electron/patches/ffmpeg/link_with_loader_path.patch
electron-roller[bot] 061e2e5e73
chore: bump chromium to 113.0.5651.0 (main) (#37553)
* chore: bump chromium in DEPS to 113.0.5645.0

* chore: update patches/chromium/mas_avoid_usage_of_private_macos_apis.patch

Xref: https://chromium-review.googlesource.com/c/chromium/src/+/4300129

Fix simple code shear

* chore: update patches/chromium/build_only_use_the_mas_build_config_in_the_required_components.patch

Xref: https://chromium-review.googlesource.com/c/chromium/src/+/4297496

Xref: https://chromium-review.googlesource.com/c/chromium/src/+/4327491

patch-fuzz update; no manual changes

* chore: remove patches/chromium/fix_x11_window_restore_minimized_maximized_window.patch

Xref: https://chromium-review.googlesource.com/c/chromium/src/+/4252946

Upstreamed by zcbenz, so local patch is no longer needed

* chore: update chromium/printing.patch

Xref: https://chromium-review.googlesource.com/c/chromium/src/+/4313019

Remove cookie parameter from PrintViewManagerBase::UpdatePrintSettings()

* chore: remove NOTIMPLEMENTED BrowserProcessImpl::GetBreadcrumbPersistentStorageManager()

Xref: https://chromium-review.googlesource.com/c/chromium/src/+/4247145

method removed upstream, so we do not need to add a stub for it in the subclass

* chore: remove PrintViewManagerElectron::UpdatePrintSettings()

Xref: https://chromium-review.googlesource.com/c/chromium/src/+/4313019

Previously, our implementation checked to see if we recognized the
cookie param that was passed in. If so, we reported a bad message.
Otherwise, we passed it up to the base class' UpdatePrintSettings().

CL4313019 removed the cookie param, so checking for a bad cookie
param is no longer necessary / no longer possible. Since the only
remaining task was to pass the work up to the base class, this commit
removes the subclass implmentation entirely.

* chore: update patches

* chore: bump chromium in DEPS to 113.0.5647.0

* chore: bump chromium in DEPS to 113.0.5649.2

* chore: bump chromium in DEPS to 113.0.5651.0

* chore: update patches

---------

Co-authored-by: electron-roller[bot] <84116207+electron-roller[bot]@users.noreply.github.com>
Co-authored-by: Charles Kerr <charles@charleskerr.com>
Co-authored-by: deepak1556 <hop2deep@gmail.com>
2023-03-15 18:20:32 +09:00

29 lines
1.2 KiB
Diff

From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Cheng Zhao <zcbenz@gmail.com>
Date: Tue, 2 Aug 2022 11:53:00 +0900
Subject: fix: link with @loader_path/libffmpeg.dylib
Submitted to https://chromium-review.googlesource.com/c/chromium/third_party/ffmpeg/+/3803946.
When building with `is_component_build=false is_component_ffmpeg=true`,
we must manually instruct executables to link with the ffmpeg.dylib,
which is generated at the @loader_path, where for most targets is the
out/{Release,Debug} dir.
Using @rpath is wrong because the @rpath of most targets does not
include the out dir, and the linker won't be able to find ffmpeg.dylib
because of so.
diff --git a/BUILD.gn b/BUILD.gn
index 91e2f508c38b711dd517346e0e2b520879940f50..f270c07f81bf72481f029a1e791dfec9f899cb48 100644
--- a/BUILD.gn
+++ b/BUILD.gn
@@ -453,7 +453,7 @@ if (is_component_ffmpeg) {
if (!is_component_build) {
if (is_mac) {
- ldflags += [ "-Wl,-install_name,@rpath/libffmpeg.dylib" ]
+ ldflags += [ "-Wl,-install_name,@loader_path/libffmpeg.dylib" ]
} else if (is_linux) {
all_dependent_configs = [
"//build/config/gcc:rpath_for_built_shared_libraries",