| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| |
|
|
|
| |
The old fsim output wasn't the only instance: both acquirement
statistics and Xom debugging also use this filename extension.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
This is a NIH implementation of a part of autoconf, it is limited as it
would be hard to affect the Makefile from such tests. On the other hand,
even nasty hacks like grepping for sqlite features are not enough here, as
we're looking for something in system headers.
I think all of configuration in the Makefile should be nuked and rewritten,
so it's just a crutch for 0.11.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
This has the side effect that XL, char and place are now separate
columns in the Webtiles lobby.
|
| | |
|
| |
|
|
|
| |
Choosing just the name works only when no directory is specified, like it
was for lua tags.
|
| |
|
|
|
|
| |
It'd be a royal pain in the butt to have to manually generate it every time
you add a minor save tag. Also, Python is not easily available on most
platforms.
|
| | |
|
| | |
|
| |\
| |
| |
| |
| | |
Conflicts:
crawl-ref/source/Makefile
|
| | |
| |
| |
| |
| |
| | |
For now, our doclet monkey-patches itself over the standard one,
so that we can use the standard template files unaltered.
(They hard-code the full module name all over the place. Very messy.)
|
| | |
| |
| |
| |
| | |
Only one function is documented at all so far, but at least the
documentation is generated.
|
| |/
|
|
|
| |
It seems like this was the only Makefile in the whole tree that didn't
check for changes in CFLAGS between runs.
|
| | |
|
| | |
|
| |
|
|
|
| |
To update it from the wiki, please run "make rest" by hand. It does web
scraping and thus might be vulnerable to changes in doku, beware!
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The old noise propagation system alerted all monsters within `loudness' radius
of the noise source, ignoring obstacles between the observer and the noise
source.
The new noise system applies only to monsters; the player is still messaged as
before. This is for two reasons:
1. noisy() must return if the player heard the noise, which forces us to
propagate the noise immediately for each noise, an unacceptable overhead on
busy maps (such as Sprint). Therefore we cheat and use the old noise rules for
the player. This is relatively simple to work around, since not many code paths
rely on the noisy() return code, and these places can be changed to work around
this lack of information.
2. The new noise system allows loud noises to drown out quiet noises. This may
be an unacceptable loss of information for the player.
The new system keeps track of noises produced by noisy(), but does not
act on them immediately. Instead, all noises are processed in world_reacts.
Noises are processed by a simple flood-out from their sources, attenuated by
distance, terrain (such as closed doors, trees), and by twists in the noise's
path (a winding corridor attenuates noise more than a straight line). Noise
propagation and attenuation uses movement geometry, not LOS geometry, *but*
attenuation assumes it is harder for the noise to change direction than to keep
going in its old direction.
To simplify handling of noises, it is assumed that louder noises overwhelm
quieter noises. An actor exposed to both a loud and a quiet noise at the same
distance will hear only the loud noise. However, an actor closer to the quiet
noise than the loud noise will hear first the quiet noise, then the loud noise
(because the floodfill visits points at an increasing radius from each noise
source). Given two equally loud noises, the closer noise will always win.
Multiple quiet noises emitted close together do not combined into a louder
noise.
Monsters hearing a noise at a distance from its source may not correctly guess
the noise's origin if the noise did not travel in a straight line to the
monster.
Noises are currently processed before and after handle_monsters() so that
monsters can react to noises produced by the player and by each other. We could
conceivably drop one of the apply_noises() calls (probably the one after
handle_monsters()) without adverse effects.
Since the new noise system allows closed doors to cut off sound, I've dropped
the Sprint noise attenuation from 1/3 to 1/2.
If Crawl is compiled with DEBUG_NOISE_PROPAGATION, it will write a
noise-grid.html dump file showing noise sources and intensities on the map for
each call to apply_noises that processed a noise.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Since the current makefile allows all dependencies to be statically linked,
creating an app bundle is as simple as setting up the directory tree.
The baroque Xcode project shenanigans that dynamically link SDL and other libs
and then use relative framework references in the .app bundle should no longer
be necessary.
Running ./mac/mac-release-build will create universal (i386, ppc) app bundles
for tiles and console and drop the zips in mac-app-zips. I've deliberately
skipped x86-64 for now (compile-time and SDL build issues with the 10.4u SDK).
|
| | |
|
| |
|
|
| |
This should also allow us to draw clouds beneath monsters.
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
considerably reduces heap use in the first Crawl process to run in a fresh install.
|
| |
|
|
|
|
| |
Due to the set of coloured floors and walls, it was getting perilously
large. The tiledef-dngn.h file now just indirectly calls the correct
feat/floor/wall functions depending on which tile index is used.
|
| |
|
|
|
|
| |
rltiles/tiledef-unrand.cc from repository.
art-data.pl no longer modifies artefact.h, but writes a new file called art-enum.h.
|
| |
|
|
| |
Signed-off-by: Steven Noonan <steven@uplinklabs.net>
|
| |
|
|
| |
Signed-off-by: Eino Keskitalo <evktalo@users.sourceforge.net>
|
| |
|
|
|
|
| |
tests, but can be parameterised.
Add a script to generate 150 level at a named place and report on all the monsters generated there.
|
| | |
|
| | |
|
| |
|
|
|
| |
I'm committing this to master instead of the debian branch because
switching between branches doesn't purge untracked files.
|
