)]}'
{
  "log": [
    {
      "commit": "feb720f746dcecfd80e47b060d620c3ffa29a43b",
      "tree": "04bce6819912e166bb2213ebfb8b4d9c4e2bb038",
      "parents": [
        "376d8128f0fc570a6304ff5216e4a92d931dfa65"
      ],
      "author": {
        "name": "skia-autoroll",
        "email": "skia-autoroll@skia-public.iam.gserviceaccount.com",
        "time": "Thu Mar 14 15:55:17 2019 +0000"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 16:36:02 2019 +0000"
      },
      "message": "Roll third_party/externals/swiftshader ca56e8b07a1a..eb7cf3c4b0b1 (1 commits)\n\nhttps://swiftshader.googlesource.com/SwiftShader.git/+log/ca56e8b07a1a..eb7cf3c4b0b1\n\n\ngit log ca56e8b07a1a..eb7cf3c4b0b1 --date\u003dshort --no-merges --format\u003d\u0027%ad %ae %s\u0027\n2019-03-14 capn@google.com Update CONTRIBUTORS\n\n\nCreated with:\n  gclient setdep -r third_party/externals/swiftshader@eb7cf3c4b0b1\n\nThe AutoRoll server is located here: https://autoroll.skia.org/r/swiftshader-skia-autoroll\n\nDocumentation for the AutoRoller is here:\nhttps://skia.googlesource.com/buildbot/+/master/autoroll/README.md\n\nIf the roll is causing failures, please contact the current sheriff, who should\nbe CC\u0027d on the roll, and stop the roller if necessary.\n\nCQ_INCLUDE_TRYBOTS\u003dskia.primary:Test-Debian9-Clang-GCE-GPU-SwiftShader-x86_64-Debug-All-SwiftShader\nTBR\u003dbenjaminwagner@google.com\n\nChange-Id: Ie0a4da348d859e8f596ccb5ac52937cc28f8c3d5\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201341\nReviewed-by: skia-autoroll \u003cskia-autoroll@skia-public.iam.gserviceaccount.com\u003e\nCommit-Queue: skia-autoroll \u003cskia-autoroll@skia-public.iam.gserviceaccount.com\u003e\n"
    },
    {
      "commit": "376d8128f0fc570a6304ff5216e4a92d931dfa65",
      "tree": "62870dc59d12b0625845203909d6d89debca5a19",
      "parents": [
        "1b258182ad94295a79dce0b7a11b6f98f5742612"
      ],
      "author": {
        "name": "Mike Reed",
        "email": "reed@google.com",
        "time": "Thu Mar 14 11:39:02 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 16:11:46 2019 +0000"
      },
      "message": "Use \u0027\u0026\u0027 to toggle showing the gm\u0027s size in viewer\n\nBug: skia:\nChange-Id: Id294b6a48d60d95b46ce3443edf10513599c43d1\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201230\nReviewed-by: Mike Reed \u003creed@google.com\u003e\nCommit-Queue: Mike Reed \u003creed@google.com\u003e\nAuto-Submit: Mike Reed \u003creed@google.com\u003e\n"
    },
    {
      "commit": "1b258182ad94295a79dce0b7a11b6f98f5742612",
      "tree": "15727c2ccfc9b769608a755bf461bc549564ceac",
      "parents": [
        "45fe9473d1f03d80eac0468b83c04f6f758471a0"
      ],
      "author": {
        "name": "Mike Klein",
        "email": "mtklein@google.com",
        "time": "Wed Mar 13 15:49:23 2019 -0500"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 16:07:21 2019 +0000"
      },
      "message": "move in center of radial_gradient_precision\n\n{1K,1K} ought to work fine for F16 precision.  16 bit floats step by 1.0\nbetween 1024 and 2048, so they should be able to just barely represent\ndistinct coordinates for each pixel.  If precision drops below 16 bit\nwe\u0027ll start seeing banding because we can\u0027t distinguish coordinates.\n\nSwap out the colors a little so that it\u0027s clear which images are the 1K\n(black+green) and which are the old 100K (black+white).\n\nChange-Id: I92761c3aa67cab57332e6f00a421c63c1d9d990a\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201041\nCommit-Queue: Mike Klein \u003cmtklein@google.com\u003e\nReviewed-by: Michael Ludwig \u003cmichaelludwig@google.com\u003e\n"
    },
    {
      "commit": "45fe9473d1f03d80eac0468b83c04f6f758471a0",
      "tree": "96f9eaaf49d716365bf57339af8deb4130b531df",
      "parents": [
        "8a5d3626a135ff6beebb9424fb4b045494608157"
      ],
      "author": {
        "name": "Brian Salomon",
        "email": "bsalomon@google.com",
        "time": "Thu Mar 14 11:00:09 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 16:06:41 2019 +0000"
      },
      "message": "Update color type after color override in GrFillRectOp.\n\nFixes const_color_processor GM in MSAA mode after\nhttps://skia.googlesource.com/skia/+/1d835423e1c6776c4cbdb91bd435d21a7c1e267c\n\nChange-Id: I07cbe17d8986405da5211e106dca58bfd61e6bbd\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201228\nAuto-Submit: Brian Salomon \u003cbsalomon@google.com\u003e\nCommit-Queue: Michael Ludwig \u003cmichaelludwig@google.com\u003e\nReviewed-by: Michael Ludwig \u003cmichaelludwig@google.com\u003e\n"
    },
    {
      "commit": "8a5d3626a135ff6beebb9424fb4b045494608157",
      "tree": "ba9c57c971554a4f319b26f6fd047f5b14a285b1",
      "parents": [
        "44c8131b779b1a21bae524733663014d3614a5ed"
      ],
      "author": {
        "name": "Jim Van Verth",
        "email": "jvanverth@google.com",
        "time": "Thu Mar 14 10:28:01 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 15:45:41 2019 +0000"
      },
      "message": "Set RG_float to be texturable on Metal\n\nBug: skia:8243\nChange-Id: Ie463cea638691249c04c1c8a9a5295c6f4d5bf5e\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201224\nReviewed-by: Brian Osman \u003cbrianosman@google.com\u003e\nCommit-Queue: Jim Van Verth \u003cjvanverth@google.com\u003e\n"
    },
    {
      "commit": "44c8131b779b1a21bae524733663014d3614a5ed",
      "tree": "696bce83f8123bbd5764c80bd1241f2026761af6",
      "parents": [
        "f35551815fe7ed756a8e0aa83c84c1d5ca70ad43"
      ],
      "author": {
        "name": "skia-autoroll",
        "email": "skia-autoroll@skia-public.iam.gserviceaccount.com",
        "time": "Thu Mar 14 14:43:17 2019 +0000"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 15:25:40 2019 +0000"
      },
      "message": "Roll third_party/externals/swiftshader 49d2c13943ef..ca56e8b07a1a (1 commits)\n\nhttps://swiftshader.googlesource.com/SwiftShader.git/+log/49d2c13943ef..ca56e8b07a1a\n\n\ngit log 49d2c13943ef..ca56e8b07a1a --date\u003dshort --no-merges --format\u003d\u0027%ad %ae %s\u0027\n2019-03-14 bclayton@google.com Reactor: Add a PrintValue for Bool types.\n\n\nCreated with:\n  gclient setdep -r third_party/externals/swiftshader@ca56e8b07a1a\n\nThe AutoRoll server is located here: https://autoroll.skia.org/r/swiftshader-skia-autoroll\n\nDocumentation for the AutoRoller is here:\nhttps://skia.googlesource.com/buildbot/+/master/autoroll/README.md\n\nIf the roll is causing failures, please contact the current sheriff, who should\nbe CC\u0027d on the roll, and stop the roller if necessary.\n\nCQ_INCLUDE_TRYBOTS\u003dskia.primary:Test-Debian9-Clang-GCE-GPU-SwiftShader-x86_64-Debug-All-SwiftShader\nTBR\u003dbenjaminwagner@google.com\n\nChange-Id: I3ab7dc4343223505b046e724aa3fb5e6ed81e7c6\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201171\nReviewed-by: skia-autoroll \u003cskia-autoroll@skia-public.iam.gserviceaccount.com\u003e\nCommit-Queue: skia-autoroll \u003cskia-autoroll@skia-public.iam.gserviceaccount.com\u003e\n"
    },
    {
      "commit": "f35551815fe7ed756a8e0aa83c84c1d5ca70ad43",
      "tree": "aec9b8d8b96e201a187e0d218fe02f763a72e2ed",
      "parents": [
        "876a017e53f2073353866eb300fcfaf54a0106ec"
      ],
      "author": {
        "name": "Kevin Lubick",
        "email": "kjlubick@google.com",
        "time": "Thu Mar 14 10:55:10 2019 -0400"
      },
      "committer": {
        "name": "Kevin Lubick",
        "email": "kjlubick@google.com",
        "time": "Thu Mar 14 15:18:22 2019 +0000"
      },
      "message": "[canvaskit] missing extern\n\nBug: skia:\nChange-Id: I9893650b394c71ced15c0b6b7623a4c3081cedff\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201225\nReviewed-by: Kevin Lubick \u003ckjlubick@google.com\u003e\n"
    },
    {
      "commit": "876a017e53f2073353866eb300fcfaf54a0106ec",
      "tree": "42b1ac5ba63dbf655b17b9357bcb1509e4482f2f",
      "parents": [
        "90791c202dd2d943565237bf511d929e8bd19951"
      ],
      "author": {
        "name": "Brian Salomon",
        "email": "bsalomon@google.com",
        "time": "Fri Mar 08 11:12:14 2019 -0500"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 15:07:10 2019 +0000"
      },
      "message": "Reland \"Reintroduce deinstantiate lazy proxy types and use for promise images.\"\n\nThis is a reland of 8b40ac35b2ec31a61c751e610b891681df272283\n\nOriginal change\u0027s description:\n\u003e Reintroduce deinstantiate lazy proxy types and use for promise images.\n\u003e \n\u003e This reverts a fraction of b2c5dae65df952999f0a12b9df80bc1433ffa19a to\n\u003e restore the deinstantiate lazy proxy type, supporting implementation,\n\u003e and tests.\n\u003e \n\u003e Use them for promise images to avoid thread safety issues for promise\n\u003e image resources. Makes promise image instantiation callbacks do a thread\n\u003e safe unref of their fulfilled GrTexture in GrResourceCache. The\n\u003e GrResourceCache mechanism for receiving unref messages is extended to\n\u003e allow multiple pending unrefs. All this is new.\n\u003e \n\u003e \n\u003e Bug: skia:8800\n\u003e Change-Id: I7b1d4fea13c053b6fbbd39c0c6eaf567b8bf81f1\n\u003e Reviewed-on: https://skia-review.googlesource.com/c/skia/+/199002\n\u003e Reviewed-by: Brian Salomon \u003cbsalomon@google.com\u003e\n\u003e Reviewed-by: Robert Phillips \u003crobertphillips@google.com\u003e\n\u003e Commit-Queue: Brian Salomon \u003cbsalomon@google.com\u003e\n\nBug: skia:8800\nChange-Id: Ib939fc5c19edf0c6b965c9f6adf0afedd4267703\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201220\nReviewed-by: Brian Salomon \u003cbsalomon@google.com\u003e\nCommit-Queue: Brian Salomon \u003cbsalomon@google.com\u003e\n"
    },
    {
      "commit": "90791c202dd2d943565237bf511d929e8bd19951",
      "tree": "1629e0ca3e2901ec4bb69ad786001fd5c9dbf745",
      "parents": [
        "86ce5c44ab47ed73f5b1115283b05b33e99052b3"
      ],
      "author": {
        "name": "Michael Ludwig",
        "email": "michaelludwig@google.com",
        "time": "Thu Mar 14 14:57:00 2019 +0000"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 14:57:12 2019 +0000"
      },
      "message": "Revert \"Update SkCanvas\u0027 experimental SkiaRenderer API\"\n\nThis reverts commit 4bf964602ab8758f6e580aaaa69add4fb260c1a6.\n\nReason for revert: vulkan dm crashes\n\nOriginal change\u0027s description:\n\u003e Update SkCanvas\u0027 experimental SkiaRenderer API\n\u003e \n\u003e This lifts the temporary functions in SkGpuDevice into SkCanvas and\n\u003e deprecates the older experimental_DrawImageSetV1 and\n\u003e experimental_DrawEdgeAARect.  The new functions can handle paints and\n\u003e transform batching. Internally, SkCanvas routes the old functions to the\n\u003e new entry points and all device-level code is updated to handle the new\n\u003e API features.\n\u003e \n\u003e While touching all of the canvas/device/recording areas, the\n\u003e experimental functions are grouped in an \"EdgeAA\" cluster instead of being\n\u003e separated into the image category and the rectangle category.\n\u003e \n\u003e Bug: skia:8739\n\u003e Change-Id: I67c2a724873040ad5dc3307ab5b2823ba1eac54b\n\u003e Reviewed-on: https://skia-review.googlesource.com/c/skia/+/190221\n\u003e Commit-Queue: Michael Ludwig \u003cmichaelludwig@google.com\u003e\n\u003e Reviewed-by: Brian Salomon \u003cbsalomon@google.com\u003e\n\nTBR\u003dbsalomon@google.com,robertphillips@google.com,michaelludwig@google.com\n\nChange-Id: I87a5a258c5a1bd15e16389cdf91743772d6fa98a\nNo-Presubmit: true\nNo-Tree-Checks: true\nNo-Try: true\nBug: skia:8739\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201226\nReviewed-by: Michael Ludwig \u003cmichaelludwig@google.com\u003e\nCommit-Queue: Michael Ludwig \u003cmichaelludwig@google.com\u003e\n"
    },
    {
      "commit": "86ce5c44ab47ed73f5b1115283b05b33e99052b3",
      "tree": "f3d1ebb067daaa5c02b04a3d318df1212dfa0c3c",
      "parents": [
        "8a63a4d024aa5108d1553fcc74ce27199d9c673b"
      ],
      "author": {
        "name": "skia-autoroll",
        "email": "skia-autoroll@skia-public.iam.gserviceaccount.com",
        "time": "Thu Mar 14 14:01:17 2019 +0000"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 14:41:10 2019 +0000"
      },
      "message": "Roll third_party/externals/swiftshader 9b62c5ea2af3..49d2c13943ef (1 commits)\n\nhttps://swiftshader.googlesource.com/SwiftShader.git/+log/9b62c5ea2af3..49d2c13943ef\n\n\ngit log 9b62c5ea2af3..49d2c13943ef --date\u003dshort --no-merges --format\u003d\u0027%ad %ae %s\u0027\n2019-03-14 bclayton@google.com SpirvShader: Fix int DBZ failures by masking away zeros.\n\n\nCreated with:\n  gclient setdep -r third_party/externals/swiftshader@49d2c13943ef\n\nThe AutoRoll server is located here: https://autoroll.skia.org/r/swiftshader-skia-autoroll\n\nDocumentation for the AutoRoller is here:\nhttps://skia.googlesource.com/buildbot/+/master/autoroll/README.md\n\nIf the roll is causing failures, please contact the current sheriff, who should\nbe CC\u0027d on the roll, and stop the roller if necessary.\n\nCQ_INCLUDE_TRYBOTS\u003dskia.primary:Test-Debian9-Clang-GCE-GPU-SwiftShader-x86_64-Debug-All-SwiftShader\nTBR\u003dbenjaminwagner@google.com\n\nChange-Id: I78f3c56d3f6a7998c0b5c21e23831dcf873bc62e\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201169\nReviewed-by: skia-autoroll \u003cskia-autoroll@skia-public.iam.gserviceaccount.com\u003e\nCommit-Queue: skia-autoroll \u003cskia-autoroll@skia-public.iam.gserviceaccount.com\u003e\n"
    },
    {
      "commit": "8a63a4d024aa5108d1553fcc74ce27199d9c673b",
      "tree": "379baa1a49f35cae7e2e260facd1c9771b6c9638",
      "parents": [
        "4bf964602ab8758f6e580aaaa69add4fb260c1a6"
      ],
      "author": {
        "name": "Brian Osman",
        "email": "brianosman@google.com",
        "time": "Thu Mar 14 10:11:24 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 14:40:50 2019 +0000"
      },
      "message": "SkLumaColorFilter: Keep alpha in [0,1]\n\nBug: skia:\nChange-Id: I2287ae75745509269e3dee2b1d31ab513580e6c3\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201221\nReviewed-by: Mike Klein \u003cmtklein@google.com\u003e\nCommit-Queue: Brian Osman \u003cbrianosman@google.com\u003e\n"
    },
    {
      "commit": "4bf964602ab8758f6e580aaaa69add4fb260c1a6",
      "tree": "519da374db64f96ce16cde92de65fbac8fbe6280",
      "parents": [
        "539909a84a2ee9f836b2464bfc2d67dad657426a"
      ],
      "author": {
        "name": "Michael Ludwig",
        "email": "michaelludwig@google.com",
        "time": "Wed Mar 13 16:19:38 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 14:24:11 2019 +0000"
      },
      "message": "Update SkCanvas\u0027 experimental SkiaRenderer API\n\nThis lifts the temporary functions in SkGpuDevice into SkCanvas and\ndeprecates the older experimental_DrawImageSetV1 and\nexperimental_DrawEdgeAARect.  The new functions can handle paints and\ntransform batching. Internally, SkCanvas routes the old functions to the\nnew entry points and all device-level code is updated to handle the new\nAPI features.\n\nWhile touching all of the canvas/device/recording areas, the\nexperimental functions are grouped in an \"EdgeAA\" cluster instead of being\nseparated into the image category and the rectangle category.\n\nBug: skia:8739\nChange-Id: I67c2a724873040ad5dc3307ab5b2823ba1eac54b\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/190221\nCommit-Queue: Michael Ludwig \u003cmichaelludwig@google.com\u003e\nReviewed-by: Brian Salomon \u003cbsalomon@google.com\u003e\n"
    },
    {
      "commit": "539909a84a2ee9f836b2464bfc2d67dad657426a",
      "tree": "fcb699c3d52fef9b119a82ba0a168fe5148b895b",
      "parents": [
        "cfdf87ecd2ac1c43a9b8c0dbdf6efb02e429af05"
      ],
      "author": {
        "name": "Mike Klein",
        "email": "mtklein@google.com",
        "time": "Wed Mar 13 15:15:55 2019 -0500"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 13:53:19 2019 +0000"
      },
      "message": "remove plus clamp\n\nClamping for plus mode should happen only as the destination format\nrequires it, or as a consequence of the limited range we\u0027re holding\ncolors in the pipeline, or because non-normalized alpha is nuts.\n\nSo I\u0027ve taken away the clamp on float rgb, keeping it on lowp to stay\nin its legal range, and on float alpha because it\u0027s crazy not to.\n\nOpen questions:\n  - also clamp alpha to zero?\n  - what do we do with alpha \u003c0 or \u003e1 in general, beyond plus?\n\nCq-Include-Trybots: luci.chromium.try:linux-blink-rel\nChange-Id: I15660ce55cc393e312f18ffdc13bbfdef08ac6e0\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201040\nReviewed-by: Brian Osman \u003cbrianosman@google.com\u003e\nCommit-Queue: Mike Klein \u003cmtklein@google.com\u003e\n"
    },
    {
      "commit": "cfdf87ecd2ac1c43a9b8c0dbdf6efb02e429af05",
      "tree": "eb0d16070313d206ed27e4f3fc9358d7f28414fb",
      "parents": [
        "1d835423e1c6776c4cbdb91bd435d21a7c1e267c"
      ],
      "author": {
        "name": "Nathaniel Nifong",
        "email": "nifong@google.com",
        "time": "Wed Mar 13 13:31:13 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 13:50:24 2019 +0000"
      },
      "message": "fill out more of SkDebugCanvas interface\n\nAdded methods for toggling clip viz, overdraw viz, gpu bound viz.\nmethods for getting number of commands, toggling command visibility and deleting commands.\nAdded method for writing out json command list.\n\nbound methods with emscrpiten and connected up to buttons in barebones interface to prove they\u0027re working.\n\n\nBug: skia:\nChange-Id: I3f733b13db00ce5d14b1b5170de8fb66102c8b14\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/199938\nCommit-Queue: Nathaniel Nifong \u003cnifong@google.com\u003e\nReviewed-by: Kevin Lubick \u003ckjlubick@google.com\u003e\n"
    },
    {
      "commit": "1d835423e1c6776c4cbdb91bd435d21a7c1e267c",
      "tree": "aa211cdc10bcffb834cdc44227df145d2b67d93c",
      "parents": [
        "f699bf0437dbfd1b50b027751ad251335a1eb2b6"
      ],
      "author": {
        "name": "Brian Salomon",
        "email": "bsalomon@google.com",
        "time": "Wed Mar 13 16:11:44 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Thu Mar 14 13:48:14 2019 +0000"
      },
      "message": "Reland \"Try to avoid vertex colors in Texture/FillRect ops when possible.\"\n\nThis is a reland of e0b989e5e3ab64f3585fe1bf0228e964dad0678c\n\nOriginal change\u0027s description:\n\u003e Try to avoid vertex colors in Texture/FillRect ops when possible.\n\u003e \n\u003e Avoids unnecessary fragment shader color multiplication.\n\u003e \n\u003e Change-Id: I353d3ca91824ce20c9e9af1c5c84ab9953ddd8ab\n\u003e Reviewed-on: https://skia-review.googlesource.com/c/skia/+/201004\n\u003e Commit-Queue: Brian Salomon \u003cbsalomon@google.com\u003e\n\u003e Reviewed-by: Michael Ludwig \u003cmichaelludwig@google.com\u003e\n\nChange-Id: I22125cb7058f528cb368ff30c3c26e3d55056e66\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201222\nReviewed-by: Brian Salomon \u003cbsalomon@google.com\u003e\nCommit-Queue: Brian Salomon \u003cbsalomon@google.com\u003e\n"
    }
  ]
}