Issue 26931 - OOo Contact connection to Evolution is loosing accented characters
Summary: OOo Contact connection to Evolution is loosing accented characters
Alias: None
Product: Base
Classification: Application
Component: code (show other issues)
Version: OOo 1.1.1RC
Hardware: PC Linux, all
: P3 Trivial with 3 votes (vote)
Target Milestone: ---
Assignee: AOO issues mailing list
QA Contact:
Depends on:
Reported: 2004-03-24 16:09 UTC by lsafar
Modified: 2017-05-20 10:45 UTC (History)
2 users (show)

See Also:
Issue Type: DEFECT
Latest Confirmation in: ---
Developer Difficulty: ---

Output of: evolution-addressbook-export --format=csv --output=addresses.csv (673 bytes, text/txt)
2004-03-29 11:01 UTC, lsafar
no flags Details

Note You need to log in before you can comment on or make changes to this issue.
Description lsafar 2004-03-24 16:09:38 UTC
I've created new data source in OOo, which is of contact type with evolution as

Accented characters in evolution database don't survive - i.e. "á" is converted
to "/303/241", "í" is converted to "/303/255". 
I would guess it has something to do with Evolution (1.4.5) storing information
in UTF-8, but not sure.
Comment 1 marc.neumann 2004-03-26 11:06:21 UTC

I can reproduce this.

msc -> windly: the evolution-addressbook-export export the values correct, so I
believe it's the OOo driver.

Bye Marc
Comment 2 2004-03-26 15:19:52 UTC
Cah you please attachment the exported file?
Comment 3 lsafar 2004-03-29 11:01:06 UTC
Created attachment 14143 [details]
Output of: evolution-addressbook-export --format=csv --output=addresses.csv
Comment 4 lsafar 2004-03-29 11:04:26 UTC
I've attached some output where the characters appear as well.
It looks like the main problem is that UTF-8 gets escaped during export, there
are two character expected in name "á" and "ý".
Comment 5 lsafar 2004-04-19 15:38:26 UTC
I've used perl oneliner to unescape output and it works OK - output is an
unicode text, import as such to OOo works OK.

evolution-addressbook-export --format=csv | perl -pe 's/\\([0-7]{1,3})/chr oct

Would it be possible to unescape automatically during import?
Comment 6 mkretzschmar 2004-04-20 09:25:20 UTC
This was already mentioned in bug 20958

I believe that octal-escaping utf-8-encoded unicode is wrong, so I wrote a patch
for evolution last October to not do that.

At that time I was thinking that might be handled reasonably
fast. Unfortunately it wasn't.

At this point of the evolution 1.4 product lifetime (and as the db connection
for evolution 1.5/2.0 is almost done AFAIK), it's probably better to add the
unescaping code on the OOo side.
Comment 7 2005-03-23 08:59:56 UTC
working transfer
Comment 8 Frank Schönheit 2005-05-10 13:57:54 UTC
Comment 9 Marcus 2017-05-20 10:45:02 UTC
Reset the assignee to the default "".