Automatic report from sources (radiusd) between 05.12.2007 - 06.12.2007 GMT

Automatic cvs log generator cparker at
Thu Dec 6 10:01:30 CET 2007

CVS log entries from 05.12.2007 (Wed) 09:00:00 - 06.12.2007 (Thu) 09:00:01 GMT
Summary by authors
Author: aland
	File: radiusd/src/modules/rlm_pap/rlm_pap.c; Revisions: 1.60
	File: radiusd/src/main/event.c; Revisions: 1.74
	File: radiusd/src/modules/rlm_preprocess/rlm_preprocess.c; Revisions: 1.79
	File: radiusd/src/lib/valuepair.c; Revisions: 1.129
	File: radiusd/src/include/libradius.h; Revisions: 1.147

Combined list of identical log entries
	Remove 'caseless' from VALUE_PAIR flags.  It's not needed.

	Added 'unknown_attr' to VALUE_PAIR flags, which tracks if
	vp->name points to a DICT_ATTR entry name or not.

	vp->name is now a pointer, rather than a character array.

	Updated code to have "vp->name = da->name" for known attributes.
	Otherwise, the memory allocated for the VALUE_PAIR is increased
	by ~24 characters.  The name is printed there (Vendor-X-Attr-Y),
	and vp->name is pointed to the string.

	Updated paircopy() to look at vp->flags.unknown_attr,
	if set, it allocates more room for the name, and does
	a memcpy() of the VALUE_PAIR + the name.

	Updated rlm_preprocess to NOT print to vp->name.

	Nothing else in the code should now write to vp->name

	Updated paircreate() to simplify printing of Vendor-X-Attr-Y

	Updated pairmake_any() to simplify parsing of Vendor-X-Attr-Y.
	It now also checks size of attribute values (e.g. 1-octet,
	2-octet, etc).  It now parses the octet string as an octet
	string, no matter what the final type is.  So you can
	have "Attr-5  = 0x00000001", and have it show up as
	"NAS-Port = 1".
Modified files:
	File: radiusd/src/include/libradius.h; Revision: 1.147;
	Date: 2007/12/05 10:22:41; Author: aland; Lines: (+4 -4)
	File: radiusd/src/lib/valuepair.c; Revision: 1.129;
	Date: 2007/12/05 10:22:41; Author: aland; Lines: (+201 -88)
	File: radiusd/src/modules/rlm_preprocess/rlm_preprocess.c; Revision: 1.79;
	Date: 2007/12/05 10:22:41; Author: aland; Lines: (+10 -5)
Log entries
	Delay a little longer in wait_a_bit.  It can't hurt.

	rcode is ssize_t, not size_t
Modified files:
	File: radiusd/src/main/event.c; Revision: 1.74;
	Date: 2007/12/06 08:53:11; Author: aland; Lines:  (+5 -5)
	Signed / unsigned issues

	And oracle LM/NT password headers, as posted to the list
Modified files:
	File: radiusd/src/modules/rlm_pap/rlm_pap.c; Revision: 1.60;
	Date: 2007/12/05 14:44:22; Author: aland; Lines:  (+18 -16)
Summary of modified files
File: radiusd/src/include/libradius.h
Revisions: 1.147
Authors: aland (+4 -4)
File: radiusd/src/lib/valuepair.c
Revisions: 1.129
Authors: aland (+201 -88)
File: radiusd/src/main/event.c
Revisions: 1.74
Authors: aland (+5 -5)
File: radiusd/src/modules/rlm_pap/rlm_pap.c
Revisions: 1.60
Authors: aland (+18 -16)
File: radiusd/src/modules/rlm_preprocess/rlm_preprocess.c
Revisions: 1.79
Authors: aland (+10 -5)
Automatic cron job from /web/pages/

More information about the Freeradius-Devel mailing list