]>
jfr.im git - irc/atheme/atheme.git/log
William Pitcock [Wed, 6 May 2015 10:50:44 +0000 (05:50 -0500)]
atheme 7.2.6.
Aaron Jones [Tue, 5 May 2015 17:25:42 +0000 (17:25 +0000)]
pbkdf2v2: Save CPU cycles in the verifying case
libathemecore has been verified to handle NULL return from crypt
providers correctly - so start returning NULL in pbkdf2v2 in the
case of trying to verify hashes that were not produced by itself
Mantas Mikulėnas [Sun, 26 Apr 2015 00:25:27 +0000 (03:25 +0300)]
contrib: update
Mantas Mikulėnas [Sun, 26 Apr 2015 00:15:09 +0000 (03:15 +0300)]
core/explicit_bzero: fix the type… again
Mantas Mikulėnas [Sun, 26 Apr 2015 00:11:08 +0000 (03:11 +0300)]
core/ptasks: remove another redundant srand()
No need to feed it not-very-random junk after init already called
srand(arc4random()).
William Pitcock [Sun, 26 Apr 2015 00:09:28 +0000 (19:09 -0500)]
Merge pull request #464 from grawity/arc4random-openssh
Update internal arc4random to a more recent version
Mantas Mikulėnas [Sun, 26 Apr 2015 00:03:52 +0000 (03:03 +0300)]
arc4random: adapt new code to libathemecore
s/bzero/memset/ as the former is no longer in POSIX.1
Mantas Mikulėnas [Sat, 25 Apr 2015 23:55:20 +0000 (02:55 +0300)]
arc4random: add new code to buildsys & headers
Mantas Mikulėnas [Sat, 25 Apr 2015 23:53:10 +0000 (02:53 +0300)]
import arc4random & explicit_bzero from openssh-portable
Mantas Mikulėnas [Sat, 25 Apr 2015 22:56:11 +0000 (01:56 +0300)]
crypto/pbkdf2v2: use arc4random()
Mantas Mikulėnas [Sat, 25 Apr 2015 22:54:32 +0000 (01:54 +0300)]
core: don't reseed every 5 minutes
If the RNG was crap, feeding it |cnt| which isn't even particularly
random wouldn't do much good...
Mantas Mikulėnas [Sat, 25 Apr 2015 22:51:52 +0000 (01:51 +0300)]
crypto/pbkdf2v2: don't srand(time)
atheme_init already does srand(arc4random()).
Aaron Jones [Fri, 24 Apr 2015 05:58:50 +0000 (05:58 +0000)]
Use new contrib module tree
Aaron Jones [Fri, 24 Apr 2015 05:42:04 +0000 (05:42 +0000)]
Don't send duplicate K-Lines for a user (if they induce more than 1)
Aaron Jones [Tue, 7 Apr 2015 00:09:04 +0000 (00:09 +0000)]
Avoid possible NULL dereferences in language code
William Pitcock [Sun, 5 Apr 2015 09:43:55 +0000 (04:43 -0500)]
protocol/inspircd: properly recognize CSTATUS_IMMUNE (+Y)
William Pitcock [Sun, 5 Apr 2015 09:40:34 +0000 (04:40 -0500)]
libathemecore: try_kick(): add support for inspircd-style per-user kick immunity the right way
Jeff Horelick [Wed, 1 Apr 2015 23:12:29 +0000 (19:12 -0400)]
Add pointless whitespace to README to test git commit hook
William Pitcock [Mon, 30 Mar 2015 23:09:38 +0000 (18:09 -0500)]
Merge pull request #461 from errietta/errietta/multimark-oldmark-restoration
Migrate any restored old-style marks at user creation.
Errietta Kostala [Mon, 30 Mar 2015 22:19:32 +0000 (23:19 +0100)]
Migrate any restored old-style marks at user creation.
Mantas Mikulėnas [Mon, 30 Mar 2015 06:19:48 +0000 (09:19 +0300)]
protocol/unreal: bandaid fix for sending numerics to 3.2
The module requires having a SID configured, so numeric_sts(me.me, …)
always sends it, breaking quite a lot of stuff.
-> :grawity MOTD :chanserv
<- :00A 375 grawity :- services.cluenet.org Message of the Day -
-> :virgule.cluenet.org KILL 00A :virgule.cluenet.org (00A(?) <- services.cluenet.org)
In the long term, I think it would be better to have two protocol
modules, unreal34 with UIDs/SIDs and unreal32 without.
William Pitcock [Wed, 25 Mar 2015 04:06:38 +0000 (23:06 -0500)]
Merge pull request #460 from Renegade334/master+ns-list-tweaks
NickServ LIST things
Renegade334 [Tue, 24 Mar 2015 23:07:34 +0000 (23:07 +0000)]
Add NOPASSWORD criterion to nickserv/list helpfile
Renegade334 [Tue, 24 Mar 2015 23:03:47 +0000 (23:03 +0000)]
nickserv/list helpfile: conditionals
Make all criteria defined outside the nickserv/list module
conditional on the defining module being loaded.
Renegade334 [Tue, 24 Mar 2015 22:41:51 +0000 (22:41 +0000)]
new nickserv/list criterion VACATION
William Pitcock [Mon, 23 Mar 2015 20:22:59 +0000 (15:22 -0500)]
Revert "chanserv: remove obsolete FTRANSFER module (pointed out by Renegade334)."
This reverts commit
8607f230f98925715ff88507ca88d5b3d2dcb241 .
William Pitcock [Mon, 23 Mar 2015 15:12:32 +0000 (10:12 -0500)]
chanserv: remove obsolete FTRANSFER module (pointed out by Renegade334).
Mantas Mikulėnas [Mon, 23 Mar 2015 07:18:20 +0000 (09:18 +0200)]
contrib/cap_sasl.pl: fix function redefinitions
apparently sub <name> {} is done at compile time even if inside an if {}
Mantas Mikulėnas [Mon, 23 Mar 2015 07:09:24 +0000 (09:09 +0200)]
contrib/cap_sasl.pl: explain why '/sasl keygen' is missing
Max Teufel [Tue, 17 Mar 2015 19:46:26 +0000 (20:46 +0100)]
nickserv/set_nopassword: remove FORCEON, still check for CertFP/pubkeys
Janik Kleinhoff [Fri, 20 Mar 2015 19:58:32 +0000 (19:58 +0000)]
crypt_verify_password: fix possible NULL dereference
Check if the crypted string is NULL before passing it to strcmp; some
crypto modules (notably POSIX-style crypt(3)) return NULL on error.
Fixes shalture/shalture#50.
William Pitcock [Thu, 19 Mar 2015 22:19:02 +0000 (17:19 -0500)]
protocol/inspircd: remove obsolete comment (pointed out by Renegade334).
Max Teufel [Thu, 19 Mar 2015 19:24:25 +0000 (20:24 +0100)]
protocol/inspircd: use DELLINE for XLine removal
Max Teufel [Fri, 27 Feb 2015 15:02:35 +0000 (16:02 +0100)]
protocol/inspircd: add ZLINE/UNZLINE implementation
Renegade334 [Sun, 15 Mar 2015 13:19:20 +0000 (13:19 +0000)]
modules/nickserv/ghost.c: Restore edge case check for lastlogin
William Pitcock [Sat, 14 Mar 2015 15:16:37 +0000 (10:16 -0500)]
Merge pull request #458 from Renegade334/master
Fix some problems with unregistered target nicks in NickServ GHOST
Renegade334 [Sat, 14 Mar 2015 14:22:34 +0000 (14:22 +0000)]
modules/nickserv/ghost.c: slight flow re-ordering
Renegade334 [Sat, 14 Mar 2015 13:53:24 +0000 (13:53 +0000)]
modules/nickserv/ghost.c: problems with unregistered targets
Fixed frozen and nopassword checks not matching unregistered targets
Fixed inability to use password authentication for unregistered targets
William Pitcock [Thu, 5 Mar 2015 20:08:28 +0000 (14:08 -0600)]
libathemecore: do not allow entities under restriction to take +F at all (closes #439)
Jeff Horelick [Thu, 5 Mar 2015 18:16:58 +0000 (13:16 -0500)]
Merge pull request #456 from attilamolnar/master+resetpass-use-after-free
nickserv/resetpass: Fix use after free
Attila Molnar [Thu, 5 Mar 2015 18:12:21 +0000 (19:12 +0100)]
nickserv/resetpass: Fix use after free
William Pitcock [Sun, 1 Mar 2015 20:03:55 +0000 (14:03 -0600)]
exttarget: explicitly disallow foundership for exttargets (closes #427)
William Pitcock [Sun, 1 Mar 2015 19:59:48 +0000 (13:59 -0600)]
groupserv/main: allow groups to take +F (ref #427)
William Pitcock [Sun, 1 Mar 2015 19:53:56 +0000 (13:53 -0600)]
chanserv: use myentity_allow_foundership() to control whether or not an entity can take +F (ref #427)
William Pitcock [Sun, 1 Mar 2015 19:08:03 +0000 (13:08 -0600)]
libathemecore/entity: add new entity validator for taking +F (ref #427)
Mantas Mikulėnas [Fri, 27 Feb 2015 16:37:02 +0000 (18:37 +0200)]
time_format: show the damn timezone
Lets scripts which parse '/ns info' work correctly regardless of
services running "local" time or UTC.
William Pitcock [Fri, 27 Feb 2015 05:28:37 +0000 (23:28 -0600)]
protocol/ts6-generic: add DLINE/UNDLINE implementation
William Pitcock [Fri, 27 Feb 2015 05:17:42 +0000 (23:17 -0600)]
libathemecore: add dline/undline core interface
Mantas Mikulėnas [Wed, 25 Feb 2015 05:34:48 +0000 (07:34 +0200)]
contrib/cap_sasl.pl: whitespace fix
Fixed spaces used for indent.
William Pitcock [Fri, 20 Feb 2015 15:13:13 +0000 (09:13 -0600)]
Merge pull request #455 from Renegade334/master
Stylistic edits
William Pitcock [Thu, 19 Feb 2015 19:51:09 +0000 (13:51 -0600)]
update contrib modules snapshot
Mantas Mikulėnas [Wed, 18 Feb 2015 22:32:07 +0000 (00:32 +0200)]
modules/resetpass: revert the frozen check
Mantas Mikulėnas [Wed, 18 Feb 2015 22:28:24 +0000 (00:28 +0200)]
nickserv/sendpass: I'm an idiot
Mantas Mikulėnas [Wed, 18 Feb 2015 22:24:59 +0000 (00:24 +0200)]
nickserv/sendpass: only set the key/password after sending
Mantas Mikulėnas [Wed, 18 Feb 2015 22:17:00 +0000 (00:17 +0200)]
nickserv/resetpass: should *add* metadata, not delete it
It resets the password just like the other modules, so that should be
recorded accordingly.
Mantas Mikulėnas [Wed, 18 Feb 2015 22:15:28 +0000 (00:15 +0200)]
nickserv/resetpass: reduce duplicated code, honor frozen flag
Mantas Mikulėnas [Wed, 18 Feb 2015 22:10:31 +0000 (00:10 +0200)]
nickserv/{set,send}pass: copy the NOPASSWORD text from set_nopassword
Mantas Mikulėnas [Wed, 18 Feb 2015 22:07:03 +0000 (00:07 +0200)]
nickserv/sendpass: clear MU_NOPASSWORD when changing the password
Mantas Mikulėnas [Wed, 18 Feb 2015 22:05:54 +0000 (00:05 +0200)]
nickserv/sendpass: command_fail() if the account is frozen
Mantas Mikulėnas [Wed, 18 Feb 2015 22:02:25 +0000 (00:02 +0200)]
nickserv/sendpass: wrap the SENDPASS version in an else
These are mostly equivalent code paths, alternatives.
(Also who the hell needs an explicit return in a void function)
Mantas Mikulėnas [Wed, 18 Feb 2015 21:56:42 +0000 (23:56 +0200)]
nickserv/sendpass: unconditionally log SENDPASS usage
If sendmail failed, you'd end up with an accout whose password has been
reset without any trace. That's kinda bad.
Also fix truncated "Overriding MARK" message, and add sender metadata in
the SETPASS case.
Mantas Mikulėnas [Wed, 18 Feb 2015 21:53:41 +0000 (23:53 +0200)]
nickserv/sendpass: never send the current password
Most of the time this code won't be reached (SETPASS present, passwords
hashed by default), but I think it's stupid for it to even exist in the
first place.
Mantas Mikulėnas [Wed, 18 Feb 2015 21:36:41 +0000 (23:36 +0200)]
nickserv/resetpass: clear the NOPASSWORD flag
Mantas Mikulėnas [Wed, 18 Feb 2015 21:34:52 +0000 (23:34 +0200)]
nickserv/resetpass: use same order and messages as SETPASS
Mantas Mikulėnas [Wed, 18 Feb 2015 21:32:54 +0000 (23:32 +0200)]
nickserv/setpass: clear the NOPASSWORD flag
Mantas Mikulėnas [Wed, 18 Feb 2015 21:08:29 +0000 (23:08 +0200)]
nickserv/setpass: flip the key check
Max Teufel [Sat, 14 Feb 2015 16:29:04 +0000 (17:29 +0100)]
nickserv: actually add set_nopassword.c to Makefile
Max Teufel [Sat, 14 Feb 2015 16:05:23 +0000 (17:05 +0100)]
nickserv/set_password: require FORCEON if there are no fingerprints etc
Max Teufel [Sat, 14 Feb 2015 14:14:45 +0000 (15:14 +0100)]
nickserv/set_password: updates to help text and error messages
Max Teufel [Sat, 14 Feb 2015 14:06:49 +0000 (15:06 +0100)]
nickserv/set_nopassword: new module
It should be discussed whether this also disables XMLRPC and JSONRPC
authentication.
Renegade334 [Fri, 6 Feb 2015 10:39:30 +0000 (10:39 +0000)]
Replace remaining instances of 'struct gflags' with typedef gflags_t
Renegade334 [Wed, 11 Feb 2015 11:37:14 +0000 (11:37 +0000)]
Convert MYCHAN_FROM to inline function mychan_from
Mantas Mikulėnas [Sun, 15 Feb 2015 11:46:44 +0000 (13:46 +0200)]
contrib/cap_sasl.pl: SET PUBKEY is now supported
Stephen Bennett [Wed, 11 Feb 2015 18:06:11 +0000 (18:06 +0000)]
Revert "libathemecore: ensure a `false` response from subclassed myentity_can_register_channel is respected (closes #427)"
While fixing the issue in #427, this commit broke channel registration
for normal users, and even if that were fixed the changed logic is no
longer able to accommodate existing functionality of users being
exempted from registration limits. A more thorough fix for #427 is
required.
This reverts commit
c50f5276ca06a68e9dd9dae2f676af492c85e936 .
Stephen Bennett [Wed, 11 Feb 2015 18:04:48 +0000 (18:04 +0000)]
Merge pull request #453 from shockkolate/perl-api-forward-compat
Make Perl API forward compatible
Shockk [Mon, 9 Feb 2015 20:40:22 +0000 (20:40 +0000)]
scripting/perl: Forward compatibility for hooks
Currently the perl script to generate the hooks for the perl API must be
updated whenever a new hook is added, either to support the hook or to
explicitly unsupport it, as doing neither results in a compilation error
of the generated file perl_hooks.c.
This patch updates the script to prevent hooks from being added to the
hooks list if the argument type is unsupported and emits a warning if
the hook has not been explicitly marked as unsupported.
Shockk [Wed, 11 Feb 2015 17:06:23 +0000 (17:06 +0000)]
scripting/perl: Fix lack of perl_hooks.c ordering
This patch resolves the contents of perl_hooks.c being different between
identical compilations due to Perl hashes being looped over in an
intentionally unique order each execution.
Renegade334 [Wed, 11 Feb 2015 10:42:50 +0000 (10:42 +0000)]
libathemecore/cidr.c: Fix minor type mismatch
Change throwaway buffer type to u_char in valid_ip_or_mask()
Gets rid of an implicit cast warning when passed to inet_pton*
William Pitcock [Mon, 9 Feb 2015 17:46:07 +0000 (11:46 -0600)]
perl: add the new hook type definition
William Pitcock [Mon, 9 Feb 2015 17:04:27 +0000 (11:04 -0600)]
libathemecore/cmode: use c99 initializers to enforce constness
William Pitcock [Mon, 9 Feb 2015 16:52:19 +0000 (10:52 -0600)]
chanserv: remove set_founder
William Pitcock [Mon, 9 Feb 2015 16:42:36 +0000 (10:42 -0600)]
nickserv/set_email: relax verification requirements so that typo'd email addresses can be fixed (closes #441)
As far as I can see, this showed up between Atheme 0.1 and 0.2, most likely it was just added with an automatic
script to add the e-mail check in a lot of places. I can't see any rational reason for having it the way it presently
is.
William Pitcock [Mon, 9 Feb 2015 16:29:04 +0000 (10:29 -0600)]
chanserv: move libathemecore component of bouncing mode changes on secure channels to chanserv (closes #449)
Janik Kleinhoff [Thu, 8 Jan 2015 20:42:45 +0000 (20:42 +0000)]
user_is_channel_banned(): check for voice/op/etc.
ircds typically allow ignoring bans/quiets for voiced users and above,
so we shouldn't consider them in this case either.
Janik Kleinhoff [Wed, 7 Jan 2015 21:36:59 +0000 (21:36 +0000)]
user_is_channel_banned(): respect +e if applicable
Janik Kleinhoff [Mon, 9 Feb 2015 15:58:10 +0000 (09:58 -0600)]
nickserv REGAIN: the target user's bannedness shouldn't matter
Whether the source user's bannedness should matter is another thing to
be discussed IMHO. The target user's bannedness definitely should not.
/ns release doesn't respect it either, and there's no good reason to.
William Pitcock [Mon, 9 Feb 2015 15:50:16 +0000 (09:50 -0600)]
libathemecore: ensure a `false` response from subclassed myentity_can_register_channel is respected (closes #427)
Max Teufel [Mon, 9 Feb 2015 15:08:18 +0000 (16:08 +0100)]
saslserv: use message source to get the source server
William Pitcock [Mon, 9 Feb 2015 13:04:17 +0000 (07:04 -0600)]
Merge pull request #450 from maxteufel/feature/sasl_bad_password_friendly
saslserv: make bad_password description more user-friendly
Max Teufel [Sun, 8 Feb 2015 18:52:13 +0000 (19:52 +0100)]
saslserv: make bad_password description more user-friendly
William Pitcock [Tue, 3 Feb 2015 23:08:07 +0000 (17:08 -0600)]
Merge pull request #448 from ilbelkyr/sasl-bad_password
saslserv: optionally call bad_password on SASL authentication failure
Janik Kleinhoff [Tue, 3 Feb 2015 20:01:28 +0000 (20:01 +0000)]
saslserv: call bad_password on SASL authentication failure
William Pitcock [Mon, 2 Feb 2015 06:08:49 +0000 (00:08 -0600)]
run autogen in git too.
William Pitcock [Mon, 2 Feb 2015 02:27:27 +0000 (20:27 -0600)]
atheme 7.2.5.
William Pitcock [Mon, 2 Feb 2015 02:23:56 +0000 (20:23 -0600)]
abirev: the ASA fix broke our ABI, so increment it
William Pitcock [Mon, 2 Feb 2015 02:20:49 +0000 (20:20 -0600)]
Merge pull request #440 from Renegade334/master
modules/groupserv/register.c: Add unverified user check
Ryan Schoppmeyer [Sun, 1 Feb 2015 10:20:14 +0000 (02:20 -0800)]
Alter several modules to use `numsvcmembers` in `struct channel`, instead of inferring or guessing how many services users should be on the channel.
A number of bugs have been corrected, mostly caused by incorrect checking of emptying channels:
- chanserv/main: not kicking users who create channels that are MLOCK'd +i if a botserv bot is present
- groupserv/main/hooks: permission changes not taking effect immediately if a botserv bot is present
- chanserv/sync: rare crash caused by use-after-free when removing the last non-services user from a channel
Ryan Schoppmeyer [Sun, 1 Feb 2015 09:07:07 +0000 (01:07 -0800)]
Add a field to channel structure to track the number of internal clients in the channel. This will make it easier to determine if removing a user will cause the channel to become empty (since, in practice, services users will often part channels when the last normal user leaves)
Janik Kleinhoff [Sat, 31 Jan 2015 00:04:49 +0000 (00:04 +0000)]
helpserv/services: skip BotServ bots
Fixes atheme/atheme#445; botserv bots are useless clutter in a list
meant for users looking for help, and the instruction to use
/msg service help doesn't apply to them either.
Renegade334 [Wed, 21 Jan 2015 10:28:38 +0000 (10:28 +0000)]
modules/groupserv/register.c: Add unverified user check
Add group registration to restricted activity list in RESTRICT helpfile