* fix: do not cancel CORS preflight request on proxy auth. (#29266)
* fix: do not cancel CORS preflight request on proxy auth.
If connecting via proxy, preflight request can receive 407
header response from proxy. This does not mean request
was finished even though it received headers (from proxy,
not the destination server), so prevent "completing"
and most importantly deleting it, which causes request
to be canceled in network layer. Just continue to monitor it
and await proper response from server. Also add circut breaker
to cancel request if proxy auth failed 3 times (for example
user keeps cancelling auth). This behavior happens only
when app registered WebRequest api listeners.
* Port chromium webrequest changes to electron code.
Move relevant parts of chromium WebRequestProxyingURLLoaderFactory from
https://chromium-review.googlesource.com/c/chromium/src/+/2011781
into electron ProxyingURLLoaderFactory.
* Update code to upstreamed version and remove retyr count failsafe.
Co-authored-by: Milan Burda <milan.burda@gmail.com>
* chore: add required header
Co-authored-by: marekharanczyk <48673767+marekharanczyk@users.noreply.github.com>
Co-authored-by: Milan Burda <milan.burda@gmail.com>
Co-authored-by: Cheng Zhao <zcbenz@gmail.com>
* fix: microtasks policy in CreateEnvironment
Microtasks policy should not be updated for the renderer because
`NodeBindings::CreateEnvironment` might be entered with or without
`UvRunOnce()` on stack. One of the examples of such calls is
`window.open()` which is possible to invoke while `uv_run()` is still
running (e.g. with `setImmediate()`).
All in all, it doesn't matter that much which policy we use since
`v8::MicrotasksScope` has a check for the policy in its destructor and
no commits will be made if the policy is `kExplicit`. It is important,
however, to not change the policy in the middle of `UvRunOnce()` so we
should respect whatever we currently have and move on.
Fix: #29463
* Move test to a better place
* Update spec-main/fixtures/crash-cases/setimmediate-window-open-crash/index.html
Co-authored-by: Jeremy Rose <nornagon@nornagon.net>
* Update spec-main/fixtures/crash-cases/setimmediate-window-open-crash/index.html
Co-authored-by: Jeremy Rose <nornagon@nornagon.net>
* simplify crash-case
* comment
* fix comment
Co-authored-by: Fedor Indutny <fedor@indutny.com>
Co-authored-by: Fedor Indutny <79877362+indutny-signal@users.noreply.github.com>
Co-authored-by: Jeremy Rose <nornagon@nornagon.net>
Co-authored-by: Fedor Indutny <indutny@signal.org>
* ci: run linux arm tests on CircleCI
* cleanup electron dirs after testing
(cherry picked from commit 1c0a6045fb)
* use start-stop-daemon to kill Xvfb
(cherry picked from commit 1d10a68c31)
* Update represented-file fiddle.
* add index and code back to guide
Co-authored-by: Kevin Hartman <kevin@hart.mn>
Co-authored-by: Ethan Arrowood <ethan.arrowood@gmail.com>
* build: restore workspace on osx builds to get docs-only-change file
* build: clean up attached workspace before checking out from cache
Co-authored-by: Samuel Attard <samuel.r.attard@gmail.com>
Unlike the other files, this file had its executable bit set in its file
mode. This change removes the executable bit to align its file mode with
the rest of the files.
Signed-off-by: Darshan Sen <raisinten@gmail.com>
Co-authored-by: Darshan Sen <raisinten@gmail.com>
* chore: bump chromium in DEPS to 93.0.4536.0
* chore: update patches
* Make InkDrop a View class property
Refs: https://chromium-review.googlesource.com/c/chromium/src/+/2918475
Replace old views::Button::ink_drop() calls with views::InkDrop::Get()
* Convert use of gfx::ImageSkia to ui::ImageModel for WidgetDelegates
Refs: https://chromium-review.googlesource.com/c/chromium/src/+/2935942
Update GetWindowIcon(), GetWindowAppIcon(), and GetDevToolsWindowIcon()
to return ui::ImageModel instead of gfx::ImageSkia.
Co-authored-by: electron-roller[bot] <84116207+electron-roller[bot]@users.noreply.github.com>
Co-authored-by: PatchUp <73610968+patchup[bot]@users.noreply.github.com>
Co-authored-by: Charles Kerr <charles@charleskerr.com>
* chore: bump chromium in DEPS to 93.0.4533.0
* chore: update patches
* chore: bump chromium in DEPS to 93.0.4534.0
* chore: update patches
* chore: bump chromium in DEPS to 93.0.4535.0
* chore: remove PdfViewerDocumentProperties from feature-flag conditional
https://chromium-review.googlesource.com/c/chromium/src/+/2938363
This feature is now enabled by default upstream.
* chore: remove kPdfViewerPresentationMode from feature-flag conditional
https://chromium-review.googlesource.com/c/chromium/src/+/2938037
This feature is now enabled by default upstream.
Co-authored-by: electron-roller[bot] <84116207+electron-roller[bot]@users.noreply.github.com>
Co-authored-by: PatchUp <73610968+patchup[bot]@users.noreply.github.com>
Co-authored-by: Charles Kerr <charles@charleskerr.com>
* remove version information from html
* change format for readability
* clarify which console the message should appear in
* minor changes to renderer.md
* update UI on click instead of developer console
* remove node-integration and fix md
* update content
* chore: remove ****
Co-authored-by: Jeremy Foster <jeremy.foster@live.com>
Co-authored-by: Ethan Arrowood <ethan.arrowood@gmail.com>
Co-authored-by: Cheng Zhao <github@zcbenz.com>
Right now, if executing `xcrun` fails, then the error message prints the
second argument to the `xcrun.py` script, which is the first argument to
the tool that `xcrun` is executing, making the whole error message quite
confusing.
Consider the following error:
```
python ../../third_party/squirrel.mac/build/xcrun.py dtrace -h -s /private/tmp/20210531211008-def376dc/src/third_party/squirrel.mac/vendor/ReactiveObjC/ReactiveObjC/RACSignalProvider.d -o /private/tmp/20210531211008-def376dc/src/out/release/gen/third_party/squirrel.mac/dtrace/RACSignalProvider.h
xcrun script '-h' failed with code '71':
xcrun: error: can't exec '/tmp/20210531211008-def376dc/dtrace' (errno=Permission denied)
```
The command that `xcrun` is executing is `dtrace`, but the error just
mentions the `-h` flag.
Notes: none
Signed-off-by: Juan Cruz Viotti <jv@jviotti.com>
Co-authored-by: Juan Cruz Viotti <jv@jviotti.com>
* fix: correctly handle shifted char in accelerator
* test: use actual accelerator of NSMenuItem
* chore: simplify KeyboardCodeFromStr
* chore: GetAcceleratorTextAt is testing only
Co-authored-by: Cheng Zhao <zcbenz@gmail.com>
