X-Git-Url: http://git.meshlink.io/?a=blobdiff_plain;f=todo;h=5d2346dd0319b6a19d3cc77f78fa884b0a8af646;hb=1e4fbfdf6b633cc3e3af9ba01b7baaa6994ffd38;hp=5576c7e8973eed679398410581aaacdf637c84f2;hpb=07b93a6d49a3252e3f5cacadd15083949241b071;p=catta

diff --git a/todo b/todo
index 5576c7e..5d2346d 100644
--- a/todo
+++ b/todo
@@ -1,12 +1,50 @@
 todo:
-* Unicast responses/queries
-* Known-Answer suppression
-* Truncation
-* Probing/Conflict resolution
-* Legacy unicast
-* uniqueness
-* defend our entries on incoming goodbye
+* finish DBUS stuff:
+	- wrap missing objects
+	- introspection
+* release!
+
+later:
+* support for special domain PTR records based on local IP subnet address
+* Changes resulting in updated RFC of 7th June 2005:
+  * Defer responses to replies with TC bit set by 400-500msec
+  * Defer unicast responses the same way as multicast responses 
 
 done:
+* Probing/Conflict resolution
+* uniqueness
+* respect escaping in name serialization
 * really send goodbye packets
-
+* refresh subscribed records only
+* FLX_DNS_TYPE_ANY support
+* Known-Answer suppression client part
+* Known-Answer suppression server part
+* make flx_server_add_text() and flx_server_add_service() variadic functions
+* name compression
+* remove expression "rrset" from source files
+* defend our entries on incoming goodbye
+* allow NULL bytes in TXT records
+* add flx_server_add_service_strlst() and friends
+* change flx_* to avahi_*
+* Unicast responses/queries
+* Legacy unicast
+* no flush bit in known answer
+* always set AA
+* check: TC bit is valid for queries ONLY
+* add SRV and TXT records referenced from PTR records automatically to packet
+* add A and AAAA records referenced from SRV records automatically to packet
+* support known answer suppresion for incoming unicast queries
+* check wether RRsets are supported correctly (i.e. that all records of an
+  RRset are really sent if it is requested) (rfc 2181)
+* case insensitve comparison
+* drop records from cache only one second after flush cache bit entry was received
+* either send entire RRSET or don't set flush cache bit!
+* mantain flush cache bit correctly in psched
+* Return to probing state on conflict
+* response job dependencies
+* enlarge packet in case a record/query is too large to fit in a normal packet
+* reflector
+* test against apple test suite
+* sensible logging
+* c++ support
+* drop trailing dot on avahi_normalize_name()