Milter-greylist

http://ftp.espci.fr/pub/milter-greylist/milter-greylist-4.1.11.tgz
MD5 (milter-greylist-4.1.11.tgz) = 642317c9e686293e676bb5cddd1d0def

Changes since 4.1.10:
        Evaluate list clases in forward order (Rudy Eschauzier)
        Fix crashes on regex substring matches (Rudy Eschauzier)

Anyone has minot changes to check in? IMO this is time to fork the 4.2
stable branch.

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

hmmm... seems like there is a typo somewhere:

./configure --enable-spamassassin --enable-dnsrbl --with-libGeoIP=/usr/ 
local --w
ith-libcurl

[root@vps milter-greylist-4.1.11]# make
gcc -g -O2 -Wall -Iyes/include -DUSE_CURL -I/usr/local/include - 
DUSE_GEOIP -DUSE_SPAMD -D_XOPEN_SOURCE=500 -DUSE_DNSRBL -D_BSD_SOURCE - 
I. -I.    -c -o milter-greylist.o milter-greylist.c
In file included from milter-greylist.c:85:
list.h:42: error: syntax error before "STAILQ_ENTRY"
list.h:42: warning: no semicolon at end of struct or union
list.h:49: error: syntax error before "STAILQ_HEAD"
list.h:49: warning: no semicolon at end of struct or union
make: *** [milter-greylist.o] Error 1


Linux vps.pugetworks.com 2.6.18-028stab059.6 #1 SMP Fri Nov 14  
14:01:22 MSK 2008 i686 i686 i386 GNU/Linux

On Feb 8, 2009, at 5:12 AM, manu@... wrote:

> http://ftp.espci.fr/pub/milter-greylist/milter-greylist-4.1.11.tgz
> MD5 (milter-greylist-4.1.11.tgz) = 642317c9e686293e676bb5cddd1d0def
>
> Changes since 4.1.10:
>        Evaluate list clases in forward order (Rudy Eschauzier)
>        Fix crashes on regex substring matches (Rudy Eschauzier)
>
> Anyone has minot changes to check in? IMO this is time to fork the 4.2
> stable branch.
>
> -- 
> Emmanuel Dreyfus
> http://hcpnet.free.fr/pubz
> manu@...
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>

Bill Levering <idbill@...> wrote:

> In file included from milter-greylist.c:85:
> list.h:42: error: syntax error before "STAILQ_ENTRY"
> list.h:42: warning: no semicolon at end of struct or union
> list.h:49: error: syntax error before "STAILQ_HEAD"
> list.h:49: warning: no semicolon at end of struct or union
> make: *** [milter-greylist.o] Error 1

Rudy introduced STAILQ_* macros, which seems to be a fairly new beast. I
fixed it in CVS. Can you check out the latest code and check if it
builds? You will have to run configure again.

If you confirm it builds, I will do 4.1.12

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

hmmm...

gcc -g -O2 -Wall -Iyes/include -DUSE_CURL -I/usr/local/include - 
DUSE_GEOIP -DUSE_SPAMD -D_XOPEN_SOURCE=500 -DUSE_DNSRBL -D_BSD_SOURCE - 
I. -I.    -c -o milter-greylist.o milter-greylist.c
In file included from dump.h:36,
                  from milter-greylist.c:82:
queue.h:37:22: sys/null.h: No such file or directory
make: *** [milter-greylist.o] Error 1

On Feb 8, 2009, at 11:08 AM, manu@... wrote:

> Bill Levering <idbill@...> wrote:
>
>> In file included from milter-greylist.c:85:
>> list.h:42: error: syntax error before "STAILQ_ENTRY"
>> list.h:42: warning: no semicolon at end of struct or union
>> list.h:49: error: syntax error before "STAILQ_HEAD"
>> list.h:49: warning: no semicolon at end of struct or union
>> make: *** [milter-greylist.o] Error 1
>
> Rudy introduced STAILQ_* macros, which seems to be a fairly new  
> beast. I
> fixed it in CVS. Can you check out the latest code and check if it
> builds? You will have to run configure again.
>
> If you confirm it builds, I will do 4.1.12
>
> -- 
> Emmanuel Dreyfus
> http://hcpnet.free.fr/pubz
> manu@...
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>

> 
> Rudy introduced STAILQ_* macros, which seems to be a fairly new beast. I
> fixed it in CVS. Can you check out the latest code and check if it
> builds? You will have to run configure again.
> 

The queue.h file on my system looks like

http://gel.sourceforge.net/examples/queue_8h-source.php

stailq is the single-linked list version of tailq (double linked).

When changing the code from stailq to tailq, you'll need to take
special care of the REMOVE macro, wich has a different number of
parameters between the two:

STAILQ_REMOVE(STAILQ_HEAD *head, TYPE *elm, TYPE, STAILQ_ENTRY NAME);

and 

TAILQ_REMOVE(TAILQ_HEAD *head, TYPE *elm, TAILQ_ENTRY NAME);

i.e., you'll need to remove the third parameter in the call.

Rudy.

A patch to change from stailq to tailq:

http://files.eschauzier.org/milter-greylist/sqtail-4.1.11.patch

Bill Levering <idbill@...> wrote:

> gcc -g -O2 -Wall -Iyes/include -DUSE_CURL -I/usr/local/include - 
> DUSE_GEOIP -DUSE_SPAMD -D_XOPEN_SOURCE=500 -DUSE_DNSRBL -D_BSD_SOURCE -
> I. -I.    -c -o milter-greylist.o milter-greylist.c
> In file included from dump.h:36,
>                   from milter-greylist.c:82:
> queue.h:37:22: sys/null.h: No such file or directory
> make: *** [milter-greylist.o] Error 1

Ok, let's roll back this one and go with TAILQ_*, as Rudy just proposed.

Do you have more success now (with latest CVS)?

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

Ok, the latest cvs compiles without a problem.

Bill

On Feb 8, 2009, at 12:56 PM, manu@... wrote:

> Bill Levering <idbill@...> wrote:
>
>> gcc -g -O2 -Wall -Iyes/include -DUSE_CURL -I/usr/local/include -
>> DUSE_GEOIP -DUSE_SPAMD -D_XOPEN_SOURCE=500 -DUSE_DNSRBL - 
>> D_BSD_SOURCE -
>> I. -I.    -c -o milter-greylist.o milter-greylist.c
>> In file included from dump.h:36,
>>                  from milter-greylist.c:82:
>> queue.h:37:22: sys/null.h: No such file or directory
>> make: *** [milter-greylist.o] Error 1
>
> Ok, let's roll back this one and go with TAILQ_*, as Rudy just  
> proposed.
>
> Do you have more success now (with latest CVS)?
>
> -- 
> Emmanuel Dreyfus
> http://hcpnet.free.fr/pubz
> manu@...
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>

And here is milter-greylist-4.1.12, which is just 4.1.11 with a build
fix

http://ftp.espci.fr/pub/milter-greylist/milter-greylist-4.1.12.tgz
MD5 (milter-greylist-4.1.12.tgz) = 062c1108b9ed4907c971146d8350687f

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

I have to agree (4.1 should become the standard install), I upgraded  
to 4.1 because there are some very desirable features not available in  
4.0.

Bill

On Feb 9, 2009, at 12:32 PM, Greg Troxel wrote:

>
>  Changes since 4.1.10:
>          Evaluate list clases in forward order (Rudy Eschauzier)
>          Fix crashes on regex substring matches (Rudy Eschauzier)
>
> Are there plans for 4.2.0?  It seems 4.1.x is getting enough new
> features that people should be running it instead of 4.0.x, and that's
> setting off my 'release is overdue' detector.
>
> Or am I totally confused and 4.1.11 is considered a stable release and
> recommended to all users?  I am guessing not because pkgsrc is at  
> 4.0.1.
>
>

Greg Troxel <gdt@...> wrote:

> Are there plans for 4.2.0?  It seems 4.1.x is getting enough new
> features that people should be running it instead of 4.0.x, and that's
> setting off my 'release is overdue' detector.

Yes, I plan to branch this week.

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