hubertf [Wed, 22 Jun 2005 20:11:03 +0000 (20:11 +0000)]
* Change program logic to not nest needlessly
* Search all acronyms databases, and don't force the user to
know in which category to look (-t is gone)
* If an acronym is not found in the database or by whatis(1),
also check pkg_info(1). Per PR bin/30539 by Geert Hendrickx
(geert.hendrickx@ua.ac.be)
Fixed wrong use of datatypes. In wait_for(), a char was used together
with getchar(), in readchar(), a char was used to read input, which in
getuchar() is used as an argument to islower() and toupper(). Also removed
a condition which terminated the process if the user typed in character
255. Approved by christos.
jwise [Mon, 21 Mar 2005 18:45:18 +0000 (18:45 +0000)]
Fix a long-standing bug in wump(6) whereby hitting ^d to exit would first
print ``Care to play another game? (y-n)'' (but then exit anyway without
waiting for an answer).
jsm [Tue, 15 Feb 2005 12:58:21 +0000 (12:58 +0000)]
Don't presume phantasia's internal bool (char) is the same as that in
curses.h: define phbool and use it where necessary to avoid declaring
functions with one bool and defining them with the other.
jmc [Thu, 9 Dec 2004 05:15:59 +0000 (05:15 +0000)]
Redo how void gets init'd.
setup now just creates an empty file. When main starts and loads it in,
it stats and if zero size init's a new location and saves it back out.
Now games.tgz can be shared among MACHINE_ARCH's
christos [Sun, 2 May 2004 21:31:23 +0000 (21:31 +0000)]
- don't catch STOP, TSTP, QUIT; the first you cannot catch and the other two
you should not.
- don't accept delay == 1000, because it turns to 0 and fix the number parsing
- misc KNF cleanups.
he [Sun, 11 Apr 2004 13:35:06 +0000 (13:35 +0000)]
Undef bool before including <curses.h>, since it tries to typedef
bool. Otherwise we get (at least on gcc 2.95.3) an empty declaration
warning from "typedef char char;" due to the #define of bool in
phantstruct.h.
If the current player has no cards, skip to the other player.
While we're here, make this compile with -DDEBUG.
Fixes PR bin/12530 from Thomas Klausner.
jsm [Fri, 13 Feb 2004 23:16:11 +0000 (23:16 +0000)]
Encode and decode more characters. Don't put commas after every
character. Don't encode whitespace as "...". Don't fail decoding
across a 1024-byte boundary. Mark end of message appropriately. From
OpenBSD but with punctuation taken from ITU-T Recommendation F.1
(03/98).
jsm [Sun, 8 Feb 2004 11:47:36 +0000 (11:47 +0000)]
Check large factor for being prime before applying Pollard's
algorithm; fixes "factor 2147483647111311". Correct comment;
algorithm is Pollard p-1, not Pollard rho. Increase base if p-1
algorithm reaches 1; fixes "factor 99999999999991". Testcases from
David A Bagley <bagleyd@tux.org>.
jsm [Thu, 1 Jan 2004 16:03:46 +0000 (16:03 +0000)]
When cleaning up, only close files that have been opened successfully.
When reporting errors for failing to open files, give error text
rather than errno number. Patch from Joey Hess <joey@kitenet.net> for
Debian bug 187251 from Mikael Hedin <micce@debian.org>.
jsm [Thu, 1 Jan 2004 16:02:07 +0000 (16:02 +0000)]
Return proper fortune length from fortlen for fortune files without
STR_RANDOM or STR_ORDERED rather than comparing it to SLEN. Bug
reported by Simon Williams <simon@no-dns-yet.org.uk>.
jsm [Thu, 1 Jan 2004 16:01:25 +0000 (16:01 +0000)]
Show residual probability rather than non-residual in "No place to put
residual probability" error message. Distribute residual probability
correctly rather than all to one file. Patch from Mike Castle
<dalgoda@ix.netcom.com>.
jsm [Thu, 1 Jan 2004 15:59:58 +0000 (15:59 +0000)]
Only score as a Gammon or a Backgammon if the opponent hasn't removed
any pieces, not if the opponent has removed pieces but has had a piece
knocked back to the bar afterwards. Fixes Debian bug 185101 from
Konstantin Seiler <list@kseiler.de>.
jsm [Thu, 1 Jan 2004 15:57:12 +0000 (15:57 +0000)]
Document that pressing return allows you to "fast forward" the game
clock. Patch from Peter Maydell <pmaydell@chiark.greenend.org.uk> in
Debian bug 216980.
lukem [Tue, 21 Oct 2003 10:01:19 +0000 (10:01 +0000)]
Rework how MAKEVERBOSE operates:
* Don't bother prefixing commands with a line of ${_MKCMD}\
and instead rely upon "make -s". This is less intrusive on
all the Makefiles than the former. Idea from David Laight.
* Rename the variables use to print messages. The scheme now is:
_MKMSG_FOO Run _MKMSG 'foo'
_MKTARGET_FOO Run _MKMSG_FOO ${.TARGET}
From discussion with Alistair Crooks.