Re: New assoc-let package

[Ted Zlatanov <tzz@lifelogs.com>]

On Sat,  6 Dec 2014 14:58:36 -0500 (EST) Christopher Schmidt <ch@ristopher.com> wrote: 

CS> Artur Malabarba <bruce.connor.am@gmail.com> writes:
>> On 6 Dec 2014 11:36, "Christopher Schmidt" <ch@ristopher.com> wrote:
>>> The inability to meaningfully nest different assoc-let forms makes
>>> the macro inflexible.
>> 
>> What would you consider meaningful nesting? Being able to access the
>> variables of the outer macro inside the inner macro?

CS> Yeah, that is what I meant.  Maybe the macro can support optional args
CS> that specify the keys to be used.

CS>     (assoc-let .foo .bar alist1
CS>       (assoc-let .rms .gpl alist2 ; foo and bar bound to alist1
CS>         ...                       ; rms and gpl bound to alist2

I think the CL package does this kind of magic using gensyms.  I
personally find it hard to examine in a stack trace, but really
convenient as a user.

CS> Maybe the name alist-let is more appropriate than assoc-let or assq-let
CS> as this facility is of the same kind as alist-get.

`let-assoc' so it starts with the functionality it provides, not with
the data format it uses?

Ted