Re: case-table functions clobbering extra slots

[Richard Stallman]

      So the "dotted-uppercase-I to i" mapping could have a
    "non-reversible" flag, and the downcasing code would notice this when
    changing it to a normal "i", and put an `uppercase' text property on
    the result character.  Then if the user subsequently did an upcase,
    the upcasing code could notice the `uppercase' property and properly
    change the normal "i" to a dotted-uppercase-I.

This is not a useful way to handle the problem of `i' in Turkish.
Every `i' has to upcase to dotted-uppercase-I, in Turkish.  Doing it
only for an `i' that has a special 'upcase' property would not be
correct.

Perhaps you are thinking of this property as a kind of cache
mechanism.  If so, the most it could achieve is to increase
efficiency.  We would still need some other mechanism to get correct
results in general.

So I think we should first design the other mechanism.  If that
mechanism turns out not to be fast enough, and if caching could make
it faster, we could add caching.


Something like the "suspicious flag" might work for that mechanism.