Re: case-table functions clobbering extra slots

[Dave Love]

Richard Stallman <address@hidden> writes:

>     Things like set-case-syntax-pair clobber the extra slots of the case
>     table.  Is there a good reason for that?  It isn't documented.
>
> The reason was documented in some of the previous commands in this file.
> I will copy these comments into the functions that didn't have them.

I saw comments, but I don't understand it.  I think it should be
documented other than in comments if it stays like that.

>     It defeats treating cases where there isn't one-to-one correspondence
>     between upper and lower (which is even mentioned as a possibility in
>     (elisp)Case Tables).
>
> If that is true, it could be because automatically recomputing the
> three derived tables does not know how to handle this case correctly.
> Is that it?

I don't know.  I don't understand what's done and why.

> Can you show the code that would handle this case correctly?

No.  It can't be done reliably at present as far as I can tell.

In the case of sigma, it should be that both σ and ς upcase to Σ, and
Σ downcases to σ.  [I hope I got those right, typing unicodes without
being able to display them.  Obviously likewise for the iso8859-7
charset.]  I set up entries in the case table interactively to get
that right and then found they were clobbered by subsequent
case-table.el functions being called.  I guess it's always been like
that, so it's probably not important to address, at least now.