The one `anything any' that was triggering a warning was shadowing
another `anything any' in the same function; no need to rename it, just
remove the unnecessary declaration. Also, mark the couple of arrays with
initializers that I'd noticed as static instead of letting them default
to auto. The abil_to_spfx()::abil2spfx[] one might need to be redone in
code as a switch if some compilers/linkers have trouble initializing it.
Three years ago <email deleted> reported that
stepping off the end of inventory via typing space to go to the next menu
page wasted his identify scroll. He suggested that some people might do
that because they don't know how to back up in a multi-page menu. I
pointed out the Guidebook section that describes < and ^ to go back one
page or back to start and left things at that. However, traditional mode
reprompts if you step through all of inventory without choosing something,
so this changes identify-via-menu to do likewise. You can dismiss the
menu with ESC to really avoid choosing anything.
This also makes identification of N items when you're carrying N or
fewer unID'd things behave the same as identify all: identify everything
without any prompting.
The characteristics display checked for cursed rings of sustain
ability but neglected to check for uncursed ones locked in place by
cursed gloves or weapon. And the half-hearted attempt to check for
future items conferring Fixed_abil couldn't handle an uncursed thing
covered by something cursed either, so just get rid of it.
Use sentences for the characteristics (instead of "Strength = 15") to
match the rest of the enlightenment display. When showing ATTRMAX, phrase
as "limit" rather than "maximum" since magical ajdustments can exceed max.
And be more selective about which alternate attribute numbers to show:
Your strength is 14 (current; base:13). => +1 ring of gain strength
Your strength is 13 (current; peak:14). => lost a point to poison or abuse
Your strength is 14 (current; base:13, peak:14). => combo of both of those
Your strength is 14 (current; limit=18/50). => you're a gnome
instead of always showing all of base & peak & limit when any of the three
are interesting. During play, a limit which is different from normal human
(18 or 18/100) is considered to be interesting; for end of game disclosure,
it's interesting iff the final value hasn't reached maximum, regardless of
what that max is.
I'm not a contender to win any spelling bees. (Mimicker does't
seem to even be a real word; I'm not sure if it ought to end in "or"
instead of "er". But changing it to "mime" would be too weird.)
Augment killer reason when slain by a shapechanged creature:
"killed by a foo" becomes "killed by a chameleon immitating a foo" or
"killed by a vampire in foo form" or "killed by the Wizard of Yendor
disguised as a foo" (after double-trouble, when the clone starts out
mimicking something).
I put the fixes entry in the new features section.
I was testing something with #monpolycontrol enabled and got a
series of "Change it into what kind of monster? [type the name]" prompts
when I went to a new level. It was asking about monsters that were being
created (in this case, multiple vampires for top level of Vlad's Tower)
and naturally I couldn't see them since the level wasn't finished yet.
This switches to noit_mon_nam() to at least be informed about the type
of creature you're being asked to specify a shape for, and adds the map
coordinates so that it doesn't appear to be reprompting for the same
monster over and over when multiple similar ones are being created.
In the process I discovered that #monpolycontrol would let you give
any type of monster for the shape of vampires. Unlike chameleons, they
don't change into arbitrary shapes so that was inappropriate. [And
trying to test the fix for this is what led to the previous ^G patch.]
A while back there was a change in how the initial shape for a
shapechanging monster gets chosen during monster creation, and a side-
effect of that lets/makes you choose the shape when #monpolycontrol is
enabled. But ^G was overriding your choice by forcing the shapechanger
to start out looking like the type of monster that you specified (to ^G,
not to subsequent #monpolycontrol prompting), hence always in its natural
shape. The intent for ^G was that if asked for a unique monster but got
a doppelganger instead, it would initially look like the requested unique
monster. Post-3.4.3 code, so no fixes entry needed.
Three years ago <Someone> reported that even though exercise of
attributes other than wisdom is suppressed while the hero is polymorphed,
attribute gains or losses due to pre-existing exercise can still take
place in that situation. Since it's an entirely different set of attrs
which will be replaced upon rehumanizing, there's not much point. (The
same is actually true for wisdom, but I didn't change how exercise works
for it.) Adopt his one-liner fix: old exercise won't cause attribute
changes while polymorphed; it will silently fade as it does when its
magnitude is insufficient to trigger a change.
While checking that out, I noticed that exerchk() was using `/= 2'
to fade out old exercise/abuse. That will produce platform-dependent
results for negative values (ie, for the abuse case) since C's integer
division doesn't specify whether to truncate towards zero or towards
negative infinity. In particular, -1 / 2 could yield -1 rather than 0
as the code expected. (Its impact on play was negligible though.)
This reduces the code for displaying the messages which accompany
attribute gain/loss. A few were repharsed in order to simplify that.
The recent SYSCF patch introduced a build problem even though I
haven't attempt to use that new stuff yet. My compiler complained that
`out' in build_english_list() was used without being initialized, which
in turn caused make to quit. The compiler was right; only the words==1
case actually set up the output buffer. Once that buffer was fixed, the
routine to copy a single word was overwriting it on each call instead of
building up via appending as intended.
I changed the 3 or more case to yield "A, B, or C" like Keni wrote
in his description rather than the "A, B or C" which was being produced.
I'm pretty sure that both forms are considered acceptible; I've always
used the first one with an extra comma in front of and/or.
infrastructure for "system options" - things currently specified at build
time that should be changeable at install time or run time but not really
under user control
generalize contact info so it can be localized and it doesn't have to be
an email address
move recently introduced WIZARDS into sysopt
drop bogus OPTIONS=wizards possibility
new function build_english_list() to comma-ize and add 'or' from a whitespace separated list: A. A or B. A, B, or C.
syscf file now handles: WIZARDS SUPPORT RECOVER
SUPPORT specifies local support information
RECOVER will eventually supply port-specific and/or localized info on how
to run recover (or get it run for you).
Note: in sys/msdos I changed sys.o (generated from pcsys.c) to pcsys.o
Note: sys/msdos/Makefile.GCC has 2 rules for sys.o (now pcsys.o)
Picking up and putting on a +1 ring of protection while blind
resulted in having a "+1 ring (on {left|right} hand)" in inventory and
having ring of protection show up in the discoveries list. The problem
is the same as the one for wands which has been previously addressed
(but not 100% fixed...): when using an item whose effect is observable,
the item's type became discovered based upon that observation even if the
item itself had never been seen.
The code for removing ring of protection lacked its break statement
and fell into the case for removing ring of protection from shapechangers,
but didn't cause any noticeable problem.
1) the change yesterday to mark inventory as seen when recovering from
blindness neglected to handle blindfold removal;
2) fix grammar when using 'R' or 'a' to try to remove cursed lenses
(now "they are cursed" instead of "it is cursed");
3) attempting to put on a wielded or quivered blindfold, amulet, or ring
and failing because of already wearing one (or two in the case of
rings, where cursed gloves and ESC at "which ring-finger?" prompt got
similar result) unwielded/unquivered item rather than leave it as is.
I suspect--but have no way to test--that there is a subtle difference
in game play between perm_invent and !perm_invent involving object merging
or other activity that depends on whether or not object->dknown is set.
For objects pickup up while blind, where object->dknown is left as is,
the perm_invent config would have such items marked as seen sooner (since
there are umpteen places that call update_inventory() which will end up
setting the seen bit while formatting objects). Non-perm_invent wouldn't
have that done until the user examines invent (or asks to see a list of
objects at the "which object?" prompt for various commands). This patch
effectively examines inventory whenever blindness ends, so both modes get
dknown set as soon as possible. And if we ever add an "effect known" flag
for unseen objects which are used while blind, this would be a suitable
place to perform deferred object discovery.
..\src\options.c(565) : warning C4101: 'opts' : unreferenced local variable
The code in initoptions_init() that uses opts seems to now be isolated to
UNIX and VMS.
I did some research, and technically, a jellyfish does indeed have no head
(and, technically, no brain, just a "nerve net"). The part on top is a
"bell" used for buoyancy, not a head.
The post-3.4.3 ring removal bug also applied to suit when wearing a
cloak and shirt when wearing a suit or cloak or both, although it would
never give "you have nothing else to take off" because the covering item
was always present as a likely candidate. The ring fix also fixed armor.
When testing that fix, I saw "you can't take that off" for trying to take
off a suit while wearing a cloak. That isn't new or a bug, but it seemed
awfully terse so I've changed it to give "you can't take that off without
taking off your cloak first". (Likewise, "cloak", "suit", or "cloak and
suit" as appropriate when trying to take off a shirt.)
This adds new routine `suit_simple_name()', comparable to 3.4.3's
cloak_simple_name() and post-3.4.3 helm_simple_name(). No doubt there
are other places besides "without taking off your X first" that could
make use of it but I haven't attempted to track them down. The "you are
already wearing _some armor_" one doesn't quite fit. It would need to
adjust "some" to "a"/"an" at times ("some mail" or "some dragon scales"
vs "a suit" or "a jacket").
A bug in some post-3.4.3 code made attempting to remove a worn ring
when wearing two skip inventory the selection and give "you don't have
anything else to remove" if you were wearing gloves and weren't wearing
an amulet or blindfold. The intent was that rings wouldn't be included
as likely candidates in the list of inventory letters to choose from if
you were wearing cursed gloves, but had a copy+paste mistake in the
argument controlling whether to consider the gloves curse state and was
in the wrong section of getobj()'s "ugly hacks". Also, it makes more
sense to require that the player know that those gloves are cursed, so
inaccessibe_equipment() got changed too. [Not knowing that the gloves
are cursed leaves rings listed as candidates; picking a ring will get
removal failure feedback later and mark the gloves as known cursed then.]
Make being tethered to a buried iron ball be a bit more interesting
than just being stuck in the floor. You can now move within a shortened
chain radius (1 step from the ball's burial location rather than normal 2
that applies when it's on the floor). This has only been modestly tested,
and the message gets to be annoying if you move around a bit while monster
attack and object pickup/drop messages are being delivered. There ought
to be some better way to override Norep.
This would work nicer if the chain was actually visible at the ball's
location instead of becoming implicit at the time of burial. But I don't
want to try to figure out all the ramifications of having an object on
the floor which can't be moved--perhaps uchain handling already suffices,
but I suspect that there's a can of worms waiting to have its lid opened.
Pull the u.trap portion of domove() into a separate routine. The
only significant change is to the wording of a couple of TT_BURIEDBALL
messages. Like with enlightenment, say "you are <steed> are chained to
a buried ball" instead of "<steed> is chained to a buried ball" if you
try to move while mounted and punished and attached ball has been buried.
Unlike floor trap activation, punishment doesn't directly affect steed.
Yesterday's changes made it possible to use ^X to recognize an
unknown ring as one of sustain ability by checking whether wearing the
ring blocks display of base/peak/maximum characteristics when those are
considered to be interesting. This limits that to cases where the ring
is cursed, so using ^X to figure out the type will leave an undesireable
ring stuck on your finger. (Both before and after are unnoticeable in
wizard mode, where items which block access to base strength, dex, &c
are overridden by magical disclosure.)
From a bug report, putting on a helm of opposite
alignment wipes out divine protection (regardless of which god originally
bestored it) but leaves piety level towards the old god in place for the
new one. Reset alignment record to 0 when changing alignment type. Yes,
that means that the helm can now be used to clear a negative alignment,
which I think is in keeping with having subtle positive uses for bad items.
Permanent alignment change did try to tweak alignment record, but
I'm not sure that it worked as intended. Lawfuls changing away from law
got a bonus for their new alignment, neutrals retained their old record,
and chaotics changing away from chaos took a penalty to their record, all
without regard to what the new alignment became. The magnitude of the
bonus or penalty was half the record limit (which gradually increases with
turn count) so could be very big, but I don't think it made much sense.
Reset alignment record to 0 for permanent conversion same as for temporary
conversion. It's simpler and reflects the fact that you haven't yet done
anything to bolster or weaken your new alignment.
Replace the old ^X output with integrated enlightenment feedback.
Expands enlightenment output into four sections: Background and
Characteristics in addtion to previous Status and Attributes. Final
disclosure and ^X in wizard or explore modes show all four; potion/wand
of enlightenment and throne insight show the last two to give traditional
enlightenment, near enough; and ^X in normal mode shows the first three,
encompassing the 3.4.x ^X display in streamlined form plus providing
details about bottom line information.
I think I'm done with enlightenment for the time being, except for
the desire to move it (plus associated conduct feedback) to end.c or
perhaps its own file. It has grown very long and doesn't particularly
fit in cmd.c; I don't think it ever belonged there at all unless wizard
mode ^X predates potion of enlightenment.
Split enlightenment output into two sections: Status and Attributes.
Status began life as the set of bottom line status conditions and steadily
grew to include many things which ought to be such conditions but aren't
because of lack of space there. It also adds a few missing enlightenment
entries for Underwater (separate from magical breathing), Punished,
trapped, and stuck to cursed saddle. Most of the lesser status conditions
with bottom line tags (Confusion, Hallucination, &c) used to be shown only
for end-of-game disclosure but now get included all the time. They aren't
necessarily redundant if/when a very long bottom line is being truncated.
The ^X changes are still to come.
Groundwork for re-doing ^X so that it'll be more integrated with
enlightenment and display bottom line information without abbreviations
or long-line truncation. `mode' doesn't do anything yet so may provoke
lint complaints.
Add options SYSCF (to add a system-wide configuration file) and SYSCF_FILE
(for a file-based implementation of SYSCF) - this allows a binary distribution
to be configured at install time. The only option supported at this time is
WIZARDS - a list of usernames which can use -D; currently only for unix-likes
but should be extendable to anything that has a concept of multiple users.
Add mac tty multiuser config using sgid.
From a bug report.
Pointer subtraction yields a result of type ptrdiff_t which can be
bigger than type long, potentially resulting in warnings about value
truncation--in these cases, when assigning to a long.
During some recent newsgroup discussion, <Someone> posted an entry from
his personal bug list: energy draining damage from ordinary attacks is
implemented even though there are no monsters with that capability and it
was not implemented for engulf attacks even though energy vortices have
the capability. This implements energy drain engulf attacks against the
hero and also both modes of energy drain attacks for monsters and poly'd
hero against spellcasting monsters. Since monsters don't have energy,
they lose access to their special abilities (their spells, that is) for a
few turns, same as a post-3.4.3 change done for anti-magic traps.
There was an issue reported where save files between different
versions of a manufacturer's compiler were incompatible because the time_t
ubirthday field was changed from 32 bits to 64 bits.
32 bit time_t implementations will break at 19:14:07 on January 18, 2038.
64 bit time_t implementations will break at 23:59:59 on December 31, 3000.
This removes the dependency on the size of time_t from the save file.
The ubirthday field is no longer embedded in struct you.
This also adds two general purpose routines to hacklib.c, one to convert a time
value to a 14 character char representation and the other to convert that
back to time_t. Those are used by the save/restore routines.
This is a savefile breaking change, so editlevel in patchlevel.h was
incremented.
From the newsgroup: when a monster throws a potion at the hero and
it hits a long worm's tail, you'd get "The bottle|flask|&c crashes on the
long worm's _head_ and breaks into shards." The relevant code changes
"head" to "body" when appropriate, but monster throwing wasn't setting up
`notonhead' so that alteration only kicked in for player throwing.
Add a hints file for an ubuntu tty build and skip xset if no fonts.dir
file exists (this test will probably need another tweak, but I need a
linux x11 build first to test it).
build system.
Anyone who wants to do a build from sys/unix and doesn't want to figure this
out just needs to do:
sh setup.sh hints/unix
instead of:
sh setup.sh
and then continue on as usual.
New files:
sys/unix/NewInstall.unx - the new directions
sys/unix/hints/* - the hints files. There will be more later.
sys/unix/mkmkfile.sh - helper for setup.sh
Summary of changes:
see NewInstall.unx for info on the new build system
introduction of various preprocessor symbols to turn options off that
are defaulted on historically
comment out nethackrc (and similar) entries that still use the old symbol
syntax.
commenting out of Makefile.* lines that now come from hints/unix
GAMEDIR is replaced with HACKDIR so the Makefiles and the C source agree.
Note that I have NOT changed the docs and/or Makefiles for be, msdos, os2,
vms, or winnt. If port maintainers don't then I will, but I can't test
those ports.
nethack.sh now handles the font path automatically
Some post-3.4.3 code in done() cleans up thrownobj and kickobj to
plug a potential memory leak (unnoticeable one since the game is over),
but they have to be free objects rather than on any list. toss_up() was
leaving thrownobj defined after putting the thrown object on the floor,
leading to "obj not free" panic from dealloc_obj() if threw you an object
upward and it dealt fatal damage when it fell back down. (For non-fatal
damage, toss_up()'s caller cleaned things up later.)
Get rid of most of the vestiges of the old warning code that was
replaced over 7 years ago. I left the list of colors which were used for
warning flashes.
In the newsgroup about three weeks ago someone described trying to
use the Bell of Opening to find the magic portal on the Plane of Water
and not succeeding. It's supposed to work like a wand of secret door
detection to mark nearby traps as known. And does, but it turns out
that the wand wasn't working as expected there either. They both require
line of sight, and since the water outside of the bubbles blocks that
they only found the portal if it was within the same bubble as the hero.
(Clouds on the Plane of Air posed a similar problem, although monster
activity usually reveals the portal on that level so this wasn't much of
an issue there.) Since the detection magic doesn't require the hero to
see the traps--wand and Bell both work while blind--this patch overrides
the line of sight requirement on the Planes of Water and Air. As long as
hero is within the detection magic's range, the portals on those levels
will get marked as having been seen and when the hero gets into the right
bubble or out of the clouds the portal traps will be shown on the map.
The line-of-sight override code is simple-minded and lets players
find traps through boulders when/if those are present (but the found
traps won't be seen yet since vision still controls the map display).
Also, it assumes that only water/air/cloud terrain is present so could
potentially yield strange results if any other terrain gets introduced on
either of those two levels.
From the newsgroup: hangup save while picking up gold from the
floor in a shop would duplicate that gold in the save file. First the
gold amount was being added to hero's gold, then two messages were given
[pline() or prinv() about pickup followed by one from costly_gold() about
shop credit], and lastly the floor gold would be removed. The second
message could trigger --More-- and provide a controllable interruption
point between giving the gold to the hero and removing it from the floor.
Change this to do the removal step before feedback.
From a bug report, attempting to open or close
a door while blind didn't always update the map display after telling the
player that a remembered-closed door was already open or a remembered-open
door was already closed. And in the cases where it did update the map to
reflect a change in door state, it didn't use up a turn. Now always update
the display and use up a turn when 'o' or 'c' reveals any new information.
