GNOME Bugzilla – Bug 162044
Changing encoding after loading
Last modified: 2006-05-19 05:50:22 UTC
one function button for re-decoding the loaded file with other character set. one function button for temparally auto-line-wrap switch.
> one function button for re-decoding the loaded file with other character set this is fixed, since the only valid case when you want to reload with another charset is if the autodetction failed. With gedit 2.13 in this case you get offered the possibility of choosing another encoding. > one function button for temparally auto-line-wrap switch this is still there, but this request is traked in bug 119428, so I am marking this as a duplicate. *** This bug has been marked as a duplicate of 119428 ***
can it also handle the situation, where the autodetection is not "failed" but just "wrong" with a temperary file located somewhere under /tmp ?
Sorry but I don't see the difference between 'failed' and 'wrong'... you open a file in gedit (in /home, in /tmp, wherever else): if gedit picks the right encoding the file is loaded and you can read it, otherwise it shows a message along the lines of "Could not detect the encoding of the file, retry with another encoding?" and a widget to manually choose the encoding with which to retry
There are files that can be opened with different encodings and so, as Wei-Lun said, autodetection can be "wrong". I don't see how we can solve this problem in an easy-way since encoding conversion happens only at loading/saving time since gedit uses UTF-8 internally. BTW, autodetection can be easily customized changing the /apps/gedit-2/preferences/encodings/auto_detected Actually the translators should customize that gconf key.
When I view an attached text file in webmail, gedit will open it somewhere in /tmp. If this file contains some UTF-8 coded characters like U-FFFFF or U-FFFFE, it would be detected as iso8859-15. if I save it to a file and open it manully with UTF-8, the file can not be loaded, and I don't know on which position it failed.