L177
At the very minimum allow an option to disable this when the application sets a menu anyway.
### Actual Behavior
Electron seems to install a default menu which consumes time for no good reasons it seems:
<img width="364" alt="image" src="https://user-images.githubusercontent.com/900690/187595296-0380cf75-2b92-4c69-be7c-0172124a66e7.png">
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35512 | https://github.com/electron/electron/pull/35525 | b3a744db8a1aa85f27f4467ae2eb1a8c2eff0233 | 6ec4c7e563c4bf59abb684e593bc6c0e0ee921c6 | 2022-08-31T04:51:53Z | c++ | 2022-09-12T22:29:42Z |
closed | electron/electron | https://github.com/electron/electron | 35,508 | ["docs/api/session.md", "shell/browser/api/electron_api_session.cc", "spec/spellchecker-spec.ts"] | [Bug]: Spellcheck API can cause errors in macOS applications despite supposedly being a "no-op" on macOS | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.10
### What operating system are you using?
macOS
### Operating System Version
macOS Monterey 12.5.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
The `setSpellCheckerLanguages` method should have no effect when invoked on macOS, other than possibly generating a warning that this API is not supported on macOS.
### Actual Behavior
On macOS, invoking `ses.setSpellCheckerLanguages` with a language array containing one or more unsupported languages causes an error which can prevent the app from rendering. ("Unsupported languages" here means languages not supported by the Electron spellcheck API, which is not actually used on macOS.)
This issue is further complicated by the fact that the `ses.getSpellCheckerLanguages` method returns the actual locale(s) that the macOS spellcheck API is checking against, regardless of whether the Electron spellcheck API supports those locales.
As such, even something as seemingly innocuous as setting the spellcheck languages to the current spellcheck languages, which should be a no-op on any platform, can cause errors on macOS.
### Testcase Gist URL
https://gist.github.com/71ad54ff04a1a5b8f1a3ceed905683a0
### Additional Information
The [Electron docs](https://www.electronjs.org/docs/latest/api/session#sessetspellcheckerlanguageslanguages) state the following:
> Note: On macOS the OS spellchecker is used and will detect your language automatically. This API is a no-op on macOS.
This issue can only be reproduced on a macOS device where the current system locale is one that the Electron spellcheck API does not support. For English-speaking users, you can accomplish this by setting your region to "India" in the "Language & Region" settings menu. | https://github.com/electron/electron/issues/35508 | https://github.com/electron/electron/pull/35514 | bf20aabb9ee7985d966ae6cb2e92f635a8d85a6f | 38a7da692aaae01dae97e96b7a5eb648765efe33 | 2022-08-30T21:52:21Z | c++ | 2022-09-14T22:10:16Z |
closed | electron/electron | https://github.com/electron/electron | 35,471 | ["shell/browser/api/electron_api_screen.cc", "shell/browser/api/electron_api_screen.h"] | [Bug]: Segfault when calling `screen.getCursorScreenPoint()` under Wayland | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
v20.0.3 and v22.0.0-nightly.20220824
### What operating system are you using?
Other Linux, and Sway WM (1.7)
### Operating System Version
ArchLinux, kernel 5.19.3
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
Should provide a Point to the cursor position and not segfault.
### Actual Behavior
Get a segmentation fault when calling `screen.getCursorScreenPoint();` under Wayland (sway 1.7).
### Testcase Gist URL
https://gist.github.com/weedz/b8002775cd1a23c46d38b5f10de957e4
### Additional Information
Works fine under Xwayland. But will crash when launched with `--ozone-platform-hint=wayland`.
Tested every version down to electron@16 (with `--enable-features=UseOzonePlatform --ozone-platform=wayland`:
- Electron@16: "interface 'wl_output' has no event 4" and no window or other errors.
- Electron v17-v22.0.0-nightly.20220824: "electron exited with signal SIGSEGV" | https://github.com/electron/electron/issues/35471 | https://github.com/electron/electron/pull/35503 | 98c0fa1c8b4759eff243807b625094d5ef2840a8 | 2db0f7f8d44be93190323a94c4a1fc255251bfa5 | 2022-08-26T21:19:42Z | c++ | 2022-09-05T08:10:50Z |
closed | electron/electron | https://github.com/electron/electron | 35,462 | ["patches/chromium/.patches", "patches/chromium/fix_crash_loading_non-standard_schemes_in_iframes.patch", "spec/api-protocol-spec.ts", "spec/fixtures/pages/iframe-protocol.html"] | [Bug]: Electron crashes when loading <iframe> from custom URL protocol | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.1.0
### What operating system are you using?
Windows
### Operating System Version
Windows 10
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
I want to load an iframe from a custom:// url protocol inside my main window.
`<iframe src="custom://test.html">`
### Actual Behavior
Application crashes when using custom:// url protocol. The protocol is registered with `protocol.registerFileProtocol()`. When loading the exact same file with the file:// protocol, everything is OK.
### Testcase Gist URL
https://gist.github.com/75a34a8f7177e616146a8e5f1f4edddc
### Additional Information
The same issue hase also been observed on Ubuntu 20.04
| https://github.com/electron/electron/issues/35462 | https://github.com/electron/electron/pull/35485 | bfced8cbfe6bad6d119ea9276d13aa67accc9c9a | e0fb5cbe1fd84d9651e6030ebab683dd9e4af42d | 2022-08-25T18:21:15Z | c++ | 2022-08-31T08:08:11Z |
closed | electron/electron | https://github.com/electron/electron | 35,442 | ["patches/node/feat_add_uv_loop_interrupt_on_io_change_option_to_uv_loop_configure.patch", "shell/common/node_bindings.cc"] | [Bug]: Upgrading to 19.0.13 break webpack hot reloading | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] #35489
### Electron Version
19.0.13
### What operating system are you using?
macOS
### Operating System Version
Monterey
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
19.0.12
### Expected Behavior
After upgrading to `19.0.13` webpack dev server reloading breaks.
<img width="486" alt="CleanShot 2022-08-24 at 23 45 20@2x" src="https://user-images.githubusercontent.com/1192452/186594068-a984939d-dce8-4534-8dd0-634472275e60.png">
### Actual Behavior
The render process does not crash
### Testcase Gist URL
_No response_
### Additional Information
Not sure how to capture the crash log but happy to provide it with some instructions!. | https://github.com/electron/electron/issues/35442 | https://github.com/electron/electron/pull/35422 | 2f23bdb19ea76029907ae247cc8eb47114df1d0c | 18475818486be806e891e8e6cb13b83c620ec0aa | 2022-08-25T06:46:37Z | c++ | 2022-08-29T13:45:48Z |
closed | electron/electron | https://github.com/electron/electron | 35,429 | ["shell/browser/electron_download_manager_delegate.cc"] | [Bug]: When downloading a file using the video tag, an extra character dot appears when saving the file! | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.3
### What operating system are you using?
Windows
### Operating System Version
Windows 11
### What arch are you using?
x64
### Last Known Working Electron version
19.0.6
### Expected Behavior
### When the video tag was used to download files, the save type was normal.
### Actual Behavior
## When downloading a file using the video tag, an extra character dot appears when saving the file! It has two character dot in the file type.
### Testcase Gist URL
https://gist.github.com/liulinboyi/2e687fbf07f0d9c836933817046ea043
### Additional Information
### I have a feeling that this bug exists from version 20. | https://github.com/electron/electron/issues/35429 | https://github.com/electron/electron/pull/35618 | 9f97c3e50a11b364f02f57209854bb4000ac1136 | 12a7d7eea52409ce02af6d5f51f8208f483f24d2 | 2022-08-24T16:35:28Z | c++ | 2022-09-13T18:47:04Z |
closed | electron/electron | https://github.com/electron/electron | 35,403 | ["shell/browser/native_window_mac.mm", "shell/browser/ui/cocoa/electron_ns_window.h", "shell/browser/ui/cocoa/electron_ns_window.mm"] | [Bug]: macOS - Using frame, roundedCorners, and fullscreen settings together automatically causes a crash | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.3
### What operating system are you using?
macOS
### Operating System Version
Catalina (confirmed on Big Sur too)
### What arch are you using?
x64
### Last Known Working Electron version
Not working in version 17-21
### Expected Behavior
Code such as the following should not crash Electron:
```
win = new BrowserWindow({
title: "Media Window",
frame: false, // These 3 options can't be
roundedCorners: false, // used concurrently on macOS
fullscreen: true // without causing a crash
});
```
### Actual Behavior
The following code crashes Electron on macOS:
```
win = new BrowserWindow({
title: "Media Window",
frame: false, // These 3 options can't be
roundedCorners: false, // used concurrently on macOS
fullscreen: true // without causing a crash
});
```
Any of the following 3 variants does **not** cause the crash:
```
win = new BrowserWindow({
title: "Media Window",
// frame: false,
roundedCorners: false,
fullscreen: true
});
```
```
win = new BrowserWindow({
title: "Media Window",
frame: false,
// roundedCorners: false,
fullscreen: true
});
```
```
win = new BrowserWindow({
title: "Media Window",
frame: false,
roundedCorners: false,
// fullscreen: true
});
```
### Testcase Gist URL
https://gist.github.com/sircharlo/c78e81a20c3eb4167037197b3803d76e
### Additional Information
Error looks like this:
Electron[785:12045] *** Assertion failure in -[ElectronNSWindow titlebarAccessoryViewControllers], /AppleInternal/BuildRoot/Library/Caches/com.apple.xbs/Sources/AppKit/AppKit-1894.60.100/AppKit.subproj/NSWindow.m:3337
| https://github.com/electron/electron/issues/35403 | https://github.com/electron/electron/pull/39747 | 2324c4d8fd6f7b2cd336d4e93591464544e12953 | ab185c058f894ddc556f67b2853de4adad2bb10c | 2022-08-22T15:16:24Z | c++ | 2023-09-11T07:38:10Z |
closed | electron/electron | https://github.com/electron/electron | 35,391 | ["shell/browser/api/electron_api_web_contents.cc", "shell/browser/api/electron_api_web_contents.h", "spec/chromium-spec.ts"] | [Bug]: sandboxed iframe can modify history of parent browsing context | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.3, same behavior going back to 17.0.0
### What operating system are you using?
macOS
### Operating System Version
macOS 12.5.1
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
n/a
### Expected Behavior
I expect that calling `history.back()` from a sandboxed iframe should do nothing, which is the behavior in Chrome 104.0.5112.101.
### Actual Behavior
Create three .html files:
page1.html:
```
<html>
<body>
Page 1
<a href="/page2.html">Go to Page 2</a>
</body>
</html>
```
page2.html:
```
<html>
<body>
Page 2
<iframe src="page_iframe.html" sandbox="allow-scripts"></iframe>
</body>
</html>
```
page_iframe.html:
```
<html>
<body>
<button onclick="history.back()">Go Back</button>
</body>
</html>
```
Serve them:
```
npx http-server .
```
Load page 1 in Electron:
```
mainWindow.loadURL('http://localhost:8080/page1.html')
```
and click the link to page2.html. Then click the `Go Back` button that is shown in the iframe on Page 2 and notice that the window is navigated back to page1.html.
### Testcase Gist URL
https://gist.github.com/7d7b3f33d4a6283e9e3cc000da2ab6f9
### Additional Information
This was previously reported in https://github.com/electron/electron/issues/3734 and https://github.com/electron/electron/issues/21680 but not with a reproducible case. | https://github.com/electron/electron/issues/35391 | https://github.com/electron/electron/pull/35420 | b0036ea43afda71e33c1450a0dc2c96c1f8f4ce4 | 730d9181b3de8561c1d87d82ccb923b1859122ae | 2022-08-20T01:02:16Z | c++ | 2022-09-09T00:08:56Z |
closed | electron/electron | https://github.com/electron/electron | 35,389 | ["patches/chromium/.patches", "patches/chromium/fix_revert_emulationhandler_update_functions_to_early_return.patch", "spec/api-web-contents-spec.ts"] | [Bug]: using debugger's Emulation.setDeviceMetricsOverride and changing origins(?) crashes Electron 21 | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
21.0.0-alpha.1
### What operating system are you using?
Ubuntu
### Operating System Version
Ubuntu 20.04
### What arch are you using?
x64
### Last Known Working Electron version
20.0.3
### Expected Behavior
When setDeviceMetricsOverride is active and a different URL is loaded, the application does not crash
### Actual Behavior
Electron crashes with a SIGSEGV, in all current Electron 21 alpha versions (1 through 4)
### Testcase Gist URL
https://gist.github.com/25c3fea282cc3a95af280ed1db54ab3d
### Additional Information
Run the application with Electron 21 and Electron 20. Click one of the buttons below the webview to change the URL
- On Electron 21, the app now crashes
- On Electron 20, the new URL is loaded
On Electron 21, when showing the electron site, you can navigate to a few other pages like docs, releases and apps. Going to Blog crashes the app. Blog is still the same origin so that's why the "(?)" is in the title.
Comment out these lines to prevent Electron 21 from crashing: https://gist.github.com/Kilian/25c3fea282cc3a95af280ed1db54ab3d#file-renderer-js-L18-L21 | https://github.com/electron/electron/issues/35389 | https://github.com/electron/electron/pull/35466 | f6bbad287a7b15a760a018b85a3bb667d384e3c0 | 4744674e933ebbd294878b60c93732500fd4217a | 2022-08-19T17:56:47Z | c++ | 2022-08-29T12:53:24Z |
closed | electron/electron | https://github.com/electron/electron | 35,384 | ["patches/node/feat_add_uv_loop_interrupt_on_io_change_option_to_uv_loop_configure.patch", "shell/common/node_bindings.cc"] | [Bug]: Mac crash on WebWorker destructor | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.3
### What operating system are you using?
macOS
### Operating System Version
12.3
### What arch are you using?
x64
### Last Known Working Electron version
20.0.2
### Expected Behavior
When destructing WebWorkers, it should not crash
### Actual Behavior
When destructing a WebWorker that uses WebAssembly, it crashes the v8 context.
### Testcase Gist URL
_No response_
### Additional Information
It seems linked to the recent change to the "stop_and_close_uv_loop" change, seen here: #35322
If I revert that single change, it does not crash.
Attached is a minidump
[minidump.zip](https://github.com/electron/electron/files/9382529/e9384a04-0947-4e26-b452-c85e84309bf9.zip)
| https://github.com/electron/electron/issues/35384 | https://github.com/electron/electron/pull/35422 | 2f23bdb19ea76029907ae247cc8eb47114df1d0c | 18475818486be806e891e8e6cb13b83c620ec0aa | 2022-08-19T13:30:36Z | c++ | 2022-08-29T13:45:48Z |
closed | electron/electron | https://github.com/electron/electron | 35,312 | ["shell/browser/ui/views/win_caption_button_container.cc", "shell/browser/ui/views/win_frame_view.cc"] | [Bug]: WCO separator colour incorrect in v20 | Alright, I guess I wasn't running it in the right electron version, but with fiddle it also is off-center. This issue doesn't appear in v20.0.0, but in v20, the WCO just looks weird. Maybe that bug occurs only with some specific settings, I'll look into it later.
This is what I mean by "weird":
_Originally posted by @wheezard in https://github.com/electron/electron/issues/35038#issuecomment-1204061897_ | https://github.com/electron/electron/issues/35312 | https://github.com/electron/electron/pull/35569 | 17d501616368e2a0c709cbc2174de38e1781e8c3 | 4fb4167b8b42276710b658a3c62f0ebe9fffe446 | 2022-08-11T16:12:49Z | c++ | 2022-09-08T13:11:51Z |
closed | electron/electron | https://github.com/electron/electron | 35,304 | ["patches/chromium/.patches", "patches/chromium/fix_allow_guest_webcontents_to_enter_fullscreen.patch", "patches/chromium/fix_on-screen-keyboard_hides_on_input_blur_in_webview.patch", "patches/chromium/webview_fullscreen.patch"] | [Bug]: Unable to exit pdf present mode in WebView | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.4
### What operating system are you using?
Windows
### Operating System Version
Windows 10
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
Should be able to exit pdf present mode, in Chrome it is possible to exit present mode by pressing Esc.
### Actual Behavior
Unable to exit pdf present mode once it is selected. The Esc key only affects the windows full-screen mode, but the pdf view is unchanged.
### Testcase Gist URL
https://gist.github.com/4868233713a4119d96c9273494157843
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35304 | https://github.com/electron/electron/pull/39616 | 3bdc7ce64a79120c4001d0c614e4ad037246229e | d42a94dddeb86d349c050c2ee59b135bd6c8245c | 2022-08-11T09:03:54Z | c++ | 2023-08-25T18:11:58Z |
closed | electron/electron | https://github.com/electron/electron | 35,282 | ["shell/browser/serial/electron_serial_delegate.cc", "shell/browser/serial/serial_chooser_context.cc", "shell/browser/serial/serial_chooser_context.h"] | [Bug]: WebSerial - SerialPort doesn't open in Electron 20 | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.1
### What operating system are you using?
Windows
### Operating System Version
Win 10 Pro 19043
### What arch are you using?
x64
### Last Known Working Electron version
19.0.11
### Expected Behavior
WebSerial API should open a serial port.
### Actual Behavior
The ``getInfo`` method works fine, but ``open`` fails to open the port.
### Testcase Gist URL
https://gist.github.com/17afc65be515f3a1aec6256cb69f069d
https://github.com/rafaelpimpa/electron-webserial-bug-repro
### Additional Information
Error thrown is ``NetworkError: Failed to open serial port.``
I didn't use an Arduino to test, but it should work the same, since it works in v19.
Reproduction has the same code from the docs. | https://github.com/electron/electron/issues/35282 | https://github.com/electron/electron/pull/35306 | cbc1ee5775375b7ec92f8c38bcbb157833b1b8a0 | 672539187c779cf874dfcecc198bcfbdec85aa97 | 2022-08-09T13:47:55Z | c++ | 2022-08-15T15:49:20Z |
closed | electron/electron | https://github.com/electron/electron | 35,264 | ["shell/browser/ui/drag_util_mac.mm"] | [Bug]: WebContents.startDrag feature seems to have changed unexpectedly from electron version 19.0.4 to 19.0.5 | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.5
### What operating system are you using?
macOS
### Operating System Version
Mac OS Montrey 12.2.1 (Apple M1)
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
19.0.4
### Expected Behavior
1. This feature should copy the target file.
2. With reference to the sample code:
```js
console.timeEnd('drag2') // line 38 of main.js
```
This line should be called at the end of dragging.
### Actual Behavior
1. This feature is moving the target file instead of copying. If you run the sample code in the attached Gist URL, you will find that you are unable to drag the same target more than once.
2. With reference to the same sample code as [2.] in Expected Behavior, above:
```js
console.timeEnd('drag2') // line 38 of main.js
```
This line is being called at the start of dragging, instead of the end.
### Testcase Gist URL
https://gist.github.com/34bc977bc272030760b586d53961e0f1
### Additional Information
The sample code I provided in the Gist URL was created from the official [Sample Code](https://www.electronjs.org/docs/latest/tutorial/native-file-drag-drop) with an addition of `console.time` and `console.timeEnd`
I have a feeling this bug was introduced when merging the following Pull Request:
https://github.com/electron/electron/pull/34615
### Questions:
How can I do the following, like before (in 19.0.4 and earlier versions)
1. Copy the target file on drag
2. Act on the end of the user's dragging
### Behavior on Windows:
In Windows 11, the behavior is as expected, and identical across all Electron versions, including 19.0.4 and 19.0.5. | https://github.com/electron/electron/issues/35264 | https://github.com/electron/electron/pull/35963 | 9006f0e0c5a8cd529dabc0de822e9098e574c799 | b3fd5eb2585aabb9182d6941c6da3f688fa3f5bb | 2022-08-08T14:19:45Z | c++ | 2022-10-11T16:19:59Z |
closed | electron/electron | https://github.com/electron/electron | 35,255 | ["docs/api/session.md"] | [Bug]: Wrong TypeScript type of ses.getStoragePath | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.1
### What operating system are you using?
macOS
### Operating System Version
macOS Monterey 12.5
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
The return type of `ses.getStoragePath()` should be `string | null`. This is returned when actually calling the function and is also described in https://github.com/electron/electron/blob/91f9436ad8e6028bf2909008e5814867289b9310/docs/api/session.md#sesgetstoragepath.
### Actual Behavior
The generated TypeScript type definition says `ses.getStoragePath()` returns `void`, as can also be seen in https://unpkg.com/browse/[email protected]/electron.d.ts#L7763.
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35255 | https://github.com/electron/electron/pull/35288 | 8646bf8d304d8dc97988227cdf76007c6fa4bfdf | 1d95b98cc86a4b4ed757931fd38105b06e3e2556 | 2022-08-08T09:07:38Z | c++ | 2022-08-10T05:39:36Z |
closed | electron/electron | https://github.com/electron/electron | 35,223 | ["shell/browser/native_window_views_win.cc"] | [Bug]: 'restore' event not emitted after minimize and maximize application | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.1
### What operating system are you using?
Windows
### Operating System Version
Windows 10 Pro 21H1
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
Event 'restore' should notify to application
### Actual Behavior
After clicking maximize application, the event 'restore' does not notify after minimize
`
BrowserWindow.on( 'restore', () => { // nothing happend });
`
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35223 | https://github.com/electron/electron/pull/35342 | 0ff6508f5bffd84829749c97c7346fc2fc776563 | 3ce35f224efa3a6e9bfc9b642816961b8691491b | 2022-08-04T11:04:47Z | c++ | 2022-08-23T01:32:42Z |
closed | electron/electron | https://github.com/electron/electron | 35,219 | ["patches/node/fix_crash_caused_by_gethostnamew_on_windows_7.patch", "patches/node/fix_suppress_clang_-wdeprecated-declarations_in_libuv.patch"] | [Bug]: Win7 can't open electron app after upgrade to v19.x.x | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.x.x
### What operating system are you using?
Windows
### Operating System Version
win7
### What arch are you using?
x64
### Last Known Working Electron version
17.x.x
### Expected Behavior
Can open and work.
### Actual Behavior
Can not open program in win7 operate system.
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35219 | https://github.com/electron/electron/pull/35702 | fa3cd1747556b66a3e714f99e988e47334187e44 | 7d3f22dd32365d5432583d936fa4c7f3d480ef45 | 2022-08-04T08:54:15Z | c++ | 2022-09-16T05:33:01Z |
closed | electron/electron | https://github.com/electron/electron | 35,170 | ["build/fuses/fuses.json5", "docs/tutorial/fuses.md", "patches/chromium/.patches", "patches/chromium/create_browser_v8_snapshot_file_name_fuse.patch", "shell/app/electron_main_delegate.cc", "shell/app/electron_main_delegate.h"] | [Feature Request]: Allow custom v8 snapshots to be used in the main process and the default snapshot in the browser process | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a feature request that matches the one I want to file, without success.
### Problem Description
We are using `electron-mksnapshot` to generate a custom v8 snapshot for our application that is desired to run **only in the main process**, not the renderer process. While debugging our app, we noticed that the snapshot is available to the renderer process **even with nodeIntegration disabled**.
We noticed that electron is [patching](https://github.com/electron/electron/blob/main/patches/chromium/load_v8_snapshot_in_browser_process.patch?rgh-link-date=2022-08-02T00%3A24%3A21Z) chromium to always load the v8 snapshots, with a comment indicating that this is required. Is loading the v8 snapshot required even when nodeIntegration is disabled?
We would like for there to be a way to use our custom snapshot in the main process but use the default snapshot (or no snapshot with nodeIntegration disabled) in the browser process.
### Proposed Solution
This could be done a variety of ways, but one possibility would be to update the [patch](https://github.com/electron/electron/blob/main/patches/chromium/load_v8_snapshot_in_browser_process.patch?rgh-link-date=2022-08-02T00%3A24%3A21Z) to not load the custom snapshot when node integration is disabled. Or if it's desired sometimes to have node integration disabled **and** a v8 snapshot included, you could have an additional parameter that specifies as such.
### Alternatives Considered
One alternative for us would be to have separate main electron processes: one for the bulk of our main code that uses the snapshot, and one to purely launch the browser that doesn't launch the snapshot. This is undesirable as it significantly complicates the code and potentially bloats the size of our binary.
### Additional Information
_No response_ | https://github.com/electron/electron/issues/35170 | https://github.com/electron/electron/pull/35266 | 5cf15cdab7f3d055a225dbc50c53d62614250468 | f25c87dc7024397526bbe221efdaa0260392c243 | 2022-08-02T00:24:21Z | c++ | 2022-09-15T20:21:16Z |
closed | electron/electron | https://github.com/electron/electron | 35,157 | ["docs/tutorial/tutorial-3-preload.md"] | [Docs] Wrong usage of `ipcMain.handle` in `/tutorial-preload` | https://github.com/electron/electron/blob/b67c69d741c648569d86c5a7846f2559219f9b14/docs/tutorial/tutorial-3-preload.md?plain=1#L198
1. Follow tutorial on a Mac
2. close only window
3. open another window
4. notice error
It should not be called for every window but just once inside main.js
<img width="283" alt="Screen Shot 2022-08-01 at 13 48 44" src="https://user-images.githubusercontent.com/163636/182133832-033f31ee-423f-4ab9-95d7-8cbbb5f7b1da.png">
| https://github.com/electron/electron/issues/35157 | https://github.com/electron/electron/pull/35479 | 87145c393c08203e6e18f0615a13baff59fb3e30 | c16baa063a404c9afefe452ef88a28cfa7cc2241 | 2022-08-01T10:56:00Z | c++ | 2022-09-07T13:37:29Z |
closed | electron/electron | https://github.com/electron/electron | 35,115 | ["docs/tutorial/mac-app-store-submission-guide.md"] | [Bug] Update Mac App Store submission documentation | https://www.electronjs.org/docs/latest/tutorial/mac-app-store-submission-guide#upload
References "Application Loader" for submitting to Mac App Store. Apple no longer supports Application Loader and it's no longer part of Xcode. | https://github.com/electron/electron/issues/35115 | https://github.com/electron/electron/pull/35116 | 15540975ff3f3979ba93ed560fdb18489c458b98 | 71b8804fd0b53c4b87de8863a1be627732f60c89 | 2022-07-28T14:09:02Z | c++ | 2022-11-01T21:07:50Z |
closed | electron/electron | https://github.com/electron/electron | 35,046 | ["shell/browser/api/electron_api_web_contents.cc", "shell/browser/electron_browser_context.cc", "shell/browser/electron_browser_context.h", "spec-main/api-web-contents-spec.ts"] | [Bug]: webContents.getUserAgent() is empty since 20.0.0-alpha.6 | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.0-alpha.6
### What operating system are you using?
Ubuntu
### Operating System Version
Ubuntu 20.04
### What arch are you using?
x64
### Last Known Working Electron version
20.0.0-alpha.5
### Expected Behavior
`webContents.getUserAgent()` returns a string with a user Agent
### Actual Behavior
webContents.getUserAgent() returns an empty string
### Testcase Gist URL
https://gist.github.com/33fd4a9ac6db6b28b1dcc5b765edb622
### Additional Information
In the console the user agent is printed in alpha.5, but empty in alpha.6. Looking at the diff between alpha.5 and alpha.6, this PR seems the likely cause: https://github.com/electron/electron/pull/34524 / #34481
It's worth noting that the renderer does have a filled-in userAgent. | https://github.com/electron/electron/issues/35046 | https://github.com/electron/electron/pull/35069 | 8004cb8722c8d52f4cd553bbaea86750149e8ca4 | 9028bb79a85fedf64230c2f1c6a29ab072c98c20 | 2022-07-25T14:12:09Z | c++ | 2022-07-29T15:09:47Z |
closed | electron/electron | https://github.com/electron/electron | 34,996 | ["shell/browser/electron_browser_main_parts.cc"] | [Bug]: SIGTRAP when starting electron >= 20.0.0-beta.9 | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
20.0.0-beta.9 and 20.0.0-beta.10
### What operating system are you using?
Other Linux
### Operating System Version
Fedora 35
### What arch are you using?
x64
### Last Known Working Electron version
All older versions, including 19 and 20.0.0-beta.8
### Expected Behavior
Electron quickstart app should work, as should ANY other electron app, I have tested a bunch of them, but all gave the same results, after you install a recent 20 beta.
### Actual Behavior
They don't start if I use version 20.0.0-beta.9 or 20.0.0-beta.10, older versions *do* work. It only happens on *one* of the devices I own, *all* others are fine, but this one does not handle these new versions, while older ones work. This must have been introduced somewhere in [these commits between those releases](https://github.com/electron/electron/compare/v20.0.0-beta.8...v20.0.0-beta.9).
### Testcase Gist URL
https://github.com/electron/electron-quick-start/
### Additional Information
Running electron with `gdb` does not give a lot of info, unless I'm running it incorrectly, please see below and let me know if this should be done differently:
```gdb
jelmerro ~/electron-quick-start (master)$ gdb --args node node_modules/.bin/electron .
GNU gdb (GDB) Fedora 12.1-1.fc35
Copyright (C) 2022 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Type "show copying" and "show warranty" for details.
This GDB was configured as "x86_64-redhat-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<https://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from node...
This GDB supports auto-downloading debuginfo from the following URLs:
https://debuginfod.fedoraproject.org/
Enable debuginfod for this session? (y or [n]) y
Debuginfod has been enabled.
To make this setting permanent, add 'set debuginfod enabled on' to .gdbinit.
Reading symbols from /home/jelmerro/.cache/debuginfod_client/adca03b9b7e322bf173d4a6972adea6c254bf914/debuginfo...
(gdb) run
Starting program: /usr/bin/node node_modules/.bin/electron .
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
[New Thread 0x7ffff23c0640 (LWP 12930)]
[New Thread 0x7ffff1bbf640 (LWP 12931)]
[New Thread 0x7ffff13be640 (LWP 12932)]
[New Thread 0x7ffff0bbd640 (LWP 12933)]
[New Thread 0x7fffebfff640 (LWP 12934)]
[New Thread 0x7ffff7fbe640 (LWP 12935)]
[Detaching after fork from child process 12936]
/home/jelmerro/electron-quick-start/node_modules/electron/dist/electron exited with signal SIGTRAP
[Thread 0x7fffebfff640 (LWP 12934) exited]
[Thread 0x7ffff0bbd640 (LWP 12933) exited]
[Thread 0x7ffff13be640 (LWP 12932) exited]
[Thread 0x7ffff1bbf640 (LWP 12931) exited]
[Thread 0x7ffff23c0640 (LWP 12930) exited]
[Thread 0x7ffff7fbe640 (LWP 12935) exited]
[Inferior 1 (process 12927) exited with code 01]
``` | https://github.com/electron/electron/issues/34996 | https://github.com/electron/electron/pull/35075 | ff804e3a748c72f4812af8ccf3b1d7442a468f4a | 62001dc6cbfd78a82d2fca849ca5eb0504cd0b71 | 2022-07-20T15:23:25Z | c++ | 2022-07-27T04:44:44Z |
closed | electron/electron | https://github.com/electron/electron | 34,939 | ["shell/browser/ui/views/menu_bar.cc", "shell/browser/ui/views/menu_bar.h"] | [Bug]: Dark pink hover/focus background for menus in menu bar | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.8
### What operating system are you using?
Windows
### Operating System Version
Windows 10 build 19044
### What arch are you using?
x64
### Last Known Working Electron version
Electron v17.4.11
### Expected Behavior
Standard menu bar menus should use a neutral gray hover/focus background (this used to be the case in the past).
### Actual Behavior
Not clear why, but it seems that standard menus in electron now (at least on my machine) have a dark pink hover/focus background. Not sure where this is being picked up from. None of my other apps on Windows use this colour, and it's certainly not part of my OS-wide theme.
I get this when running the vanilla quick-start example https://github.com/electron/electron-quick-start
first spotted on this electron-based application I sometimes contribute to https://github.com/ThePacielloGroup/CCAe/issues/251
### Testcase Gist URL
from @jediwade https://gist.github.com/eb8cfee3a6ae8fd88bb808f8b12e5f62
### Additional Information
_No response_ | https://github.com/electron/electron/issues/34939 | https://github.com/electron/electron/pull/37766 | a75b892e9080753dcc5e1344f2a59fa641a65be1 | 601217218759d088f12680a428fead8c8b26a5e0 | 2022-07-17T14:37:19Z | c++ | 2023-03-31T16:06:09Z |
closed | electron/electron | https://github.com/electron/electron | 34,887 | ["shell/browser/serial/electron_serial_delegate.cc", "shell/browser/serial/electron_serial_delegate.h", "shell/browser/serial/serial_chooser_context.cc", "shell/browser/serial/serial_chooser_context.h", "shell/browser/serial/serial_chooser_controller.cc", "shell/browser/serial/serial_chooser_controller.h"] | Web Serial API removed and added events are never called | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.8
### What operating system are you using?
macOS
### Operating System Version
macOS Monterey 12.4
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
I want to update UI when picking a port if a serial port is plugged in or unplugged.
Test Steps:
1, Click button to call `navigator.serial.requestPort()`
2, Show a modal to show port list in the `select-serial-port` event handler
3, Plug in or unplug a device to check the `serial-port-added` and `serial-port-removed` events
```
modal.webContents.session.on('select-serial-port', (event, portList, webContents, callback) => {
//Add listeners to handle ports being added or removed before the callback for `select-serial-port`
//is called.
modal.webContents.session.on('serial-port-added', (event, port) => {
console.log('serial-port-added FIRED WITH', port)
//Optionally update portList to add the new port
})
modal.webContents.session.on('serial-port-removed', (event, port) => {
console.log('serial-port-removed FIRED WITH', port)
//Optionally update portList to remove the port
})
// modal.webContents.openDevTools();
modal.webContents.send('update-list', portList);
modal.show();
event.preventDefault();
ipcMain.once('serial-port', (_event, portId = '') => {
callback(portId);
modal.hide();
});
});
```
### Actual Behavior
But the `select-serial-port` and `select-serial-removed` events never called. Did I use these api in a wrong way?
### Testcase Gist URL
https://gist.github.com/0bbeeaa1c9e8f5affb6b59663870616e
### Additional Information
_No response_ | https://github.com/electron/electron/issues/34887 | https://github.com/electron/electron/pull/34958 | aeba6ca973c8ed8a8b75cceb4e0dde33f3595850 | 648c9934c01c12ea082a18fdf690c37e3112ce12 | 2022-07-12T03:31:13Z | c++ | 2022-07-25T14:50:19Z |
closed | electron/electron | https://github.com/electron/electron | 34,852 | ["shell/browser/native_window_views.cc", "shell/browser/notifications/linux/libnotify_notification.cc", "shell/common/platform_util.h", "shell/common/platform_util_linux.cc"] | [Bug]: Electorn 18 app is missing desktopFileName window property | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
18.2.4
### What operating system are you using?
Other Linux
### Operating System Version
Manjaro Linux
### What arch are you using?
x64
### Last Known Working Electron version
17.4.3
### Expected Behavior
`desktopFileName` window property should not be empty
### Actual Behavior
`desktopFileName` window property is empty
With Electron 17.4.3 - OK
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/34852 | https://github.com/electron/electron/pull/34855 | 8f3fb8db09ac049894c8ab099979810899f9435b | f63bba8ce24917f1c78a8804496fe0f5b461b0af | 2022-07-08T12:02:26Z | c++ | 2022-07-11T18:26:18Z |
closed | electron/electron | https://github.com/electron/electron | 34,822 | ["shell/browser/native_window_mac.mm"] | [Bug]: `setTrafficLightPosition` is reset when calling `setRepresentedFilename` | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.7
### What operating system are you using?
macOS
### Operating System Version
12.4
### What arch are you using?
arm64 (including Apple Silicon)
### Last Known Working Electron version
_No response_
### Expected Behavior
The traffic light position should remain stable.
### Actual Behavior
The traffic light position changes after the presented file name is changed for the window.
### Testcase Gist URL
https://gist.github.com/9ad043673ca425c92bbb4e1f951c455c
### Additional Information
| https://github.com/electron/electron/issues/34822 | https://github.com/electron/electron/pull/34834 | e83c3ec74442f0533e2d66c53fa84a075ebc20ae | 1941c88442b61f3e83125f197becc83cf6975a9c | 2022-07-05T04:24:03Z | c++ | 2022-07-08T06:33:42Z |
closed | electron/electron | https://github.com/electron/electron | 34,820 | ["patches/chromium/.patches", "patches/chromium/add_maximized_parameter_to_linuxui_getwindowframeprovider.patch", "shell/browser/ui/views/client_frame_view_linux.cc", "shell/browser/ui/views/client_frame_view_linux.h"] | [Bug]: wayland window decorations mangled and not clickable | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
19.0.7
### What operating system are you using?
Ubuntu
### Operating System Version
22.04
### What arch are you using?
x64
### Last Known Working Electron version
19.0.0-alpha.3
### Expected Behavior
Running under wayland natively (`--ozone-platform=wayland`) and enabling window decorations (`app.commandLine.appendSwitch('enable-features', 'WaylandWindowDecorations');` in main.js) should show a title bar which is clickable.
Below is from 19.0.0-alpha.3 where it still worked:
### Actual Behavior
Running under wayland natively (`--ozone-platform=wayland`) and enabling window decorations (`app.commandLine.appendSwitch('enable-features', 'WaylandWindowDecorations');` in main.js) shows mangled transparent border on the left side instead of a title bar and is not clickable:
On **electron 19** is broke during the alpha and beta cycle:
Below is from 19.0.0-alpha.4 where it first broke, here the buttons are still clickable:
Starting with 19.0.0-beta.4 the buttons are also no longer clickable:
With 19.0.0 up and until 19.0.7 the same, broken titlebar and not clickable:
On **electron 18-x-y** it also broke between 18.2.0 and 18.2.1, but only the buttons cannot be clicked any longer:
18.2.0 still everything ok, titlebar there and clickable:
18.2.1: Titlebar still there, but no longer clickable:
Up and including 18.3.5 the titlebar stays there, but not clickable:
### Testcase Gist URL
https://gist.github.com/csett86/5d45c47cb1d5331b9ede0788ff5ae284
### Additional Information
To reproduce with the gist, you need to add ```--ozone-platform=wayland``` to the user-provided flags, as I could find a way to add this to the fiddle (the WaylandWindowDecorations are part of the fiddle):
| https://github.com/electron/electron/issues/34820 | https://github.com/electron/electron/pull/34955 | 4e8480b15b63cf71648f8f9357713b8827dec27c | 7b8fb2b07418c7bad6ce811ca75c926ee53663fc | 2022-07-04T18:59:46Z | c++ | 2022-08-03T08:51:52Z |
closed | electron/electron | https://github.com/electron/electron | 34,813 | ["shell/browser/printing/print_view_manager_electron.cc"] | [Bug]: Can not print contents of browser window silently in Ubuntu linux | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
18.2.3
### What operating system are you using?
Ubuntu
### Operating System Version
Ubuntu 20.04
### What arch are you using?
x64
### Last Known Working Electron version
none
### Expected Behavior
I want to print the contents of browser window silently on a POS printer. The code attached at Gist section works fine on windows, but doesn't work in Ubuntu. My targeted platform is ubuntu linux 20.04 and above
### Actual Behavior
The output at the console for the Gist is :
Console ready 🔬
Saving files to temp directory...
Saved files to /tmp/tmp-17209-xBE03rN99Kka
Electron v18.2.3 started.
libva error: /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so init failed
[19927:0702/212214.938092:ERROR:sandbox_linux.cc(377)] InitializeSandbox() called with multiple threads in process gpu-process.
[
{
name: 'DCP-7030',
displayName: 'DCP-7030',
description: 'Brother DCP-7030',
status: 5,
isDefault: false,
options: {
copies: '1',
'device-uri': 'usb://Brother/DCP-7030?serial=000D0N969645',
finishings: '3',
'job-cancel-after': '10800',
'job-hold-until': 'no-hold',
'job-priority': '50',
'job-sheets': 'none,none',
'marker-change-time': '0',
'number-up': '1',
'printer-commands': 'none',
'printer-info': 'Brother DCP-7030',
'printer-is-accepting-jobs': 'true',
'printer-is-shared': 'true',
'printer-is-temporary': 'false',
'printer-location': 'embeddedSaugat',
'printer-make-and-model': 'Brother DCP-7030, using brlaser v6',
'printer-state': '5',
'printer-state-change-time': '1651417533',
'printer-state-reasons': 'paused',
'printer-type': '4164',
'printer-uri-supported': 'ipp://localhost/printers/DCP-7030',
system_driverinfo: 'Brother DCP-7030, using brlaser v6'
}
},
{
name: 'POS80QMS',
displayName: 'POS80QMS',
description: '80mm printer for QMS Project',
status: 3,
isDefault: true,
options: {
copies: '1',
'device-uri': 'usb://Unknown/Printer',
finishings: '3',
'job-cancel-after': '10800',
'job-hold-until': 'no-hold',
'job-priority': '50',
'job-sheets': 'none,none',
'marker-change-time': '0',
'number-up': '1',
'printer-commands': 'none',
'printer-info': '80mm printer for QMS Project',
'printer-is-accepting-jobs': 'true',
'printer-is-shared': 'false',
'printer-is-temporary': 'false',
'printer-location': 'holus',
'printer-make-and-model': 'POS-80',
'printer-state': '3',
'printer-state-change-time': '1655307054',
'printer-state-reasons': 'none',
'printer-type': '2248708',
'printer-uri-supported': 'ipp://localhost/printers/POS80QMS',
system_driverinfo: 'POS-80'
}
}
]
(node:19895) electron: Deprecation Warning: getPrinters() is deprecated. Use the asynchronous and non-blocking version, getPrintersAsync(), instead.
(Use `electron --trace-warnings ...` to show where the warning was created)
failed
[19895:0702/212216.612479:ERROR:print_view_manager_base.cc(85)] Invalid printer settings The selected printer is not available or not installed correctly. <br> Check your printer or try selecting another printer.
Electron exited with code 0.
### Testcase Gist URL
https://gist.github.com/saugatdai/5ba400d42e3fa4be66d323b2d826fe50
### Additional Information
My project is not going to production because of this problem. Please help with some solutions and suggestions. | https://github.com/electron/electron/issues/34813 | https://github.com/electron/electron/pull/34893 | eb8c9452cb38b6bd32bdc92724cc1fb42a3d43dc | 05d4966251580cd4ae95479ccbbfdd5e8a70702f | 2022-07-02T15:55:59Z | c++ | 2022-07-19T12:46:08Z |
closed | electron/electron | https://github.com/electron/electron | 34,801 | ["default_app/main.ts"] | [Bug]: Electron icon shown instead of app's on Wayland | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searched the [issue tracker](https://www.github.com/electron/electron/issues) for a bug report that matches the one I want to file, without success.
### Electron Version
18.2.4
### What operating system are you using?
Other Linux
### Operating System Version
Linux 5.18.7-arch1-1 #1 SMP PREEMPT_DYNAMIC
### What arch are you using?
x64
### Last Known Working Electron version
_No response_
### Expected Behavior
* I use Linux with a KDE environment
* I run an electron app with Wayland using chromium flags --enable-features=UseOzonePlatform,WebRTCPipeWireCapturer --ozone-platform=wayland
* The default Electron icon is shown, minimize to tray doesn't work and many other problems
* The problem is similar as described here
https://nicolasfella.de/posts/fixing-wayland-taskbar-icons/ but instead of the Wayland icon, it is the Electron icon.
### Actual Behavior
* The proper icon is shown as when running without Wayland flags
* Can minimize/close to the system tray
### Testcase Gist URL
_No response_
### Additional Information
_No response_ | https://github.com/electron/electron/issues/34801 | https://github.com/electron/electron/pull/35997 | 76880be6d2cc466a44cf437bf34bf32f94a9f6ad | 295c5331eeb2cce40419b6a54875fba88d61422d | 2022-06-30T01:32:30Z | c++ | 2022-10-17T08:34:24Z |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.