[webkit-changes] [WebKit/WebKit] 663954: Cherry-pick 260875 at main (afe779a29470). https://bu...

Tim Nguyen noreply at github.com
Wed Mar 1 07:15:52 PST 2023


  Branch: refs/heads/webkitglib/2.40
  Home:   https://github.com/WebKit/WebKit
  Commit: 663954fac5a2c77dc16aa46c8bb9050bd18fb3d6
      https://github.com/WebKit/WebKit/commit/663954fac5a2c77dc16aa46c8bb9050bd18fb3d6
  Author: Philippe Normand <philn at igalia.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    M Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp
    M Source/WebCore/platform/graphics/gstreamer/mse/WebKitMediaSourceGStreamer.cpp
    M Source/WebCore/platform/mediastream/gstreamer/GStreamerMediaStreamSource.cpp

  Log Message:
  -----------
  Cherry-pick 260875 at main (afe779a29470). https://bugs.webkit.org/show_bug.cgi?id=252958

    REGRESSION(260082 at main): [GStreamer][1.20] YT broken
    https://bugs.webkit.org/show_bug.cgi?id=252958

    Reviewed by Xabier Rodriguez-Calvar.

    In 260082 at main we made the MSE source element act as an adaptive stream, triggering a different code
    path in urisourcebin, that turned out to be broken in GStreamer versions older than 1.22 and
    triggering caps negotiation errors during typefinding in urisourcebin. The commit that fixed this in
    GStreamer 1.22 is too big for backporting, so fallback to previous behavior in runtimes older than 1.22.

    * Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
    (WebCore::MediaPlayerPrivateGStreamer::configureElement):
    * Source/WebCore/platform/graphics/gstreamer/mse/WebKitMediaSourceGStreamer.cpp:
    (webkit_media_src_class_init):
    * Source/WebCore/platform/mediastream/gstreamer/GStreamerMediaStreamSource.cpp:
    (webkit_media_stream_src_class_init):

    Canonical link: https://commits.webkit.org/260875@main


  Commit: 7f5d1e7ae1d1c8c57d9f122bea2028bb642bef0d
      https://github.com/WebKit/WebKit/commit/7f5d1e7ae1d1c8c57d9f122bea2028bb642bef0d
  Author: Wu Yu Wei <wusyong9104 at gmail.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    M Source/WebCore/platform/network/soup/ResourceRequest.h
    M Source/WebCore/platform/network/soup/ResourceRequestSoup.cpp
    M Source/WebKit/UIProcess/API/glib/WebKitURISchemeRequest.cpp
    M Source/WebKit/UIProcess/API/glib/WebKitURISchemeRequest.h.in
    M Tools/TestWebKitAPI/Tests/WebKitGLib/TestWebKitWebContext.cpp

  Log Message:
  -----------
  Cherry-pick 260949 at main (5fe923aa166a). https://bugs.webkit.org/show_bug.cgi?id=252564

    [GLib] New API to get the request body of WebKitURISchemeRequest
    https://bugs.webkit.org/show_bug.cgi?id=252564

    Reviewed by Carlos Garcia Campos, Michael Catanzaro and Adrian Perez de Castro.

    * Source/WebCore/platform/network/soup/ResourceRequest.h:
    * Source/WebCore/platform/network/soup/ResourceRequestSoup.cpp:
    (WebCore::ResourceRequest::createInputStream const): Added new function
    API
    * Source/WebKit/UIProcess/API/glib/WebKitURISchemeRequest.h.in:
    * Source/WebKit/UIProcess/API/glib/WebKitURISchemeRequest.cpp:
    (webkit_uri_scheme_request_get_http_body): Added new function API
    * Tools/TestWebKitAPI/Tests/WebKitGLib/TestWebKitWebContext.cpp:
    (testWebContextURIScheme): Expanded 'post' URI scheme to test
    webkit_uri_scheme_request_get_http_body

    Canonical link: https://commits.webkit.org/260949@main


  Commit: 2e5ad799ee861fdc2f6e423a794dc1bc3caf6b51
      https://github.com/WebKit/WebKit/commit/2e5ad799ee861fdc2f6e423a794dc1bc3caf6b51
  Author: Michael Catanzaro <mcatanzaro at redhat.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    M Source/WebCore/platform/gtk/po/sv.po

  Log Message:
  -----------
  Cherry-pick 260973 at main (19b33ac9c769). https://bugs.webkit.org/show_bug.cgi?id=253084

    [GTK] Updated Swedish translation
    https://bugs.webkit.org/show_bug.cgi?id=253084

    Unreviewed translation update.

    * Source/WebCore/platform/gtk/po/sv.po:

    Canonical link: https://commits.webkit.org/260973@main


  Commit: 60a5d6e443e4deb1ca73e1b4beee49dc5995867a
      https://github.com/WebKit/WebKit/commit/60a5d6e443e4deb1ca73e1b4beee49dc5995867a
  Author: Michael Catanzaro <mcatanzaro at redhat.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    M Source/WebCore/platform/glib/UserAgentGLib.cpp

  Log Message:
  -----------
  Cherry-pick 260989 at main (dfbd862000d2). https://bugs.webkit.org/show_bug.cgi?id=253107

    [GLib] Bump Safari version in user agent header
    https://bugs.webkit.org/show_bug.cgi?id=253107

    Reviewed by Carlos Garcia Campos.

    * Source/WebCore/platform/glib/UserAgentGLib.cpp:
    (WebCore::buildUserAgentString):

    Canonical link: https://commits.webkit.org/260989@main


  Commit: b98c17eae1e76e976e47255fb459126c58862c89
      https://github.com/WebKit/WebKit/commit/b98c17eae1e76e976e47255fb459126c58862c89
  Author: Carlos Garcia Campos <cgarcia at igalia.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    M Source/WebKit/UIProcess/API/glib/WebKitDownload.cpp
    M Source/WebKit/UIProcess/API/glib/WebKitDownload.h.in
    M Source/WebKit/gtk/migrating-to-webkitgtk-6.0.md
    M Tools/TestWebKitAPI/Tests/WebKitGLib/TestDownloads.cpp

  Log Message:
  -----------
  Cherry-pick 261002 at main (6fea66849a84). https://bugs.webkit.org/show_bug.cgi?id=253146

    [GTK][WPE] WebKitDownload destination should be a path instead of a URI
    https://bugs.webkit.org/show_bug.cgi?id=253146

    Reviewed by Michael Catanzaro and Adrian Perez de Castro.

    Destination is handled as a path internally, so using URIs in the API
    means we have to convert to/from URI all the time. We can't change this
    in the current API, but for simplicity we can allow passing a path,
    while still support URIs for backwards compatibility.
    WebKiDownload::created-destination always returns a URI in the old API
    for compatibilty, but webkit_download_get_destination() only returns a
    URI when a URI was set with webkit_download_set_destination(). The new
    API only allows to pass a path, and a path is also passed to
    WebKiDownload::created-destination signal.

    This patch is based on previous patch by Michael Catanzaro.

    * Source/WebKit/UIProcess/API/glib/WebKitDownload.cpp:
    (webkitDownloadDecideDestination):
    (webkit_download_class_init):
    (webkitDownloadDecideDestinationWithSuggestedFilename):
    (webkitDownloadDestinationCreated):
    (webkit_download_get_destination):
    (webkit_download_set_destination):
    * Source/WebKit/UIProcess/API/glib/WebKitDownload.h.in:
    * Tools/TestWebKitAPI/Tests/WebKitGLib/TestDownloads.cpp:
    (testDownloadRemoteFileError):
    (testWebViewDownloadURI):
    (testPolicyResponseDownload):
    (testDownloadEphemeralContext):
    (testDownloadDestinationURI):
    (testContextMenuDownloadActions):
    (testBlobDownload):
    (beforeAll):

    Canonical link: https://commits.webkit.org/261002@main


  Commit: ccba5695c2326647f410096c3b891e54258747cb
      https://github.com/WebKit/WebKit/commit/ccba5695c2326647f410096c3b891e54258747cb
  Author: James Hilliard <james.hilliard1 at gmail.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    M Source/WebCore/platform/graphics/gstreamer/GStreamerVideoFrameHolder.cpp
    M Source/WebCore/platform/graphics/gstreamer/GUniquePtrGStreamer.h

  Log Message:
  -----------
  Cherry-pick 260995 at main (397546bb68cc). https://bugs.webkit.org/show_bug.cgi?id=253138

    [GStreamer][GL] Fix includes with USE_WPE_VIDEO_PLANE_DISPLAY_DMABUF
    https://bugs.webkit.org/show_bug.cgi?id=253138

    Reviewed by Philippe Normand.

    Fixes:
    Source/WebCore/platform/graphics/gstreamer/GStreamerVideoFrameHolder.cpp:58:9: error: ‘gst_is_gl_memory_egl’ was not declared in this scope; did you mean ‘gst_is_gl_memory_pbo’?
       58 |     if (gst_is_gl_memory_egl(memory)) {
          |         ^~~~~~~~~~~~~~~~~~~~
          |         gst_is_gl_memory_pbo

    WebCore/PrivateHeaders/WebCore/GUniquePtrGStreamer.h:40:10: fatal error: wpe/extensions/video-plane-display-dmabuf.h: No such file or directory
       40 | #include <wpe/extensions/video-plane-display-dmabuf.h>
          |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    * Source/WebCore/platform/graphics/gstreamer/GStreamerVideoFrameHolder.cpp:
    * Source/WebCore/platform/graphics/gstreamer/GUniquePtrGStreamer.h:

    Canonical link: https://commits.webkit.org/260995@main


  Commit: af6e203ebcd5d59f1bd09609c2ce52ad29fa469f
      https://github.com/WebKit/WebKit/commit/af6e203ebcd5d59f1bd09609c2ce52ad29fa469f
  Author: Carlos Garcia Campos <cgarcia at igalia.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    M Source/WTF/wtf/unix/LoggingUnix.cpp
    M Source/WebCore/platform/unix/LoggingUnix.cpp
    M Source/WebKit/Platform/unix/LoggingUnix.cpp
    M Source/WebKit/Shared/WebKit2Initialize.cpp

  Log Message:
  -----------
  Cherry-pick 260601 at main (173e3ebecf4f). https://bugs.webkit.org/show_bug.cgi?id=252558

    [GTK][WPE] UI process release log not shown
    https://bugs.webkit.org/show_bug.cgi?id=252558

    Reviewed by Michael Catanzaro.

    The problem is that logLevelString() implementation in WebKit layer
    doesn't take RELEASE_LOG_DISABLED into account. Actually the
    implementation of logLevelString() in all layers does the same, getting
    th value of WEBKIT_DEBUG env var, but with a few differences. We can
    have a single implementation in WTF and use WTF::logLevelString() from
    WebCore and WebKit.

    * Source/WTF/wtf/unix/LoggingUnix.cpp:
    (WTF::logLevelString):
    * Source/WebCore/platform/unix/LoggingUnix.cpp:
    (WebCore::logLevelString):
    * Source/WebKit/Platform/unix/LoggingUnix.cpp:
    (WebKit::logLevelString):
    * Source/WebKit/Shared/WebKit2Initialize.cpp:
    (WebKit::InitializeWebKit2): Remove log initialization here since it's already done in UIProcessLogInitialization.cpp.

    Canonical link: https://commits.webkit.org/260601@main


  Commit: 2c2f63cc7d76ecb94d4dd1fb38ab06d577011f06
      https://github.com/WebKit/WebKit/commit/2c2f63cc7d76ecb94d4dd1fb38ab06d577011f06
  Author: ChangSeok Oh <changseok at webkit.org>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    M Source/WebCore/platform/gstreamer/PlatformSpeechSynthesizerGStreamer.cpp

  Log Message:
  -----------
  Cherry-pick 260600 at main (5e6a4c1f5660). https://bugs.webkit.org/show_bug.cgi?id=252653

    Use an rvalue reference argument for the GstSpeechSynthesisWrapper constructor
    https://bugs.webkit.org/show_bug.cgi?id=252653

    Reviewed by Philippe Normand.

    Using an rvalue reference argument, we avoid an unnecessary copy of
    PlatformSpeechSynthesizer in the GstSpeechSynthesisWrapper constructor.

    * Source/WebCore/platform/gstreamer/PlatformSpeechSynthesizerGStreamer.cpp:
    (WebCore::GstSpeechSynthesisWrapper::GstSpeechSynthesisWrapper):

    Canonical link: https://commits.webkit.org/260600@main


  Commit: f602746644ac2a1f4cc7100871c49ae948edad34
      https://github.com/WebKit/WebKit/commit/f602746644ac2a1f4cc7100871c49ae948edad34
  Author: Tim Nguyen <ntim at apple.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    A LayoutTests/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/modal-dialog-backdrop-opacity-expected.html
    A LayoutTests/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/modal-dialog-backdrop-opacity-ref.html
    A LayoutTests/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/modal-dialog-backdrop-opacity.html
    M Source/WebCore/rendering/RenderLayerBacking.cpp

  Log Message:
  -----------
  Cherry-pick 260556 at main (903fc90f8a6a). https://bugs.webkit.org/show_bug.cgi?id=252536

    REGRESSION(257538 at main): CSS opacity doesn't work on dialog::backdrop
    https://bugs.webkit.org/show_bug.cgi?id=252536
    rdar://105636947

    Reviewed by Simon Fraser.

    The background layer functionality is to avoid flashes of white when rotating the screen on mobile device.
    However, it breaks the opacity property, so only restrict it to the fullscreen element's backdrop, which is the only main
    case that needs this polish (e.g. rotating while watching a fullscreen video).

    * LayoutTests/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/modal-dialog-backdrop-opacity-expected.html: Added.
    * LayoutTests/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/modal-dialog-backdrop-opacity-ref.html: Added.
    * LayoutTests/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/modal-dialog-backdrop-opacity.html: Added.
    * Source/WebCore/rendering/RenderLayerBacking.cpp:
    (WebCore::RenderLayerBacking::RenderLayerBacking):

    Canonical link: https://commits.webkit.org/260556@main


Compare: https://github.com/WebKit/WebKit/compare/a517f533fd16...f602746644ac


More information about the webkit-changes mailing list