summaryrefslogtreecommitdiff
path: root/ai
AgeCommit message (Collapse)Author
2005-12-14(svn r3298) Remove unused and write-only variablestron
2005-12-14(svn r3297) Staticisetron
2005-12-13(svn r3294) - Fix: use INVALID_STRING_ID instead of -1.Darkvater
- Fix: savegames only give back one message, show this by ignoring the first argument. Perhaps make the message more verbose in the future by adding STR_ equivalents next to the already existing debug messages.
2005-12-10(svn r3285) -Codechange: Tron was confused by a name, so let's make him happy, truelight
prefixed it with 'tmp', so now we know it is a temp char* :)
2005-12-10(svn r3284) -Fix: DoCommand always cleans up _cmd_text (tnx to Tron)truelight
2005-12-10(svn r3283) -Fix: decode_parameters was still used, while _cmd_text is the waytruelight
-Fix: _cmd_text is cleared after a command, so we need to store it temporaray in order to first test the command, before executing
2005-12-10(svn r3282) - Codechange: Replace tests against CMD_ERROR with CmdFailed()peter1138
2005-12-09(svn r3279) -Fix: do not try to execute something that is not set to executetruelight
-Fix: do not increase an uid number if it isn't returned
2005-12-09(svn r3278) -Fix: lets update the tail-pointer if we add a new item, else ↵truelight
the linkedlist never grows ;)
2005-12-09(svn r3277) -Fix: lets use the right DoCommand for testing a build (tnx to ↵truelight
Igor2Code)
2005-12-08(svn r3272) -Fix: [AI] fixed issue in AI that prevented compilation without ↵bjarni
network support
2005-12-07(svn r3269) -Fix: return CMD_ERROR instead of -1 if AI_DoCommandChecked failstruelight
2005-12-06(svn r3266) -Codechange: added comments for why the revert of r3625 inside ↵truelight
the code. Also, I forgot to tnx GraphiteNL for his help with this problem. Tnx GraphiteNL :)
2005-12-06(svn r3265) -Fix: partly reverted back r3263, because MSVC, who else, ↵truelight
doesn't support variadic macros............
2005-12-06(svn r3263) -Fix: use () around 'variables' in macros, gives less crashes ;)truelight
-Fix: va-args in macros are supported via __VA_ARGS__ in C99, so changed to that; now GCC 2.95 likes it too :)
2005-12-05(svn r3260) -Add: add events for AIs to check if a command execution failed ↵truelight
or succeeded
2005-12-02(svn r3253) -Add: add BuildRoad event for AIstruelight
2005-12-01(svn r3251) -Fix: report errors from GPMI in a more detailed way (Igor2Code)truelight
2005-11-30(svn r3250) -Fix: AIs weren't uninitialized when a new game was loadedtruelight
2005-11-29(svn r3246) -Fix: small glitch in ai_network_client code (network_client.c)truelight
-Fix: make sure this glitch can never happen again (ai.c)
2005-11-29(svn r3245) -Add: allow OpenTTD to give his GPMI-params to the GPMI modulestruelight
2005-11-28(svn r3244) -Fix: [GPMI] Even more GPMI based AI-code cleanup, bug fixes, ↵truelight
and you can now control the AI that is going to boot
2005-11-27(svn r3242) -Fix: small glitch in AllowAI controltruelight
2005-11-26(svn r3238) -Fix: always save if you want GPMI-based AIs, this on request by truelight
Darkvater, so we can, in the future, save patch options in savegames without any problems.
2005-11-26(svn r3236) - Fix: warnings about 'CDECL must be used with ...' on VS6Darkvater
2005-11-23(svn r3232) -Add: implemented the event-system for AIstruelight
-Add: added several hooks (event-callbacks) for road-related-stuff
2005-11-22(svn r3229) -Add: add more GPMI support. Now GPMI-based AIs can be loaded ↵truelight
(doesn't change a thing if you didn't enable GPMI)
2005-11-22(svn r3226) -Fix: GPMI implementation had minor glitchestruelight
-Fix: the AI speed control is done by the AI-core, individual AIs don't have to do it (so, AIs were delayed twice ;) -Add: Support for AI-network-clients (an AI, connecting to a remote server) -Fix: minor AI-core problems
2005-11-21(svn r3224) -Add: Allow the NewAI to work in Multiplayer Games (switchable ↵truelight
via patch settings, off by defaut). An other step to AIScripts. WARNING: this is still highly experimental and has known bugs!
2005-11-15(svn r3190) Turn some loops into canonical for loop formtron
2005-11-15(svn r3185) consttron
2005-11-14(svn r3181) -Bracingtron
-Indentation -Whitespace -DeMorgan's Law -Test with NULL or 0 for non-booleans -'\0' instead of 0 for chars -Remove redundantly redundant comments (like DoFoo(); // Do foo) -Join multiple short lines with a single statement -Split single lines with multiple statements -Avoid assignments in if
2005-11-14(svn r3177) GB, CLRBIT, HASBIT, TOGGLEBITtron
2005-11-13(svn r3172) static, consttron
2005-10-24(svn r3080) byte -> PlayerID, int -> EngineID, -1 -> INVALID_ENGINEtron
2005-10-23(svn r3078) Some more stuff, which piled up:tron
- const, whitespace, indentation, bracing, GB/SB, pointless casts - use the trinary operator where appropriate - data types (uint[] -> AcceptedCargo, ...) - if cascade -> switch - if (ptr) -> if (ptr != NULL) - DeMorgan's Law - Fix some comments - 0 -> '\0', change magic numbers to symbolic constants
2005-10-22(svn r3077) static, const, bracing, indentation, 0 -> '\0'/NULL, typos in ↵tron
comments, excess empty lines, minor other changes nothing spectacular, just some stuff, which piled up
2005-10-20(svn r3073) int/byte -> EngineID/PlayerID/VehicleIDtron
0/-1 -> INVALID_ENGINE 0x10 -> OWNER_NONE
2005-10-18(svn r3058) The default AI may not build dual headed trains under certain ↵tron
circumstances. Simplify this check by testing the rail vehicle info flags instead of passing a global variable around. Note: This is not exactly the original behaviour, because the Lev4 was always allowed to be built, but i guess that was a glitch.
2005-10-16(svn r3052) Fix last commit *grml*tron
2005-10-16(svn r3051) Remove the unused function RemoteSubsidyAdd()tron
2005-09-23(svn r2973) Move a function declaration somewhere where it belongstron
2005-09-23(svn r2972) Fix the speed of the AItron
2005-09-14(svn r2951) - Fix: [ 1259345 ] Changing engine in netgame opens train window ↵Darkvater
for everyone - Add IsLocalPlayer() which substitutes _local_player == _current_player
2005-09-13(svn r2949) The AI no longer needs to 'cheat' to build aircraft; eg it ↵Darkvater
builds them now from a hangar. Also, to query the price of a new aircraft tile information is not needed
2005-09-13(svn r2948) -Fix: the old AI needs a special flag that triggers all kind of ↵truelight
special abilities you really don't want to know about (free bridges, etc..) I removed this flag some revisions ago, but the Aircraft part depends on it, so I re-enabled it again..
2005-09-07(svn r2921) -Codechange: moved all AI-code to 1 central place (ai/ai.c)truelight
-Fix: removed the ability for the oldAI to cheat (this will criple him somewhat) -Add: base-code for many improvements to come in the AI-system -Add: added base-code for multiplayer AIs (DOES NOT WORK YET!)
2005-09-02(svn r2907) -Codechange: splitted the AIs to their own directory. AINew ↵truelight
becomes 'trolly', AIOld becomes 'default', both in their own dir in the 'ai' dir. More AIs to come.