The report that parse_config_line() could overflow its buffer
when passed a long value from parse_config_file() was already
fixed for 3.6.1, but longer config lines mean that later stages
of option parsing need to be aware of the possibility of lines
longer than BUFSZ. This fixes the three special options which
deal with regular expressions. Autopickup exceptions, message
types, and menu coloring all could exceed BUFSZ when formatting
a value to put in the menu for 'list' or 'remove' of existing
entries. Menu colors could overflow by an arbitrary amount,
message types by up to 13 bytes, and autopickup exceptions by
just 1 byte.
Both tty and X11 seem to cope with long menu entries (wider
than can be displayed but not exceeding BUFSZ) by truncating.
For menu colorings, that hides the color and highlight attribute
since those are placed after the regexp.
Menu line truncation for tty was straightforward, just chopped
a big chunk off the end of the long string. For X11, it did
that too, taking off a lot less but ending up with much of the
menu hidden off the left edge of the screen. Dragging it into
view showed that the width fit the whole screen (or possibly
fit the width of the map, which was also as wide as the screen).
So the initial position is being miscalculated.
I glanced over the options parsing looking for issues, and it
seems that paranoid_confirm:long-string-of-junk might have been
vulnerable. So this fixes that.
...to make it more interesting. Using #vanquished in wizard mode
or answering 'a' to the "disclose vanquished monsters?" prompt will
put up a menu to choose how the list of vanquished monsters should
be ordered. Right now there are 6 choices:
Traditional: by monster level, by internal index within level;
by monster toughness, by internal index within monstr[] rating;
alphabetically, first unique monsters, then others;
by monster class, low to high level within class;
by count, high to low, by internal index within tied count;
by count, low to high, by internal index within tied count.
Two other orderings are implemented but suppressed from the menu
since they seemed uninteresting (alphabetical with uniques
intermixed with other monsters, and by-class high to low within
class). The first two are very similar to each other and one of
them should probably be discarded too. The by-class order(s) have
class-name separator lines between classes.
Options parsing for end of game disclosure has extended current
+v (always show vanquished monsters)
-v (never show vanquished monsters)
yv (prompt about them, with default response 'y')
nv (prompt about them, with default response 'n')
to include
#v (always show vanquished monsters and choose the ordering)
?v (prompt about them, with default response 'a' to choose ordering)
The 'a' response was picked because it's easy to use ynaq()
instead of ynq(), but it can be considered to mean "ask about sort
order". (Neither of the two new option values could be "av"; 'a'
for disclosing attributes would become ambiguous.)
+v or answering 'y' for any of yv, nv, or ?v uses the most recent
sort ordering (if #vanquished has been used in wizard mode) or the
traditional one (normal mode, or #vanquished not used). Players
will probably want to specify a default order and then use +v
rather than choose the final order from a menu. That hasn't been
implemented here. Count high to low might be a better default than
level high to low.
While looking through Guidebook.tex to try to determine whether
the new text needed special handling, I spotted multiple mistakes
in the existing text. Probably all from earlier updates of mine;
this attempts to fix them. As usual of late, Guidebook.mn has been
tested and Guidebook.tex hasn't.
flags.sortloot was changed from boolean to xchar, but proper type
is plain char. (Presumeably it was originally off or on, then got
changed to 'n' for off, 'l' for on, plus 'f' for super-on.)
Also, make the sortloot menu for 'O' mark the current value as
preselected when interactively setting the value. (I've been
meaning to do this for various other options but haven't gotten
around to it. The need to workaround PICK_ONE+MENU_SELECTED menu
behavior is a nuisance.)
Add support for character enclosed within single quotes. Single
character without quotes would work for most characters, but not
'#' and possibly not '^' or '\\'. All the values in dat/symbols
are specified via backslash+digits so it isn't obvious that some
other form of value is allowed.
I think this parsing accepts all valid values. It doesn't reject
all invalid ones: opening quote followed by valid escape sequence
followed by junk followed by closing quote will ignore the junk.
I don't think there's any pressing need to worry about that.
Noticed when testing tty menucolor changes recently, using the 'O'
command to interactively add a new entry would accept one and then
quit, unlike remove which accepts multiple at a time and then goes
back to the menu where you could choose 'remove' all over again.
Adding is still done one entry at a time, but instead of finishing,
it goes back to the menu where you can choose to add another.
When removing items from a container via menu, list gold as '$'
instead of 'a' when it is the first item. Requested during beta
testing last year....
When gold isn't first ('sortpack' false, or custom 'inv_order[]'),
it uses the next letter in sequence instead of '$', otherwise it
would be the only item out of sequence.
Another OS upgrade (OSX 10.6.8 -> 10.8.5) with different toolset,
another change in compiler behavior. Earlier 'gcc -Wwrite-strings'
didn't complain about passing string literals as 'char *' paremeters
if there was no prototype in scope. This one found one or two of
those in options.c and several in makedefs.c (fix coming soon in a
separate commit...). This adds some missing prototypes and reorders
the existing ones to match their order within the file. There were
also several functions which were declared static in their advance
declarations but not in the definitions, which can be confusing when
reading the source.
Force a screen redraw if the tiled_map option is toggled via the 'O'
command. The X11 interface switches map modes even without this, but
conceptually it's something that must be done when the option setting
is changed.
I upgraded from OSX 10.5.8 via 10.6.3 to 10.6.8, plus Xcode to whatever
version was on the 10.6 dvd, and ended up with a more recent version of
gcc that is configured to use 64 bit longs and 64 bit pointers (by
default; presumably that can be changed if necessary). It triggered
several warnings about converting int to pointer of different size or
vice versa even when explicit casts were in use, and a couple of other
things.
For !menu_tab_sep, the boolean options were formatted with a different
width than the compound and 'other' options when the menu for 'O' was
active. Use the same format string for all of them, consider them all
when constructing it, and construct it just once instead of every time.
This is more robust than the previous hack. The issue of whether to
use it in other places is still unexplored. Ultimately it's the user's
fault if overzealous message suppression hides something important.
[For an eerie game, try 'MSGTYPE=hide .'.]
Rename the option for adding coordinates to autodescribe feedback for
the '/' and ';' commands from 'getpos_coord' to 'whatis_coord', after
the '/' command that uses it instead of after the internal routine
that implements it. The 'whatis' name was only in dat/hh as far as I
could find, so this changes it to 'what-is' and also updates dat/help
and the Guidebook to mention the name too.
Add a 'screen' choice to the option to show coordinates as row,column
rather than x,y or compass direction(s). Revise the /m, /M, /o, /O
operations of 'what-is' to honor the whatis_coord option (mostly; a
value of 'none' gets overridden by 'map' to force coordinates).
Also, update the description of the functionality of the '/' command
in the Guidebook. The .mn version is tested, the .tex one isn't.
Several people have asked if 3.6.0 supports playing with tiles on
a public server. Because there's no way for the user's end to know
what that white @ is, this adds special console escape codes for
tile hinting.
The idea was originally a patch called TelnetTiles by Justin Hiltscher,
but this expanded version comes via NAO, where it's been in use for years.
This is basically an interim hack, which should go away when/if
we support actual client-server model.
This was a request from a blind player. It's hard to find
the left edge of the menu when it's drawn on the map, so
clear the screen and align menus to the left edge of the screen
when this option is turned off.
Originally this was called the window edge patch.
Allow 'msgtype=show' for messages that nethack uses Norep() for.
I don't know whether anyone will ever want to do that, but if felt
strange to have two different message suppression mechanisms that
were completely disconnected from each other.
For a user with no msgtype filter, there'll be no difference in
behavior.
Move autopickup exception, menucolor, and message type entries
in the options menu under "Other settings" -header, and reorders
them into alphabetical order.
Also use enums instead of hardcoded values.
Some people are confused by the boulder -option, and
SYMBOLS=S_boulder, so allow defining symbols with
OPTION-lines in addition to the SYMBOLS.
So these are the same thing:
SYMBOLS=S_boulder:0
OPTIONS=S_boulder:0
Some people try to set boolean options in the config file
by giving the option a parameter, so allow that:
OPTIONS=color:true
Allowed parameters are "true", "yes", "false", and "no".
Negating an option and giving it a parameter is an error.
Options parsing didn't support "default" (shown by the 'O' command)
or "Default symbols" (menu entry for choosing a symbol set via the
'O' command. Symbol handling is somewhat confusing, but this seems to
do the trick. They can't be truncated, but they're case-insensitive,
and "Default" and "symbols" can be separated by dash or underscore as
well as space, or run-together with no separator.
Fix some more of the complaints from clang's static analyzer. The one
in options.c (manipulating warnings symbols) appears to be an actual bug.
All the rest are either because the analysis isn't quite sophicated
enough or outright bogus.
Two of them appear to be because a static routine is attempting to guard
against callers in the same file failing to pass in required output
pointers. Stripping away the check for missing pointer should convince
the analyzer that those output parameters always receive a value. We'll
see once the analysis is eventually re-run....
This isn't urgent, but I figure that until the mac build stuff gets
merged in, the core is still fair game....
'O' command's autopickup_exceptions was freeing a menu pick-list even
when it hadn't been allocated (for the list case, and for the remove
case if nothing was chosen for removal). That code was evidently used
as the model for msgtype and menucolors; they had the same situation.
I think ANSI and ISO sanction free(NULL) as a no-op, but pre-ANSI free
implementations don't necessarily handle that benignly. Even if they
all do, freeing something--even if that 'something' is nothing--which
hasn't been allocated is a bug on our end.
Replace several 'foo = alloc(strlen(bar)+1), strcpy(foo,bar)' sequences
with 'foo = dupstr(bar)' calls.
Change 'free(foo)' into 'free((genericptr_t) foo)' to possibly pacify
'lint' and/or really old compilers.
Add braces around 'if something;' when 'else { otherwise; }' has braces.
Simplify option value formatting for 'sortloot'.
Explicitly combine adjacent string literals so that pre-ANSI compilers
still have a chance to compile the code. I thought these had already
been dealt with, but I kept stumbling across them while reformatting,
so am trying to get them all out of the way now.
Fixing up mis-indented block comments, but hit some files that hadn't
had the earlier mixture of tab replacement, etc, so it's bigger than I
expected. If I get to it, they'll be another round of this tomorrow.
Move this small utility routine to hacklib.c where other such things
live and where it's feasible to find them if you need the functionality
elsewhere.
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.
As reported by Pat:
> After leaving a level and then returning, I'm seeing lit
> room squares as blank
> Options all have their default settings
OPTIONS=nocolor is the default; we should probably change this
sometime, because nearly everyone plays with color.
