Re: query-replace-interactive

[Stefan]

>     but M-% invoked not from isearch could provide the last search string
>     by M-n.

> That sounds like a good and unproblematical idea.  Right now there is
> no default for the from-string, so it can't cause any trouble to offer
> the last search string as a default.

> We could then perhaps make M-% no longer special in a search.
> Instead, the default value would do the whole job.

But then

    C-s foo M-% bar RET

becomes

    C-s foo M-% M-n RET bar RET

and you lose the fact that M-% obeys the regexp/delimited/case-fold
attributes of the isearch.


        Stefan