I was asked how a window-port controls which options are
set to SET_IN_FILE, DISP_IN_GAME, or SET_IN_GAME.
This provides a run-time way to change an option's SET_IN_FILE,
DISP_IN_GAME, or SET_IN_GAME status through code, rather
than clog up options.c with a lot of compile-time #ifdefs
for different ports to offer different default option settings.
Update the documentation to reflect this.
- added support for new options
- fixed couple of typos in options.c
- added sunken edge around message window
- added capability to replace font in the font table (if we will want to
change font sizes dynamically later)
- visual feedback after a prompt: append an empty line to the message
window when clear_nhwindow(WIN_MESSAGE) is called. Filter
out empty lines in the buffer except for the active slot.
- append ellipses to the status line text if the text is truncated
- get rid of message boxes at the end of the game (ignore empty lines in
raw_print as Yitzhak suggested)
Add absent prototypes to some core routines.
Also add some port function() to function(void) in some win32 routines.
Also updates the Borland C Makefile for win32.
As Warwick suggested, instead of having fixed tile
sizes as options, allow specification of the size
explicitly.
Also, at Yitzhak's suggestion, provide a hook for
overriding the port's tile file name. That name,
and the contents of the file it points to, will
be window-port specific of course.
<Someone> wrote:
>I also note that it's weird for your nose to stop tingling and your
>sense of smell to return to normal, if flags.beginner prevented your
>realising that your nose was tingling in the first place.
The following fixes several bugs:
1) Mismatch between docs and game in definition of what '+' resolved in
favor of docs...
2) When game needs to be recovered a message box is shown. This is a very
deprecated fix. It pretty much answers just the conditions that require
this (a yn question to an erroneous winid), and is not useful for other
purposes.
3) The score file is written.
to allow common parsing in the core, and direct access to the
results by the window port.
Notes:
o Adds a new field, wincap, to the window_procs
structure for setting bits related to the preference
features that the window port supports. This allows
run-time determination of whether a particular option
setting is applicable to the running window port. A
window-port is free to support as many, or as few,
of the available options as it wants. Ensure that
only the ones supported have their corresponding bit
set in window_proc.wincap. [see chart in
doc/window.doc for help with that.]
o The settings I stuck into wincap for each window
port are almost certainly not accurate, so each port
team should review them. You should only include
the ones that you will actually react to and make
adjustments for if the user changes that option.
Without the setting in wincap, the option won't even
show up in the 'O'ptions menu.
o preference_update() added to the window-port
interface, so that the window-port can be notified
if an option of interest (an option with its
corresponding bit set in wincap field) is
changed.
o provided a genl_preference_update() routine in
windows.c and used it for all the existing
window ports since they don't have a functional
one of their own yet.
o this messes around heavily with iflags and the options
arrays in options.c
o I hope I didn't break any port's existing code. I
tried not to. The Mac however, in particular, should
be looked at because it suffered a namespace collision
with what I was working on around fontname. It had
Mac specific font stuff in options.c. Please test
the Mac.
I'm surprised that you credit me with the ninja yumi-ya thing as a
"patch" in README; I didn't submit any code for it, it's trivial, and
it's vanishingly unlikely ever to come into play :-) If I'm to be
credited with anything on that list, I'd have thought it'd be the
no starting pet option.
Apparently, " is part of an accented character in US(international) keyboard
layout (additional character specific to the language of an origin - for
example, left and right double quotation marks). The code did not handle it
too well since it maps to 2 ASCII characters instead of one (one of them is
so called "dead" character). We can ignore the dead character as a
workaround for this problem. The patch is attached.
<Someone>.
The patch is attached. Array bounds went unchecked in
the menu page operations (, \ ~) This resulted in
memory corruption.
The actual crash depends on your luck actually. It will
only crash if heap headers are corrupted, otherwise it
can go unnoticed. When you do "Du," the list page size
is 18 (on my screen) with only 2 items in the menu. The
program assigned count of -1 to 18 items in the array
of 2. Ka-boom. I put bounds checking code in several
places. The window size does not have anything to do
with it.
<Someone>.
- allow spaces between the different possibilities
- add a missing null at the end of the list of prefixes to prevent index()
from going crazy
- slight re-wording of the Guidebook
- B1026 [confirmed] win32: setting multiple options
- bugfix: menu: set focus to the item and make it visible when it is
selected with accelerator key.
- increased menu font size by 1 point
- layout the main window on WM_MOVE message so the menu windows stays with
the main window (it was not changing the location before)
- got rid of WinCE specific define (ARM)
- handle all keyboard input via WM_KEYDOWN instead WM_CHAR
Files affected: config1.h mhmsg.h mhmain.c mhmenu.c mswproc.c resource.h
winhack.rc
Notes:
That overrides Yitzhak's mhmenu patch - I don't like the menu prompt on the
title bar since it can be null and in most cases it is. It looks much better
as a header column in the list and is displayed only if present.
I moved ARM-related (processor) defines to wceconf.h You cannot undef ARM
before windows.h is processed - ARM version won't compile.
.rc files are generated by IDE - you cannot edit them directly or your
changes will be gone next time the file is saved. If anything needs to be
added to winhack.rc file manually it has to be marked by the following
defines or via menu option View->Resource Includes:
#define APSTUDIO_READONLY_SYMBOLS
...
#undef APSTUDIO_READONLY_SYMBOLS
[...]
One of them deals with the patch you just sent, that changes the name but
does so more extensively, particularly, also in the Save dialog in
mhmain.c. Another is one of the changes to winhack.rc to make it compile
under Borland.
This also (beyond patches I sent): fixes no Options prompt bug, and fixes
up the menu bar (which I broke in one of my patches).
Tested on msc/bcc tty/win makefiles and vc ide.
B2001 <Someone> [reported] win32?: giant ant is black square
Where is the open doorway with the giant ant standing in it?
There used to be a closed door there, if I recall correctly.
[now there is just a black square]
I'm saving the level files at this point, and I will be saving
the game after I did that, in case you're interested in any save
files.
>
> I'd like the default for "Would you like to see your <whatever>"
> at the end of a game to be "y" instead of "n". I haven't asked
> for full disclosure in order to have it skipped if I press the
> space bar once too often by mistake.
This changes the way the flags.end_disclose array is used to
allow what this request is asking for. It should be backward
compatible with previous "disclose" options.
The order that the end_disclore options are stored:
inventory, attribs, vanquished, genocided, conduct
There is an array in flags:
end_disclose[NUM_DISCLOSURE_OPT];
with option settings for the each of the following:
iagvc [see disclosure_options in decl.c]:
Legal setting values in that array are:
DISCLOSE_PROMPT_DEFAULT_YES ask with default answer yes
DISCLOSE_PROMPT_DEFAULT_NO ask with default answer no
DISCLOSE_YES_WITHOUT_PROMPT always disclose and don't ask
DISCLOSE_NO_WITHOUT_PROMPT never disclose and don't ask
Those setting values can be used in the option
string as a prefix to each disclosure option
to get the desired behaviour for that option.
For backward compatibility, no prefix is actually required,
and the presence of a i,a,g,v, or c without a prefix sets
the corresponding value to DISCLOSE_YES_WITHOUT_PROMPT;
The actual prefixes used are controlled by the following in flag.h:
#define DISCLOSE_PROMPT_DEFAULT_YES 'y'
#define DISCLOSE_PROMPT_DEFAULT_NO 'n'
#define DISCLOSE_YES_WITHOUT_PROMPT '+'
#define DISCLOSE_NO_WITHOUT_PROMPT '-'
As far as the docs go, I don't know if I've got the *roff
stuff right. The TeX stuff looks okay when I converted it to .pdf.
This increments EDITLEVEL. If that is a problem, I can
add a routine to restore.c to perform a conversion of the old
values in flags. Let me know.
was added, specific to menustyle:combination in
menu_drop(), which short-circuited a redundant
prompt that <Someone> had complained about by returning
immediately if a special code came back from
ggetobj() in ggoresults.
The top of menu_drop, however, had created a
u_gold object and inserted it into the invent
chain, so you can't just return. You have
to use a "goto" to get to drop_done, so that
the object is cleaned up.
- Fix the tty build so it wasn't trying to link with GUI switches.
- Don't make graphical the default for the Makefile. It is for the
IDE projects.
- Be explicit about compiler versions required/tested.
- added counting to the menu window. I changed the way it works in the TTY
port though ("you hear the rumble of distant thunder..." :) The count
applies to the currently selected item instead of being aplied to the item
to be selected (in other words a12 instead of 12a). It works better this
way in the graphical port since there is a notion of current menu item.
I also notes that the count is ignored for PICK_ONE menu - is this by
design?
- somebode mention that text is not scrollable by default - this is fixed.
- fixed player selection dialog to use initXXXX options
-----------------------
B1028 [reported] win32: popup window is too short
The text in the popup window's top line "What do you want to
call the ..." is cut off. Could the window be bigger, so you
can read the entire line?
Fixed.
---------------------
B1027 [reported] win32: select via traditional class char
In a shop (and perhaps elsewhere), I can autoselect a class of
items to drop: a for all types, b for coins, c for weapons, ...,
A for every item, B for blessed items, ... I like the new
possibilities, but I would like to be able to use the old way of
using ! for potions, ) for weapons, etc., as well.
Fixed.
---------------------
B1026 [reported] win32: setting multiple options
I want to set packorder and pickup_types. The first comes up
first, and I type in the order. Then comes the second, without
any hint what is asked of me. A little explanation here would
be appropriate, like "pick up what things automatically?" or
somesuch. When more than one option is set where extra input
is required, some explanation is useful.
Looking into. Apparently something end_menu() call is ignored by nethackw.
---------------------
B1025 [reported] win32: menu navigation
I press O. Now, when I press "a" to turn on "autodig", not only
is the mark set at "a", but there also appears a dotted
rectangle around option "A" (capitalized). Shouldn't this
rectangle also appear at "a"? Pressing "a" repeatedly toggles
the checkmark, but the rectangle switches between "A" and "a".
My CAPS LOCK is not set. The same goes for B, C, D, ... N,
by the way. Pressing "A" works the other way around: the first
press checks "A" and "rectangles" "a", the next unchecks "A"
and rectangles "A" as well.
Fixed.
---------------------
B1024 [reported] win32: no text in final window
After I clicked away the last screen with information
(challenges, I believe it was), I got an empty text screen, as
big as my screen, without any content. It looked like the text
screens that appear on the right, only it was larger and had
only a large OK button over the full width of the window, and
no cancel button. Why this window?
Fixed. The program did not set window text for RIP window.
---------------------
B1010 [known issue] win32: cosmetic
When I get my list of spells, the columns aren't properly
aligned: if the spell name is longer, the rest of the line
shifts right.
Fixed. Fixed-pitch font is used for menu window.
---------------------
B1004 [reported] win32: numeric keyboard Del/.-key
The Del/.-key on my numeric keyboard doesn't make me rest a turn
Fixed. Added rest on Del.
---------------------
B1003 [reported] win32: focus
When the window has the focus, the window bar isn't highlighted.
This is by design. Menu and text widnows are popup and the main
window is disabled when they are up.
Not using FDECL to declare the prototype for dogmove.c:can_reach_food()
causes the Digital UNIX C compiler to throw a prototype mismatch error
when compiling dogmove.c.
1. The switch statement was using the material "GOLD"
rather than GOLD_CLASS.
2. If getobj() had been working for gold when it
came to touchstones, there would have been a
memory leak here because the object returned
would have been from mkgoldobj(). The goldobj
was not being freed anywhere, nor was it being
put on a chain. You also would have had zero
gold after rubbing it on the stone. The intent
was clearly to allow gold since there was a
case in the switch statement.
3. getobj() wasn't working properly for gold
selection here anyway, so this was
not the cause of <Someone>'s gold obj in inventory.
You ended up dropping through to code that
was supposed to print "You cannot verb object."
For touchstones that came out as:
"You cannot rub on the stone gold."
because there was an extraneous argv[1] that was
a repeat of the module name, but in quotation
marks. The processing in pcmain stops on the first
argument that doesn't start with '-' so my other
arguments got ignored.
argv[0] 0x00b40800 "C:\test\binary\nethackw.exe"
argv[1] 0x00b40878 ""C:\test\binary\nethackw.exe""
argv[2] 0x00b408f0 "-uwizard"
argv[3] 0x00b40938 "-D"
Fixes:
- menu shortcuts implemented
- most windows close on space (except for menus with
PICK_ANY style)
- "hilite_pet" option is implemented
- map scrolling is improved somewhat (it now scrolls if
the char is within 5 spaces from the edge of the map -
configurable by #define CLIPAROUND_MARGIN)
- added 3 winhack-specific options:
win32_map_mode:[tiles|ascii4x6|ascii6x8|ascii8x8
|ascii16x8|ascii7x12|ascii8x12|ascii15x12
|ascii12x16|ascii10x18|fit_to_screen]=20
win32_align_status:[left|top|right|bottom]
win32_align_message:[left|top|right|bottom]
Note: aligning status window to left or right edge of the screen does
not look good.
over the place.
Often they would use
"%ld zorkmid%s", amt, plur(amt)
but not consistently, so some of the hard-coded usage
could result in "1 zorkmids"
This adds the function
currency(long)
to return the name of the currency, either plural
or singular depending on the argument passed to it.
That eliminates the need for the extra %s in the
format string and the use of the plur() macro.
so that it will not be included in the diff between
the versions.
Also note that the -ko option in effect for that file
causes it to leave the value at whatever is
checked in from now on.
Regarding the Shrieker source,
http://www.peldor.com/chapters/story678.html:
" The player characters depicted in these stories are copyright
1991-2000 by Thomas A. Miller. Any resemblance to any persons
or characters either real or fictional is utterly coincidental.
Copying and/or distribution of these stories is permissible under
the sole condition that no money is made in the process. In that
case, I hope you enjoy them!
Thomas A. Miller"
This was caused by character set incompatibility between message and map
windows. Apparently, Nethack is using IBM character set (CP 437) but fonts
were create for Windows ANSI codepage. I fixed most of it by changing
character set of the fonts except for the Rogue level. I had to make changes
to src/drawing.c for Rogue level since most of Windows fonts are not capable
of displaying control characters (char. code<0x20)
And of course, disabling IBMGraphics option fixes it all.
If you have an unidentified gray stone in your pack,
you just had to use the 'a'pply command. If it was
a touchstone, it would be included in the list,
otherwise it wouldn't.
This allows any stone to be included in the 'a'pply
command.
Also, adds missing punctuation to the use_stone()
messages.
Also prevents rubbing a stone on itself.
nethack.allison