Erich/openttd - Unnamed repository; edit this file 'description' to name the repository.

Age	Commit message (Collapse)	Author
2011-05-02	(svn r22410) -Document: some more bits ;)	rubidium

2011-04-19	(svn r22350) -Change: make YAPF hash tables much smaller, it should improve ↵	smatz
	performance significantly (about 15%)
2011-04-19	(svn r22348) -Codechange: split ship-specific part of yapf_node.hpp to ↵	smatz
	yapf_node_ship.hpp
2011-02-25	(svn r22144) -Codechange: Unify 'while (true)' to 'for (;;)'	alberth

2011-02-07	(svn r22017) -Codechange: move MarkTileDirtyByTile to viewport_func.h	rubidium

2011-02-06	(svn r21996) -Fix [FS#4472]: [YAPF] Under some circumstances vehicles could ↵	rubidium
	be lost
2011-01-15	(svn r21815) -Fix [FS#3908]: [YAPF] Apply a pathfinder penalty for back of ↵	rubidium
	one-way path signals so those aren't preferred over other possibilities
2011-01-15	(svn r21803) -Cleanup: Remove unnecessary semicolons.	terkhen

2010-12-28	(svn r21652) -Cleanup: one semicolon is enough	smatz

2010-12-27	(svn r21651) -Fix [FS#4338] (r21229): infinite loop in the road pathfinder ↵	rubidium
	due to bouncing around in an "one way" trap; two one ways pointing towards eachother making it impossible to leave
2010-12-22	(svn r21594) -Codechange: Whitespace fixes in ifndef/define lines.	alberth

2010-12-15	(svn r21524) -Fix [FS#4302]: Do not apply the last signal red pathfinder ↵	michi_cc
	penalty when the signal is a path signal.
2010-12-14	(svn r21521) -Codechange: Unify some cached values that were present in both ↵	terkhen
	road vehicles and trains.
2010-12-13	(svn r21511) -Feature: vehicle lost message for ships	rubidium

2010-12-13	(svn r21510) -Feature [FS#1956]: vehicle lost message for road vehicles	rubidium

2010-12-13	(svn r21509) -Codechange: rename the rail pathfinders "path_not_found" ↵	rubidium
	parameter to "path_found" and remove the ! where the variables are set / read
2010-11-20	(svn r21273) -Codechange: Return values should start at the same line.	alberth

2010-11-20	(svn r21271) -Change: use the last red instead of last red exit penalty for ↵	rubidium
	making sure other waypoint entries are evaluated as well when they are occupied, e.g. when there are no signals before the waypoint but a train just beyond the waypoint is stopped (like for stations)
2010-11-09	(svn r21121) -Fix (r21100): Use the cached max speed when calculating YAPF ↵	terkhen
	penalties for road vehicles.
2010-11-07	(svn r21107) -Fix: Don't use the maximum track speed where the maximum ↵	michi_cc
	vehicle speed is meant.
2010-10-30	(svn r21060) -Doc: Improved wording of comments (mostly by __ln__)	alberth

2010-08-26	(svn r20632) -Cleanup: remove "a few" unneeded includes, and add them to ↵	rubidium
	some headers
2010-08-12	(svn r20475) -Fix (r20474): also remove the include of crc32.hpp	yexo

2010-08-01	(svn r20286) -Codechange: Unify end of doxygen comments.	frosch

2010-08-01	(svn r20283) -Codechange: Unify start of doygen comments.	frosch

2010-08-01	(svn r20281) -Codechange: unify case scope closure + break coding style	rubidium

2010-07-24	(svn r20211) -Codechange: Indented code should have curly braces around it.	alberth

2010-05-26	(svn r19896) -Fix [FS#3803] (r18648): [YAPP] Inform the pathfinder as well ↵	michi_cc
	about the fact that the backside of an one-way path signal can be a safe waiting point.
2010-05-21	(svn r19873) -Fix [FS#3817]: if the (guessed initial) destination tile of a ↵	rubidium
	road stop wasn't a road stop but was a T-junction or turn, the road vehicles would jump around in circles
2010-05-13	(svn r19810) -Fix: [YAPF] Improperly formatted (global) constant	rubidium

2010-04-18	(svn r19670) -Codechange: Add CeilDiv() and RoundDiv() to simplify integer ↵	frosch
	divisions with rounding.
2010-03-23	(svn r19507) -Codechange: remove semicolon after DECLARE_POSTFIX_INCREMENT ↵	smatz
	and DECLARE_ENUM_AS_BIT_SET
2010-02-25	(svn r19247) -Codechange: Rename methods to fit better to common style (skidd13)	yexo

2010-02-25	(svn r19242) -Codechange: Perfer pointer instead of reference (skidd13)	yexo
	-Cleanup: merge PopHead() and RemoveHead() into Shift()
2010-02-18	(svn r19161) -Codechange: make the default size of the open list used for ↵	yexo
	yapf a lot lower to reduce memory usage
2010-02-18	(svn r19160) -Codechange: Enlarge a CBinaryHeapT if the heap is full instead ↵	yexo
	of dropping the added item -Fix: CBinaryHeapT::CheckConsistency compared pointers instead of the actual items (skidd13)
2010-02-15	(svn r19141) -Fix [FS#3619] (r18421): look-ahead for multitile waypoints ↵	rubidium
	'made up' data that shouldn't go into the cache, causing desyncs in MP
2010-01-16	(svn r18826) -Codechange: Unifiy return value of ↵	frosch
	(SmallArray\|FixedSizeArray)::(Append\|AppendC) with other containers. (skidd13)
2010-01-16	(svn r18822) -Codechange: Rename YAPF-related container classes and their ↵	frosch
	members to better fit other container classes. (skidd13)
2010-01-15	(svn r18809) -Codechange/Cleanup: remove unneeded headers from some files, ↵	rubidium
	if a header require a header make it include that header
2009-12-19	(svn r18535) -Fix [FS#2722]: don't account for path reservation costs when ↵	rubidium
	entering a signal block via a 'block' signal. This way you won't get double penalties, both red signals and reservation costs, for the block signalled tracks
2009-12-13	(svn r18481) -Codechange: unify the curve pathfinder penalty defaults; 0.01 ↵	rubidium
	tile won't make a dent, 3 tiles might be a bit too much -Feature-ish: make maximum pathfinder penalties for finding depots customisable, also increase it slightly to 20 tiles worth of penalties.
2009-12-07	(svn r18421) -Fix [FS#3244]: pathfinders wouldn't consider the 'other' ↵	rubidium
	reachable waypoint tile if the closest one is free but there is no safe waiting point directly after it. Now check for a free safe waiting point beyond the waypoint unless there are junctions before the first safe waiting point.
2009-12-07	(svn r18420) -Codechange: split YAPF's track follower from the actual YAPF code	rubidium

2009-12-04	(svn r18404) -Codechange: link drive through stops better together	rubidium
	-Feature: make penalty for road stop occupancy user configurable -Fix [FS#1944]: road vehicles would not pick an empty drive through stop. Now they will if the penalty for driving around is less than the occupancy penalty -Fix [FS#1495]: long (articulated) road vehicles could block loading of others when the following road vehicle already got 'permission' to go to the next bay even when it could not reach it -Change: improve the throughput of the drive through road stops by letting them stop closer together
2009-12-04	(svn r18394) -Fix (r18373): one couldn't manually send a RV to a depot... ↵	rubidium
	but why compilers haven't complained about this?
2009-12-02	(svn r18388) -Codechange: move entering/leaving of road stops to functions ↵	rubidium
	of RoadStop
2009-12-02	(svn r18385) -Cleanup: remove the now unneeded multistop slot management code	rubidium

2009-12-02	(svn r18383) -Fix (r18375): triggering NOT_REACHED for some waypoints	rubidium

2009-12-02	(svn r18382) -Codechange: make road vehicles behave more like trains ↵	rubidium
	'around' stations and use pathfinder penalties to determine to which 'part' to go. Note that the pathfinder penalties for drive through stops are currently only looking at the occupation of the first in a row, but this is to change later on.