* Understanding macroexp.el
@ 2015-11-18 10:32 Michael Heerdegen
2015-11-18 11:37 ` Andreas Schwab
2015-11-18 13:40 ` Stefan Monnier
0 siblings, 2 replies; 5+ messages in thread
From: Michael Heerdegen @ 2015-11-18 10:32 UTC (permalink / raw)
To: Emacs Development; +Cc: Stefan Monnier
[-- Attachment #1: Type: text/plain, Size: 136 bytes --]
Hello,
yesterday I tried to understand the helpers in macroexp.el. I have a
question:
Would it make sense to install the following:
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: macroexp.patch --]
[-- Type: text/x-diff, Size: 740 bytes --]
*** /tmp/ediff139941AX 2015-11-18 11:12:07.788012282 +0100
--- /home/micha/software/emacs/lisp/emacs-lisp/macroexp.el 2015-11-18 10:47:41.041352909 +0100
***************
*** 331,337 ****
(cond
((null bindings) exp)
((eq 'let* (car-safe exp)) `(let* (,@bindings ,@(cadr exp)) ,@(cddr exp)))
! (t `(let* ,bindings ,exp))))
(defun macroexp-if (test then else)
"Return an expression equivalent to `(if ,TEST ,THEN ,ELSE)."
--- 331,337 ----
(cond
((null bindings) exp)
((eq 'let* (car-safe exp)) `(let* (,@bindings ,@(cadr exp)) ,@(cddr exp)))
! (t `(let* ,bindings ,@(macroexp-unprogn exp)))))
(defun macroexp-if (test then else)
"Return an expression equivalent to `(if ,TEST ,THEN ,ELSE)."
[-- Attachment #3: Type: text/plain, Size: 479 bytes --]
Here is an example where it matters:
(defmacro test (exp)
(macroexp-let2 nil res exp
`(progn
(f ,res)
(g ,res))))
(note that the `progn' is mandatory here.) Without the patch:
(macroexpand '(test (+ 1 2)))
==>
(let*
((#:res #1=(+ 1 2)))
(progn
(f #1#)
(g #1#)))
After patch:
(macroexpand '(test (+ 1 2)))
==>
(let*
((#:res #1=(+ 1 2)))
(f #1#)
(g #1#))
(And is it currently allowed to push this to master?)
Thanks,
Michael.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Understanding macroexp.el
2015-11-18 10:32 Understanding macroexp.el Michael Heerdegen
@ 2015-11-18 11:37 ` Andreas Schwab
2015-11-18 12:03 ` Michael Heerdegen
2015-11-18 13:40 ` Stefan Monnier
1 sibling, 1 reply; 5+ messages in thread
From: Andreas Schwab @ 2015-11-18 11:37 UTC (permalink / raw)
To: Michael Heerdegen; +Cc: Stefan Monnier, Emacs Development
Michael Heerdegen <michael_heerdegen@web.de> writes:
> Here is an example where it matters:
>
> (defmacro test (exp)
> (macroexp-let2 nil res exp
> `(progn
> (f ,res)
> (g ,res))))
>
> (note that the `progn' is mandatory here.) Without the patch:
>
> (macroexpand '(test (+ 1 2)))
> ==>
> (let*
> ((#:res #1=(+ 1 2)))
> (progn
> (f #1#)
> (g #1#)))
>
>
> After patch:
>
> (macroexpand '(test (+ 1 2)))
> ==>
> (let*
> ((#:res #1=(+ 1 2)))
> (f #1#)
> (g #1#))
For byte-compiled code there should be no difference between the two
expansions.
Andreas.
--
Andreas Schwab, SUSE Labs, schwab@suse.de
GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7
"And now for something completely different."
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Understanding macroexp.el
2015-11-18 11:37 ` Andreas Schwab
@ 2015-11-18 12:03 ` Michael Heerdegen
2015-11-18 13:36 ` Stefan Monnier
0 siblings, 1 reply; 5+ messages in thread
From: Michael Heerdegen @ 2015-11-18 12:03 UTC (permalink / raw)
To: Andreas Schwab; +Cc: Stefan Monnier, Emacs Development
Andreas Schwab <schwab@suse.de> writes:
> For byte-compiled code there should be no difference between the two
> expansions.
What's the use of `macroexp-unprogn' then at all? Or does it make a
difference in other cases?
Michael.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Understanding macroexp.el
2015-11-18 12:03 ` Michael Heerdegen
@ 2015-11-18 13:36 ` Stefan Monnier
0 siblings, 0 replies; 5+ messages in thread
From: Stefan Monnier @ 2015-11-18 13:36 UTC (permalink / raw)
To: Michael Heerdegen; +Cc: Andreas Schwab, Emacs Development
>> For byte-compiled code there should be no difference between the two
>> expansions.
> What's the use of `macroexp-unprogn' then at all?
It's when you care to generate slightly cleaner code it's easy to do so.
It usually doesn't really matter much.
> Or does it make a difference in other cases?
I can't remember exactly what motivated me to introduce
macroexp-unprogn, but I don't think it was a strong need.
I can retroactively come up with scenarios where it could make
a difference, but they all seem reasonably infrequent.
E.g. (+ 1 2) will be optimized to 3, but (+ 1 (progn 2)) won't.
Stefan
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Understanding macroexp.el
2015-11-18 10:32 Understanding macroexp.el Michael Heerdegen
2015-11-18 11:37 ` Andreas Schwab
@ 2015-11-18 13:40 ` Stefan Monnier
1 sibling, 0 replies; 5+ messages in thread
From: Stefan Monnier @ 2015-11-18 13:40 UTC (permalink / raw)
To: Michael Heerdegen; +Cc: Emacs Development
> Would it make sense to install the following:
>
> *** /tmp/ediff139941AX 2015-11-18 11:12:07.788012282 +0100
> --- /home/micha/software/emacs/lisp/emacs-lisp/macroexp.el 2015-11-18 10:47:41.041352909 +0100
> ***************
> *** 331,337 ****
> (cond
> ((null bindings) exp)
> ((eq 'let* (car-safe exp)) `(let* (,@bindings ,@(cadr exp)) ,@(cddr exp)))
> ! (t `(let* ,bindings ,exp))))
>
> (defun macroexp-if (test then else)
> "Return an expression equivalent to `(if ,TEST ,THEN ,ELSE)."
> --- 331,337 ----
> (cond
> ((null bindings) exp)
> ((eq 'let* (car-safe exp)) `(let* (,@bindings ,@(cadr exp)) ,@(cddr exp)))
> ! (t `(let* ,bindings ,@(macroexp-unprogn exp)))))
>
> (defun macroexp-if (test then else)
> "Return an expression equivalent to `(if ,TEST ,THEN ,ELSE)."
That'd be fine, I think, yes. It probably won't make much difference,
if any, tho.
Stefan
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2015-11-18 13:40 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-11-18 10:32 Understanding macroexp.el Michael Heerdegen
2015-11-18 11:37 ` Andreas Schwab
2015-11-18 12:03 ` Michael Heerdegen
2015-11-18 13:36 ` Stefan Monnier
2015-11-18 13:40 ` Stefan Monnier
Code repositories for project(s) associated with this public inbox
https://git.savannah.gnu.org/cgit/emacs.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).