GNOME Bugzilla – Bug 110547
status-item-view text isn't UTF-8
Last modified: 2006-06-18 05:10:55 UTC
While downloading an article whose subject header has the letter 'a' with an accent over it, status-item-view filled up the error log with thousands of `sending invalid utf-8 sequence to pango' messages.
nntp_download_headers() runs the subject through pan_header_to_utf8 (), so it should be in UTF-8 already. The only case where this would not happen is if the conversion fails (faillback to g_strdup ()). I guess status-item-view could do a g_validate_utf8() and fall back to a generic description in this case ?
Backtrace of this event:
+ Trace 36170
Note the original_len: 135927072.
Eh, that seems to be a bug in RedHat9's gdb and/or gcc. Disregard that ... :-) The problem is that string_shorten_nolock() treats the string as a binary character stream and can truncate it at any position, including in the middle of a multibyte character. This needs to be enhanced to be UTF8-aware.
Fixed in CVS: http://cvs.gnome.org/bonsai/cvsview2.cgi? diff_mode=context&whitespace_mode=show&subdir=pan/pan&command=DIFF_FR AMESET&file=status-item-view.c&rev1=1.82&rev2=1.83&root=/cvs/gnome
You are the wind beneath my wings.