Mail Archives: djgpp-workers/2001/01/31/13:00:34
> Why not make it use the same mode consistently? Thus, a command such
> as "cvs diff foo.c --binary" will use binary I/O both for the
> temporary files and in computing the diffs.
I'll look into this. Shouldn't be too hard.
> > I could try fiddling with the diff library, but I'm hesitant
> > to do so; there's too much scope for breaking it.
> I don't think you need to fiddle with the diff code: it already does
> The Right Thing (has been doing that for a long time). What you need
> is make the diff library's mode consistent with the mode used by CVS
> elsewhere. Then it will DTRT in general, I think.
Probably.
> > In any case, I believe the current way does the right thing. Seeing as
> > how cvs is supposed to be able to handle (and preserve) different styles
> > of line endings when not using --text, it seems logical for the diff
> > subcommand to see them as different.
> I'm not sure. "cvs diff" is mostly used locally, to see how is the
> workfile different from the repository. Therefore, it won't do any
> harm to make it default to text I/O: it will work for Unix-style
Yes, but if you're using the 'default' mode, that suggests you want to
be able to see the difference between DOS and Unix text files (regardless
of whether it's a local repository you're using). Still, I have no problem
having diff default to text mode, provided that a) I can get the --binary
option for diff to work, and b) cvs is able to easily override this (so
-kb files are still read in binary mode by diff).
> files, and will allow users who keep their local files with DOS-style
> EOLs to not bother with --text unless they want to.
People who want DOS EOLs for their local files SHOULD use --text (as for
them not having to bother with --text: they can just add it to their
.cvsrc).
> Too bad. WinCVS is just too much of a PITA, but I guess I'll have to
> continue living with it...
'fraid so. Unless you can find someone who's willing to finish adding
winsock2 support to libsocket...
- Raw text -