Defined strbuf_t and related routines to support dynamically sized
strings. Modified strip_newline() to strip the last newline in a string
instead of the first.
Simplified splash window code using new strbuf_t.
Prior to exiting game, re-enable getreturn and call wait_synch() in
case there is buffered raw prints that must be displayed to user.
The extended command added to test handling for adjacent mouse clicks
had a description which was too long. In the list from '#?', white
space for column alignment got squeezed out to make it fit (at least
for tty, where it ended up looking awful).
The new description isn't a complete sentence any more, but I don't
think anyone will care.
The code that checked for and complained about having more than one
'prize' object on the mines' end or sokoban end level uses static
counters and would complain if you used #wizmakemap to recreate the
level. (Or if a game got far enough that either of those levels was
created and then started over--I'm not sure what state support for
that has reached.) So re-init those counters each time any special
level gets created; that's sufficient for what they track.
I also changed several variables in sp_lev.c from global to file
scope since they aren't used anywhere else.
The followup message about the fix for #5056 was trapped by the spam
filter so didn't reach us for a while.
xlogfile has an extra field to track various achievements made during
the game it logs, two of which are fully exploring the gnomish mines
and fully exploring sokoban. Those are accomplished by finding the
special 'prize' item on the final level of their branch: luckstone
for mines and bag of holding or amulet of reflecition for sokoban.
3.6.0 had a bug where any item of the target type found anywhere in
the dungeon resulted in achieving the relevant goal. A post-3.6.1 fix
for that required that the item be found on the end level of the branch
and attempted to require that it an item explicitly placed there by the
special level loader, but the latter aspect had a bug which meant that
random items of the appropriate type placed on final level would count
as the prize. Chance of extra luckstones on mines' end is fairly high,
so potential for false completion of the achievement was also high.
The second complaint was that since the achievement was only recorded
if the special prize item was found on final level, then if a monster
took it to another level then the achievement became impossible. (Not
true, the player could take it back, drop it, and pick it up again, but
that is admittedly a pretty silly hoop to jump through.) On the other
hand, if a monster removed the item before the hero found it, then a
case could be made that the hero hadn't really fully explored the
level. However, this fix records the achievement no matter where the
hero picks up the item. The final level must be entered--otherwise no
monster could possibly acquire and transport the item--but it isn't
guaranteed to have been fully explored. Big deal....
The prize could also be acquired in bones data. Before the second
portion of this fix, that wouldn't have mattered. But now it does, so
clear the prize indicator when saving bones unless it happens to be the
same level where that item is created (impossible for sokoban, where no
bones are left; not sure offhand about mines' end). The former prize
stone or bag or amulet becomes an ordinary one of its type.
This can all be done in a much cleaner fashion once we give up on the
current save file compatability. Putting obj->o_id values into new
context.mines_prize and context.soko_prize, plus a hack to mkobj() to
not reuse those two values if the o_id counter ever wraps back to 0,
would cover most of the details. Adding an achievement tracking flag
to lev_comp's object handling for use by the special level loader
would cover most of the rest.
The recent fix to prevent #adjust from letting the player move things
into slot '-' (if compactify() reduced any sequence of consecutive
letters to x-y, introducing dash into the string of characters that
could be chosen from) was triggering a complaint about mixing &&
and || without parentheses. Fixing that was trivial, but I ended up
making a much more substantial change.
If the '#' overflow slot is in use, you can move something into it
even when you no longer have all 52 regular slots in use. (When it
isn't already in use, you can't access it. Previously you could swap
from '#' to any letter but not vice versa.) If you manage to get
gold in multiple slots or in some slot other than '$', you can move
or merge it into the '$' slot. And when that situation isn't present
(if even possible--I had to force it with a debugger to test), then
gold will no longer be listed among the inventory letters to adjust.
(That became an issue when GOLDINV came into use, but either nobody
ever noticed or at least never reported. "Adjust what? [$abd]",
then pick '$' and be told you can't adjust gold. Prior to GOLDINV,
'$' wasn't included in the list of candidates.)
Poly'd into a giant with a full inventory that already contains at
least one boulder, moving onto a boulder (that can't be pushed due
to a wall or other obstacle) yielded
You try to move the boulder, but in vain.
However, you can easily pick it up.
You are carrying too much stuff to pick up another boulder.
You see here a boulder.
The second and third statements contradict each other. Make the
code that dishes out the second message smarter. If autopickup is
set for it and you will pick up the boulder:
However, you easily pick it up.
If autopickup is not set for it but would have worked if it was:
However, you could easily pick it up.
If your inventory is full and you have a boulder (or are in Sokoban)
However, you easily push it aside.
That last one is instead of "however, you can squeeze yourself into
a small opening" that you'd get if not a giant and not carrying much.
The #adjust command allowed, due to compactified buf, putting
items into the '-' inventory letter, which is usually reserved
for "empty hands", zeroobj. This caused problems down the line
when user was allowed to pick that letter for dropping.
Reading a scroll while blind is permitted if you know its label, but
message is "as you pronounce the words, the scroll vanishes" unless
you are poly'd into a form which can't make sounds, in which case you
"cogitate" rather than "pronouce". Switch to the cogitate variant if
you are suffering from strangulation.
Casting spells didn't even have the distinction; you could cast them
without regard to speech capability. Check for that. Unlike with
scrolls, now you can't cast if you can't speak (or grunt or bark or
whatever) instead of having a variant description of the action, so
this is a bigger change.
A wishlist/TODO item: when "female" is highlighted, change "caveman"
to "cavewoman" and "priest" to "priestess". If it gets toggled to
"male", change them back.
Rearrange the tests for edibility of non-food so that touching an
artifact won't happen unless the object could be eaten.
Add a bit of bulletproofing for rust monsters trying to eat a
rustproofed item and spitting it out. Wishing for rustproof iron
ring, cursing it, wearing it, and attempting to eat it as a rust
monster would remove the rustproofing and spit it onto the floor,
ignoring the cursed state as far as taking it off goes. That was
an issue in 3.4.3 and probably in 3.6.0, but in current code the
'rustproof' part of the wish would be ignored for an item which
isn't subject to erosion damage, so hero-as-rust monster will
successfully eat the ring instead of spitting it out.
This eliminates nearly 40 warnings, most by suppressing complaints of
used function arguments but a few for unused local variables. There's
also some reformatting thrown in....
There are still 18 warnings about uses of XtSetArg(), about assigning
const to non-const.
The dialog shows the player's name, race, role, gender, and
alignment in a single window, similar to the Qt4 dialog.
Also allows randomizing the character selection.
Use the dialog by setting OPTIONS=player_selection:dialog
X11 supports both ascii map and tiled map and is able to switch back
and forth during play. 'O' shows both of them as boolean options, but
toggling ascii_map did nothing since tiled_map retained whatever value
it had at the time. For core options handling, make toggling either
ascii_map or tiled_map also set the other one to the opposite value,
so ascii on forces tiled off and vice versa.
The X11 interface reads file NetHack.ad (after cd'ing to the playground
directory, where 'make install' puts a copy) and feeds the contents to
X Windows for use as default resources to override the compiled in
defaults. When use of #define was introduced into NetHack.ad (back in
September, 2016) this was severely hobbled and startup spit out a lot
complaints to stderr about invalid resource values. This implements
rudimentary macro expansion for '#define name value' within the data
stream that's fed to X, getting back decent default values and
eliminating the invalid value complaints.
Fix several warnings about using 'void *' for a function pointer and
a couple of unused variables. Add a_nfunc for 'int NDECL((*func))'
alternative for union anything. Make the enum list of union anything
types actually match the alternatives (field a_uchar was missing from
enums, enum mask32 had no corresponding a_mask32 field).
Add another command, #therecmdmenu, so that the context menu for an
adjacent spot can be tested without mouse support. It revealed that
you could get an empty menu if nothing applicable was at target spot.
Add a few adjacent actions: lock/unlock door if carrying suitable
implement, search door for traps, examine known trap (door/ceiling,
not door), #untrap known trap, mount saddled critter, remove saddle.
Make "kick door" be the last choice for closed door instead of first.
Add one 'here' action: dismount.
Both #herecmdmenu and #therecmdmenu interact strangely with ^A, but
differently from each other. I didn't make any attempt to solve this.
There's no documentation for #therecmdmenu.
