On Monday, 17.07.2006 at 22:19 +0200, Adam Borowski wrote: > --- colordiff~ 2006-07-17 15:01:56.564810563 +0200 > +++ colordiff 2006-07-17 15:01:56.564810563 +0200 > @@ -31,6 +31,8 @@ > use Getopt::Long qw(:config pass_through); > use IPC::Open2; > > +exec "/usr/bin/diff",@ARGV unless -t STDOUT; > + > my $app_name = 'colordiff'; > my $version = '1.0.4'; > my $author = 'Dave Ewart'; > > (Of course, the exact place doesn't really matter.)
It's not as simple as that, though: the above will fail if colordiff (or
'diff' if aliased to colordiff) is used in a pipe. In other words,
this is an incomplete fix.
In detail...
i.e. (assuming 'alias diff=colordiff')
diff file1 file2 - this will work and colour-highlight the output
diff file1 file2 > my.patch - this will *not* colour-highlight, as intended
but
someprog | diff > my.patch - this will give an error
/usr/bin/diff: missing operand after `/usr/bin/diff'
/usr/bin/diff: Try `/usr/bin/diff --help' for more information.
Given this, and the fact that colordiff's entire raison d'être is to
colour highlight, my feeling is that a better option is this: rather than try
to make colordiff *avoid* highlighting in certain circumstances, this
limitation is explained in the man page.
A documentation change might actually be a more thorough solution: it
will make the user know "what they're letting themselves in for" by
aliasing 'diff' to 'colordiff'.
Thoughts on that, Adam?
> > > I already reported this bug upstream over three years ago, but it
> > > appears to be left unfixed. At least, the bug bit me today again :p
> >
> > [Odd - I've got no record of you ever contacting me, which is strange
> > since I'm usually fairly obsessive about keeping that sort of email.
> > Did I ever reply?! Apologies if not, but I don't think I got your
> > message.]
>
> I'm afraid I didn't use mail, just the SourceForge bug report system:
>
> https://sourceforge.net/tracker/?func=detail&aid=745456&group_id=65829&atid=512406
>
> I'm not sure if they mail you any new bug reports by default; I would
> probably fail to notice such reports on my projects for years, too.
Well, that's news to me - never got any notifications of those bug
reports. I'll go take a look at them ...
Dave.
--
Dave Ewart - [EMAIL PROTECTED] - jabber: [EMAIL PROTECTED] - freenode: davee
All email from me is now digitally signed, key from http://www.sungate.co.uk/
Fingerprint: AEC5 9360 0A35 7F66 66E9 82E4 9E10 6769 CD28 DA92
signature.asc
Description: Digital signature
