Add spec link to beforematch-infinite-loop.html (web-platform-tests#5…

…4114)

wpt: Ensure that faulty JSON testharness output does not crash the Py…

…thon test harness

When a testharness test also prints debugging output, sometimes the
output can be mixed with the JSON output printed via an alert. This
causes a JSON decoding error in the output. Instead of crashing the
harness and printing many lines of Python stack trace output, print a
nice error. This makes the test output easier to read.

Signed-off-by: Martin Robinson <mrobinson@igalia.com>

[WebNN] Use MLNumber for padding value

Bug: 335909582, 328567884
Change-Id: Ie8528b25a3a5a50fa3a1b21d8d82850542fdf817
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6791467
Commit-Queue: Wei4 Wang <wei4.wang@intel.com>
Reviewed-by: Alex Gough <ajgo@chromium.org>
Reviewed-by: Rafael Cintron <rafael.cintron@microsoft.com>
Reviewed-by: ningxin hu <ningxin.hu@intel.com>
Reviewed-by: Phillis Tang <phillis@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1495871}

[WebNN] Use MLNumber for padding value

Bug: 335909582, 328567884
Change-Id: Ie8528b25a3a5a50fa3a1b21d8d82850542fdf817
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6791467
Commit-Queue: Wei4 Wang <wei4.wang@intel.com>
Reviewed-by: Alex Gough <ajgo@chromium.org>
Reviewed-by: Rafael Cintron <rafael.cintron@microsoft.com>
Reviewed-by: ningxin hu <ningxin.hu@intel.com>
Reviewed-by: Phillis Tang <phillis@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1495871}

[WebNN] Use MLNumber for padding value

Bug: 335909582, 328567884
Change-Id: Ie8528b25a3a5a50fa3a1b21d8d82850542fdf817
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6791467
Commit-Queue: Wei4 Wang <wei4.wang@intel.com>
Reviewed-by: Alex Gough <ajgo@chromium.org>
Reviewed-by: Rafael Cintron <rafael.cintron@microsoft.com>
Reviewed-by: ningxin hu <ningxin.hu@intel.com>
Reviewed-by: Phillis Tang <phillis@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1495871}

Add spec link to beforematch-infinite-loop.html (web-platform-tests#5…

…4114)

[WebNN] Use MLNumber for padding value

Bug: 335909582, 328567884
Change-Id: Ie8528b25a3a5a50fa3a1b21d8d82850542fdf817
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6791467
Commit-Queue: Wei4 Wang <wei4.wang@intel.com>
Reviewed-by: Alex Gough <ajgo@chromium.org>
Reviewed-by: Rafael Cintron <rafael.cintron@microsoft.com>
Reviewed-by: ningxin hu <ningxin.hu@intel.com>
Reviewed-by: Phillis Tang <phillis@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1495871}

[text-spacing-trim] Support fallback (part 2)

As a follow up to crrev.com/c/6772946, this patch supports
when font fallback occurs, and punctuation pairs ended up with
different fonts. The first patch supported when the characters
to apply kerning come later. This patch supports when they
come earlier.

In that case, characters to apply kerning were already shaped,
and that applying the `halt` feature isn't possible. This
patch applies kerning by adjusting spacing instead.

Bug: 431660829
Change-Id: Ifa08e054fae4d01ad507385ebae97570939c4768
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6800652
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Auto-Submit: Koji Ishii <kojii@chromium.org>
Commit-Queue: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1495753}

[FedCM] Adding mark_signed_in to fedcm_test

Adding mark_signed_in to the beginning of fedcm_test ensures that user
is in a signed in state for each test. Tests calling mark_signed_out but
never sign back in cause flakiness in repeated test runs. This change
fixes that.

Bug: 41482163
Change-Id: Ib533bba63e73f59da32ce597f6a690873d0cf884
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6804547
Reviewed-by: Christian Biesinger <cbiesinger@chromium.org>
Commit-Queue: Nicolás Peña <npm@chromium.org>
Reviewed-by: Nicolás Peña <npm@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1495722}

[Gap Decorations]: Avoid expanding GapDataList

This CL avoids expanding GapDataList for the purpose of painting. It
starts off by introducing a new class `GapDataListIterator` to traverse
a `GapDataList` without expanding integer repeaters. The crux of the
implementation is given that at paint time we know the number of gaps
(which is fixed), we can calculate the amount of slots auto repeaters
will need.

This allows us to treat each gap data entry as occupying a fixed number
of "slots", which makes it possible to segment the list into three
logical regions based on the presence and position of an auto repeater:
* Leading: entries before auto repeater
* Auto: auto repeating values
* Trailing: entries after auto repeater

Each region is assigned a slot count, and the iterator advances within a
region until its slots are exhausted. When slots are exhausted, it
transitions to the next region.

Internally, the iterator maintains key traversal states:

* `list_idx_`: index into the original gap data list
* `repeats_left_`: how many times the current entry should repeat
* `repeated_value_idx_`: index into the repeated values of the current entry.

These states are updated as the iterator moves through the list,
ensuring the correct decoration is returned for each gap without
expanding the full list.

Bug: 357648037, 431815998
Change-Id: I6fcc8267c9d21ac7c2eea15e4b47412840ef589d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6773829
Reviewed-by: Javier Contreras <javiercon@microsoft.com>
Reviewed-by: Alison Maher <almaher@microsoft.com>
Reviewed-by: Kevin Babbitt <kbabbitt@microsoft.com>
Commit-Queue: Sam Davis Omekara <samomekarajr@microsoft.com>
Cr-Commit-Position: refs/heads/main@{#1495573}