... 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.
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.
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.
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:
melt_ice -> minliquid -> mondead -> m_detach -> relobj -> mdrop_obj
-> flooreffects -> water_damage -> erode_obj
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 swallowed by an air elemental, going down into a pit
placed the attached iron ball on the floor. Saving (or
using #wizmakemap) then deallocated the iron ball.
Check being swallowed before trying to go down.
Instead of forgetting maps and objects, make amnesia forget skills.
Forgetting maps and objects could be circumvented with taking notes,
or by using an external tool to remember the forgotten levels.
Forgetting skills allows the player to optionally go down another
skill path, if they trained the wrong weapon in the early game.
Amnesia still forgets spells.
As a replacement for the deja vu messages when entering a forgotten
level, those messages will now indicate a ghost with your own name
existing on the level, given only when the level is entered for
the first time.
These changes based on fiqhack, with some adjustments.
In wizard mode, if you angered your quest leader and got expelled from
the quest (which removes the quest portal from the dungeon), then used
the wizmode level teleport menu to teleport back to the quest start and
talked with the angry leader, you would get booted back to the dungeon,
and the game panicked because the dungeon end of the portal didn't
exist anymore.
Eliminate the cache that was supporting rndmonst() and pick a random
monster in a single pass through mons[] via "weighted reservoir
sampling", a term I'm not familiar with.
It had a couple of bugs: if the first monster examined happened to
be given a weighting of 0, rn2() would divide by 0. I didn't try
to figure out how to trigger that. But the second one was easy to
trigger: if all eligible monsters were extinct or genocided, it
would issue a warning even though the situation isn't impossible.
Aside from fixing those, the rest is mostly as-is. I included a bit
of formatting in decl.c, moved some declarations to not require C99,
and changed a couple of macros to not hide and duplicate a call to
level_difficulty().
Fixes#286
Mounting a steed while legs are wounded would offer to cure them
but wasn't going through the heal_legs() routine so didn't update
the status line when Wounded_legs condition is enabled.
Move some common code for describing left/right/both legs into a
new routine used for feedback by jumping, kicking, and ridiing.
For ^X, distinguish between one wounded leg and both but don't
bother with left vs right when it is just one.
Setting or clearing u.ustuck now requires that context.botl be set,
so make a new routine to take care of both instead of manipulating
that pointer directly.
Several conditions result in stale data on the status line when
starting or stopping because things which didn't used to affect it
haven't been setting context.botl to force an update. This wasn't
systematic; there are bound to be lots more.
Introduce eight achievements that can be attained by more players.
Entered Gnomish Mines - self explanatory
Entered Mine Town - the town portion, not just the level
Entered a shop - any tended shop on any level
Entered a temple - likewise for temple
Consulted the Oracle - bought at least one major or minor oracle
Read a Discworld Novel - read at least one passage
Entered Sokoban - like mines
Entered the Big Room - not always possible since not always present
The novel and bigroom ones aren't always achieveable since novels are
only guaranteed if a book or scroll shop gets created and bigroom is
only guaranteed in wizard mode. No one ever claimed that every
possible achievement can be attained in a single game. (If one for
entering the Fort Ludios level--or perhaps entering the Fort itself--
eventually gets add, that won't be possible in every game either.)
The mine town one probably needs some tweaking. Two of the town's
seven variants have no town boundary (despite a rectangular area of
pre-defined map) and at present simply arriving on either of those
levels is enough to be credited with the entered-town achievement.
Bump EDITLEVEL because u.uachieved[] has increased in size. This
time it has been expanded to the maximum that xlogfile's bitmask of
achievements can handle, enough for up to 9 more achievements without
another EDITLEVEL increment.
Instead of an assortment of bits, assign numeric indices to the
potential achievements and keep an array of those in the order they
were attained. So disclosure might show the same subset occurring
differently in different games depending on the player's actions.
The encoded field in xlogfile doesn't care about that and remains
the same.
Modifies 'struct u', so EDITLEVEL has been incremented and existing
save files are invalidated.
While testing the changes to dungeon and special level handling, I got
|A mysterious force prevents you from descending!
|You materialize on a different level!
The mystery force is handled by goto_level() so level_tele() doesn't
know that the failure is going to happen when it sets up the message
for deferred delivery. Suppress the message if you don't change levels.
If level teleport took you somewhere special, things like shop entry
or quest summons or valley atmosphere messages were being given first,
then "you materialize on a different level" after. schedule_goto()
takes before-change-message and after-change-message arguments; it
ought to be extended to have a mid-change-message one too but I didn't
think of that until just now.
Subtracting one dungeon depth value from another had the subtraction
backwards and that yielded a negative value where a positive one is
expected. If NH_RELEASE_STATUS were to be set to NH_STATUS_RELEASED
then this was at risk of crashing (if the bad subtraction yields -2,
rn2(diff+2) would divide by 0) since rn2()'s argument isn't validated
for released version.
fixes37.0 was confused, listing a couple of things that aren't bugs
in 3.6 as general fixes. I suspect that the DLB one was fixed before
being exposed via git, so shouldn't be there at all.
Poly'd hero hiding on the ceiling was told "you can't go down here"
if using '>' at a spot that didn't have down stairs, trap door, hole,
or pit. Let '>' bring a ceiling hider out of hiding; lurker above
resumes flying, piercer falls to floor or whatever is underneath it.
Izchak implemented the mysterious force and as far as I'm concerned,
it's here to stay. But it can be fine tuned. This is an experimental
attempt to make it happen less. Each time it happens, the chance for
it happening again later will usually go down by an amount proportional
to how far it sent the hero back. So chaotics will be sent back--or
"side to side"--less often than in 3.6.x but the tapering off of such
occurrences will be slower for them. Lawfuls will also be sent back
less often--still potentially farther down than others--but tapering
off of send backs for them will be quicker.
I'll let somebody else figure out the before and after values for
number of attempts to climb up it takes to finally get out of Gehennom.
The numbers might need tuning.
do.c:1005:54: warning: address of function 'uescaped_shaft' will always
evaluate to 'true' [-Wpointer-bool-conversion]
if (trap && (uteetering_at_seen_pit(trap) || uescaped_shaft)) {
~~ ^~~~~~~~~~~~~~
A reddit thread about an unaligned altar in an aligned temple was
a tipoff that mimics posing as altars didn't have any particular
alignment. The look-at code was misusing an operloaded field of the
underlying terrain. Pick an alignment at random when taking on the
appearance of an altar, store it in the mimic's mon->mextra->mcorpsenm
field, and have look-at use that.
Also, dropping a ring of polymorph into a sink can transform it, and
one possible outcome is an altar. In this case, the alignment is
part of the location's topology, but code setting that up was using
Align2amask(rn2(foo)). That's a macro which evaluates its argument
more than once. The first evaluation was effectively a no-op. If
the second evaluation picked lawful then the result was lawful as
intended. But if the second picked non-lawful and the third picked
lawful, the result would end up as none-of-the-above (a value of 3
when it needs to be a single-bit mask of 1, 2, or 4).
Make some progress on a couple of next minor release checklist
items, hopefully without introducing too many new bugs. This
is just the initial commit, and work continues.
Checklist items:
Savefiles compatible between Windows versions, whether 64-bit
or 32-bit in little-endian field format.
Selection of file formats:
historical (structlevel saves),
lendian (little-endian, fieldlevel saves),
and just for proof-of-concept, ascii fieldlevel saves
(the ascii is huge! 10x bigger than little-endian).
For the fieldlevel save, all complex data structures recursively
get broken down until until it is one of the simple types that
can't be broken down any further, and that gets when it gets
written to the output file.
New files needed for this build:
hand-coded:
include/sfprocs.h
src/sfbase.c - really a dispatcher to one of the
output/input format routines.
src/sflendian.c - little-endian output writer/reader.
src/sfascii.c - ascii text output writer/reader.
auto-coded (generated):
include/sfproto.h
src/sfdata.c
This is just one approach. I'm sure there are countless others
and they have different pros and cons.
For producing the auto-coded files a utility called
universal-ctags, that is actively maintained and evolving,
was used to do all the heavy-lifting of parsing the
NetHack C sources to tabulate the data fields, and store
them in an intermediate file called util/nethack.tags
(not required for building NetHack if you already have a
generated include/sfproto.h and src/sfdata.c)
util/readtags (also not required for building NetHack
itself) will decipher the nethack.tags file and produce
the functions that can deal with the NetHack struct data
fields.
You can obtain the source for universal-ctags by cloning it
from here:
https://github.com/universal-ctags/ctags.git
The combination universal-ctags + util/readtags has been
tried and tested under both Windows and Linux, so it is
not tied to a particular platform.
Note: util/readtags will work only with universal-ctags
output, so other ctags are unlikely to work as-is.
Universal-ctags can be build from source very easily
under Linux, or under Windows using visual studio.
Changing an inventory item's bknown flag wasn't followed by a call to
update_inventory() in many circumstances, so information which should
have appeared wasn't showing up until some other event triggered an
update.
Make the Plane of Water be water all the way to edge instead of having
stone on left, top, and right. The Plane of Air already has air all
the way to edge (including unused/unuseable column #0) but does so via
code rather than the level description file so Water does that now too.
The edges of the Plane of Air were cloudless (3 columns on the left,
2 rows on the top, and 2 columns on the right; don't recall about the
bottom) and that looked pretty strange. Those rows and columns are
beyond the range of bubble/cloud movement so just make some of those
spots randomly be sight-blocking cloud terrain instead of all open air.
It isn't integrated with the moving clouds but looks fairly good when
the hero moves along the edge of the level.
Using wizard mode to leave Water or Air and later return resulted in
no clouds on the Air level and bubbles as usual on the Water level.
I still don't understand why, but on return to those levels run the
bubble creation routine as if the old discarded bubbles or clouds were
being restored.
Leaving the Plane of Water to return to a previously visited endgame
level didn't free the air bubbles unless/until you visit a new level.
Returning to that level creates a new set of air bubbles, losing track
of the previous set. Likewise with Plane of Air and its clouds. (Not
an issue with actual save and restore when on those levels, or when
just moving forward to not-yet-visited levels.)
Not applicable to normal play where it isn't possible to return to a
previously visited endgame level.
For 3.7, bubble save/restore ought to become part of savlev() instead
of being handled by savegamestate().
Lock context wasn't being cleared if it was for a container and that
container got destroyed. Case discovered was forcelock() ->
breakchestlock() -> delobj() (sometimes the container is destroyed
rather than just breaking its lock) followed by #wizmakemap (replace
current level) and maybe_reset_pick() trying to check whether
xlock.box was being carried. But being interrupted, destroying the
container or dropping it down a hole to ship it to another level, then
attempting to resume picking the lock would also find a stale pointer.
This is branched from Alex's hallu-rng-stability branch,
with two build corrections (detect.c, zap.c), and merged
with the isaac64 branch that we have ready to go.
Alex's dual rng is supported by setting up the array
of multiple isaac64 contexts.
I stuck with Alex's approach of passing the rng function
name around as the parameter (rng or rn2_on_display_rng)
for the new additional parameter needed for
set_random(), init_random(), reseed_random(),
and init_isaac64().
For platforms that read from the system's random number generator,
reseed during level change, before the map of a new level is created and
after level creation has finished.
