Re: iterating over arguments

[Sam Steingold]

> * Sam Steingold <address@hidden> [2009-09-12 23:57:51 -0400]:
>>
>> AC_DEFUN([CL_CLISP_REQUIRE_FEATURE],
>> [m4_foreach_w([cl_feat], [$1],
>> [m4_pushdef([CL_FEAT], m4_toupper(cl_feat))dnl
>> AC_CACHE_CHECK([for CL_FEAT in CLISP], [cl_cv_clisp_]cl_feat,
>>   [CLISP_SET([cl_cv_clisp_]cl_feat,
>>     [[#+]]cl_feat[[ "yes" #-]]cl_feat[[ "no"]])])
>> test $cl_cv_clisp_]cl_feat[ = no && AC_MSG_ERROR(
>>   CL_FEAT[ is missing in CLISP])
>> m4_popdef([CL_FEAT])])])
>
> m4 on this also seems to loop forever in aclocal.
> (just like in 
> http://article.gmane.org/gmane.comp.sysutils.autoconf.general:12051)
> would you like an strace?

replacing "m4_toupper" with "upcase" resulted in a "good" configure with,
e.g., "upcase(ffi)" instead of "FFI".
replacing it with "m4_upcase" resulted in

configure:4249: error: possibly undefined macro: m4_upcase
      If this token and others are legitimate, please use m4_pattern_allow.
      See the Autoconf documentation.

so, it looks like if I sacrifice the aesthetics of upcase features, I am
good to go with

AC_DEFUN([CL_CLISP_REQUIRE_FEATURES],
[m4_foreach_w([cl_feat], [$1],
[AC_CACHE_CHECK([for cl_feat in CLISP], [cl_cv_clisp_]cl_feat,
 [CLISP_SET([cl_cv_clisp_]cl_feat,[[#+]]cl_feat[[ "yes" #-]]cl_feat[[ "no"]])])
test $cl_cv_clisp_]cl_feat[ = no && AC_MSG_ERROR(cl_feat[ is missing in 
CLISP])])])

however, it would be nice to have the right case...

-- 
Sam Steingold (http://sds.podval.org/) on Ubuntu 9.04 (jaunty)
http://www.PetitionOnline.com/tap12009/ http://camera.org
http://memri.org http://dhimmi.com http://thereligionofpeace.com
NY survival guide: when crossing a street, mind cars, not streetlights.