commit 871eaac0121ec68280a1829b06b4ac8151ccf15c
author Alan Screen <awscreen@chromium.org> Mon Nov 18 20:44:14 2024
committer Chromium LUCI CQ <chromium-scoped@luci-project-accounts.iam.gserviceaccount.com> Mon Nov 18 20:44:14 2024
tree 3dc477b4752fc04305612aa63250ca8a06ea4dfe
parent 9e176251d76eab8cf9073a970ceacbdc0885a8ce

[M132][PDF Ink Signatures] Fix PdfInkModule invalidation areas

When CanonicalInkEnvelopeToInvalidationScreenRect() was initially added
it intentionally ignored the origin of the page content rectangle.
This was due to callers making an extra offset modification to the
returned result before actually doing the invalidation.  This omission
was a mistake, because the callers add an offset for the available area
but not for the page content origin.  This means that there could be a
shift in the resulting invalidation area that didn't account for that
origin.  This shift becomes much more noticeable once a user scrolls
the document.

Update the CanonicalPositionToScreenPosition() helper to account for
that offset, which now makes it a real inverse operation of the
EventPositionToCanonicalPosition() function.  This is because available
area offsets are also already accounted for in event positions.

This removes the need for the prior workaround function
CanonicalInkEnvelopeToExpandedInvalidationScreenRect().

(cherry picked from commit f483be856a9a5eb2f86e6a0f8cf9263b9f8fc70f)

Fixed: 375445386, 376301209
Change-Id: I6df4303e803ead43b1d54d503350b8ee11545ec7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6020758
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Alan Screen <awscreen@chromium.org>
Code-Coverage: findit-for-me@appspot.gserviceaccount.com <findit-for-me@appspot.gserviceaccount.com>
Cr-Original-Commit-Position: refs/heads/main@{#1383301}
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6031936
Cr-Commit-Position: refs/branch-heads/6834@{#449}
Cr-Branched-From: 47a3549fac11ee8cb7be6606001ede605b302b9f-refs/heads/main@{#1381561}