1. 11 Sep, 2019 15 commits
    • Dominik Röttsches's avatar
      Roll src/third_party/harfbuzz-ng/src/ bbad1b829..170b5dd85 (99 commits) · fcdcf033
      Dominik Röttsches authored
      Define HAVE_ROUNDF according to recommendations in
      https://github.com/harfbuzz/harfbuzz/pull/1953#issuecomment-527424349
      
      https://chromium.googlesource.com/external/github.com/harfbuzz/harfbuzz.git/+log/bbad1b829812..170b5dd856b1
      
      $ git log bbad1b829..170b5dd85 --date=short --no-merges --format='%ad %ae %s'
      2019-09-08 behdad [aat] Minor
      2019-09-08 behdad [aat] Minor
      2019-09-09 ebrahim [number] Add static to hb-number-parser.hh functions
      2019-09-09 ebrahim [number] Optimize _pow10 constants (#1963)
      2019-09-08 behdad Change HB_VAR_ARRAY to 1 again
      2019-09-07 ebrahim [number] Add overflow checks used to be done in parse_bcd
      2019-09-06 ebrahim Replace strtod with a ragel implementation
      2019-09-07 ebrahim Minor, remove unused header
      2019-09-06 behdad [kerx] Fix offset issue in kern format 2
      2019-09-06 behdad Change HB_VAR_ARRAY from 1 to 0
      2019-09-06 behdad Rename VAR to HB_VAR_ARRAY
      2019-09-06 behdad Fix a few struct size declarations
      2019-09-06 razrfalcon Remove duplicated tests from test-ot-tag (#1958)
      2019-09-06 behdad Revert "[null] Silence undefined-behavior complaints with too-small null bytes"
      2019-09-06 behdad [null] Silence undefined-behavior complaints with too-small null bytes
      2019-09-01 behdad [subset] Use newer iter tools in SinglePosFormat1
      2019-09-01 behdad [iter] Partialize hb_zip()
      2019-09-01 behdad Minor
      2019-09-04 ebrahim Use roundf to fix cast to int difference of msys2 w64
      2019-09-04 ebrahim Check roundf availibity in autotools
      2019-09-04 ebrahim Use hb_parse_double in CFF::dict_opset_t::parse_bcd
      2019-09-04 ebrahim [number] Turn hb_parse_float into hb_parse_double
      2019-09-03 ebrahim [number] Add whole buffer check and test it
      2019-09-03 ebrahim Don't check null terminaion of source in hb_codepoint_parse
      2019-09-03 ebrahim [test] resolve msvc complain on signedness comparing
      2019-09-03 ebrahim Use hb_parse_uint in deserializer token parser
      2019-09-03 ebrahim [number] Add test
      2019-09-03 ebrahim [number] Minor tweak on parser related codes
      2019-09-03 ebrahim [number] Remove parsing code duplication of the calls using lambda
      2019-09-03 ebrahim Move hb_codepoint_parse to hb_parse_uint
      2019-09-03 ebrahim Merge and aggregate number parsing logics to form hb-number
      2019-09-03 ebrahim Fallback if roundf didn't exist, like in dietlibc (#1953)
      2019-09-02 ebrahim Revert in-house pow10 (d80a3ea) and fix oss-fuzz/16922
      2019-09-02 ebrahim [subset] Use internal API of hb_set_t
      2019-09-02 ebrahim [subset] Run once for a tag
      2019-09-01 khaledhosny [ft] Prefer symbol cmap subtable if found (#1948)
      2019-09-01 ebrahim [editorconfig] Treat Ragel files also as C sources
      2019-09-01 ebrahim [cff] Implement in-house power of 10
      2019-09-01 ebrahim Don't allow reference blob be accessed using empty tag in hb_face_reference_table (#1947)
      2019-08-31 behdad [iter] Simplify hb_chop()
      2019-08-31 behdad [iter] Add hb_chop()
      2019-08-31 behdad [iter] Add hb_take() specialization for arrays
      2019-08-31 behdad [iter] Add hb_len()
      2019-08-31 behdad [iter] Add hb_take()
      2019-08-31 behdad [iter] Make iota() accept invokable for increasing to next item
      2019-08-30 behdad [algs] Fix hb_inc/dec signature
      2019-08-30 behdad [iter] Add hb_repeat()
      2019-08-30 behdad [iter] Remove some &&
      2019-08-30 behdad [iter] Use hb_ridentity instead of lazy +
      2019-08-30 behdad [iter] Separate hb_iota implementation from hb_range
      2019-08-30 behdad [iter] Rename
      2019-08-30 behdad [algs] Add hb_inc() and hb_dec())
      2019-08-30 behdad Revert "[array] Add chop()"
      2019-08-31 ebrahim [subset] minor
      2019-08-29 behdad [array] Add chop()
      2019-08-29 behdad [iter] Change item_size to get_item_size()
      2019-08-29 behdad [array] Add truncate() method
      2019-08-29 behdad [array] Use injected class name more
      2019-08-29 behdad Minor
      2019-08-29 48925186+qxliu76 bug fix in optimizing coverage table format (#1942)
      2019-08-29 behdad [subset] in SingleSubst subsetting, check for substitute in glyphset
      2019-08-29 ebrahim [gsub] Fix clang's semicolon complains
      2019-08-29 ebrahim [test] Avoid alloca use as clang complain
      2019-08-29 ebrahim [subset] Don't allow malicious fonts to insert unlimited table headers
      2019-08-28 behdad [subset] Implement AlternateSubst subsetting
      2019-08-28 behdad Minor
      2019-08-28 behdad [subset] Implement MultipleSubst subsetting
      2019-08-28 behdad [GSUB] Use dagger in Ligature::intersects()
      2019-08-28 behdad Minor
      2019-08-28 behdad Rename
      2019-08-28 behdad [subset] Subset ligature substitutes!
      2019-08-28 behdad Beef up HeadlessArrayOf<>
      2019-08-28 behdad Add ArrayOf::pop()
      2019-08-28 behdad Add arithmetic operators to IntType<>
      2019-08-28 behdad [debug] Minor
      2019-08-28 behdad Add ArrayOf::serialize_append
      2019-08-28 adrianwjw Adjustments to the generated Indic table output (#1936)
      2019-08-27 ebrahim Use hb_bytes_t as<T> in hb_blob_t
      2019-08-27 ebrahim [glyf] Move GlyphHeader::from_bytes to hb_bytes_t, introduce .as<T> ()
      2019-08-27 ebrahim [glyf] Revert the way indexToLocFormat is set
      2019-08-27 ebrahim [subset] Remove subset table size assertion
      2019-08-24 ebrahim Turn 8 spaces to tab across the project
      2019-08-25 ebrahim [subset] Raise the bar in new vs old table size
      2019-08-25 ebrahim [glyf] minor
      2019-08-25 ebrahim [glyf] Move GlyphHeader size checking to a static method
      2019-08-24 ebrahim [glyf] Move GlyphHeader related logics to its, making its fields protected
      2019-08-24 ebrahim [glyf] format source
      2019-08-24 ebrahim [subset] Consider instruction length place itself
      2019-08-23 ebrahim [subset] Fail on table grow more than 16x+4096
      2019-08-23 ebrahim [subset] Fix blob leak of _subset2 when returns early
      2019-08-23 behdad [debug] Fix extra semicolon issue
      2019-08-23 behdad Whitespace
      2019-08-23 behdad [debug] Fix build with HB_DEBUG
      2019-08-23 khaledhosny [doc] Update list of default features in the manual
      2019-08-22 behdad 2.6.1
      2019-08-21 behdad [ot-font] Prefer symbol cmap subtable if found
      2019-08-21 ebrahim Remove hb_directwrite_shape_experimental_width public API
      2019-08-20 behdad [test] Add test for aaa85931f3542639cd9d0dfb92fd6baab5b0298d
      2019-08-20 behdad [font] Update multipliers when creating sub_font
      
      Created with:
        roll-dep src/third_party/harfbuzz-ng/src
      R=bashi@chromium.org,behdad@chromium.org,bungeman@chromium.org,drott@chromium.org,eae@chromium.org,jshin@chromium.org
      
      Change-Id: I658578a297968bfdb5f7069ae86bae00cd92031c
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1795003
      Auto-Submit: Dominik Röttsches <drott@chromium.org>
      Commit-Queue: Behdad Esfahbod <behdad@chromium.org>
      Reviewed-by: default avatarBehdad Esfahbod <behdad@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695410}
      fcdcf033
    • Alex Gough's avatar
      Adds sandbox API for policy diagnostics · 9a32130d
      Alex Gough authored
      This will eventually support a chrome://sandbox WebUI for Windows.
      
      Adds PolicyList and PolicyInfo types, the GetPolicyInfo() interface
      to the sandbox (broker) and a basic implementation which simply
      returns an empty list.
      
      The intention is that GetPolicyInfo() will be called on a low
      priority thread, and will then wait until the tracking thread
      can complete the work of cloning policy information before
      returning.
      
      PolicyInfo is merely stubbed out at the moment and will be expanded
      to describe more of the policy once this gets plugged into the
      webui later.
      
      Change-Id: Ie6ea8180e26ad71fa9e1628a2e9eef8f46a93fbd
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1756952
      Commit-Queue: Alex Gough <ajgo@chromium.org>
      Reviewed-by: default avatarWill Harris <wfh@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695409}
      9a32130d
    • Kurt Horimoto's avatar
      [iOS] Add support for child view controller overlay UI. · 02df911f
      Kurt Horimoto authored
      A presentation context cannot present a view controller until its
      coordinator's view has been added to a window.  However, overlay UI
      supported using child view controllers can be presented immediately.
      This is preferable to waiting for the context to have a window because
      the |-didMoveToWindow| signal is sometimes received later than
      expected after a transition animation, allowing for a flicker between
      when the view is visible and when overlay UI can be presented upon it.
      
      This CL adds support for OverlayRequestCoordinators implemented using
      child view controllers to be presented immediately, rather than waiting
      for the view to be added to the window.  This is tracked by an
      activation state enum that is updated according to the context's
      coordinator's presentation capabilities.  This means that now a
      presentation context's coordinator is set the entire time the
      coordinator is started, and signals from its view moving to a new
      window are used to update the state.
      
      Follow-up CLs will update alerts in OverlayModality::kWebContentArea
      to use child view controllers where we have more control over the
      presentation style.
      
      Bug: 1001235
      Change-Id: If6a016a42fe9f1478fdd034b4457790c98a0870c
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1787562Reviewed-by: default avatarMike Dougherty <michaeldo@chromium.org>
      Commit-Queue: Kurt Horimoto <kkhorimoto@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695408}
      02df911f
    • Ahmed Fakhry's avatar
      Desks: Update the style of the bar items border · 63ad2ad6
      Ahmed Fakhry authored
      This CL implements the new specs of the mini_views and the new desk
      button borders, which are designed to have a gap between the view's
      contents and the border so it's more obvious against white/light
      contents or background.
      
      Only the active, highlighted, or hovered-over mini_views will have borders.
      
      Demo: https://bugs.chromium.org/p/chromium/issues/detail?id=996490#c9
      
      BUG=1001230, 996490
      TEST=Manually
      
      Change-Id: Iedd05a55ec36b9c6193c80f36e1d0d252ef93412
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1796112
      Commit-Queue: Ahmed Fakhry <afakhry@chromium.org>
      Reviewed-by: default avatarSammie Quon <sammiequon@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695407}
      63ad2ad6
    • James Cook's avatar
      Settings: Fix cursor when hovering on People > Profile row · 5b32d37e
      James Cook authored
      The row is sometimes clickable, depending on whether you are looking at
      browser vs. OS settings, and whether ChromeOSAccountManager is enabled
      or not.
      
      When it is clickable, ensure the <div> elements are "actionable" so we
      get the right cursor.
      
      Also fix a problem where the avatar icon wasn't clickable in the
      OS settings people section.
      
      Screenshot from Chrome OS showing actionable row in OS settings:
      https://screenshot.googleplex.com/ygC6Bn208dB
      
      Screenshot from Linux (unchanged):
      https://screenshot.googleplex.com/efW97d7diRY
      
      Bug: 1002024
      Test: added to browser_tests
      Change-Id: Ib3a9f17e2d0f9b72eb6793e845c4679c141ccc1b
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1797122Reviewed-by: default avatarXiyuan Xia <xiyuan@chromium.org>
      Commit-Queue: James Cook <jamescook@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695406}
      5b32d37e
    • Renjie Tang's avatar
      Add histograms to log details of race_stale_dns_on_connection experiment. · 97d71b6f
      Renjie Tang authored
      Also removed some unused logs.
      
      Change-Id: Ie23768f8f9fd5ceb6814df9dadd5f896bc809aa3
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1793444
      Commit-Queue: Renjie Tang <renjietang@chromium.org>
      Reviewed-by: default avatarAlexei Svitkine <asvitkine@chromium.org>
      Reviewed-by: default avatarRyan Hamilton <rch@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695405}
      97d71b6f
    • Alan Cutter's avatar
      Add src/results to .gitignore · a5351ff9
      Alan Cutter authored
      src/tools/run-swarmed.py uses src/results as the output directory by
      default. We should avoid treating them as source files in the repository.
      
      Change-Id: I69ee66aa7c65382d0068bbd949fa761e638a5dc1
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1781925
      Auto-Submit: Alan Cutter <alancutter@chromium.org>
      Reviewed-by: default avatarElly Fong-Jones <ellyjones@chromium.org>
      Commit-Queue: Alan Cutter <alancutter@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695404}
      a5351ff9
    • Kristi Park's avatar
      [NTP] Fix max number of shortcuts error for grid layout · 02a7394c
      Kristi Park authored
      If the page was loaded with Most Visited shortcuts, the maximum number
      of shortcuts was set to 8. This was not updated if users swapped to
      custom links, so an error was thrown.
      
      Do update the maximum number of shortcuts, and remove the error since
      we can ignore any shortcuts over the max.
      
      Change-Id: I95e3941ed2c9f2b266075edad7f8a60bddddd93d
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1793230
      Commit-Queue: Kristi Park <kristipark@chromium.org>
      Reviewed-by: default avatarKyle Milka <kmilka@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695403}
      02a7394c
    • Alex Cooper's avatar
      Fix WebXR input browser test usage of RunJavaScriptAndExtractBoolOrFail · a36a08e0
      Alex Cooper authored
      Bug: 1002737
      Change-Id: I74070a0ad0f0633521247e1c74ea0147ccc0505d
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1797248
      Commit-Queue: Alexander Cooper <alcooper@chromium.org>
      Commit-Queue: Brian Sheedy <bsheedy@chromium.org>
      Auto-Submit: Alexander Cooper <alcooper@chromium.org>
      Reviewed-by: default avatarBrian Sheedy <bsheedy@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695402}
      a36a08e0
    • Matt Falkenhagen's avatar
      service worker: Disable two unit tests when ServiceWorkerOnUI is on. · fbcffd3c
      Matt Falkenhagen authored
      The code path these test doesn't happen when the feature is on:
      * EmbeddedWorkerInstanceTest.DetachDuringProcessAllocation
      * EmbeddedWorkerInstanceTest.StopDuringProcessAllocation
      
      Bug: 824858
      Change-Id: Iadfc474fc1a0acb56ec9562a17fed28023927931
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1795530Reviewed-by: default avatarMakoto Shimazu <shimazu@chromium.org>
      Commit-Queue: Matt Falkenhagen <falken@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695401}
      fbcffd3c
    • Andres Calderon Jaramillo's avatar
      cc: Refactor flags for image decode acceleration · 6c5267fb
      Andres Calderon Jaramillo authored
      This CL refactors how we use the feature flags for image decode
      acceleration. The goal is that for a given image type (i.e., JPEG or
      WebP), we only check the corresponding hardware decode acceleration
      feature flag after we know that OOP-R is enabled and the device supports
      hardware decoding images of that type. The reason is that we only want
      hardware decode acceleration finch experiments to include users from
      that population (otherwise, both the control and enabled groups would
      always avoid accelerated decodes).
      
      To achieve this, we compute two bool members in the GpuImageDecodeCache
      ctor, |allow_accelerated_jpeg_decodes_| and
      |allow_accelerated_webp_decodes_|. These are computed by first checking
      for OOP-R and hardware decoding support for the corresponding image
      type.
      
      We also remove the feature flag checks in
      VaapiImageDecodeAcceleratorWorker::Create() (which happens during the
      the GPU process startup). This implies that the supported decoder
      profiles are always advertised to the renderers. However, the checks in
      [1] remain intact and they ensure that the service side only accepts
      requests for accelerated decoding if one of the feature flags (JPEG or
      WebP) is on. Furthermore, we check the flags again in
      VaapiImageDecodeAcceleratorWorker::GetDecoderForImage() to ensure that
      if a flag for a specific image type is off, we don't use the decoder for
      that type. This is all to prevent compromised renderers from triggering
      hardware decodes forbidden by the feature flags.
      
      [1] https://cs.chromium.org/chromium/src/gpu/ipc/service/image_decode_accelerator_stub.cc?l=118-123&rcl=4fbe89686751aa0c90da185c82d796f9f01cf2aa
      
      Bug: 995155
      Test: Adds GpuImageDecodeCache test coverage for the new behavior.
      Change-Id: I4957f9e2606b012aef337c93b45732eaf010248a
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1787594
      Commit-Queue: Andres Calderon Jaramillo <andrescj@chromium.org>
      Reviewed-by: default avatarSunny Sachanandani <sunnyps@chromium.org>
      Reviewed-by: default avatardanakj <danakj@chromium.org>
      Reviewed-by: default avatarKhushal <khushalsagar@chromium.org>
      Reviewed-by: default avatarDaniele Castagna <dcastagna@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695400}
      6c5267fb
    • Saman Sami's avatar
      Fix 3D Canvas failure with Vulkan · 14da1984
      Saman Sami authored
      Currently for accelerated 3D resources we skip SharedImage provider
      and fall through to Texture provider which is not compatible with
      Vulkan.
      
      Bug: 985366
      Change-Id: Ibc7a6ad38ba62c9e82b86822dcf1dff06a7d2096
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1796307
      Commit-Queue: Saman Sami <samans@chromium.org>
      Reviewed-by: default avatarBrian Sheedy <bsheedy@chromium.org>
      Reviewed-by: default avatarKhushal <khushalsagar@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695399}
      14da1984
    • Chromium WPT Sync's avatar
      Import wpt@d550f113244b9df3ad17369c0111f0ce9d43b062 · 7571008d
      Chromium WPT Sync authored
      Using wpt-import in Chromium eaa5aa87.
      With Chromium commits locally applied on WPT:
      cf88186a "[LayoutNG] Fix `vertical-align: top` and `bottom`"
      
      
      Note to sheriffs: This CL imports external tests and adds
      expectations for those tests; if this CL is large and causes
      a few new failures, please fix the failures by adding new
      lines to TestExpectations rather than reverting. See:
      https://chromium.googlesource.com/chromium/src/+/master/docs/testing/web_platform_tests.md
      
      Directory owners for changes in this CL:
      foolip@chromium.org, lpz@chromium.org, robertma@chromium.org:
        external/wpt/tools
      kojii@chromium.org, mstensho@chromium.org:
        external/wpt/css/CSS2
      
      NOAUTOREVERT=true
      TBR=lpz
      
      No-Export: true
      Change-Id: I08c071c7915b4428a9b35ec803e049a4e76340dd
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1796904Reviewed-by: default avatarWPT Autoroller <wpt-autoroller@chops-service-accounts.iam.gserviceaccount.com>
      Commit-Queue: WPT Autoroller <wpt-autoroller@chops-service-accounts.iam.gserviceaccount.com>
      Cr-Commit-Position: refs/heads/master@{#695398}
      7571008d
    • ziyangch's avatar
      [Chromecast] Fix |audio_video_pipeline_impl_unittest| failure. · c21a52a1
      ziyangch authored
      Bug: internal b/140761503
      
      Test: audio_video_pipeline_impl_unittest.
      
      Change-Id: I77f8f44dc8fe71e018599e927e48a915c76093d8
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1796643Reviewed-by: default avatarYuchen Liu <yucliu@chromium.org>
      Reviewed-by: default avatarKenneth MacKay <kmackay@chromium.org>
      Commit-Queue: Ziyang Cheng <ziyangch@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695397}
      c21a52a1
    • Tibor Goldschwendt's avatar
      Properly pass bundle smoke test flags to bundle under test · d6817702
      Tibor Goldschwendt authored
      Before, the flags were passed to the test APK.
      
      + Fix flags to make bundle smoke test run.
      
      Bug: 989654
      Change-Id: I175b57af7b0c0ad9badb5d16b341f0eae4b8369a
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1795311Reviewed-by: default avatarJohn Budorick <jbudorick@chromium.org>
      Commit-Queue: Tibor Goldschwendt <tiborg@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#695396}
      d6817702
  2. 10 Sep, 2019 25 commits