Re-use the array allocated for iterating over all monsters during
monster movement much of the time. It was being allocated from
scratch for each round of monster movement, then freed after they
moved, then repeated the next round.
While hunting for a memory leak in object allocation--which I haven't
found yet--I discovered one in monster movement. iter_mons_safe()
allocates an array of (monst *) pointers for the monsters on the
current level, loops over that array to call a function for each
one, then frees the array. But if the game ends while that called
function is running, execution never returns to iter_mons_safe() so
it wasn't able to free the memory.
Since that can happen at most once per game, it wasn't a signifcant
leak. This fixes it anyway.
There was a second issue: make sure that iter_mons_safe() doesn't
call alloc(0) to make the temporary array for zero monsters when
there aren't any on the level. That might not be able to happen for
monster movement but the routine is written to be more general than
just movement. alloc(0) could confuse the MONITOR_HEAP code. In
C89/C90 I think malloc(0) is allowed to return NULL (don't recall
for sure; maybe that was just known pre-standard behavior for some
implementations). Null return would trigger a panic even without
MONITOR_HEAP. Don't know about C99 and later.
Some functions are passed an obj or monst chain,
and the callers typically don't check them
against 0, so mark them explicitly as NO_NONNULLS
(NO_NONNULLS expands to nothing, but it flags that
some null arg analysis has been done)
The tracks left by hero were cleared when player saved and
restored the game, or changed levels. Now the tracks are
saved in the dungeon level, so changing levels keeps the tracks
left by hero in that level.
Also increased the length of tracks from 50 to 100, and
simplify the tracking function.
Thing not done: fade out old tracks when returning to a level.
Breaks saves and bones.
Adds a new lua command
des.exclusion({ type = "teleport", region = { x1,y1, x2,y2 } });
which allows defining "exclusion zones" in the level, areas where
random teleports (or falling into the level) will never place the hero.
Does not prevent targeted teleportation into the area.
Breaks saves and bones.
When returning to play from within the tutorial, remove the level files
similar to how they're discarded for the rest of the dungeon when going
into the endgame. It turned out to be a bit messier than anticipated.
The dungeon.c bit is sufficient for #overview, which now hides regular
level 1 while in the tutorial and hides all tutorial levels once exited.
Those will still appear in end-of-game disclosure.
Reported five months ago, a save was performed while a mounted hero
was engulfed. Restore issued a warning about the engulfer being
placed on top of the steed (who shouldn't have been on the map).
The report arrived at about the same time as engulfing a riding
hero was changed to force a dismount instead of engulfing both hero
and steed so nothing further was done about it. This changes
restore to not put a steed on the map and then take it off again.
It also attempts to simplify usteed and ustuck handling during save
and restore.
Testing so far indicates that things are still working correctly.
Keep makeplural(body_part(FINGER)) crossed.
Existing save and bones files are invalidated.
A number of C compiler suites have a math.h library that includes a yn()
function name that conflicts with NetHack's yn() macro:
"The y0(), y1(), and yn() functions are Bessel functions of the second kind,
for orders 0, 1, and n, respectively. The argument x must be positive. The
argument n should be greater than or equal to zero. If n is less than zero,
there will be a negative exponent in the result."
At one point, isaac64.h included math.h, although that has since been removed.
Some libraries used in NetHack (Qt for one) do include math.h and that required
build work-arounds to avoid the conflict.
Rename the NetHack macro from yn() to y_n() and avoid the math.h conflict
altogether, eliminating the need for that particular work-around.
Keep track of how a role|race|gender|alignment option got its value
so that role:!Tourist in .nethackrc and role:!Priest in NETHACKOPTIONS
yield 'role:!Priest' rather than merging into 'role:!Priest !Tourist'.
It also doesn't write the value into new config file for #saveoptions
if that value comes from environment or command line (not applicable
since the command line arguments for role,&c don't go through options
handling). Also, the old config file value takes precedence over
the current game's value file so that 'role:random' doesn't become
'role:Healer' or such in a new config after the random value gets
picked for play.
This only tracks the role, race, gender, and alignment options but the
concept could be extended to all options. The data would need to be
saved and restored if values set interactively need to be retained in
restore sessions (doesn't apply to role,&c since those don't change
during play).
The consolidation of global variables from scattered source
files into decl.c and declared in decl.h was begun in 3.7.0.
Their placement in common files was done for centralized
initialization and potential re-initialization during a
"play again" scenario.
It wasn't really necessary for all of them to be housed in a
single huge structure to meet the "play again" requirement,
and the single huge structure has been a little unwieldy when
it comes to maintenance.
Following this commit, instead of one single extremely large structure
named 'g' to house all of the relocated global variables, they
are distributed into several ga through gz.
To make things easy for the developer, each variable is placed
into the struct corresponding to the starting letter of the variable.
That way, no lookup is required in order to know which struct houses
a particular variable, it is a simple match to the starting letter
for all the centralized global variables.
A global variable named 'amulets', would be found in ga.
ga.amulets
^ ^
A global varable named 'move', would be found in gm.
gm.moves
^ ^
A global variable named 'val_for_n_or_more' would be found in gv.
gv.val_for_n_or_more
^ ^
A global variable named 'youmonst' would be found in gy.
gy.youmonst
^ ^
Trap doors saved their destinations as an absolute level, rather than a
relative one, so if you loaded bones from a special level their
destinations would reflect the dungeon layout from the bones player's
game. For example, die on the Oracle level, on dlvl5, with a trap door
that goes to dlvl6. Another player gets those bones on their Oracle
level, which is dlvl8... the trap door would still go to dlvl6. Pretty
amazing trap door -- something you might see in a funhouse!
Include relative rather than absolute destinations in save and bones
files, much like stairs do, to avoid this problem.
I bumped EDITLEVEL because although this won't break save files in an
obvious way, it will interpret the (absolute) destinations in existing
save and bones files as relative, leading to some crazy long falls. :)
The pull request included some changes that were neither accidental nor
unintentional, so only a subset of the changes from pull request #869
submitted by klorpa were manually applied.
behaviour -> behavior
speach -> speech
knowlege -> knowledge
incrments -> increments
stethscope -> stethoscope
staiway -> stairway
arifact -> artifact
extracing -> extracting
The uses of "iff" were left alone.
Close#869
Make sure u.uswallow is cleared when u.ustuck gets set to Null so
that they won't be out of sync with each other. Having u.uswallow
be non-zero does imply that u.ustuck is non-Null.
Running #panic while swallowed didn't produce any anomalies for me,
either before or after this change.
Add a #saveoptions extended command, to allow saving configuration
settings from within the game. This is still highly experimental,
and gives plenty of warnings before asking to overwrite the file.
Lack of option saving is one of the biggest complaints new players
have, so this should help with it. More experienced players with
highly customized config file should not use this feature, as it
completely rewrites the file, removing all comments and non-config
lines.
Stair dlevels weren't being restored with the correct values when
recovered after the game crashed, apparently because they weren't being
reset back to their 'absolute' level from a 'relative' level. I'm not
totally sure of why this affected only recovered games (maybe that's the
only time when the 'relative' stair values are used?) but this fix seems
to work.
Fixes#812
One of the drivers of this change was that screen coordinates require a
type that can hold values greater than 127. Parameters to the window
port routines require a large type in order to be able to have values
a fair bit larger than COLNO and ROWNO passed to them, particularly for
their use to the right of the map window.
This splits the uses of xchar into 3 different situations, and adjusts
their type and size:
xchar
|
-----------------------
| | |
coordxy xint16 xint8
coordxy: Actual x or y coordinates for various things (moved to 16-bits).
xint16: Same data size as coordxy, but for non-coordinate use (16-bits).
xint8: There are only a few use cases initially, where it was very
plain to see that the variable could remain as 8-bits, rather
than be bumped to 16-bits. There are probably more such cases
that could be changed after additional review.
Note: This first changed all xchar variables to coordxy. Some were
reviewed and got changed to xint16 or xint8 when it became apparent that
their usage was not for coordinates.
This increments EDITLEVEL in patchlevel.h
Reported by entrez: fix memory being accessed after having been
freed by trying to keep ball&chain data up to date when they're
processed by the save code. This fix is a little more elaborate
than this suggested one. I'm crossing my fingers on this one....
savelev() gets run to clean up memory even if the player quits before
level 1 is created, and a change made yesterday panicked if it couldn't
figure out what level the hero is on. Caught by entrez, again....
If not actually writing a level's file, don't panic if both u.uz and
g.uz_save are 0. Having one of those be non-zero is only essential
when the level being processed is the Plane of Water or Plane of Air.
The air bubbles on the Plane of Water and the clouds on the Plane of
Air were being saved and restored as part of the current level's state
(which is the 'u' struct and invent and such) rather than with the
current level itself. That was ok for normal play, but for wizard
mode's ^V allowing you to return to a previously visited endgame level
after moving to a different one it meant a new set of bubbles for
Water and new set of clouds for Air. Even that was ok since it only
applied to wizard mode, but using #wizmakemap to recreate Water or Air
while you were on it added a new set of bubbles or clouds to the
existing ones. If repeated, eventually there wouldn't be much water
or air left.
Instead of just adding a hack to #wizmakemap, change save/restore to
keep the bubbles/clouds with the level rather than with the state.
That wasn't trivial and now I know why the old odd arrangement was
chosen. Saving hides u.uz by zeroing it out for levels that the hero
isn't on and it is zero during restore so simple checks for whether a
given level is water or air won't work.
This also adds another non-file/non-debugpline() use of DEBUGFILES:
DEBUGFILES=seethru nethack -D
will make water and clouds be transparent instead of opaque. It also
makes fumaroles and other light-blocking gas clouds be transparent
which wasn't really intended, but avoiding it would be extra work that
doesn't accomplish much.
Increments EDITLEVEL for the third time this week....
Handle thrown or kicked object that's in transit for hangup save and
panic save in addition to normal end of game. Affects ball and chain
placement too, if they've been temporarily taken off the map.
MONITOR_HEAP+heaputil pointed out some unreleased memory. The livelog
stuff wasn't being freed. Not surpringly the data used for collecting
and formatting build-options that just got changed from strdup() to
dupstr() wasn't being freed. And a couple of date/version bits.
Log game events, such as entering a new dungeon level, breaking
a conduct, or killing a unique monster, in a new "Major events"
chronicle. The entries record the turn when the event happened.
The log can be viewed with #chronicle -command, and the entries
also show up in the end-of-game dump, if that is available.
This feature is on by default, but can be disabled by
defining NO_CHRONICLE compile-time option.
This also contains "live logging", writing the events as they
happen into a single livelog-file. This is mostly useful for
public servers. The livelog is off by default, and must be
compiled in with LIVELOG, and then turned on in sysconf.
Mostly this a version of livelogging from the Hardfought server,
with some changes.
... instead of hard-coding them to 50. New allocated value is
(COLNO*ROWNO)/30, which is slightly higher (56), and that formula
seems to work for hypothetical larger maps too.
Instead of returning 0 or 1, we'll now use ECMD_OK or ECMD_TURN.
These have the same meaning as the hardcoded numbers; ECMD_TURN
means the command uses a turn.
In future, could add eg. a flag denoting "user cancelled command"
or "command failed", and should clear eg. the cmdq.
Mostly this was simply replacing return values with the defines
in the extended commands, so hopefully I didn't break anything.
Function the() wasn't supposed to be used for monsters because many
of the ones with capitalized names confuse it, but over time multiple
instances of the(mon_nam()) have crept into the code. Instead of
ripping those out, modify the() to handle that situation better.
Pull request #636 by entrez dealt with this with one extra line of
code, but could end up scanning all the names in mons[] repeatedly
if the("Capitalized string") gets called a lot. This uses a similar
one line fix but calls a whole new routine that scans through mons[]
once collecting all the relevant special case names. As a bonus,
it does the same for hallucinatory monster names which name_to_mon()
couldn't handle.
Fixes#626
It's redundant with g.moves, so there is no more need for it.
Way, way back, it looks like g.moves and g.monstermoves can and did
desync, where g.moves would track the amount of moves the player had
gotten (and would therefore increase faster if the player were hasted)
and g.monstermoves would track the amount of monster move cycles, aka
turns. But this has not been the case for a long time, and they both
increment together in the same location in allmain.c. There are no
longer any cases where they will not be the same value.
This is a save-breaking change because it changes struct
instance_globals, but I have not updated the editlevel in this commit.
The code has been assuming that time_t is some number of seconds.
That's valid for traditional Unix systems and for Posix compliant
systems but is not something guaranteed by the C standard. (We ran
into a long time ago when trying out an alternate way to calculate
phase of moon. That code made a similar assumption and broke one
of the ports.)
'ubirthday' also warrants being re-done but I've run out of energy.
Add des.finalize_level() used for testing in conjunction with
des.reset_level().
Add nhc.DLB to return 0 or 1 if DLB was defined at compile-time.
Change the test_lev.lua to give more informative messages instead of
just lua error when required file doesn't exist.
Add bigrm-11 to the level tests.
Restoring a level cleared residence pointer from those subrooms
which were in the restored level, but if the previous level had
more subrooms, those pointers weren't cleared.
This caused weird problems when a shopkeeper data was looked
up based on the subroom number.
Just to be safe, clear all the room data when freeing the level.
add MALE, FEMALE, and gender-neutral names for individual monster species
to the mons array. The gender-neutral name (NEUTRAL) is mandatory, the
MALE and FEMALE versions are not.
replace code uses of the mname field of permonst with one of the three
potentially-available gender-specific names.
consolidate some separate mons entries that differed only by species into a
single mons entry (caveman, cavewoman and priest,priestess etc.)
consolidate several "* lord" and "* queen/* king" monst entries into
their single species, and allow both genders on some where it makes some
sense (there is probably more work and cleanup to come out of this at some
point, and the chosen gender-neutral name variations are not cast in stone
if someone has better suggestions).
related function or macro additions:
pmname(pm, gender) to get the gender variation of the permonst name. It
guards against monsters that haven't got anything except NEUTRAL naming
and falls back to the NEUTRAL version if FEMALE and MALE versions are
missing.
Ugender to obtain the current hero gender.
Mgender(mtmp) to obtain the gender of a monster
While the code can safely refer directly to pmnames[NEUTRAL] safely in the
code because it always exists, the other two (pmnames[MALE] and
pmnames[FEMALE] may not exist so use:
pmname(ptr, gidx)
where -ptr is a permonst *
-gidx is an index into the pmnames array field of the
permonst struct
pmname() checks for a valid index and checks for null-pointers for
pmnames[MALE] and pmnames[FEMALE], and will fall back to pmnames[NEUTRAL] if
the pointer requested if the requested variation is unavailable, or if the
gidx is out-of-range.
Allow code to specify makemon flags to request female or male (via MM_MALE
and MM_FEMALE flags respectively)to makedefs, since the species alone doesn't
distinguish male/female anymore. Specifying MM_MALE or MM_FEMALE won't
override the pm M2_MALE and M2_FEMALE flags on a mons[] entry.
male and female tiles have been added to win/share/monsters.txt.
The majority are duplicated placeholders except for those that were
separate mons entries before. Perhaps someone will contribute artwork in the
future to make the male and female variations visually distinguishable.
tilemapping via has the MALE tile indexes in the glyph2tile[]
array produced at build time. If a window port has information that the
FEMALE tile is required, it just has to increment the index returned
from the glyph2tile[] array by 1.
statues already preserved gender of the monster through STATUE_FEMALE
and STATUE_MALE, so ensure that pmnames takes that into consideration.
I expect some refinement will be required after broad play-testing puts it to
the test.
consolidate caveman,cavewoman and priest,priestess monst.c entries etc
This commit will require a bump of editlevel in patchlevel.h because it alters
the index numbers of the monsters due to the consolidation of some. Those
index numbers are saved in some other structures, even though the mons[] array
itself is not part of the savefile.
Window Port Interface Change
Also add a parameter to print_glyph to convey additional information beyond
the glyph to the window ports. Every single window port was calling back to
mapglyph for the information anyway, so just included it in the interface and
produce the information right in the display core.
The mapglyph() function uses will be eliminated, although there are still some
in the code yet to be dealt with.
win32, tty, x11, Qt, msdos window ports have all had adjustments done to
utilize the new parameter instead of calling mapglyph, but some of those
window ports have not been thoroughly tested since the changes.
Interface change additional info:
print_glyph(window, x, y, glyph, bkglyph, *glyphmod)
-- Print the glyph at (x,y) on the given window. Glyphs are
integers at the interface, mapped to whatever the window-
port wants (symbol, font, color, attributes, ...there's
a 1-1 map between glyphs and distinct things on the map).
-- bkglyph is a background glyph for potential use by some
graphical or tiled environments to allow the depiction
to fall against a background consistent with the grid
around x,y. If bkglyph is NO_GLYPH, then the parameter
should be ignored (do nothing with it).
-- glyphmod provides extended information about the glyph
that window ports can use to enhance the display in
various ways.
unsigned int glyphmod[NUM_GLYPHMOD]
where:
glyphmod[GM_TTYCHAR] is the text characters associated
with the original NetHack display.
glyphmod[GM_FLAGS] are the special flags that denote
additional information that window
ports can use.
glyphmod[GM_COLOR] is the text character
color associated with the original
NetHack display.
Support for including the glyphmod info in the display glyph buffer
alongside the glyph itself was added and is the default operation.
That can be turned off by defining UNBUFFERED_GLYPHMOD at compile time.
With UNBUFFERED_GLYPHMOD operation, a call will be placed to map_glyphmod()
immediately prior to every print_glyph() call.
