[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Your bug report about string-search

    (string-search "" "banana") returns 0, while (string-search-char "" "banana")
    and (string-search-set "" "banana") return NIL.  Seeing that the zeroeth element
    of "banana" is "b" and not the null string, it seems that string-search should do
    what the other two functions do, namely return NIL for this somewhat pathalogical

string-search doesn't search for elements, it searches for substrings.
Indeed there is a null string in characters 0 up to but not including 0
of "banana", hence string-search is returning the right value.
There is no reason that it should be doing what the other two functions
do, since they (each) do something different.

string-search-char should be giving an error, since the null string is
not a character.  In fact it has a bug; string-search-char does no
type checking on its first argument.  Should it coerce symbols and
strings into characters, as the character function does?

string-search-set could consistently return either 0 or NIL, depending
on whether the glass is half-full or half-empty (whether it is looking
for a character in the (empty) set, or looking for a place where there
isn't a character not in the (empty) set.)  Evidently it is doing the