Re: broken font locking in gdb-script-mode

[Nick Roberts]

Stefan Monnier writes:
 > > AFAICS it never took line continuation into account.  I generalised the
 > > character set that a keyword can have.  It happens to include - / and =
 > > (and [0-9]).  If you had done:
 > 
 > > set args --foo --bar= \
 > >     baz --xyzzy=/some/path.ext
 > 
 > > baz would have had a keyword face.
 > 
 > Hmm... so keywords like /foo are possible?  Or is it only foo/bar (i.e. the
 > first char can't be one of / - or =)?  If the latter, maybe the patch below
 > will help.

Sure.  I just wanted to generalise it to include [0-9] and "-", so I copied
from the regexp for font-lock-function-name-face.  However a proper solution
to OPs problem would need to take line continuation into account.

We could also distinguish between commands and keywords in the same way that
sh-mode distinguishes between builtins (umask, alias, limit etc) and keywords
(if, then, else etc).  But I'm not sure how much time people spend editing
their .gdbinit and whether its worth the effort.

As an aside, I notice that the 'd' in end doesn't get fontified like it used
to.  I've just made a couple of changes to gdb-script-mode and I'm not sure
how I could have bust that.


-- 
Nick                                           http://www.inet.net.nz/~nickrob