summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2014-09-03Validate mode parameters from ModeParser::ProcessSingle()Attila Molnar
2014-09-03Split out ModeParser::ProcessSingle() from Process()Attila Molnar
This applies up to one MODE line's worth of mode changes from a Modes::ChangeList
2014-09-03Add mode process flag MODE_CHECKACCESSAttila Molnar
2014-09-03Remove ModeHandler::m_paramtype and GetTranslateType()Attila Molnar
2014-09-03Remove ModeParser::LastParseTranslate and GetLastParseTranslate()Attila Molnar
2014-09-03Remove ModeParser::LastParseParams and GetLastParseParams()Attila Molnar
2014-09-03Compare Limits.MaxModes to the size of LastChangeList in ModeParser::Process()Attila Molnar
2014-09-03Remove ProtocolInterface::SendMode()Attila Molnar
2014-09-03m_spanningtree Send MODE/FMODE from the OnMode hookAttila Molnar
If the MODE_LOCALONLY flag is set the mode change is not propagated
2014-09-03m_spanningtree Add function to serialize parameters in a ↵Attila Molnar
Modes::ChangeList::List into MODE/FMODE format
2014-09-03Pass the modes to the OnMode hook as a Modes::ChangeList, pass ↵Attila Molnar
ModeProcessFlags too
2014-09-03Populate a Modes::ChangeList object in ModeParser::Process()Attila Molnar
2014-09-02Change mode_sequence to be a const ref in ModeParser::Process()Attila Molnar
Also change the signature of DisplayListModes() to accept a const ref
2014-09-02Replace stringstream with a std::string in ModeParser::Process()Attila Molnar
There is no benefit in using a stringstream here
2014-09-01m_spanningtree Add optional key parameter to SVSJOIN and allow it to do ↵Attila Molnar
override joins Slightly more readable version of #892 by @ShutterQuick
2014-08-31m_blockamsg Update coding style, use User::WriteNotice()Attila Molnar
2014-08-31m_blockamsg Remove redundant checkAttila Molnar
2014-08-31m_blockamsg Uncrook channel counting logicAttila Molnar
2014-08-31m_blockamsg Prevent double copy of the target stringAttila Molnar
2014-08-31m_blockamsg Remove useless variableAttila Molnar
2014-08-30Move the intrusive list containers into the insp namespaceAttila Molnar
2014-08-06m_ojoin Change allocation of NetworkPrefix to be physically part of the ↵Attila Molnar
object containing it
2014-08-06m_ojoin Leave it to the core throw an exception if the prefix char is in useAttila Molnar
2014-08-06Migrate ModeChannelOp and ModeChannelVoice constructors to builtinmodes.h, ↵Attila Molnar
remove cmode_{o,v} The default value for levelrequired is HALFOP_VALUE, meaning the ModeChannelVoice class is not required but it will be useful in case the default changes
2014-08-06Pass prefix rank and prefix char to PrefixMode constructorAttila Molnar
2014-08-06Remove a bunch of useless classes representing simple core modesAttila Molnar
2014-08-06m_exemptchanops Fix parameter validationAttila Molnar
2014-08-06m_spanningtree Remove redundant checks from HandleRemoteWhois()Attila Molnar
The size of the vector is checked in OnPreCommand() and that hook only runs for local users
2014-08-06core_who Remove redundant NULL checks from CanView()Attila Molnar
2014-08-04m_pbkdf2 Less string copyingAttila Molnar
2014-08-04m_pbkdf2: Added moduleDaniel Vassdal
Bugfixes and minor changes are by @attilamolnar, original PR #767
2014-08-04m_bcrypt: Add moduleDaniel Vassdal
bcrypt updated to 1.3 by @attilamolnar, original PR #767
2014-08-04Hashing: Redo APIDaniel Vassdal
* Don't assume the printable output of hashes is hex * Add virtual Compare() function, usable for KDFs like BCrypt Some changes and bugfixes are by @attilamolnar, original PR #767
2014-08-04m_password_hash: Use out_size as HMAC-key length, as RFC2104 recommendsDaniel Vassdal
"In any case the minimal recommended length for K is L bytes (as the hash output length)."
2014-08-04Remove InspIRCd::BindSocket()Attila Molnar
2014-08-04threadengine_win32 Use a stripped-down, local version of ↵Attila Molnar
InspIRCd::BindSocket() for setting up the ITC pipe
2014-08-04Fix off by one in ping timeout.md-5
Been bugging me that users ping timeout at 121 rather than 120 seconds by default.
2014-07-29Differ database error messages for m_permchannels and m_xline_dbJustin Crawford
m_permchannels and m_xline_db both have the same error messages as well as not informing the server operator what file is at fault for the error. This makes it hard to find which module is causing the error and what file the module is referring to. The new messages explain which database is trying to be written and the path used to write the file.
2014-07-28Do not attempt to detect non-culled users at destruction by looking for ↵Attila Molnar
their uuid, leave it to Extensible This does not take uuid/sid reuse into consideration and Extensible implements this detection already in a proper way
2014-07-28Allow fast sid reuse by erasing fake users from UserManager::uuidlist when ↵Attila Molnar
the netsplit is handled, not in cull()
2014-07-28m_spanningtree Throw a ProtocolException if a server SQUITs a server it ↵Attila Molnar
shouldn't
2014-07-28m_spanningtree Squit dead servers immediately when they die, not at the next ↵Attila Molnar
timer tick
2014-07-28m_spanningtree Remove the now needless SplitInProgress workaroundAttila Molnar
2014-07-28m_spanningtree Check TreeServer::IsDead() to learn whether a user is being ↵Attila Molnar
quit due to a netsplit in OnUserQuit()
2014-07-28m_spanningtree Change type of num_lost_* to be unsignedAttila Molnar
2014-07-28m_spanningtree Quit all split users in one goAttila Molnar
2014-07-28m_spanningtree Mark servers awaiting destruction after being lost in a ↵Attila Molnar
netsplit, add TreeServer::IsDead()
2014-07-28m_spanningtree Remove now unused TreeServer::Tidy()Attila Molnar
2014-07-28m_spanningtree Add the TreeServer that split to the cull list and destroy ↵Attila Molnar
everything under it recursively at cull time instead of at squit processing time
2014-07-28m_spanningtree Remove split servers from the hash maps in ↵Attila Molnar
TreeServer::SQuitInternal()