| Age | Commit message (Collapse) | Author |
|---|
|
|
|
TileLayoutSpriteGroup instead of allocating it separately.
|
|
spritelayouts allocated on the heap, and make use of constructors and destructors.
|
|
resolving custom station foundations.
|
|
|
|
|
|
|
|
some headers
|
|
|
|
the southern bridgehead as tile.
|
|
|
|
|
|
|
|
RAIL_GROUND_HALF_SNOW. That is, resolve the sprites for upper and lower part of the foundation independently.
|
|
railstation selection callback.
|
|
industries from construction callbacks 17, 28 and 2F. That is: The randombits the house/industry will start with, if construction succeeds.
|
|
industries and houses during construction callbacks.
|
|
temporary/persistent storage. (Spotted by yexo)
|
|
|
|
rail types. (To be documented...)
|
|
if a header require a header make it include that header
|
|
num_sprites (as in number of sprite sets) is not the same as num_sprites (as in number of building sprites)
|
|
including of newgrf_cargo.h
|
|
production callback version 0.
|
|
procedure and main chain.
-Fix (r8075): 'scope' and 'count' are not.
|
|
it is only needed for them.
|
|
anymore.
|
|
which were sometimes missing and sometimes just typos
|
|
|
|
can be reused for Waypoints.
|
|
town_type.h to separate files
|
|
when the new pool was added, which mean there weren't enough spritegroups when you have more than about a dozen ECS vectors.
-Change: increase the spritegroup pool's maximum size to something more than the number of real sprites that OpenTTD can handle; for example ECS has about 30 spritegroups per real sprite. With the 'old' limit that would mean 'only' about 4000 real sprites worth of spritegroups could be loaded.
|
|
|
|
'features' of 0.3.2.1, which was released only 5 years ago.
|
|
instead of using a big union
|
|
|
|
time, binary size and run time (with asserts disabled) should be improved
|
|
|
|
|
|
something more than whitespace as description of files that don't have a description.
|
|
(especially because newgrf_station.h includes a lot of stuff).
|
|
|
|
were 15 bit, and vice versa.
Var 0x7E procedure-results are always 15 bit.
Callbacks 0x2A, 0x2C, 0x36 (sometimes), 0x39 and 0x145 are 15 bit.
Non-varaction2-calculated callback-results are also affected by the 8bit masking.
|
|
with some parts for AI use.
|
|
doing industry callback.
Some variables used were linked to invalid industry.
|
|
|
|
logical place and remove about 50% of the includes of 'functions.h'
|
|
-Refactor: some code so there is less code duplication.
|
|
to registers 0x100 to 0x10F.
|
|
12 and 13.
|
