Re: Can't compile `or' pcase pattern

[Michael Heerdegen <michael_heerdegen@web.de>]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> Looking at the definition of `l` I think you're pushing pcase a bit
> further than it can go.
>
> A bit like a DFA representation of a regexp can explode, a pcase's
> expansion can explode in some cases, and your `l` has all the tools
> needed to triggger such explosions.

Hmm, somehow I expected the technical term "explode" in your answer.

> Maybe you can reduce the pain, by looking at the expansion and trying
> to see how pcase-mutually-exclusive-p could be improved to get a
> better result.

I guess you mean `pcase--mutually-exclusive-p'.  When I get the time
I'll try to learn how that works.  For now, I have narrowed the
semantics of the lpats so that they don't automatically also try quoted
and function-quoted symbols.

Thanks,

Michael.