aboutsummaryrefslogtreecommitdiff
path: root/advtrains/path.lua
Commit message (Collapse)AuthorAge
* Occupation system: store multiple indices for the same train, introduce ↵orwell962023-06-24
| | | | reverse_lookup_sel() to select appropriate index out of multiple based on a heuristic
* Add function to get wagon in train from indexorwell962023-05-27
| | | | | Can be used to determine which wagon is at a certain world position Testing: use debugitems.lua
* Rework train same-track collision system and deterministic couplingorwell962021-07-07
| | | | | | | - Adds a separate collision system for trains sharing a path - Moved some coupling-related code to couple.lua and refactor it - Handle coupling in a way that the initiating train always keeps its ID - As a side effect, engine has its direction reversed after coupling
* Don't try to invalidate_ahead when path has been regularily invalidated beforeorwell962021-02-10
|
* Start using path_invalidate_ahead()orwell962021-02-10
|
* Implement a LZB speed lookup table for the path and rewrite velocity controlsorwell962021-02-10
|
* optimize path_get_index_by_offsettechniX2020-07-10
| | | | | | | Instead of calling path_get_adjacent twice (which calls path_get twice, i.e. 4 times overall), we call path_get directly only 2 times with min and max indices to generate all the path we need for calculations.
* comment out annoying debug infoGabriel Pérez-Cerezo2020-06-14
|
* Workaround to handle split points (dt. aufgefahrene Weichen), fixing H#77orwell962020-06-10
|
* Simplify and optimize path_get_index_by_offset (cherrypicked)orwell962019-12-09
|
* Possible fix for braking problems + debug outputsorwell962019-01-23
|
* Warn when something clears the path winthin train steporwell962019-01-22
|
* Revert "Correct last commit"orwell962019-01-22
| | | | This reverts commit 5f290819cdb78303396f9f89907ebbc66a9d74b3.
* Revert "Debugging code to trace down path_invalidate within callbacks error"orwell962019-01-22
| | | | This reverts commit 1965e846b6c61958063ea13e2ac88ae18b701d09.
* Correct last commitorwell962019-01-22
|
* Debugging code to trace down path_invalidate within callbacks errororwell962019-01-22
|
* Additional info print for path_print()orwell962019-01-21
|
* Fix H#69orwell962018-10-17
| | | | Cause was a cascaded crash, original cause was something different
* Add signal safety control override, restructure control systemorwell962018-10-10
|
* Add off-track train flag, fixes wagon object property reset on reversingorwell962018-07-04
|
* Fix bugs found while testingorwell962018-06-14
|
* Fix final bugs and to-do's (u.a.save/load system)orwell962018-06-14
|
* Mainly make collisions and coupling workorwell962018-06-14
| | | | Missing: ATC stuff, yaw problems
* Implement a reverse path lookup for trains instead of an occupations window ↵orwell962018-06-14
| | | | system
* Fix path_dir to actually be an angle, path item deletion and orientation of ↵orwell962018-06-14
| | | | | | | | wagons The occupation system as it is now will change. For each position, I will save the index in the train's path, and implement a callback system. I need this because the occupation window system will not be enough to cover all use cases (e.g. to make a train stop with it's center or back at a certain position, I need 3 different brake distances, which doesn't fit into the scheme)
* Bugfixes part 1orwell962018-06-14
| | | | There's something wrong with the new paths, next time build a path validity checker to trace the issue
* Occupation System, new train steps, still incompleteorwell962018-06-14
|
* Adapt wagons to new path systemorwell962018-06-14
| | | | Still outstanding: trains
* Restructure path systemorwell962018-06-14
| | | | The changes to the path system have not yet been integrated...
* Rewrite rail connection system...orwell962017-12-18
...to support an arbitrary number of connections for rails, which leads to these new features: - switches now get recognized by the trackworker correctly - ability to add real rail crosses During this, I also rewrote the rail registering system and the conway function (important part of path prediction) Note, developers: the track preset format changed, you might need to rewrite them according to the presets in tracks.lua if you wrote your own (possibly breaks advcarts)