Apache OpenOffice (AOO) Bugzilla – Full Text Issue Listing |
Summary: | Chars from "Symbol" font incorrectly exported to PDF | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Product: | Writer | Reporter: | tab <tab> | ||||||||
Component: | open-import | Assignee: | michael.ruess | ||||||||
Status: | CLOSED FIXED | QA Contact: | issues@sw <issues> | ||||||||
Severity: | Trivial | ||||||||||
Priority: | P3 | CC: | h.ilter, issues, jbf.faure | ||||||||
Version: | OOo 3.3 Beta 1 | Keywords: | regression | ||||||||
Target Milestone: | --- | ||||||||||
Hardware: | All | ||||||||||
OS: | All | ||||||||||
Issue Type: | DEFECT | Latest Confirmation in: | --- | ||||||||
Developer Difficulty: | --- | ||||||||||
Issue Depends on: | |||||||||||
Issue Blocks: | 92011, 113586, 111112 | ||||||||||
Attachments: |
|
Description
tab
2010-08-13 02:21:23 UTC
Created attachment 71084 [details]
file with special chars
Created attachment 71085 [details]
pdf export by OO3.2
Created attachment 71086 [details]
pdf export by OO3.3 --bad
MRU->PL: the document contains some characters from Symbol font. These are exported incorrectly to PDF. Regression from 3.2.1 where it worked fine. MRU->HI / PL: Showstoppper? @hdu: please have a look looks like "Symbol" matches another font now; in 3.2 the PDF builtin font gets used. Are you sure that the CheckBox "Embed standard fonts" was not checked in the PDF export dialog ? Bad effect on Mac, too. Here the builtin Symbol font is actually taken, but the glyphs output are wrong. Maybe a separate problem though. Changing platform to "All" The problem on Mac could be issue 92011, the original problem on other platforms could be the new OpenSymbol (issue 105084, issue 105085, issue 106815) and the subsequent issue 113586. Complex as these matters always are I'll have to debug into it to confirm that suspicion. It would be interesting to know if it helps if the system-wide OpenSymbol gets deinstalled. It showed that the problem still existed even on systems where the issues mentioned where not immanent. The remaining root cause was that the text-encoding converters for single byte charsets such as the ones employed for the PDF base14 fonts did not handle "symbol aliasing" from U+F0XX -> S+00XX Extending these converters to allow symbol-aliasing for the symbol-charsets (symbol and dingbats) solves the problem. Done in CWS ooo33gsl06. @mru: please verify in CWS ooo33gsl06 Verified fix in CWS ooo33gsl06. Checked fix in OOO330m6. *** Issue 114629 has been marked as a duplicate of this issue. *** *** Issue 114629 has been marked as a duplicate of this issue. *** |