http://qa.mandrakesoft.com/show_bug.cgi?id=3961
------- Additional Comments From [EMAIL PROTECTED] 2003-04-08 23:56 -------
So, testing with the latest version from Cooker, I don't see much improvement.
It's even worse than before :
1. In the "text import/export" parameter dialog, the default locale is "English:
ASCII" and needs to be manually changed to either "iso-8859-1" or "utf-8" -
although my system locale already is "fr_FR", hence should automatically imply
one of those charsets.
2. Even if I change the locale in the text import parameter form, it will revert
back to "English: ASCII" on a subsequent operation, which makes it very tedious
and error-prone.
3. Exporting French accents to utf-8 does work as before, but it does not work
in iso-8859-1* : it creates a borked file with French accents and adjacent
letters replaced with garbage (try it out yourself with the word "�l�phant").
4. If I try to reimport a borked iso-8859-1 CSV file as created by Gnumeric,
Gnumeric crashes !
5. When opening a file, automatic detection of CSV/TSV files fails ("unsupported
file type")
--
Configure bugmail: http://qa.mandrakesoft.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.
------- Reminder: -------
assigned_to: [EMAIL PROTECTED]
status: UNCONFIRMED
creation_date:
description:
The current Gnumeric package is broken when you want to import text files using
iso-8859-1 characters, and these text files contain non-ascii characters. It
works when using utf-8 characters.
Simple test case (assuming iso-8859-1 or iso-8859-15 is your default locale)
- launch Gnumeric
- type some text with non-ascii characters in a cell, e.g. "�l�phant"
- export the sheet as a text file Book1.csv
- close the file
- re-import Book1.csv as a text file : it fails (wrong character)
Now the same test case but using utf-8 :
- launch "LC_ALL=fr_FR.UTF-8 Gnumeric" (or LC_ALL=en_GB.UTF-8, etc.)
- type some text with non-ascii characters in a cell, e.g. "�l�phant"
- export the sheet as a text file Book1-utf8.csv
- close the file
- re-import Book1-utf8.csv as a text file : it succeeds
This bug was first filed as a generic Gnumeric bug
(http://bugs.gnome.org/show_bug.cgi?id=110182), but it seems the mainline
version works ok, so I think this bug is Mandrake-related. It only appeared with
Mandrake 9.1, it was fine before. Also, I have no locale problems in other
programs. Using native Gnumeric files is ok.
I am attaching files for the two test cases mentioned above.
