On Mon, Dec 08, 2014 at 03:28:30PM -0500, Richard Hipp wrote: > > > On Mon, Dec 8, 2014 at 3:22 PM, bch <brad.har...@gmail.com> wrote: > > Does the "mode" (text/binary) factor into the sha1 fingerprint hash of > an artifact (single file) or commit ? > > > > No. The "mode" is only used to determine how to display the content on a > webpage. Also, independent changes can only be merged in text files, not > binary files. And "diff" will only work for text files. > > For the last two points - merge and diff - you can tell Fossil that the file > is > text all you want, but if it really is binary that isn't going to make the > merge and/or diff algorithms work any better. You will still end up with a > mess. >
> So what is the point of this? For what I understand from the OP, it's only for convenience to be able to call "fossil diff/gdiff" on a file that is mostly text but is detected as binary by fossil. >Why is the default text/binary detection not working for Tony? Andy analyse the OP file: https://www.mail-archive.com/fossil-users@lists.fossil-scm.org/msg18304.html Seems to be because there's some control and null characters in the file. For myself, I like the fact that fossil detect the binary files by itself, it save me a lot of time, especially on repository with a lot files. But I think it would be good to have either a kind of text-glob setting or at least a flag to the diff/gidff command to force the diff to happens. -- Martin G. _______________________________________________ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
