Hi, all, i just saw something interesting and thought i'd report it to those more knowledgeable than i in the mysteries of dealing with mangled UTF8. i know some work was recently done on getting the www diff view to deal with that, and i _think_ i just found a related case from the CLI.
My to-be-committed changes included the introduction of some multi-byte chars in one UTF8-encoded file[1]. When i did (fossil diff) before committing, some of those characters were shown mangled on my console (my console is UTF8-capable - i rely on that all the time). However, i figured it was just a minor glitch in the CLI version of the diff output and made my commit. Now, when i view the diff either with the WWW or CLI (fossil diff --from ... --to ...) it all looks perfectly fine. My _suspicion_ (without having looked at it) is that i hit a corner case involving diffing of local files with UTF8 (fossil diff) vs a diff of in-repo content (fossil diff --from ... --to ...). Platform = Linux x86. Anyway, i'll watch out for it and report the details if it happens again (too tired to try to reproduce it on other files right now). Until then... well, just ignore this and i'll refer to it again later if i can reproduce the problem. Ah, they're still in my scrollback buffer. Screenshots: Before commit: http://www.wanderinghorse.net/tmp/diff-utf8-local.png After: http://www.wanderinghorse.net/tmp/diff-utf8-committed.png Not a big deal, just thought it might interesting the character encoding fans out there. Happy Hacking! [1]: strange coincidence: the MBCs were added explicitly to test that the fossil script bindings correctly deal with mangled UTF8 in deltas/patching. -- ----- stephan beal http://wanderinghorse.net/home/stephan/ http://gplus.to/sgbeal
_______________________________________________ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
