Re: info-lookup-symbol index case sensitivity

[Kevin Ryde <user42@zip.com.au>]

Richard Stallman <rms@gnu.org> writes:
>
> It is a good thing for index search to ignore case.  That should make
> it far easier to use, especially for beginners.  As for occasional
> conflicts, that is less of an issue.

But I would assert that if I ask for "C-h C-i FOO" then I should be
shown the doco for "FOO" and not for "foo".

I'm pretty sure this is just a matter of the second visit to the index
nodes being case insensitive whereas the first visit (building the
cache) was sensitive.

(Arguably the cache would be better off holding the actual target node
name, rather than re-scanning the indexes to get it.  Might use up a
bit more memory though.)


A nice way to establish some leeway for case, in a case sensitive
mode, might be to look first for a symbol in the case given, and if
that fails then look again with a case-fold.