Reported nearly four years ago for 3.4.3, original subject was
\#H3013: NetHack grammar bug when taking unpaid fruit from chest
Player used OPTIONS=fruit:Quorn and the capitalized value confuses
the() into thinking it's a proper name which shouldn't be preceded
by an article, resulting in "Quorn will cost you N zorkmids" when
removing it from a chest in a shop, followed by "X - a Quorn (unpaid)"
as it went into inventory. It is a product name, but when used as a
fruit it shouldn't be treated as a proper name. (Quorn is a meat
substitute rather than anything related to fruit.) Teach the() about
named-fruits, so that we'll get "The Quorn will cost you N zorkmids."
Unfortunately, it means that someone who names their fruit after a
proper name used by the program, for example Mjollnir, can probably
induce other poorly worded messages (about the item rather than the
named-fruit). the() is used all over the place and all it has to work
with is text, not the object whose formatted name produced that text.
I looked through a bunch of old cvs log messages last night, and
spotted one I wrote (in objnam.c) a dozen years ago where I suggested
forcing named-fruit values into lower case as they're being set up.
I don't remember that, but if we'd done it, this bug would have been
avoided.
The object name formatting routines operate using a pool of buffers to
hold intermediate and/or final result. Some routines consume multiple
intermediate buffers, so use releaseobj() to try to reuse just one in
order to avoid churning through too many and maybe clobbering live
data. It worked as intended for routines that use nextobuf() directly
but wouldn't haved worked right for xname(), also doname() and other
xname() callers. This fixes that.
There have never been any reports of garbled messages which could be
traced to clobbering of formatted object names, so this fix is mostly
academic.
Add some new routines for dealing with fruit. I had hoped they would
let the existing fruit handling be simplified quite a bit, but the
improvement wasn't great. However, they're also groundwork for fixing
an old bug.
A couple of days ago when verifying the report about being forced to
pay for a second tin when eating one from a stack on a shop's floor,
wishing for 'tins' (rather than 'tins of foo meat') was repeatedly
producing tin wands instead. The name "tin" and the wand description
"tin" in objects[] were being given 50:50 chance for either one by
the 3.6.1 wishing code. Wishing for "tins of spinach" also gave me
a tin wand on the first attempt. Handle 'tin(s)' more explicitly.
This also does some reformatting.
It's apparently somehow possible to have a leash "attached" to
a monster not present on the current level. Complain instead of
segfaulting. (This is a post-3.6.0 bug)
This segfault happened when a detected monster such as a garter snake was
hiding under an unknown/unseen armor of a type that has no description
field (eg. a leather armor), and you farlooked at the monster.
At the same time, simplifies the code a tiny bit.
farlook was changed (end of December) to use doname instead of xname
to yield more info for items which had already been seen up close,
but it gave away info about ones which hadn't. So doname was changed
(end of April) to use "some" instead of precise quantity (when the
quantity is greater than 1) for the latter, but that doesn't work
well with corpse_xname() when the hero is blind, yielding "a some
<foo> corpses". While testing the first fix attempt, I noticed that
pickup gave "you can only lift some of the some <foo> corpses".
This fix is far from perfect. farlook can still say "some <item>s"
but lookhere and pickup always say "N <item>s". Picking up a stack
while blind will show "N <item>s" in inventory display, but dropping
it while still blind will revert to "some <item>s" for farlook.
"The the Wizard of Yendor corpse tastes terrible."
"The" was supplied when formatting the sentence, "the" when formatting
the corpse object. Same applied to several other monsters, those
flagged as unique and without personal name (Oracle, Cyclops, others).
This also changes the corpse eating message to not always specify the
taste as "terrible" when the eater is omnivorous.
It also makes corpses (and tins) of unique monsters be wishable in
wizard mode. I had killed the Wizard at least half a dozen times
without a corpse being left, so used wishing instead. Later retries
did finally leave one--lack of corpse turned out to just be by chance.
"Your pair of speed boots glow silver for a moment." should be
"Your pair of speed boots glows silver for a moment.". The fix
reverses a post-3.6.0 change to is_plural(). Also, add new
pair_of() to test for object formatted as "pair of Bars". For verb
usage, that's definitely singular, but for pronoun usage, sometimes
plural seems better (although it might actually be incorrect).
I fixed up the formatting of a block comment in obj.h, but it is
still a candidate for tab cleanup.
Make wishing for an artifact and not getting it because it already
exists break never-wished-for-artifact conduct. The wish was made
even if the result wasn't what the player wanted.
rnd_otyp_by_namedesc() had an off by one error when choosing which
matching item to return, making it impossible to successfully wish
for the Amulet of Yendor, always yielding the plastic imitation.
n == 2, maxprob == 2
prob = rn2(maxprob); => 0 or 1
i = 0; while (i < n - 1 && (prob -= ... + 1) > 0) i++;
always exited the loop on the first test of the condition because
subtracting 1 from 0 or 1 never yielded a result greater than 0.
It's still suboptimal: "amulet of yendor" should find an exact match
and should never return "cheap plastic imitation of amulet of yendor"
from the partial match.
I think biasing the choice among multiple candidates based on their
generation probabilities only makes sense when all the candidates are
within the same class. If scroll of light occurred in 5% of scrolls
and spellbook of light occurred in 10% of spellbooks (both percentages
pulled out of thin air), having "light" get a 2 out 3 chance to be a
spellbook doesn't seem right because scrolls are four times more
likely than spellbooks (in most of the dungeon; books aren't randomly
generated at all on the rogue level or in Gehennom).
Globs of pudding maintain an item quantity of 1 but vary in size
based on weight. xname() formats with prefix of "small", <nothing>,
"large", and "very large" depending upon weight but wish processing
didn't recognize the prefix strings and reported "nothing fitting
that description exists". Make wishing handle "small" (which is
actually the default), "medium" (since <nothing> isn't the default),
"large", and "very large". They get ignored if the item isn't a
"glob of {gray ooze, brown pudding, black pudding, or green slime}".
Also, clean up recently added rnd_otyp_by_namedesc().
At the end of December, farlook was changed to use doname() instead
of xname() in order to give more information when looking at stuff
the player had already seen. But it ended up giving away precise
stack size for mergable objects too, even if they hadn't been seen
up close. Changing doname() to be vague when dknown wasn't set
only worked for items in particular classes; dknown is pre-set for
a lot of things. So this changes mksobj()'s dknown handling to not
do that for stackable items.
The change to objclass.h is just comment formatting (for the first
part of the file only), provoked by the second line of the one for
oc_pre_discovered.
There have been two or three reports on getting feedback about
amulets rusting. Object formatting doesn't display erosion for
them, so being told about damage then not seeing that damage
feels like a bug. Even if damage was displayed, it has no effect
on them so would still feel somewhat strange. It does display
erosion for wands and rings, which is strange too.
This limits erosion damage--and its feedback--to items which are
actually impacted by erosion: armor, weapons and weapon-tools;
also heavy iron balls and iron chains since they've traditionally
shown rust even though it has little effect.
A side-effect of this change is that flammable items (other than
armor and weapons) which don't burn up immediately will no longer
become burnt, then very burnt, thorougly burnt, and finally be
destroyed. Since the player couldn't see or possibly repair the
erosion state, it seemed incomplete. It could be reinstated by
making other flammable items be subject to erosion and displayed
as such by xname() & co.
Wishing now avoids applying erosion and erosion-proofing to items
that aren't affected by it, regardless of material. It also now
allows wishing for "rusty rustproof <iron-object>" which used to
suppress "rusty" in that combination and triggered a couple of
old bug reports.
Heavy iron balls and iron chains can have rust repaired and can
be made rustproof by wielding, then reading enchant weapon while
confused, as if they were weapons.
Sometimes you can see a hidden monster without bringing it out of
hiding (wand of probing, blessed potion of monster detection) but
look_at wasn't mentioning the fact that the monster was hidden and
probing described mimics accurately but lumped all hiders together
as "concealed". Describe all hidden monsters more consistently.
'Poisoned by a rotted gray ooze corpse' should have been
'Poisoned by a rotted glob of gray ooze'.
eatcorpse() is called for non-corpse globs and then corpse_xname()
is called for them too to set up death reason for make_sick(), but
it didn't know anything about globs. Now it does. Blob size is
ignored since it's not relevant for cause of death.
Change "unlockable" to "broken" so that it won't be misunderstood to
mean "capable of being unlocked". The accompanying suggestion to omit
"broken" unless/until a lock or unlock attempt is made is no good since
the main reason for describing the broken lock is to avoid unnecessary
attempts to lock or unlock a container that the hero knows to be broken
but the player may have forgotten.
I also changed remote look-at for objects to use distant_name(doname)
instead of distant_name(xname) so that qualifiers like "empty" and
"broken" will show up on chests you've investigated before but aren't
standing on now. Monster type for corpse also gets shown, instead of
just 'food (corpse)'. Other remote items will become more verbose,
but only those that the hero has already seen up close.
Avoid "spellbook of novel" after novel becomes discovered. Now it will
just be "novel". Prior to discovery, it might be on the list as "book
called whatever" if the player assigns a type name.
Also, make novel become discovered after reading one instead of only via
object identification. It already shows up as "novel" in inventory, but
changing its definition to designate it as not-interesting-to-discover
feels disrespectful to the tribute.
Make a fix suggested during beta testing: you can read scrolls while
blind if you know the label, and you can write a scroll with a magic
marker while blind, but the result was flagged as description unknown
so you couldn't read the newly written scroll until regaining sight
or obtaining object identification. So change writing a previously
discovered scroll while blind to set dknown since a successful write
always yields the type of scroll requested. Getting lucky while
attempting to write an undiscovered scroll--which has to be done by
scroll's type name (for instance "food detection") rather than by its
label ("YUM YUM")--still leaves the description flagged as unknown
since hero hasn't seen the what sort of label the new scroll has.
Along the way I got side-tracked by the possibilty of writing a scroll
of mail. It's allowed and yielded the same result as finding such a
scroll in bones, or wishing for one: when read, it was junk mail from
Larn. Make one written via marker give different feedback since it
comes from creation of a stamped scroll without any stamps available.
Also, suppress an "argument not used" warning for readmail().
Entered in bugzilla prior to release: "slice of birthday cake" became
"slouse of birthday cake" when made plural. "slice of pizza" used to
work, but adding an entry for "louse" <-> "lice" to one of the special
handling lists for singular/plural broke "slice" since only a trailing
substring match is performed for entries in that particular list.
distant_name() temporarily blinded the hero before calling xname() or
doname() in order to prevent the object being formatted from having
its dknown flag set. The Eyes of the Overworld override blindness, so
that bit got set for heros wearing them regardless of intention. This
switches to a file-scope global instead of blindness as the way that
distant_name() tells xname() not to set dknown.
This bug has been present ever since the Eyes were added (3.3.0?).
Replace the code that uses strcat with two pointers into the same buffer.
Treated separately, they point at distinct strings (no overlap possible),
but the C standard does disallow that in order to enable optimizations
using block transfer or such, so the tool that complained about it isn't
wrong. The characters getting appended to the output pointer can end
up overlapping the beginning of the other input pointer, conceivably
breaking an implementation that didn't use simple left-to-right byte-at-
a-time copying.
Also, I noticed that wishing for "luck stone" gave me a random gem.
There's code to strip off " stone" and compare against gem types, but it
prevents other code that accepts "foo bar" as a match for "foobar" and
vice versa from finding a match, since "luck" doesn't match anything
once "stone" is gone. So add the four gray stones into the array of
alternate spellings.
Another bit: it now accepts " gem" in addition to " stone" as optional
gem suffix, so "ruby", "ruby stone", and "ruby gem" all yield ruby.
("luck gem" won't work; you'll end up with a random gem-class object.)
And a last other bit: wishing for "lamp (lit) named foo" would yield
an unlit, unnamed lamp because "(lit)" followed by anything didn't match
"(lit)" and threw away everything past the opening paren. Now it will
produce "lamp named foo (lit)"--a lit lamp named "foo". (Wishing for
"lamp named foo (lit)" produces an unlit lamp named "foo (lit)". That's
acceptable to me... I'm not crawling any farther down this hole. Maybe
object formatting should be changed to keep the lit attribute in front
of the name?)
While the topic of strprepend() is current, make good use of it.
Simplify code which inserts "the Nth " in front of "<arrow,&c>".
I'm pretty sure there are one or two other places where I assumed that
the outpuf of xname() was a char array which is BUFSZ in length rather
than BUFSZ-PREFIX, and reused the buffer, but I don't know where they
occur. (BUFSZ-PREFIX is still big enough to hold most things, so it
might not lead to trouble.)
Author: PatR <rankin@nethack.org>
Date: Fri Oct 30 00:50:52 2015 -0700
more formatting
Fix up the files containing '[?:] */' to get trailing trinary operator
followed by end-of-line comment. Tab replacement and removal of excess
parentheses on return statements also done.
Somewhere along the line I started removing redundant parentheses from
return statements, but only in files that needed continuation fixups
so it's not comprehensive.
Flesh out wet towels a bit:
1) wielding a wet towel--or a dry one which becomes wet--won't give a
"you begin bashing with your wet towel" message when attacking;
2) if a formerly wet towel dries out completely while wielded, *do* give
"you begin bashing with your towel" on the next attack;
3) successfully hitting with a wet towel no longer always loses wetness;
4) water damage to dry towel always confers at least 1 point of wetness;
5) taking fire damage (via burnarmor() which is used for most types of
fire damage) has a chance to partially or fully dry a wet towel
(regardless of whether it's wielded at the time; applies to monsters
as well as hero; each towel being carried is checked until one is
affected, then any others escape drying.
Not done:
-) attacking with a wielded wet towel perhaps ought to be treated as a
weapon attack using whip skill rather than an augmented arbitrary-
junk-by-weight attack;
-) throwing a wet towel should probably ignore wetness--it's just a wet
piece of cloth when not finishing with a whip snap; right now, it
loses a point of wetness when thrown and usually--#3 above--another
point if it hits...;
-) hitting burning creatures is no different than hitting anything else;
-) likewise for hitting wet creatures.
Dipping a towel into a potion, fountain, or some other water source
makes the towel wet. Hitting with a wet towel deals up to 6 points
of damage, but every hit reduces wetness, as does throwing or applying
the towel. You can also wish for a moist or wet towel.
Make novels be wishable in normal and explore modes in addition to
wizard mode. I don't think this weakens the tribute and it prevents
someone who attempts such a wish from getting misleading feedback of
"Nothing fitting that description exists in the game."
Wishing for "novel" will yield "novel named Foo" where "Foo" is a
randomly chosen Discworld title. Wishing for "novel named Bar" will
yield "novel named Bar" or "novel named The Bar" if "Bar" or "The Bar"
is a valid Discworld title, or else override "Bar" and pick random
Discworld "novel named Foo" if it isn't.
Since first read of a novel bestows some experience (once per game, no
matter how many novels become available), a pacifist with an early
wish can get a head start. I don't think that's a big deal. And it
will require an awful lot of wishes for any player who wants to acquire
all 41 titles in one game. I imagine someone will manage it.
Add "(glowing light blue)" to the formatted object description when
Sting or Orcrist is glowing due to presence of orcs or "(glowing red)"
if Grimtooth is glowing due to elves. Use "(glowing)" if blind;
assumes that some aspect of the glow (perhaps warmth or vibration) can
be noticed via touch.
Make enlightenment's "you are warned about <monster class> because of
<artifact>" catch up with Orcrist and Grimtooth. It was attributing
Orcrist's warning against orcs to Sting, and Grimtooth's warning was
against "something" rather than elves.
The glow color is now a new field in artilist[], so the biggest part
of this patch is adding an extra value to each artifact's definition.
gcc complained about mixing && with || without parantheses. After
scratching my head a bit, I think this change yields the intended
result.
'omit_buc' is a bad option name. It's cryptic and it doesn't even
describe the function. At a minimum it ought to be changed to
'omit_uncursed' to accurately describe what it does.
'implicit_uncursed' or 'explicit_uncursed' (with opposite boolean
value) would be even more precise but probably not any clearer.
