https://bugs.documentfoundation.org/show_bug.cgi?id=166497
Bug ID: 166497
Summary: Test document with RTL+LTR text looks different and
causes unit test failure
Product: LibreOffice
Version: 25.2.0.3 release
Hardware: All
OS: All
Status: UNCONFIRMED
Keywords: bibisected, bisected
Severity: normal
Priority: medium
Component: graphics stack
Assignee: [email protected]
Reporter: [email protected]
CC: [email protected]
Blocks: 126862, 162322
Created attachment 200701
--> https://bugs.documentfoundation.org/attachment.cgi?id=200701&action=edit
Screenshot in Writer
I noticed the following unit test failure (the test was added for bug 162314)
in my LO build on Windows:
test/source/xmltesttools.cxx:169:`anonymous namespace'::TestTdf162314::TestBody
equality assertion failed
- Expected: Aa aa aaaa ﷽
- Actual : Aa aa aaaa ﷽ aaaa
- In <>, attribute 'portion' of
'/root/page/body/txt/SwParaPortion/SwLineLayout[1]' incorrect value.
When opening sw\qa\extras\layout\data\tdf162314.fodt from the build, which as
the same as attachment 195665, just using Noto Sans Arabic (this matters) and
in FODT format, the Arabic text is shown compacted into two lines, but the
whole RTL+LTR text is within a single line. This doesn't conform the original,
where the Arabic text wasn't split into two lines, and part of the LTR text is
moved to the 2nd line.
See attached screenshot.
This started from the following change in 25.2:
https://git.libreoffice.org/core/commit/e6282cff7857999b031bb847a6ca8d996a090db1
author Michael Stahl <[email protected]> Fri Aug 02
12:48:05 2024 +0200
committer Michael Stahl <[email protected]> Mon Aug 12
12:18:25 2024 +0200
more_fonts: ship "hinted" instead of "full" for "Noto Sans Arabic"
Referenced Bugs:
https://bugs.documentfoundation.org/show_bug.cgi?id=126862
[Bug 126862] [META] Issues with unusual builds / unit tests, rare test failures
https://bugs.documentfoundation.org/show_bug.cgi?id=162322
[Bug 162322] [META] Right-to-Left (RTL) text layout issues
--
You are receiving this mail because:
You are the assignee for the bug.
