GNOME Bugzilla – Bug 547312
file-roller fails to read all files in archive that contains utf-8 files
Last modified: 2008-08-17 14:19:24 UTC
When file-roller is used to read a 7z zip file for example, that contains a file with a non utf8 name, file-roller stops processing at that point and we never get to see the rest of the archive contents. Example attached zip file testzip.zip. Contains the following : testzip/dir1 testzip/dir1/file1 testzip/dir1/file2 testzip/dir1/file3 testzip/dir2 testzip/dir2/file1 testzip/dir2/file2 testzip/dir2/file3 testzip/dir3 testzip/dir3/file1 testzip/dir3/file2 testzip/dir3/file3 testzip/dir4 testzip/dir4/çtest/love_to_see_this testzip/dir5 testzip/dir5/never_gets_here Notice the "ç" character in the filename under dir4. When loading this zip file with file-roller, function : g_io_chanel_read_line() is used to read the terminal chanel output from the 7z command. However when it encounters the "çtest" file name it returns G_IO_STATUS_ERROR and the error is set to : "Invalid byte sequence in conversion input" This we never get to see testzip/dir4/çtest/love_to_see_this or the dir5 and it's contents. However when you extract the contents they are extracted correctly. Would be nice of function fr-process.c:fr_channel_data_read() could be modified to handle these types of files. This bug is similar to : http://bugzilla.gnome.org/show_bug.cgi?id=306403
Created attachment 116363 [details] testzip.zip containing non utf8 character Example zip file containing non utf8 character.
*** This bug has been marked as a duplicate of 306403 ***
current trunk opens the attached archive without problems, however the general problem isn't fixed yet.