File: TODO

package info (click to toggle)
ircd 2.10.02-1
  • links: PTS
  • area: main
  • in suites: hamm
  • size: 2,228 kB
  • ctags: 2,087
  • sloc: ansic: 29,122; makefile: 664; sh: 307; perl: 18
file content (55 lines) | stat: -rw-r--r-- 2,476 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
Last updated: Sun Dec 07 18:58:16 1997

What you can do (but inform me before you start):

* Fix the bug that ISON now responds with 'No such nick' for every nick :/
* Allow to specify the interface to bind to by IP-number
  (instead of using a DNS lookup on the M: line name).
* Find and fix the 'shutdown(8, 2)' bug (I didn't get a response yet
    from the guys that reported this, on how we can reproduce this,
    so I think we'll have to drop it again).
* Fix HACK wallops (detailed report received from Xorath; done by Run/Xorath)
* Add MODE +c and +r and multiple op-levels (being done by Run (damn hard))
* Disallow MODE from server before EOB has been received (Run).
* NT port (being done by Bleep)
* Split up K-lines in multiple hashed lists (being done by Nemesi)
* Implement /kline, /unkline (after Nemesi finished the above)

More discussion:

* Implementation of /PROTOCOL
  
  The idea of this is to come to an accepted standard (by at least
    IRCnet, efnet, Undernet and dalnet), that will allow to introduce
    (major) client protocol changes.
  The goal in these cases will be to remove the support for the old
    protocol after a couple of years.

* Priority queues

  The goal of this project is to come to a server-server and server-client
    'flood control' that charges (also) the receiver of messages (being
    either server query replies or channel messages) and automatically
    divides the available bandwidth over the servers and local clients in
    a way that everyone gets an equal share, no send queue is building up
    into the extreme (avoiding severe lag), and the available bandwidth is
    used as much as possible at all times.
  Note that we are FAR from an implementation here, a lot of basic thinking
    and design needs to be done first.

Done:

* write lgline3.patch
* Fix IP-number returned by /WHO
* Suppress 'make' warning about "No suffixes" (did this work?)
* Fix /stats w 'Day   Yest. YYest.' output
* Fix UPING output (sprintf_irc "%lu" actually prints "%09lu")
* Fix umode buffer bug in gline8-9.patch
* Check switch statement in res_mkquery.c (removed file completely)
* Propagation of "/stats M target time time", time parameters
* Implement /CPRIVMSG & /CNOTICE
* Disallow an accidential KILL for +k services
* Add UNLIMIT_OPER_QUERY to config/config-sh.in and doc/Configure.help
* Fix /USERIP nonick[,ormultiple]
* Suppress the error on machines that have no 'set -h' (hopefully)
* Fix the a_kills_b_too() bug