Re: index problem

[Eli Zaretskii]

> Date: Sun, 23 Sep 2001 12:45:47 +0200 (CEST)
> From: Werner LEMBERG <address@hidden>
> 
> OK, got 4.0c and applied your patch.  I'm not completely satisfied
> yet.  For example, typing `\A<TAB><TAB>', I get
> 
>   6 completions:
>   \a
>   \A
>   \A, allowed delimiters
>   \a, and translations
>   \A, incompatibilities with Unix troff
>   \a, used as delimiter
> 
> (as you can see, I've completely revised the index of the
> groff.texinfo file :-)

Yes; and I really appreciate that.

> I would like to see this instead:
> 
>   3 completions:
>   \A
>   \A, allowed delimiters
>   \A, incompatibilities with Unix troff

My change was not intended to do that.  In fact, you are looking in
the wrong place for the effect of the change ;-)  You are looking at
the list of possible completions which pos up after two TABs, but
please look instead at the echo area after a single TAB.  With the
original code, after typing `\A<TAB>', you'd see "\a" (lower-case `a')
in the echo area.  With the new code, you see "\A" instead.  This
might sound like a very insignificant change, but now try typing
`\A, a<TAB>' and I think you will see a big difference.

As for making the search for completion candidates case-sensitive, I'm
not sure that's a good idea.  There's no single standard for the
letter-case in index entries, so if we make the search case-sensitive,
a user could easily miss entries.