Branch: refs/heads/webkitglib/2.52
  Home:   https://github.com/WebKit/WebKit
  Commit: aff2f9c58f3a81bea34579ef8e23f6c9c3dfb5f7
      
https://github.com/WebKit/WebKit/commit/aff2f9c58f3a81bea34579ef8e23f6c9c3dfb5f7
  Author: Chris Dumez <[email protected]>
  Date:   2026-03-26 (Thu, 26 Mar 2026)

  Changed paths:
    M Source/WebKit/Platform/IPC/ArgumentCoders.h

  Log Message:
  -----------
  Cherry-pick 309955@main (f9759f70c3d8). 
https://bugs.webkit.org/show_bug.cgi?id=310705

    Add missing WTF::move() to HashCountedSet's IPC decoder
    https://bugs.webkit.org/show_bug.cgi?id=310705

    Reviewed by Ryosuke Niwa.

    * Source/WebKit/Platform/IPC/ArgumentCoders.h:

    Canonical link: https://commits.webkit.org/309955@main

Canonical link: https://commits.webkit.org/305877.320@webkitglib/2.52


  Commit: 4c6d8fde4e68277fa356774f248ff798c7ff5b3f
      
https://github.com/WebKit/WebKit/commit/4c6d8fde4e68277fa356774f248ff798c7ff5b3f
  Author: Brady Eidson <[email protected]>
  Date:   2026-03-26 (Thu, 26 Mar 2026)

  Changed paths:
    M Source/WebCore/loader/FrameLoader.cpp

  Log Message:
  -----------
  Cherry-pick 309881@main (e4fec8bf903e). 
https://bugs.webkit.org/show_bug.cgi?id=310656

    Only create Navigation API history item on commit when Navigation API is 
enabled (resolves myserious WK1 issues)
    rdar://172987198
    https://bugs.webkit.org/show_bug.cgi?id=310656

    Reviewed by Rupin Mittal.

    Only create a Navigation API related history item on provisional load 
commit when the API is enabled.

    This should resolve a crash we're seeing with some very particular WK1 
usage, but we haven't been
    able to create a test case to reproduce.

    * Source/WebCore/loader/FrameLoader.cpp:
    (WebCore::FrameLoader::commitProvisionalLoad):

    Canonical link: https://commits.webkit.org/309881@main

Canonical link: https://commits.webkit.org/305877.321@webkitglib/2.52


  Commit: c74cda8fe951b05af0b52e1e8e011830c347e4f7
      
https://github.com/WebKit/WebKit/commit/c74cda8fe951b05af0b52e1e8e011830c347e4f7
  Author: Chris Dumez <[email protected]>
  Date:   2026-03-26 (Thu, 26 Mar 2026)

  Changed paths:
    M Source/WebKit/Platform/IPC/Connection.cpp

  Log Message:
  -----------
  Cherry-pick 309832@main (22b4d584f769). 
https://bugs.webkit.org/show_bug.cgi?id=310618

    Check sendMessageImpl() return value in IPC::Connection::sendSyncMessage()
    https://bugs.webkit.org/show_bug.cgi?id=310618

    Reviewed by Kimmo Kinnunen.

    sendSyncMessage() discarded the return value of sendMessageImpl(). If
    the send failed (e.g. connection became invalid between the isValid()
    check and the send), the code would fall through to waitForSyncReply()
    and block until it independently detected the failure — wasting at
    least one wait iteration and potentially longer.

    Check the error and return immediately on failure, avoiding the
    unnecessary wait. Use makeScopeExit for the signpost and
    popPendingSyncRequestID cleanup that is common to both the new
    early-return path and the normal path, which also reduces code
    duplication.

    * Source/WebKit/Platform/IPC/Connection.cpp:
    (IPC::Connection::sendSyncMessage):

    Canonical link: https://commits.webkit.org/309832@main

Canonical link: https://commits.webkit.org/305877.322@webkitglib/2.52


  Commit: f0aaca10391689fe4820f5da7e78434e2dbf7ca2
      
https://github.com/WebKit/WebKit/commit/f0aaca10391689fe4820f5da7e78434e2dbf7ca2
  Author: Chris Dumez <[email protected]>
  Date:   2026-03-26 (Thu, 26 Mar 2026)

  Changed paths:
    M Source/WebKit/Platform/IPC/Decoder.h

  Log Message:
  -----------
  Cherry-pick 309940@main (65303ca72a39). 
https://bugs.webkit.org/show_bug.cgi?id=310696

    IPC::Decoder::markInvalid() should not leave the m_bufferPosition iterator 
dangling
    https://bugs.webkit.org/show_bug.cgi?id=310696

    Reviewed by Kimmo Kinnunen.

    IPC::Decoder::markInvalid() clears m_buffer and fails to reset 
m_bufferPosition
    which is an iterator into m_buffer.

    * Source/WebKit/Platform/IPC/Decoder.h:
    (IPC::Decoder::markInvalid):

    Canonical link: https://commits.webkit.org/309940@main

Canonical link: https://commits.webkit.org/305877.323@webkitglib/2.52


  Commit: a8aa57207ca54290aad4ae87240e2c8a00e740e4
      
https://github.com/WebKit/WebKit/commit/a8aa57207ca54290aad4ae87240e2c8a00e740e4
  Author: Phinehas Fuachie <[email protected]>
  Date:   2026-03-26 (Thu, 26 Mar 2026)

  Changed paths:
    M Source/WebCore/page/Quirks.cpp
    M Source/WebCore/page/QuirksData.h

  Log Message:
  -----------
  Cherry-pick 309874@main (61f47004c2d9). 
https://bugs.webkit.org/show_bug.cgi?id=310057

    iheart.com - Radio icon is not shown in control center
    https://bugs.webkit.org/show_bug.cgi?id=310057
    rdar://171198911

    Reviewed by Jer Noble.

    iheart.com sets an "app=legacy:60" cookie that causes it to serve a player 
without Media Session
    API support, resulting in generic website icons in Control Center instead 
of station-specific
    artwork. Use the scriptToEvaluateBeforeRunningScriptFromURL quirk to 
override the cookie to
    "app=listen:60" before page scripts run, forcing the modern player with 
proper MediaSession
    metadata.

    Also remove the PLATFORM(IOS_FAMILY) guard around
    NeedsScriptToEvaluateBeforeRunningScriptFromURLQuirk so the quirk 
infrastructure is available on
    all platforms and add an early return for empty script URLs to prevent 
recursive evaluation.

    * Source/WebCore/page/Quirks.cpp:
    (WebCore::Quirks::scriptToEvaluateBeforeRunningScriptFromURL):
    (WebCore::handleIHeartQuirks):
    (WebCore::Quirks::determineRelevantQuirks):
    * Source/WebCore/page/QuirksData.h:

    Canonical link: https://commits.webkit.org/309874@main

Canonical link: https://commits.webkit.org/305877.324@webkitglib/2.52


  Commit: eec4cc90ea016da9b5c555f79b9f369b226b57f2
      
https://github.com/WebKit/WebKit/commit/eec4cc90ea016da9b5c555f79b9f369b226b57f2
  Author: Chris Dumez <[email protected]>
  Date:   2026-03-26 (Thu, 26 Mar 2026)

  Changed paths:
    M Source/WTF/wtf/text/StringView.cpp
    M Tools/TestWebKitAPI/Tests/WTF/StringView.cpp

  Log Message:
  -----------
  Cherry-pick 309716@main (8710df277928). 
https://bugs.webkit.org/show_bug.cgi?id=310468

    Fix reversed condition in StringView::reverseFind(StringView matchString, 
unsigned start)
    https://bugs.webkit.org/show_bug.cgi?id=310468

    Reviewed by Darin Adler and Ryosuke Niwa.

    When matchString is non-empty (the normal case), matchString.length() is
    truthy, so the function immediately returns std::min(start, length()),
    a position, instead of actually searching.

    Also update the function to return notFound when passed in a null
    StringView, to be consistent with the behavior of
    StringImpl::reverseFind().

    Test: Tools/TestWebKitAPI/Tests/WTF/StringView.cpp

    * Source/WTF/wtf/text/StringView.cpp:
    (WTF::StringView::reverseFind const):
    * Tools/TestWebKitAPI/Tests/WTF/StringView.cpp:
    (TestWebKitAPI::TEST(WTF, StringViewReverseFindStringView)):

    Canonical link: https://commits.webkit.org/309716@main

Canonical link: https://commits.webkit.org/305877.325@webkitglib/2.52


  Commit: e69ef5d619f6990cbf95d236916678364fbe7dca
      
https://github.com/WebKit/WebKit/commit/e69ef5d619f6990cbf95d236916678364fbe7dca
  Author: Chris Dumez <[email protected]>
  Date:   2026-03-26 (Thu, 26 Mar 2026)

  Changed paths:
    M Source/WTF/wtf/text/WTFString.cpp

  Log Message:
  -----------
  Cherry-pick 309657@main (eabb201d50ae). 
https://bugs.webkit.org/show_bug.cgi?id=310354

    Implicit narrowing conversion in String::ascii()
    https://bugs.webkit.org/show_bug.cgi?id=310354

    Reviewed by Darin Adler.

    Use static_cast to make it clearer that we're doing a narrowing conversion.

    * Source/WTF/wtf/text/WTFString.cpp:
    (WTF::String::ascii const):

    Canonical link: https://commits.webkit.org/309657@main

Canonical link: https://commits.webkit.org/305877.326@webkitglib/2.52


  Commit: d653d80165192452842924e1eecff7739d149d38
      
https://github.com/WebKit/WebKit/commit/d653d80165192452842924e1eecff7739d149d38
  Author: Ahmad Saleem <[email protected]>
  Date:   2026-03-26 (Thu, 26 Mar 2026)

  Changed paths:
    R LayoutTests/platform/mac/svg/dom/path-parser-expected.png
    M LayoutTests/svg/dom/path-parser-expected.txt
    M LayoutTests/svg/dom/path-parser.html
    M Source/WebCore/svg/SVGParserUtilities.cpp

  Log Message:
  -----------
  Cherry-pick 309608@main (ff16207ecb69). 
https://bugs.webkit.org/show_bug.cgi?id=310332

    genericParseNumber returns false instead of std::nullopt for exponent 
overflow
    https://bugs.webkit.org/show_bug.cgi?id=310332
    rdar://172984199

    Reviewed by Rob Buis.

    genericParseNumber incorrectly returned `false` instead of `std::nullopt`
    when the parsed exponent exceeded the representable float range. Since the
    return type is std::optional<FloatType>, `false` implicitly constructs a
    successful result of 0 rather than signaling a parse failure.

    This caused paths like "M0,50 h1e39" to silently produce "h0" instead of
    being treated as malformed.

    Also converts svg/dom/path-parser.html to testharness.js and adds
    additional test coverage for repeated commands, trailing content after
    close, and incomplete command sequences based on Blink's test suite
    (svg_path_parser_test.cc).

    * LayoutTests/platform/mac/svg/dom/path-parser-expected.png: Removed.
    * LayoutTests/svg/dom/path-parser-expected.txt:
    * LayoutTests/svg/dom/path-parser.html:
    * Source/WebCore/svg/SVGParserUtilities.cpp:
    (WebCore::genericParseNumber):

    Canonical link: https://commits.webkit.org/309608@main

Canonical link: https://commits.webkit.org/305877.327@webkitglib/2.52


Compare: https://github.com/WebKit/WebKit/compare/64555637d9b4...d653d8016519

To unsubscribe from these emails, change your notification settings at 
https://github.com/WebKit/WebKit/settings/notifications

Reply via email to