Hash table read syntax

Hash table read syntax

[Chong Yidong]

I think there's a problem with the design of the hash table read syntax:

(make-hash-table)
#s(hash-table size 65 test eql rehash-size 1.5 rehash-threshold 0.8 data ())

The plist ought to use `:size', `:test', `:rehash-size', etc., rather
than `size', `test', `rehash-size', etc.  Otherwise, it's inconsistent
with the arguments to `make-hash-table', which uses `:size', `:test',...

I would like to make this change now, before it becomes "official" with
the Emacs 23.2 release.  Any objections?

Re: Hash table read syntax

[Chong Yidong]

Chong Yidong <cyd@stupidchicken.com> writes:

> I think there's a problem with the design of the hash table read syntax:
>
> (make-hash-table)
> #s(hash-table size 65 test eql rehash-size 1.5 rehash-threshold 0.8 data ())
>
> The plist ought to use `:size', `:test', `:rehash-size', etc., rather
> than `size', `test', `rehash-size', etc.  Otherwise, it's inconsistent
> with the arguments to `make-hash-table', which uses `:size', `:test',...
>
> I would like to make this change now, before it becomes "official" with
> the Emacs 23.2 release.  Any objections?

Never mind, I just remembered that the absence of a colon is for
compatibility with XEmacs.  LAME.

Re: Hash table read syntax

[Ted Zlatanov]

On Tue, 22 Dec 2009 14:31:54 -0500 Chong Yidong <cyd@stupidchicken.com> wrote: 

CY> Chong Yidong <cyd@stupidchicken.com> writes:
>> I think there's a problem with the design of the hash table read syntax:
>> 
>> (make-hash-table)
>> #s(hash-table size 65 test eql rehash-size 1.5 rehash-threshold 0.8 data ())
>> 
>> The plist ought to use `:size', `:test', `:rehash-size', etc., rather
>> than `size', `test', `rehash-size', etc.  Otherwise, it's inconsistent
>> with the arguments to `make-hash-table', which uses `:size', `:test',...
>> 
>> I would like to make this change now, before it becomes "official" with
>> the Emacs 23.2 release.  Any objections?

CY> Never mind, I just remembered that the absence of a colon is for
CY> compatibility with XEmacs.  LAME.

Yeah, it definitely looks unpleasant.  The reader could accept both
forms but it's probably not worth the trouble.

Ted

Re: Hash table read syntax

[Stephen J. Turnbull]

Chong Yidong writes:

 > Never mind, I just remembered that the absence of a colon is for
 > compatibility with XEmacs.  LAME.

Actually, no, it's for compatibility with the versions of Common Lisp
that were supported by Lucid Energize in about 1992, when generic hash
tables for Emacsen were implemented.  Guy and Kent admitted that it
was a mistake, though, so I guess we could too.