Static analysis notices that
if (physical_damage)
tmp = Maybe_Half_Phys(tmp);
will never pass the test because all code paths leading to it set
'physical_damage' to False. Instead of getting rid of it, add a fake
case that leaves that True.
Another item from static analysis. If an internal error ever caused
the "bad do_look buffer" warning from checkfile(), open file 'data'
would not be closed. (The bug in checkfile()'s caller which prompted
that check was fixed long go.)
An alternate fix would be to move the input buffer check to before
the file is opened, but verifying the file first seems worthwhile.
From Jessie's old static analysis report. 'prezero' was used in 3.4.3
when processing a count for 'multi', but count handling is now done in
a separate routine and 'prezero' in parse() never changes value, so
get rid of it.
Another bug from seven years ago, sent directly to devteam so no #H
number. Report stated that throwing recoil could move a levitating
hero diagonally through a shop's doorway to exit it. If the thrown
item was unpaid, it remained unpaid after landing on shop's floor
and was an unlisted item on shop's bill. Moving diagonally out the
door seems to have been fixed, but the same effect still occurred
if you were far enough from the door to have the shopkeeper vacate
his door-blocking spot and throwing recoil took hero to that spot.
The thrown unpaid item remained unpaid, and walking out the door was
treated as shop robbery.
I don't know why we have two different functions which do exactly
the same thing (checking whether an item is unpaid or is a container
that holds at least one unpaid item), but switch the #H2504 fix to
use 'the other one' and reverse one of the changes made when using
the inventory one.
I thought that the earlier fix for #H2504 was too easy for anything
shop related. It didn't deal sensibly with containers owned by hero
but holding unpaid shop goods.
This one is only seven years old. Dropping an unpaid item inside an
engulfer leaves it unpaid and still on bill. If engulfer is killed,
it ends up unpaid when back on the shop's floor.
Treat dropping an unpaid item into engulfer's inventory as stealing
that item. You have to pay for it to leave the shop, and like any
other dying monster's inventory, the shopkeeper will take ownership
if it lands on the shop floor when the engulfer is killed.
The 'theft' doesn't anger the shopkeeper and the cost shows up on 'Ix'
as part of "usage fees/other charges" rather than as an itemized used
up item.
That #H number isn't a typo. This finally fixes--at least improves--
something reported eight years ago. The monster types chosen by
mkclass() could be way off in some circumstances. Cited example was
repeated same-race sacrifice by chaotic hero on dungeon level 20; it
produced about twice as many incubi as succubi even though they're
the same as far as difficulty goes. (No changes in the intervening
years had any discernable effect; that was still reproducible.)
The report also mentioned that ndemon() threw away the result from
mkclass() and retried quite often and suggested that mkclass() be
taught to filter by alignment when caller cared about that.
This seems to even things out, although it also made harder monsters
chosen more often. A test program generated these numbers when
picking a chaotic demon 10000 times (level 1 hero on dungeon level 20,
so not realistic; actually probably level 0 hero since the program
didn't initialize struct u.) Third column is the number of times the
monster type was chosen with the old mkclass(), fourth is same for
the new one.
mkclass() calls 27315 10000
286 succubus 2800 3309
288 incubus 5552 3262
291 marilith 973 780
292 vrock 477 1617
293 hezrou 150 626
294 bone devil 46 247
295 ice devil 2 107
296 nalfeshnee 0 23
297 pit fiend 0 15
298 sandestin 0 4
299 balrog 0 10
Note that vrock has generation frequency 2 and marilith only 1, so
getting twice as many vrocks as mariliths should be expected.
I temporarily changed ndemon() to ask for lawful demons instead of
chaotic ones and got this.
mkclass() calls 15762 10000
287 horned devil 3197 3375
289 erinys 4991 3339
290 barbed devil 1812 3286
I also ran it for dragons with any alignment (so the outcome was
never thrown away; 10000 calls were needed for 10000 picks) instead
of demons of specific alignment and am suspicious of the outcome.
mkclass() calls 10000 10000
140 baby yellow dragon 1124 0
141 gray dragon 1096 1096
142 silver dragon 1073 1099
143 red dragon 1061 1126
144 white dragon 1077 1128
145 orange dragon 1141 1118
146 black dragon 1154 1049
147 blue dragon 1137 1123
148 green dragon 1137 1154
149 yellow dragon 0 1107
There may be a flaw in the test program. Or else old mkclass() was
not very good at picking dragons.
Stop pretending that long and int are the same size when picking status
highlight rule for gold or time or experience-points.
Also, K&R compilation might lack <limits.h>, so let XXXconf.h define the
necessary macro(s) (currently just LONG_MAX) so that it can be skipped.
Salvage part of an old patch. Swapping places with any long worm,
even a baby one, always failed with "You stop. Foo is in the way!"
This lets you swap places with tame baby long worms, and adult ones
if they have no tail (which won't happen unless there are more than
32 long worms on the current level--even if a long worm appears to
be only a head, there is normally a hidden tail segment at the same
location).
Although choose_classes_menu() is only used for objects, it is written
to handle monsters too. My change to give <space> special handling
might have broken selecting ghosts if it's ever used for monsters so
fix that.
More on clearing pickup_types so that autopickup reverts to picking up
evertyhing: for menustyle:Full and Partial, add a menu entry for 'all
classes' as an alternative to unselecting every class already set.
Also, Full and Partial had no way to include venom. Now it's a choice
when in wizard mode. There's still no way--other than switching to
Traditional or Combination--during normal play (where venom objects can
exist if they were wished for in wizard mode and then left in bones).
'sz' is the size of the buffer; 'if (count < sz) buf[count++] = c;'
can fill the entire buffer, leaving count==sz, so buf[count] = '\0';
would be out of bounds.
Formatting was way off. Indentation these days should be multiples
of 4 spaces, never tabs.
To use 'O' to clear a value from pickup_types with menustyle Traditional
or Combination, you needed to give a value starting with 'a' (for 'all').
Accept space(s) too, similar to removing an object or monster name.
I watched the fuzzer run for a bit and noticed that Str and most other
characteristics were steadily dropping until they hit 3 and not being
recovered, so I gave the defenseless hero a chance to benefit from
blessed restore ability occasionally. It hasn't helped much. Str and
Con both still drop to 3. [If I had to guess, I'd go with side-effect
of polymorphing, but not an intended one.]
A change I included with the vault guard fix was triggering fuzzer
panics about dead monsters on the fmons list. I'm not quite sure why;
I couldn't reproduce it interactively. [Perhaps caused by hero killing
a monster and then getting another move before monsters get their turn,
but trying to do that still didn't trip the dead monster sanity check.]
Suppress that check so that the fuzzer can run amok.
Also, a waiting-to-exit vault guard could move extra times, uselessly
since ''hero hasn't left temporary corridor yet'' is why he's waiting,
if there were any monsters fast enough to get extra moves before the
hero's next turn.
"Placing monster over another?" warning was triggered for vault guard
by an earlier change which made m_detach() stop removing monsters at
<0,*> from level.monsters[][]. So one guard would replace another at
<0,0> for however many guards were created, and memory for all but
the last one would be lost.
This involved a lot of flailing about and the patch includes various
things would could have been discarded. One or two extended monster
sanity checks are included, plus a couple of debugpline()'s for
tracking guard movement.
The fruit name fix a couple of days ago revealed that an() didn't
pick the right article for single letters. If you set named fruit to
"X" and wished for "fruit", you'd have 'k - a X' added to inventory.
This changes an() and also doname() which bypasses an().
struct rm.flags in overloaded for a bunch of rm.typ -dependent things
(doormask, altarmask, throne/fountain/sink looted, a few others) and
wasn't being reset for various cases where rm.typ gets changed.
I've changed a lot, some no doubt unnecessarily, and probably missed
plenty. This compiles but has not been thoroughly tested.
When clairvoyance revealed an underwater object and the player wasn't
being given a chance to browse for some other reason, the map didn't
get refreshed and the object stayed displayed with the water hidden
under it. Force a chance to browse when revealing underwater objects
despite that being intrusive when clairvoyance kicks in while walking
around.
When options processing encountered OPTIONS=windowtype:X,
the code would immediately attempt to switch over to that
windowtype right in the midst of options processing.
This stores the chosen option into
chosen_window[WINTYPELEN]
thus allowing the startup code to choose it after
options processing has been completed.
Jumping or teleporting while levitating in xorn form wouldn't toggle
blocking of levitation when moving from open spots to wall/stone and
unblocking of same when moving the opposite way.
This handles those cases but there are no doubt others. The only
other one I checked was when failed #untrap moves hero onto trap.
That case works correctly--at least after this fix is in place.
Noticed while working on it: change of terrain didn't always update
the status line. When levitation became blocked, it still said Lev
and when unblocked, didn't say that. Next status update got status
condition back in sync.
Some code added for 3.6.1 tries to find the longest matching prefix
when comparing a user-supplied with previously set up fruit names.
It does so by temporarily replacing space with NUL then passing that
to makesingular(). After already having named a fruit (resulting in
something to try to compare with), attempting to assign a name
beginning with two or more spaces would yield an impossible "singular
of null?" warning.
After the warning, the name minus its leading spaces got successfully
assigned. I'm not sure why a single leading space didn't trigger it
too, nor where the leading (and trailing, if any) spaces are going
away when the name is assigned.
Fix by removing all leading and trailing spaces from a new fruit name,
and combine consecutive internal spaces to one, before any other
manipulations. (This can result in names that used to work as-is now
being simplified a bit--when consecutive internal spaces have been
given--but that shouldn't be a problem.)
Also, don't complain about "missing parameter for 'fruit:'" if user
hits <return> when prompted for fruit name by 'O'. An empty fruit
name at that stage is just a no-op.
Inadvertently reintroduced in f9f1236. It was just the conditional
that was bad: due to resolving the possible buffer underflow when
comparing to "muskox", the pluralizer now only adds -es when the length
of the string is greater than 5. So for "box" and "fox" the pluralizer
will never add the -es ending, since they are greater than 5.
This commit checks for "does not end in muskox" correctly.
Changes to be committed:
modified: include/winprocs.h
modified: src/options.c
modified: sys/share/pcmain.c
new file: sys/share/safeproc.c
modified: sys/winnt/Makefile.msc
modified: sys/winnt/stubs.c
new file: sys/winnt/windmain.c
modified: sys/winnt/winnt.c
modified: win/win32/vs2017/NetHack.vcxproj
modified: win/win32/vs2017/NetHackW.vcxproj
modified: win/win32/winhack.c
Because multiple window ports are supported on Windows
now, even in the same executable and selectable via
config file in some cases, some adjustments became
necessary. There will likely be some further refining
of this over the next day or two.
List of changes:
Move Windows startup from sys/share/pcmain.c and
into its own sys/winnt/windmain.c so that it can
be modified to fix some current breakage, and
allow altering the order of some things.
There is startup processing code that is common to
all of the Windows WindowPorts, but that startup
processing code needs to have no dependency on
any one of those WindowPorts.
Yet, during startup processing, some of the initialization
routines can end up calling NetHack functions that
expect an active Window port underneath, and if there
isn't one, routines like pline, impossible, panic can
end up invoking null function pointers.
Place a new file sys/share/safeproc.c, in which a complete
window port is available for early startup processing
purposes. It's WindowPort name field is set to
"safe-startup" just for reference. The prototypes in
include/winprocs.h require that SAFEPROCS be
Usage:
windowprocs = get_safe_procs(0);
initializes a set of winprocs function pointers that ensure
none of the function pointers are left null, but that's all it does.
windowprocs = get_safe_procs(1);
initializes a set of winprocs functions pointers that ensure
none of the function pointers are left null, but also
provides some basic output and input functionality using nothing
other than C stdio routines (no platform or OS specific code).
The conditional code related to WIN32 has been removed from
sys/share/pcmain.c
The code common to all of the Windows WindowPorts calls
get_safe_procs() almost immediately to ensure that
there is a set of WindowPort winprocs available.
Make the sequence:
be zapped by lightning,
have worn ring of levitation be destroyed,
fall onto fire trap
work better. The fire trap handling will mark everything in inventory
as already processed; anything vulnerable to lightning past the destroyed
ring would not be checked. So delay destroying such a ring until after
all of inventory has been subjected to lightning.
After consultation with the original committer, this
is being pulled out, possibly revisited later. There was
originally meant to be a follow-up piece to this that he
never had a chance to integrate for various reasons.
