bug#49534: [External] : bug#49534: 26.3; Isearch should support using filter predicates with empty search hits

[Drew Adams]

>   (= (match-beginning 0) (match-end 0))
> 
> was moved below after
> 
>   (funcall isearch-filter-predicate
>            (match-beginning 0) (match-end 0))
> 
> This means that if isearch-filter-predicate does own matching,
> it will break later (= (match-beginning 0) (match-end 0)).
> 
> What would be better: to remember its result in a let-bound variable,
> or to use save-match-data?  Probably, save-match-data:

Yes, that's probably the right thing; good catch.
___

[On the other hand, a filter predicate can really
do more than just act as a predicate (it can
perform useful side effects).  There could
presumably be reasons we'd want to let it change
the match data.

No, I don't have any concrete use case in mind;
and yes, without some other changes, that could
mess up code that follows using the predicate.
I just wonder about further restricting the
predicate.  The point of this bug fix was to
_remove_ an unnecessary restriction, but now
we're adding another restriction. ;-)]