Hi,
On 12 Nov 2001, Akim Demaille wrote:
> | Forthermore I believe the definition of m4_car is wrong. If I test
> | m4_foreach with the example in the explanation in m4sugar.m4... :
> |
> | m4_define(a, 1)dnl
> | m4_define(b, 2)dnl
> | m4_define(c, 3)dnl
> | m4_foreach([f], m4_split([a (b c)]), [echo f
> | ])dnl
> |
> | ... the output is:
> | echo 1
> | echo (2
> | echo 3)
>
> Because split has changed too.
Hmm, no. I copy&pasted the exact definition out of m4sugar.m4 (of
m4_foreach and all used macros) into a new .m4 file, added the above
test-commands, and that was the output, so I used the unchanged def of
m4_foreach and everything (of course I also used that file as my
playground for the other tests, so may be I screwed something, but I
believe not). Anyway...
> I have one question though: did you run the test suite?
... I now did with autoconf-2.52.
> How does your proposal behave?
With the below patch applied directly to autoconf-2.52 this is the result:
=============================
All 155 tests were successful
=============================
;-)
Ciao,
Michael.
--
2001-11-11 Michael Matz <[EMAIL PROTECTED]>
* m4sugar.m4 (_m4_foreach): Make it linear instead quadratic.
(m4_car2): New.
(m4_car): Properly quote arguments.
diff -urp autoconf-2.52/m4sugar.m4 autoconf-2.52-matz/m4sugar.m4
--- autoconf-2.52/m4sugar.m4 Wed Jul 18 13:11:14 2001
+++ autoconf-2.52-matz/m4sugar.m4 Tue Nov 13 02:58:01 2001
@@ -671,11 +671,12 @@ m4_define([m4_foreach],
[m4_pushdef([$1])_m4_foreach($@)m4_popdef([$1])])
# Low level macros used to define m4_foreach.
-m4_define([m4_car], [$1])
+m4_define([m4_car], [[$1]])
+m4_define([m4_car2], [[$@]])
m4_define([_m4_foreach],
[m4_if(m4_quote($2), [], [],
[m4_define([$1], [m4_car($2)])$3[]_m4_foreach([$1],
- [m4_shift($2)],
+ m4_car2(m4_shift($2)),
[$3])])])
