For the POSIX regexp interface, move local declaration to beginning
of block to avoid requiring C99. Also switch to alloc() from bare
malloc() so that MONITOR_HEAP won't log a free which doesn't match
up to any allocation. This results in a change in behavior: if
the allocation fails, nethack will panic rather than report an
option parsing error. In practice there will be no difference
because nethack is not going to run out of dynamic memory during
initial options processing.
The intent is to look for platform-specific facilities for regex
matching, to provide portable MENUCOLORS configuration files.
This is a prototype implementation being committed to see if Windows can
use the POSIX regex implementation provided with the C++11 standard
library. If this works, I will write a harness for POSIX regexes and for
pmatch(), and those can be linked in by platforms as appropriate.
pmatch() should be used only as a very last resort, because it breaks
compatibility between platforms.
Changes to be committed:
modified: include/extern.h
modified: src/bones.c
modified: src/do.c
modified: src/files.c
modified: src/music.c
modified: src/restore.c
modified: src/save.c
modified: sys/share/pcmain.c
modified: sys/share/pcsys.c
modified: sys/share/pcunix.c
In order to get level file locking correctly again post 3.4.3
with the newer compilers for windows, I had to funnel close()
calls to an intercepting routine.
I had two choices:
1. Surround every close() in at least 9 source files with messy:
#ifdef WIN32
nhclose(fd);
#else
close(fd);
#endif
OR
2. Replace every close() with nhclose() and
deal with the special code in the nhclose()
version for windows, while just calling
close() for other platforms (in files.c).
It is also possible, although not done in this commit,
to
#define nhclose(fd) close(fd)
in a header file for non-windows, rather than funnel
though a real nhclose() function in files.c.
-Add a boolean option menucolors to toggle menu color
-Add MENUCOLOR -config file option
TODO:
-Better support for win32
-Support more windowports
-Update Guidebook
-Allow changing menucolor lines in-game
Re-run nhgitset.pl to install.
"perldoc DEVEL/hooksdir/nhsub" for details. General docs still to come.
Quick notes:
- "git nhsub" lets you apply substitutions to a file without involving any
version control.
- When doing nhadd/nhcommit, the working directory WILL reflect the results
of the substitutions.
Let's see what this breaks.
Things won't build for ports that first
define SYSCF.
This moves assure_syscf_file() from unixmain.c
to files.c and adjusts extern.h to get it
out from under #ifdef UNIX.
The call to assure_syscf_file() in options.c was
only #ifdef SYSCF, SYSCF_FILE and not UNIX,
so new ports #defining SYSCF would get an erro.
assure_syscf_file() will be utilized by mswin
when SYSCF is defined.
There is a lot of code affected by this, and Pat Rankin correctly
observes that it would be better to store roguelike as a level flag
rather than just using Is_rogue_level. A note for the future.
There is a lot of code affected by this, and Pat Rankin correctly
observes that it would be better to store roguelike as a level flag
rather than just using Is_rogue_level. A note for the future.
gcc reports "comparison between signed and unsigned". One header file
uses unsigned long for tty flags, another uses [signed] short. 'unsigned'
seems like the best compromise, but this might accidentally introduce lint
for some other configuration.
Remove date.h and patchlevel.h from win/tty/wintty.c, win/X11/winX.c,
and sys/share/pcmain.c (caveat: the latter two are untested) so that they
don't get recompiled every time any other source file changes and triggers
creation of a new date.h. Only version.c needs to be recompiled in that
situation. Also, Makefile.src was missing a reference to botl.h.
Tested on the unix port; I've updated as many other ports as I can figure
out but they're not tested. See window.doc for info on the changed banner
lines. Also adds the ability to override the generic "Unix" port - used now to get
"MacOSX" into the version line instead of "Unix" (so we don't scare people who don't
know what's going on).
This is all tiny stuff - allow overriding WIDENED_PROTOTYPES from the hints
file, missing NO_SIGNAL conditionals, remove a GCC-ism, conditional indentation,
void return in a non-void function.
PORTS: Please make sure I've done the right thing for/to your code.
This patch adds a new winproc that lets the window port approve or cancel
the suspend request - this should take care of the Mac Qt lockup issue.
In addition, Unix suspend is restricted to accounts that can use the shell
if SYSCF is defined.
A few other fixes:
- bug: text window does not scroll left on left arrow
- pcmain.c: fix warning: _fileno() returns void* in Windows CE SDK
3.5.0 only:
- files.c: compile error: don't include <signal.h> if NO_SIGNAL is defined. WinMo does not support POSIX signals
- version.c: compile error: append_port_id() is undeclared
- link error: mktime() is unsupported -> define it in celib.c
-<Someone>
cvs admin -ko sys/share/*_*.c to suppress $Date$ and $Revision$ expansion,
reverting to the text checked a few days ago.
cvs commit -f sys/share/*_*.c to trigger this mail; you'll need cvs update
to get the actual text.
It always struck me as odd that x_maze_max and y_maze_max were
initialized in main(). They're only needed when making new levels, so
don't have to come before saved game restoration. They could easily have
gone into init_dungeon() (although they make well predate that), or even
mklev() (reinitializing them for each new level wouldn't have been a big
deal). That's all moot, though, since it's trivial to initialize them at
compile time.
[See cvs log for src/role.c for a much longer description.]
When picking role, race, and so forth, new menu entries allow you to
pick any of the other items before the one currently being handled. After
picking all four of race, role, gender, and alignment (or if you answered
'y' to "shall I pick for you?"), there is a followup prompt to confirm the
choices. It's a menu which also provides a chance to rename the character.
This has only been implemented in win/tty's player_selection(), with
some support code in the core that might be useful to other interfaces.
And so far, the chance to rename is only presented as a menu choice if
you've given an answer to "who are you?" prompt earlier during startup.
Also, ports that use pcmain.c aren't able to perform hero renaming yet.
More warning bits that never got committed.
More appropriate compiler flags for warning checks (macosx only for the moment).
The changes in dgn*[lc] just rename line_number to nh_line_number to avoid a
clash, so no need to regenerate the lex output.
