GNU bug report logs - #19371
25.0.50; doc of functions and macros defined in macroexp.el

Previous Next

Full log

View this message in rfc822 format

From: Lars Ingebrigtsen <larsi <at> gnus.org>
To: Michael Heerdegen <michael_heerdegen <at> web.de>
Cc: 19371 <at> debbugs.gnu.org, Drew Adams <drew.adams <at> oracle.com>
Subject: bug#19371: 25.0.50; doc of functions and macros defined in macroexp.el
Date: Sat, 03 Aug 2019 13:50:27 +0200

Michael Heerdegen <michael_heerdegen <at> web.de> writes:

> Lars Ingebrigtsen <larsi <at> gnus.org> writes:
>
>>  (defun macroexp-progn (exps)
>> -  "Return an expression equivalent to \\=`(progn ,@EXPS)."
>> +  "Return EXPS with `progn' prepended.
>> +If EXPS is a single expression, `progn' is not prepended."
>>    (if (cdr exps) `(progn ,@exps) (car exps)))
>
> That's described a bit confusingly: AFAIU EXPS should always be a list
> of expressions, and when it's _a_list_ of only one expression, `progn'
> is not prepended and the expression is returned (and not EXPS as your
> text suggests).  BTW, compared to that description I find the original
> version much better (simpler).

Yeah, the doc string is now misleading.  I've rewritten it now as:

---
Return EXPS (a list of expressions) with `progn' prepended.
If EXPS is a list with a single expression, `progn' is not
prepended, but that expression is returned instead.
---

>> Uhm...  I have no idea what it does:
>>
>> (defmacro macroexp-let2* (test bindings &rest body)
>>   "Bind each binding in BINDINGS as `macroexp-let2' does."
>>   (declare (indent 2) (debug (sexp (&rest (sexp form)) body)))
>>   (pcase-exhaustive bindings
>>     ('nil (macroexp-progn body))
>>     (`((,var ,exp) . ,tl)
>>      `(macroexp-let2 ,test ,var ,exp
>>         (macroexp-let2* ,test ,tl ,@body)))))
>
> You understand what macroexp-let2 does?

No, not really.

> It supports one binding (a var plus an expression, specified as
> separate arguments).  macroexp-let2* supports a list of such pairs
> specified as BINDINGS, similar to let*.
>
> The naming scheme `macroexp-let2' vs. `macroexp-let2*' is not ideal:
> first, because `macroexp-let2' doesn't support multiple bindings like
> `let', and secondly because, if I look at the use cases in the sources,
> most of them just want to establish multiple bindings, but parallel
> binding would suffice, so they actually want to non-stared version of
> macroexp-let2*, which is not macroexp-let2 - AFAICT it doesn't exist.

Yes, it's confusing.  But perhaps you could propose a doc string for
`macroexp-let2*'?  :-)

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.