William Pitcock
3ad21d7513
hurt: unbreak compile
2010-02-17 06:44:42 -06:00
William Pitcock
ff31db8473
Add support for client certificate fingerprints in o:lines.
2010-02-17 06:41:41 -06:00
Jilles Tjoelker
5155c03c96
Update target change documentation.
2010-02-17 00:38:08 +01:00
Jilles Tjoelker
1c5683dede
Update NEWS.
2010-02-17 00:24:12 +01:00
Jilles Tjoelker
0cce7774d5
Subject /invite to umodes +g and +R.
...
The behaviour is the same as /msg except that where
/msg would send RPL_UMODEGMSG to the user, the /invite
is instead let through. This counts as a notification
for caller_id_wait like RPL_UMODEGMSG.
Checks are on the target user's server, which means an
error message will appear after RPL_INVITING.
This must be because the accept list is not globally
known.
2010-02-15 23:07:14 +01:00
Jilles Tjoelker
40b79a3908
Add missing return value.
2010-02-15 22:18:00 +01:00
Jilles Tjoelker
890af0e7c5
Subject /invite to target change.
...
Similar to /msg, inviting a user that is not in a channel
you have op or voice in requires a free target; opers always
have a free target.
Being invited adds the source as a reply target.
2010-02-15 22:08:55 +01:00
Jilles Tjoelker
4f2685f3e1
Move target change code to src/tgchange.c,
...
so we can use it for /invite as well.
2010-02-15 21:58:34 +01:00
Jilles Tjoelker
c9f01c4f2f
target change: Allow free replies.
...
When a user receives a private message, notice or RPL_UMODEGMSG,
add the source to a special set of 5 target slots.
These slots are checked in the normal way when sending messages,
allowing a reply without using up a free target.
This feature will not be very useful if a user is being messaged
by many different users; to help this, messages blocked entirely
by +g or +R do not affect the targets. CTCP replies also remain
free in terms of targets.
2010-02-15 01:07:07 +01:00
Jilles Tjoelker
179becdf5f
target change: Overwrite the least recently used target with a new one.
2010-02-15 00:31:17 +01:00
Jilles Tjoelker
ad1d39a76f
Make the number of targets tracked for target change a #define.
2010-02-13 15:18:17 +01:00
JD Horelick
97deedc42f
English fix to fix the fix i made about 80 minutes ago.
...
I suck at a/an cometimes. Also, fix.
2010-02-08 18:04:54 -05:00
JD Horelick
bc4fccbc44
One or 2 English language cleanups. Less than I thought there were.
2010-02-08 16:28:08 -05:00
Jilles Tjoelker
033d16bd2d
Add new function rb_get_ssl_certfp() to export-syms.txt.
2010-02-07 20:38:37 +01:00
Jilles Tjoelker
96bfafc1f8
testline/testgecos: Send "not an oper" message if appropriate.
2010-02-07 17:57:36 +01:00
Jilles Tjoelker
8eda114a78
Pass certfp to other servers and show it in whois. Do not show it on connect.
...
The server protocol for this is
:<uid> ENCAP * CERTFP :<40 hex chars>
both in new user introductions and in burst.
As in oftc-hybrid, only the user themselves and opers can see the certfp.
Displaying the certfp on connect seems unnecessary to me,
the user can whois themselves if needed.
2010-02-06 00:18:27 +01:00
Jilles Tjoelker
62a360ea22
Fix serno.h for compiles from Mercurial.
2010-02-03 00:46:14 +01:00
Jilles Tjoelker
17e4b48b55
Apply ratbox3 workaround for too permissive inet_pton6():
...
Any kline (or various other things) containing * or ? is
a mask kline and not an IP kline. Ideally, rb_inet_pton_sock()
would return failure for those, but in practice this is not
always the case for IPv6.
Such a kline that is erroneously treated as an IP line
likely matches way fewer IPs than expected.
2010-02-03 00:44:51 +01:00
William Pitcock
11dff8a423
Clarify CertFP message.
2010-01-31 17:35:15 -06:00
Jilles Tjoelker
7247337afa
Add certfp support to libratbox and ssld.
...
This lets a user connect with a client certificate, and
passes the certificate's fingerprint to ircd, which
currently just notices it to the user.
A new ssld->ircd message 'F' is used to pass on the
fingerprint.
This is only for OpenSSL for now, not GNUTLS.
2010-01-31 19:04:20 +01:00
William Pitcock
47f9d6002a
Automated merge with ssh://hg.atheme.org//hg/charybdis
2010-01-30 19:21:28 -06:00
William Pitcock
7d778d510a
modreload: Causing an entire 50,000 user network to crash due to user-error is a bad idea, don't let it happen again by not allowing paths in /modreload.
2010-01-30 19:21:17 -06:00
Jilles Tjoelker
5f2df25109
Slightly tighten auth{} duplicate check:
...
If the existing auth{} block has no auth_user, it will
trump any auth{} block for the same u@h.
2010-01-31 01:40:43 +01:00
Jilles Tjoelker
3d1f32c07a
Take auth_user into account when detecting redundant/duplicate auth{}.
...
from ircd-seven (spb)
2010-01-31 01:36:37 +01:00
Jilles Tjoelker
90afc118fd
No need to be paranoid about clearing auth_user.
...
It is a username, not a password.
2010-01-31 01:30:25 +01:00
Jilles Tjoelker
fcb057992c
Fix memory leak with auth_user.
...
from ircd-seven (spb)
2010-01-31 01:29:08 +01:00
Jilles Tjoelker
8f10356217
Remove remnants of <foo>_t typedefs.
2010-01-30 15:57:10 +01:00
Jilles Tjoelker
114105b429
Fix /links buffer overflow.
2010-01-27 21:05:10 +01:00
Jilles Tjoelker
1fd171a547
Fix op-moderate (cmode +z) for channel names with '@'.
2010-01-24 19:37:00 +01:00
Jilles Tjoelker
cd5d9abf63
Do not read in help files starting with a dot (including dot and dot-dot).
2010-01-22 02:02:45 +01:00
Jilles Tjoelker
0b5ea4f865
Fix include/serno.h generation.
...
Apparently #value# is no longer special in hg templates.
2010-01-22 01:41:48 +01:00
Jilles Tjoelker
dcb22e07b9
Fix memory leak and bad error reporting with posix_spawn():
...
* an initialized posix_spawnattr_t must be destroyed
* posix_spawn() returns an error number instead of setting errno
libratbox trunk r26730
2010-01-22 00:09:56 +01:00
Stephen Bennett
d06f3da955
Don't cache files that are empty or not files. Avoids crashing on the likes of 'HELP .'
2010-01-21 16:27:34 +00:00
Jilles Tjoelker
641eb2c3c8
Put back fb7d6089158e, not setting large_ctcp_sent for CTCP ACTION.
2010-01-20 00:03:57 +01:00
William Pitcock
2e918bf515
Merge +C (no CTCP to channels) from ircd-seven.
2010-01-19 02:11:04 -06:00
William Pitcock
fdb90316e1
Automated merge with ssh://hg.atheme.org//hg/charybdis
2010-01-19 01:47:27 -06:00
William Pitcock
bde6442c47
supported: Add CLIENTVER=3.0 to indicate presence of 3.0 core capabilities (such as CAP, extended WHO, etc) as requested by LifeIsPain.
2010-01-19 01:47:21 -06:00
Jilles Tjoelker
3d0bbdcbe1
Do not set large_ctcp_sent for CTCP ACTION as it does not request a reply.
2010-01-14 01:12:16 +01:00
Jilles Tjoelker
a990586f4d
If a removed resv was temporary, mention this in the server notice.
...
This agrees with xline.
2010-01-09 22:14:53 +01:00
Jilles Tjoelker
1a9ea263fa
Add back unresv (permanent resv) notices.
2010-01-09 22:12:06 +01:00
Jilles Tjoelker
43d4d72ca9
Put back resv_forcepart.
...
This undoes erroneous revert in a3c064b3b8a2.
2010-01-09 19:08:48 +01:00
Jilles Tjoelker
70ea02ebd6
Complete the move of xlines and resvs from aconf->name to aconf->host.
2010-01-08 18:46:29 +01:00
Jilles Tjoelker
f67e7283f9
bantool: change empty oper fields to "unknown".
...
Otherwise they would cause a crash when ircd tries to load them.
2010-01-08 18:17:16 +01:00
Jilles Tjoelker
4418166c7e
Do not change \s in xlines to spaces, match_esc() treats \s correctly.
...
This undoes erroneous revert in a3c064b3b8a2.
2010-01-08 18:13:47 +01:00
Jilles Tjoelker
407e83c192
Fix build: we need -DSQLITE_THREADSAFE=0 -DSQLITE_OMIT_LOAD_EXTENSION=1
...
for the sqlite sources.
2010-01-08 16:02:30 +01:00
William Pitcock
00c036b172
Add back missing notices.
2010-01-07 18:17:08 -06:00
William Pitcock
23959371d0
Fix up more things with bandb.
2010-01-07 18:14:15 -06:00
William Pitcock
80c9ac516e
Load in new bans from bandb on startup.
2010-01-07 18:06:15 -06:00
William Pitcock
d4c273075e
Make this work again.
2010-01-07 17:50:49 -06:00
William Pitcock
f36d4fdd54
Make this work again.
2010-01-07 17:49:08 -06:00
