"were{rat,jackal,wolf}" each occur twice in mons[], once for the
beast form and second time among '@' for the human form. Wishing
for werecreature corpse or tin always matches the first entry so
yields the beast form, but all their beast forms are flagged as
no-corpse so the wish would fallback to a corpse with random monster
type. Wishing for werecreature figurine worked but always produced
one that created its beast form if/when activated.
This fix allows specifying "human werecreature" to match the second
entry. It's optional for corpse and tin; the wish code will now
switch to that implicitly if it gets a no-corpse were-form for
those. It has to be specified explicitly to get a figurine that
will activate as the human form. It works for ^G too.
name_to_mon() has a bunch of alternate monster names, such as
"gray-elf" to match "grey-elf" and "ki rin" to match "ki-rin". Those
worked as intended when they occurred at the end of a wish, but only
worked in the middle if their length was the same or one character
less than the canonical name in mons[].mname.
djinni figurine -> h - a figurine of a djinni
genie figurine -> i - a figurine of a djinni
figurine of mumak -> j - a figurine of a mumak
mumak figurine -> k - a figurine of a mumak
figurine of mumakil -> l - a figurine of a mumak
mumakil figurine -> nothing fitting that description exists
(The one-less case worked because its following space ended up being
implicitly removed when skipping ahead by the length of mons[].mname;
subsequent explicit removal didn't find a space so was a no-op.)
The bases[] array allows finding the index of the first object in
a particular class. Extend it so that bases[class + 1] - 1 is a
reliable way to find the last object in any class. The array had
to be extended by one so that the last class has a [class+1] entry
available, and object initialization now makes sure that classes
within objects[] are in ascending order so that [class+1] always
holds a higher index than [class].
Water locations on Medusa's level didn't show steam clouds. It
wasn't because the location was a moat rather than a pool, it was
because the moat location was unlit (and in line of sight) and
tested pool locations were lit. Poison gas clouds explicitly
override the lit/unlit issue but other region types weren't.
Pasi Kallinen [Sat, 18 Apr 2020 13:51:35 +0000 (16:51 +0300)]
Fix segfault on teleport while punished due to flooreffects
... deleting the ball & chain, but keeping a boulder in the pit.
Noticed a segfault when fuzzing, teleport while punished caused
a segfault via fill_pit -> flooreffects -> bury_objs -> unpunish,
and then the next line in teleds tried to look up uchain.
Guard against that particular case.
Fix the case of boulder being in a pit, triggered by you being in
a pit and a giant throwing a boulder on top of you.
This adds a superset of the code from github pull request #328
to create a short-lived cloud of steam when fire hits a pool or
fountain. The original code required C99; this doesn't. It also
allowed vapor clouds on the Plane of Water where they don't work
sanely becaure regions don't understand air bubble movement and/or
vice versa. This inhibits the clouds there [the same ought to be
done for scrolls of stinking cloud]. It also left as-is the code
that reported when fountains got used up even though conceptually
the steam should interfere with being able to see that. This adds
a glyph-check hack to augment cansee() so that fountains that boil
away entirely are hidden at the time.
Regions that block line of slight should be calling block_point()
when created and unblock_point() when removed but a naive attempt
to introduce that didn't work as expected so I'm giving up on.
Screen erasure leaves the map set to spaces. If S_unexplored is
something other than <space>, tty wasn't drawing with S_unexplored
after a menu or long message line got erased following temporary
overwrite of part of the map.
This seems to work but is not the correct way to do things.
clear_screen(), cl_eos(), and cl_end() should all be taught to
flag the map as needing to be refreshed after they erase part of it.
tty_clear_nhwindow(WIN_BASE) is also lacking since it erases the
message line, full map, and status lines but leaves their internal
windows with stale data about what is shown instead of marking them
blank.
Pasi Kallinen [Thu, 16 Apr 2020 18:01:24 +0000 (21:01 +0300)]
Dehardcode the monk minetown food shop conversion
Instead of trying to figure out in core whether to change a minetown
food shop to health food shop for monks, just figure it out in the
minetown level creation script.
Make a mimic that picks special VEGETARIAN_CLASS look like food
instead of a random object. Also, at least a couple of Minetown
variants are classified as mazes so had a 50% chance of not even
making that choice and look like a statue instead. Override maze
when 'in_town()' yields True.
When a food shop gets converted into a health food shop (minetown
when playing as a monk), the shop type and underlying room type
weren't changed to match.
This started with fixing a warning suggesting parentheses when
mixing || and &&, then started sprawling. The getpos help for x/X
was misleading. It said that using x or X would move cursor to the
next unexplored location but it actually moves the cursor to next
explored location that's adjacent to an unexplored one.
Pasi Kallinen [Wed, 15 Apr 2020 08:02:33 +0000 (11:02 +0300)]
Mimic in a health food store could cause a panic
Health food stores use a special "vegetarian class" of items,
which mkobj doesn't understand. Just make the mimic use a random
item class in that case.
copperwater [Mon, 28 May 2018 23:36:23 +0000 (19:36 -0400)]
Change logic for immobilized displacement, fixing a bug
The bug was that you could easily displace peaceful sleeping monsters.
The source of this was that there was no check for sleeping in the
attack() function that checks for immobilization.
As well as adding the sleeping check, this logic is modified so that it
makes more sense: if a monster is immobilized at all (sleeping, frozen,
or mcanmove = 0) it always "doesn't seem to move!" You can't randomly
displace it anyway 1/6 of the time.
Sessile monsters who are otherwise not immobilized don't move 5/6 of the
time, but can be displaced with the other 1/6.
copperwater [Tue, 20 Mar 2018 16:17:38 +0000 (12:17 -0400)]
Pet displacement improvements
Pets can no longer be displaced out of a trap, because that was
inconsistent with peaceful monsters refusing to be displaced out of a
trap. The untaming-via-displacement-out-of-trap code is removed.
Pets also now have a better survival instinct: they follow the code for
peaceful displacement into a bad position, and refuse to swap places.
This means it's no longer possible to accidentally kill a pet by
levitating/walking over water and displacing it.
copperwater [Mon, 12 Mar 2018 02:20:32 +0000 (22:20 -0400)]
Fix: rn2(0) when trying to displace peaceful out of trap
A side effect of making is_safepet() count peacefuls. Now checks
directly for a trapped, peaceful monster and says "they can't move out
of the trap"; this is inconsistent with pet behavior, and pet behavior
should probably be changed to be in line with it (ie they can't be
displaced out of a trap at all.)
Also refactor the code here a bit: a bunch of different if statements
have the exact same resetting code and steed resetting code in them.
Change this to a boolean flag and put the resetting code in one place
checked by that flag.
copperwater [Sat, 24 Feb 2018 04:02:29 +0000 (23:02 -0500)]
More peaceful monsters that can't be displaced: priests, Oracle
Shouldn't be able to displace priests since that could theoretically
eventually force them out of their temple, which would probably cause
problems. The Oracle doesn't usually move anyway, but seems like she
should "not want to swap places" in any circumstance.
copperwater [Fri, 8 Dec 2017 04:26:17 +0000 (23:26 -0500)]
Player can now displace peaceful monsters
Changes domove() code to allow displacing peaceful monsters.
Specifically, is_safepet() now returns true if the monster is peaceful.
Peacefuls are slightly pickier than pets about whether they consent to
being displaced: they will not displace if a goodpos() check fails for
the displaced space, or if there is a trap on the displaced space, or if
they are your quest leader. is_safepet should probably be renamed to
something else.
In the process of doing this, some other changes were made: the code now
checks whether the player and monster should be swapping places at all
first (previously it ran some code for displacing pets out of traps
first, which was a little weird if the displacement didn't actually
happen.)
In the original commit for this, I needed to guard the spoteffects()
call made in domove with a clause testing whether the player actually
moved; it was previously possible to fail to displace a monster and then
re-trigger a trap on the space you were still standing on. However, the
devteam has apparently put in an if (u.umoved) clause in the same place
and serving the same purpose.
Pasi Kallinen [Wed, 15 Apr 2020 04:15:37 +0000 (07:15 +0300)]
Fix "x" and "a" when choosing map location
Recent(ish) change to split unexplored glyphs from solid wall glyphs
resulted in getloc commands to choose unexplored and interesting
locations not work correctly.
Allow specifying a montype on figurines in lua files
This was already allowed for the other montype-compatible objects like
statues, corpses, eggs and tins. I don't see a reason why figurines
shouldn't be part of this group; perhaps it was an oversight.
Move demon summoning and were creature summoning out of mattacku()
to a separate routine. Should be no change in behavior.
Also, redo several comments that came after what they applied to
rather than before, and reorder the file-scope prototypes to be in
the same order as the corresponding functions.
When rest and search refuse to operate because a hostile monster is
adjacent, include a reminder of how to force them to operate. Every
time if 'cmdassist' is On, or just once until after some subsequent
try actually does something.
This new rest and search behavior probably needs to be optional and
default to the old behavior. It isn't uncommon to deliberately rest
while adjacent to a hostile monster if also adjacent to a peaceful
one and trying to wait for Stun or Confusion to time out, or maybe
search while next to such a monster hoping to find a secret door to
run away through. A count prefix won't work and needing an extra
keystroke each time is going to be an annoyance.
When polymorphed into a nymph and melee hit steals items from the
target, the same-gender charm message vs opposite-gender seduce
message is being chosen by hero's base gender rather than nymphs
always being female. The seduce message used dynamic pronouns for
the target monster but the charm message used hardcoded She and her
for female nymph attacking female target. I'm not sure why hero's
base gender is used so left that alone; this changes charm message
to use dynamic pronouns that correctly match the target monster.
Pasi Kallinen [Mon, 13 Apr 2020 18:49:16 +0000 (21:49 +0300)]
Scatter bag of holding contents instead of deleting them
Losing all of your items in one go is really frustrating.
If you blow up your bag of holding, make the contents scatter around
instead of outright deleting them. This will destroy fragile objects.
Pasi Kallinen [Mon, 13 Apr 2020 16:59:03 +0000 (19:59 +0300)]
Allow matching any wall map terrain in lua scripts
When matching a terrain, allow using a "w" placeholder that matches
any solid wall:
For example:
local s = selection.match([[w.w]]);
would match all floor locations with a wall to the left and right of it.
The walls can be solid stone, horizontal, vertical, etc.
This applies to selection.match(), selection.filter_mapchar(), and
des.replace_terrain()
Pasi Kallinen [Sun, 12 Apr 2020 14:21:23 +0000 (17:21 +0300)]
Prevent searching or waiting next to a hostile
Generally speaking there's no reason to wait or search next to
a hostile monster, so let's just prevent those actions. You can
still do those commands by prefixing them with the 'm' prefix.
Instead of just picking up wands of undead turning when one or more
corpses are in hero's inventory, have monsters also pick those up if
hero has a cockatrice corpse in a carried container. And instead of
only zapping such wands (as defensive measure) when hero is wielding
a cockatrice corpse, also zap them (as offensive measure) if hero is
carrying any corpse in open inventory.
Testing a forthcoming extension of monsters using wands of undead
turning revealed a couple of pre-existing bugs. Previously only
noticeable if hero zaps self or breaks a wand of undead turning so
unlikely to have happened much.
"Your <mon> corpse comes alive" was given even if it was revived
as an undead. Also, it was "your <mon> corpse" instead of "one of
your <mon> corpses" even when one from a stack was involved. If
done by hero to self that message follows "one of your <mon>
corpses glows iridescently" so the comes alive message was ok but
verbose. Change that to "it comes alive" or "it reanimates" when
following the glow message.
when hero is wielding a cockatrice corpse. Wands of undead turning
aren't generated as starting equipment but they will now be picked
up if come across while the hero is carrying any corpse, and used
in preference to any other item when carried and non-empty and hero
is wielding a petrifier's corpse.
m_lined_up() was declared 'boolean' but returned 0, 1, or 2.
The 1 case isn't actually used any more. I changed it to 'int'
rather than 2 to TRUE; it could just as easily be the other way
'round.
Avoid giving "you are back on the bottom" nearly every step when
moving around underwater.
Avoid "you are back on floor" followed by "you trip over <object>"
when fumbling in case that fumbling was due to being on ice when
taking the step to floor. Done for all fumbling rather than just
one-turn fumbling instigated by ice.
When moving from ice or water to ground, show "you are back on floor"
before listing objects at that spot instead of after.
I think there was at least one more thing but have lost track. At
any rate, 'mention_rate' potentially has a new set of bugs.
"You materialize at another location," was delivered while the
previous location still controlled line of sight. Very noticeable
if you started from underwater and landed on the surface in an area
which hadn't been mapped yet.
The fix to try to avoid messages about out-of-view objects taking
erosion damage made water_damage_chain() vulnerable to dereferencing
a null pointer, leading to a crash if you create a pool via wizard
mode wishing.
Eating a tin of one of the Riders and being life-saved or declining
to die would crash when trying to revive a non-existent corpse. An
old comment stated that since such tins were impossible it could
assume that it was dealing with a corpse, but wishing for tins of
the Riders is possible in wizard mode.
They can't be passed along to normal mode via bones because they're
changed to empty tins as bones are saved. So there doesn't seem to
be much point in allowing wizard mode wishing to create them, but
I've left that as is.
I got "The chain mail rusts." seemingly out of the blue, then when
moving around the corner of the building on Valk home level I saw a
spot of remembered ice be redrawn as water. Before that I checked
for any mapped objects (via ';' 'o' 'o' ... so I didn't overlook
anything; there were only a couple of objects shown on the map and
none of them were piles) and didn't see any remembered chail mail or
anything at all on that ice spot, so I'm assuming that it was carried
by a monster. I may be leaving out some steps in the call chain here:
erode_obj() uses bhitpos for visibility check of eroding objects not
carried by the hero or by a monster, with a comment expressing doubt
about doing that. It wouldn't have yielded the right answer for the
possible call chain here unless it got set by some monster activity.
I had been zapping a wand just before and bhitpos would have been set
to a coordinate I could see at the time, fooling erode_obj()'s check
if the value was stale.
Anyway, this only addresses objects eroded from flooreffects(),
water_damage_chain(), and fire_damage_chain(). There are lots of
other indirect calls to erode_obj().
When dual-wielding and you wield a different weapon, two weapon combat
was silently toggled off even when the new weapon was eligible to be
dual-wielded. If the verbose flag is On, explicitly tell the player
when wielding something toggles off two-weapon mode. Wielding '-' is
an exception because you already get told that you're empty handed.
Pasi Kallinen [Tue, 7 Apr 2020 14:23:18 +0000 (17:23 +0300)]
Fix monster booted into limbo trying to act
There's a chance that a tactics-using monster (eg. arch-lich)
teleporting next to you on a full level would get booted off
into limbo. After that, if the monster tried using an item
in the same action you'd get panic "relmon: mon not in list"
Fix this particular case by checking monster mstate.
Other cases may lurk in other places that eventually call mnexto.
When dual-wielding and both weapons use the same skill, the weapon
feedback by ^X was inadvertently skipping the part where it would
tell the player whether either the weapons' skill or two-weapon
skill or both had already been trained enough to to be advanced.
Add support for black&white ice (3.7.0 feature) similar to already
supported black&white lava: show in inverse video if it uses the
same character as floor (in the ice case; as water in the lava case).
Inverse for monster detection, black&white lava, and now black&white
ice was being done unconditionally but has been changed so that the
user can disable it by toggling the 'use_inverse' run-time option.
[Bug noticed in the process: if you move an inverse video cursor
onto inverse video detected monster/lava/ice (when doing farlook, for
instance), the cursor disappears. I'm not sure how to address that.]
The extra column that the core sometimes uses for bookkeeping and
that was drawn as stone until recently when changed to blank space
(an unintentional left margin) is now gone for both the tiles map
and the text map. It's still part of the internal data but the map
window width and the map rendering exclude it.
This was too easy. There are bound to be bugs lurking....
Revert the change from impossible() to panic() that I made recently
and add code to allow continuing to succeed. The raw_print issued
there doesn't seem to go anywhere, but the game switches from tiles
map to text map and runs sanely.
Also, there was a second place that initialized the text map to all
spaces rather than S_unexplored. Change it to call the routine that
was already fixed for this.
copperwater [Sat, 29 Feb 2020 02:19:50 +0000 (21:19 -0500)]
Implement dice and percent as global lua functions
Intended to simplify many of the math.random calls currently in use, and
make them more semantic and thus more readable.
The dice function d() takes either a two-argument form which is the same
as in the C source (number of dice, faces per die) or a one-argument
form that rolls a single die.
The percent(N) function returns true N% of the time.
copperwater [Sat, 29 Feb 2020 16:06:30 +0000 (11:06 -0500)]
Re-implement gradient selections
Uncomments and makes available selection.gradient in Lua. (The backend C
code for this still existed, it just wasn't used.)
The only valid way to specify a gradient is with a table. I considered
adding non-table versions, but decided that there are too many
independent variables that can be optional. A non-table version, without
named parameters, would be confusing to read, especially since most of
the arguments are ints.
Also adds an impossible in the (possibly unreachable) case that
selection_do_gradient gets called with a bad gradient type.
copperwater [Sun, 1 Mar 2020 20:20:16 +0000 (15:20 -0500)]
Print out Lua error message when encountering an error
Rather than spitting out an error code number that is not particularly
useful. The string contains the line number and the nature of the error,
which is much more useful for debugging or reporting an issue.
Report weapon skill in the ^X status section when dual-wielding,
The effective skill level is the lower of the weapon's skill and
two-weapon skill, separately for primary and secondary. It's a
much bigger chunk of code than most enlightenment/^X features so
I put it in its own routine.
Reject arrows and darts as candidates for wielding two weapons at
once.
Make the check for being able to two-weapon when polymorphed be more
robust. Instead of just testing whether the monster form's second
atttack is a weapon attack and then assuming that the first one is
too, test the first three to validate that at least two of those are
AT_WEAP. The existing code works but seemed fragile.
Get rid of a couple of unnecessary calls to set_twoweap() to clear
u.twoweap when using 'A' to 'take off' either weapon. setuwep() and
setuswapwep() don't do that, but they call setworn() which does.
Make the feedback when disarming with 'A' be more specific when you
are two-weaponing at the time either weapon is unwielded.
A couple of the new prototypes used 'char' where 'CHAR_P' is needed.
Also, move them out of middle of long block of command declarations.
I started to reorder the prototypes into the order in which those
functions appear in the file but gave that up pretty quickly.
X11 was still initializing a blank map to 'stone' instead of to
'unexplored'. When the core started forcing 'unexplored' as part
of cls(), you could see the S_stone background show up and then be
overwritten with S_unexplored.
Also, X11 is [still] drawing unused column 0. That was also 'stone'
but has been changed to 'nothing' so is now blank for both tiles map
and text map (regardless of S_unexplored value). The extra useless
column doesn't look too bad normally but does if a vertical scroll
bar is added to support a clipped map.
The impetus for this was to avoid ugly constructions such as the one
below (none of which currently appear in vanilla NetHack):
mongets(mtmp, LONG_SWORD);
struct obj* sword = m_carrying(mtmp, LONG_SWORD);
if (sword)
/* do thing to sword */
Most cases where mongets is used discard the returned value (which used
to be the created object's spe); the only places that do use it are the
series of statements that give various human monsters armor and prevent
them from getting too much armor. These statements included hardcoded
constants representing the base AC of the armor, which would have caused
discrepancies if armor's base AC were ever changed.
With mongets now returning a pointer to the created object, it can just
be passed into ARM_BONUS instead, which covers both the base AC and the
enchantment. (It will also cover erosion, if anyone ever decides that
armor should rarely generate as pre-eroded).
The overall algorithm is not changed by this; human monsters should
receive armor with the same probabilities as before.
copperwater [Thu, 31 Jan 2019 01:54:30 +0000 (20:54 -0500)]
Don't interrupt tin opening if being slimed and tin is chameleon
Another SliceHack feature. It's possible for you to eat the chameleon
tin and turn into a fiery monster that burns off the slime in its
natural form, either extremely luckily at random or if you have
polymorph control.
copperwater [Mon, 30 Oct 2017 23:22:44 +0000 (19:22 -0400)]
Blessed teleport scrolls now give a single controlled teleport
This buffs the blessed effect of the teleport scroll by providing the
reader control over their destination even if they lack teleport
control. This seems like it makes the blessed/uncursed distinction
actually meaningful, rather than mostly pointless.
copperwater [Fri, 19 Oct 2018 14:21:25 +0000 (10:21 -0400)]
Pets will not eat shapeshifter corpses except in extreme circumstances
Ported from SpliceHack, and generalized to all shapeshifters (Splice
only implemented it for chameleons). It's very aggravating when your
powerful but hungry pet chows down on a shapeshifter before you can stop
them and then turns into something much more useless, so this aims to
prevent that.
The extreme circumstances under which a pet will eat a shapeshifter are:
1. The pet is starving, and prefers polymorph to starvation
2. The pet's tameness is 1
The reasoning behind the second condition is that if you mistreat your
pet almost to the point of untaming it, it might want to take a chance
on turning into something that might get some more respect from you.
Practically, whenever this happens, this will result in the player now
owning a newly polymorphed and *still* nearly feral pet.
Compiler complained about 'ptr' possibly being used unitinialized
in meatcorpse() and in this case it was right. meatcorpse() was
cloned from meatobj() but the necessary initialization was missing.
Purple worm would devore an entire stack of corpses in one bite.
Split one off and have it eat that instead.
I'm not sure whether attempting to revive a Rider corpse can force
a monster off the level to make room. If so, meatobj() and
meatcorpse() weren't prepared to handle that, nor was their caller.
It appears that the monster (either g.cube or purple worm) will
only eat as it moves so can't revive a Rider on a completely full
level since it won't be able to move in that situation. I fixed
the caller to be prepared to handle a result of 3 (no further
action allowed) instead of just dealing with 2 (died), but I didn't
fix either of the meatfoo() routines to return 3 since bumping the
eating monster off the level seems to not be possible.
Don't let purple worms eat lichen corpses, regardless of whether
they'll swallow live ones.
Make demon lords hostile if wielding Demonbane as well as Excalibur
This makes a lot of sense. Why would they hate one artifact sword so
much and not really care about the one that is especially designed to
kill their type personally?
copperwater [Wed, 1 Nov 2017 02:15:44 +0000 (22:15 -0400)]
Scroll of remove curse becomes learned when items' curses are removed
The scroll of remove curse is trivially identified by checking inventory
after reading it to see whether anything became uncursed. This leads to
annoying tactics like remembering which scroll you just read so you can
go call it "remove curse" on the discoveries list.
This simply autoidentifies it when an item that was known to be cursed
has its curse removed.