This is to fix#601. The function used to extract the mode string
assumed that ":" would only occur in a particular spot. This lead to the
possibility that ":" could be treated as part of things like nicknames
or mode arguments, where it should have been stripped as part of
protocol escaping.
Make sure not to set connection_lost if the server was
already disconnected.
git-svn-id: file:///var/www/svn.irssi.org/SVN/irssi/trunk@5030 dbcabf3a-b0e7-0310-adc4-f8d773084564
Use g_string_append_printf() instead of g_string_sprintfa() (which is considered deprecated.)
git-svn-id: file:///var/www/svn.irssi.org/SVN/irssi/trunk@5003 dbcabf3a-b0e7-0310-adc4-f8d773084564
function to read linewise from the associated io channel.
Rewrite irc/dcc/proxy read logic on top of it.
git-svn-id: file:///var/www/svn.irssi.org/SVN/irssi/trunk@4841 dbcabf3a-b0e7-0310-adc4-f8d773084564
These mostly happen when doing remote whois and the
target server is slow.
The code uses the source server but will try to do what it
can if servers think they need to mask it.
git-svn-id: http://svn.irssi.org/repos/irssi/trunk@4638 dbcabf3a-b0e7-0310-adc4-f8d773084564
safe but generally you write lines less than 100 chars so it shouldn't
really matter. Calculated (2 + line_length/100) in seconds.
git-svn-id: http://svn.irssi.org/repos/irssi/trunk@2575 dbcabf3a-b0e7-0310-adc4-f8d773084564
the connection_lost to TRUE, the disconnection is after the "server
incoming" signal is finished.
Changed irc_parse_incoming() back to not handling any commands in server's
input buffer after server_disconnect() is called, it's not safe enough.
git-svn-id: http://svn.irssi.org/repos/irssi/trunk@2299 dbcabf3a-b0e7-0310-adc4-f8d773084564
shouldn't emit the "server disconnected" again.
We'll now handle the remaining data coming from server after disconnection.
git-svn-id: http://svn.irssi.org/repos/irssi/trunk@2290 dbcabf3a-b0e7-0310-adc4-f8d773084564
use a destroyed server record when some /command disconnects the server
(shouldn't happen really) or when irc_send_cmd() fails sending data to server
and disconnects the server (I don't know if this ever happens, but if it does,
it very well could have caused crashes)
git-svn-id: http://svn.irssi.org/repos/irssi/trunk@2243 dbcabf3a-b0e7-0310-adc4-f8d773084564
should be able to do the redirecting a lot more error-proof. Changed
lag-checking to use PINGs instead of NOTIFYs. This breaks scripts using
redirection. Hopefully this doesn't break too much things in irssi :)
git-svn-id: http://svn.irssi.org/repos/irssi/trunk@1980 dbcabf3a-b0e7-0310-adc4-f8d773084564
bind protocol-specific commands. Added #define command_bind_irc() for easier
access. CMD_IRC_SERVER(server) check should be done at the beginning of each
command requiring IRC server as active server, it handles it correctly the
cases when it is not. Did some other cleanups as well.
git-svn-id: http://svn.irssi.org/repos/irssi/trunk@1955 dbcabf3a-b0e7-0310-adc4-f8d773084564
command parameter parsers, it's probably useless now that opt.channels are
in core.
git-svn-id: http://svn.irssi.org/repos/irssi/trunk@1482 dbcabf3a-b0e7-0310-adc4-f8d773084564
they're known to all files and I don't need those stupid "void *xxx"
anymore just to avoid useless #include. Header files themselves don't
either include others as often anymore.
Added channel->ownnick to point to our NICK_REC in channel's nicks.
Gives a minor speedup in few places :)
Moved completion specific lastmsgs from channel/server core records to
fe-common/core specific records. Also changed the nick completion logic
a bit so it should work better now. Removed
completion_keep_publics_count setting, but changed the meaning of
completion_keep_publics to same as _count was before. Nick completion
doesn't have any time specific code anymore.
git-svn-id: http://svn.irssi.org/repos/irssi/trunk@1034 dbcabf3a-b0e7-0310-adc4-f8d773084564
