summaryrefslogtreecommitdiff
path: root/src/ai
AgeCommit message (Collapse)Author
2007-09-04(svn r11040) -Fix [FS#1179]: removing CMD_AUTO from some commands could ↵rubidium
remotely trigger an assertion.
2007-08-30(svn r11003) -Codechange: replace Vehicle->next to Vehicle->Next() and ↵rubidium
Vehicle->SetNext() so we can trap instances that change a next pointer and (in the future) update the first/previous pointers based on that.
2007-08-29(svn r11001) -Codechange: unify the way to determine whether a vehicle is in ↵rubidium
a depot.
2007-08-19(svn r10942) -Fix [FS#1133]: make the AI not crash when it has ships as the ↵rubidium
AI does not support them.
2007-08-10(svn r10838) -Fix [FS#1119] (r10698): double negation where only single ↵rubidium
negation was needed. Patch by frosch.
2007-08-02(svn r10759) -Codechange: make the industry struct use the pool item class ↵rubidium
as super class.
2007-08-02(svn r10756) -Codechange: use vehicle->IsValid in favour of ↵rubidium
IsValidVehicle(vehicle).
2007-07-29(svn r10733) -Codechange: change MP_STREET into MP_ROAD as we use the word ↵rubidium
"road" everywhere except in the tile type.
2007-07-26(svn r10698) -Codechange [FS#1082]: simplify the code related to ↵rubidium
foundations. Primarily removal of (duplicated|magic) code and introduction of few helper functions to ease foundation determination. Patch by frosch.
2007-07-24(svn r10672) -Codechange: typify some parameters/variables.rubidium
2007-07-23(svn r10657) -Fix r10655: commit didn't comply with coding style (tnx glx :))truelight
2007-07-23(svn r10655) -Fix [FS#1070]: Trolly AI didn't know about steep slopes, and ↵truelight
used wrong tileh in some cases (frosch)
2007-06-22(svn r10266) -Codechange: keep track of the origin, time of travel and ↵rubidium
accumulated feeder share (transfers) of individual pieces of cargo. This means that cargo isn't thrown on a big pile when it's put in a station or unloaded at a station, however the GUI does not reflect these changes yet so you will not actually see it.
2007-06-21(svn r10246) -Fix (r10297): some forgotten money conversions and truncation ↵rubidium
issues. Thanks to benc for providing the patch.
2007-06-18(svn r10205) -Codechange: refactor returning of cost, so it can be more ↵rubidium
easily modified.
2007-06-18(svn r10200) -Codechange: add "shortcut" for !CmdFailed (CmdSucceeded).rubidium
2007-06-18(svn r10197) -Codechange: replace int32 with CommandCost where appropriate.rubidium
2007-06-17(svn r10177) -Fix: Passengers has one A and two Es (ln)peter1138
2007-06-11(svn r10106) -Fix (r10098): there's always a compiler complaining about ↵rubidium
something...
2007-06-11(svn r10098) -Codechange: bilbo cared about the performance of the AI with ↵rubidium
respect to the airport placement, so he rewrote the aircraft handling of the AI; now it can actually make lots of money again :)
2007-06-10(svn r10081) -Fix [FS851]: towns/AIs didn't build bridges anymore (boekabart).rubidium
2007-06-07(svn r10058) -Codechange: give some industry variable sensible names (like ↵rubidium
not telling "last_mo_production" when it is the production of the current month).
2007-06-04(svn r10033) -Feature [FS#760]: skip to the selected order in the order list ↵rubidium
when clicking on the "skip" button while pressing CTRL.
2007-05-27(svn r9954) -Fix [FS#811]: trolley AI crashed when trying to determine what ↵rubidium
the roadbits are.
2007-05-24(svn r9914) -Codechange: prepare GTTS and the pathfinders to handle multiple ↵rubidium
road types on a single tile.
2007-05-24(svn r9912) -Fix: the AIs could build any road(station)s.rubidium
2007-05-20(svn r9892) -Codechange: lots of ground work for allowing multiple types of ↵rubidium
"road" with multiple owners on a single tile.
2007-05-18(svn r9867) -Codechange: Remove data duplication. The exact same values can ↵belugas
be found in the industry spec, so take it from there instead.
2007-05-09(svn r9817) -Fix [FS#772]: an assertion that was triggered by the AI ↵rubidium
building when its rating for the town was not high enough.
2007-04-18(svn r9672) -Cleanup: lots of coding style fixes around operands.rubidium
2007-04-12(svn r9609) -Codechange: Move some function prototypes out of functions.h ↵maedhros
and into landscape.h, and add a few where they didn't exist.
2007-04-01(svn r9540) -Fix (r1): Bankrupt AIs no longer buy themselves (usage of wrong ↵celestar
variable)
2007-03-31(svn r9530) -Codechange: Replace a direct industry type access by a behaviourbelugas
2007-03-22(svn r9400) -Codechange: Use some more representative enum names for ↵belugas
landscape types.
2007-03-16(svn r9251) -Fix(r6057, FS#644): Use a less CPU-intensive algorithm to find ↵celestar
a random industry for the AI to prevent it slowing down the game. The AI now sucks a even a bit more than before (if that's even possible).
2007-03-08(svn r9068) -Codechange: capitalize the VEH_Train etc. enums to match the ↵rubidium
coding style (and rest of the code).
2007-03-07(svn r9051) -Codechange: typedef [enum|struct] Y {} X; -> [enum|struct] X {};rubidium
2007-03-07(svn r9050) -Codechange: Foo(void) -> Foo()rubidium
2007-02-22(svn r8841) -Fixtron
Remove {,u}intswap() and replace them by Swap()
2007-02-17(svn r8767) -Fixtron
-Codechange: Do not hardcode the catchment radius of airports, but hold the information in AirportFTAClass -Fix (r979): The default AI tested possible airport locations with a fixed catchment radius instead of the radius of the to be built airport
2007-02-16(svn r8755) -Fixtron
Abbreviate GetAirport(st->airport_type) to st->Airport()
2007-02-15(svn r8750) -Fix (r8747): PLANES defined in wingdi.h caused compilation ↵KUDr
error under Win32. Changed to AIRPLANES
2007-02-15(svn r8747) -Fixtron
-Codechange: Make the encoding of accepted aircraft types of airports a bit more sensible and move the enum into struct AirportFTAClass
2007-02-14(svn r8732) -Codechange/Fix(r8705): Turned the bit-handling macros into ↵celestar
template functions. Fixes a problem with MSVC and 64-bit shifts.
2007-02-10(svn r8660) -Fixtron
-Codechange: Remove _avail_aircraft Its name is misleading. It is rather _avail_airports, but then only some of them, which leads to inconsistencies when using it. Further it is unnecessary to store it in savegams.
2007-01-30(svn r8455) -Codechange: Give a more meaningful name (railveh_type)to member ↵belugas
flags of RailVehInfo, as well as changing the code to reflect the fact that it was not a flag but rather a one value only variable. Doing so, some evaluations have been simplified. -Codechange: Add and use RAILVEH_SINGLEHEAD when railveh_type is set to 0, which was implicit before. -Cleanup: Remove some extraneous parenthesis.
2007-01-27(svn r8428) -Codechange: Add proper names to aircraft subtypes instead of ↵Darkvater
magic numbers and add a function IsNormalAircraft() which tells us whether the aircraft is in fact some flying device or a rotor/shadow.
2007-01-25(svn r8402) -Codechange: Move RoadStop-specific enums to the RoadStop class, ↵celestar
and changed a one-member enum into a static const. Simplify their naming and add some doxygen-comments to RoadStop
2007-01-24(svn r8385) -Fixtron
-Regression (r8314): I only did half the necessary changes to move railtype from Engine to RailVehicleInfo. Now do the rest.
2007-01-19(svn r8276) -Fixtron
Change the signature of Swap() to be less error prone, i.e. pass the variables to be swapped by reference instead of passing pointers to the variables. Just do Swap(x, y) instead of Swap(&x, &y). This prevents accidents when the variables are pointers.