On Mon, 18 Mar 2002, Karl Berry wrote:
> For instance, consider this index entry from the autoconf manual, which
> is currently misparsed by both Info and Emacs:
> * @%:@: Quadrigraphs.
>
> A regular expression such as (egrep syntax for simplicity):
> ^\* (.*)( <[0-9]>*)?:[ \t]*(.*)\.$
>
> will end up with \1 being @%:@ (the index entry name)
> and \3 being Quadrigraphs (the node name).
Two problems:
(a) The stand-alone Info doesn't use regular expressions to parse
references;
(b) You could have more than one colon in a row (happens a lot in C++,
for example).
The possibility of parsing references from the end was raised in the past
(IIRC). It works better in some cases, but fails in others.
_______________________________________________
Bug-texinfo mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-texinfo
