Re: [Sbcl-devel] patch: SB-EXT:DEFINE-HASH-TABLE-TEST

[Christophe R. <cs...@ca...>]

"Nikodemus Siivola" <nik...@ra...> writes:

> The attached patch provides my first on SB-EXT:DEFINE-HASH-TABLE-TEST:

The interface that (as I understand it) is provided by Allegro might
be worth a look.  There, if I remember correctly, instead of having a
name for a test/hash-value pair, there's an extra keyword argument to
make-hash-table.  This is more flexible but maybe also more
error-prone?

I vaguely wonder about having define-hash-table-test define functions
named like (hash-table-test <identifier>) and (hash-table-hash
<identifier>), so that the Allegro interface can be easily supported
while still also allowing easy defaulting based on a
define-hash-table-test name.

Sunday afternoon witterings from me,

Best,

Christophe