https://bz.apache.org/ooo/show_bug.cgi?id=128549

[email protected] changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |regression
                 CC|                            |[email protected]
           Hardware|Mac                         |All
             Status|UNCONFIRMED                 |CONFIRMED
             Latest|---                         |4.2.0-dev
    Confirmation in|                            |
     Ever confirmed|0                           |1

--- Comment #4 from [email protected] ---
Confirming based on screenshot. The character handling is a regression from
OpenOffice 2.1.

The source code for Writer's RTF parsing is in:
main/sw/source/filter/rtf
which subclasses the lower-level RTF parser in:
main/svtools/source/svrtf

In the attached RTF, the subtitle text "Mac et vidŽ o !" is encoded as:

00000240  70 36 34 20 5c 62 20 4d  61 63 20 65 74 20 76 69  |p64 \b Mac et vi|
00000250  64 5c 27 38 65 20 6f 20  21 20 5c 66 73 32 38 20  |d\'8e o ! \fs28 |

so the "\'8e " (5c 27 38 65 20) is coming through as "Ž " instead of "é ".

That trailing space is shown in OpenOffice, LibreOffice and Calibre. So let's
ignore it for now.

We need to find where the 4 characters "\'8e" (5c 27 38 65) are parsed and why
they are coming through as "Ž" (U+017D) instead of "é" (U+00E9).

-- 
You are receiving this mail because:
You are the assignee for the issue.

Reply via email to