[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: commit: ldap/servers/slapd acl.c aclparse.c slap.h main.c

Pierangelo Masarati wrote:

Howard Chu wrote:

With a fresh update I get this error when compiling:
cc -g -I../../include -I../../../head/servers/slapd/slapi -I.
-I../../../head/include       -c -o acl.o
../../../head/servers/slapd/acl.c: In function `dynacl_aci_parse':
../../../head/servers/slapd/acl.c:2671: error: structure has no member
named `si_ad_aci'
../../../head/servers/slapd/acl.c:2674: error: `SLAPD_ACI_SYNTAX'
undeclared (first use in this function)
../../../head/servers/slapd/acl.c:2674: error: (Each undeclared
identifier is reported only once
../../../head/servers/slapd/acl.c:2674: error: for each function it
appears in.)
make[2]: *** [acl.o] Error 1

sorry about that. As soon as it looks fine and acceptable, I plan to
enable this by default and move ACI stuff in a conditionally compiled
aci.c file so that also ACIs can be dynamically loaded. p.

Well, there was a #ifdef LDAP_DEVEL / #define SLAP_DYNACL in slap.h, so the new code was being compiled without all of the required definitions being present. I've commented this out in my copy of slap.h.

One thing that currently makes me unhappy with the dynacl stuff is that
ACL caching is not passed thru to the dynamic ACL infrastructure.  This is
because I quickly designed it based on ACIs, whose ACL caching was broken
and thus disabled.  However, I'm not sure the idea of ACL caching can work
at all with the current design of ACIs (and dynamic ACLs, which is
essentially the same).

I agree, it's not clear how to make it work, although it would certainly be desirable.

 -- Howard Chu
 Chief Architect, Symas Corp.       Director, Highland Sun
 http://www.symas.com               http://highlandsun.com/hyc
 Symas: Premier OpenSource Development and Support