Milter-greylist

Here is milter-greylist 4.0 alpha1

http://ftp.espci.fr/pub/milter-greylist/milter-greylist-4.0a1.tgz
MD5 (milter-greylist-4.0a1.tgz) = ef5d79fd31238e6d71410fa50f781730

There have been many feature addition since 3.0 to make milter-greylist
much more flexible. You can now use milter-greylist as a content filter,
interract with an external source of configuration (a simple CGI, a LDAP
directory) to build per-user setups. You can build time-based or
sender-location-based filters...

-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
manu@...

> Here is milter-greylist 4.0 alpha1
> 
> http://ftp.espci.fr/pub/milter-greylist/milter-greylist-4.0a1.tgz
> MD5 (milter-greylist-4.0a1.tgz) = ef5d79fd31238e6d71410fa50f781730
> 

Hello,

I have some name space poisoning on Tru64 UNIX, sorry about that:

cc -I/usr/local/src/sendmail/libmilter/include -pthread -D_XOPEN_SOURCE=500 -D_O
SF_SOURCE -D_BSD_SOURCE -c conf_yacc.c
cc: Error: /usr/include/arpa/nameser_compat.h, line 154: Invalid declarator. (de
clarator)
} HEADER;
--^

I attach two patches which fixes this platform specific problem
via replacing all the occurrences of the HEADER identifier with GLHEADER
in conf_lex.l and in conf_yacc.y .
Is there a way to include these patches into the main source distribution stream?

Bests,
Attila

> Here is milter-greylist 4.0 alpha1
> 
> http://ftp.espci.fr/pub/milter-greylist/milter-greylist-4.0a1.tgz
> MD5 (milter-greylist-4.0a1.tgz) = ef5d79fd31238e6d71410fa50f781730
> 

Hello,

My compiler on Tru64 UNIX is a bit picky about the proper type utilization:

cc -I/usr/local/src/sendmail/libmilter/include -pthread -D_XOPEN_SOURCE=500 -D_O
SF_SOURCE -D_BSD_SOURCE -c acl.c
cc: Error: acl.c, line 129: In the initializer for acl_clause_rec[0].acr_print,
"*acl_print_list" is not constant, but occurs in a context that requires a const
ant expression. (needconstexpr)
          *acl_print_list, *acl_add_list,
----------^

It looks like those extra asterisks are not needed.
I attach a patch hoping that it does break anything on other platforms.
If it doesn't, is it possible to include it into the main source distribution stream?

Bests,
Attila

> I attach a patch hoping that it does break anything on other 
> platforms.

"hoping that is does NOT break"
Sorry for the typo.

<attila.bruncsak@...> wrote:

> It looks like those extra asterisks are not needed. attach a patch hoping
> Ithat it does break anything on other platforms. f it doesn't, is it
> Ipossible to include it into the main source distribution stream?

According to the K&R's book on C language, the asterisks should not be
used, so it should not be a problem. Anyone see a reason why the
asterisks should be kept?

-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
manu@...