summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* catch up to huntd changes.dholland2014-03-293-37/+38
| | | XXX: this needs to stop promiscuously sharing .h files.
* tsort contents of filedholland2014-03-291-49/+43
|
* const in unused codedholland2014-03-291-7/+7
|
* remove stubbed-out functiondholland2014-03-293-53/+7
|
* Rip out the code that broadcasts talk requests to the localdholland2014-03-297-707/+7
| | | | | | | | | | | | | | | | | | | | | | hunt-players mailing list. In this day and age it doesn't even begin to work: even if you were to have such a mailing list locally, sendmail wouldn't expand it for you, the addresses you got back wouldn't be talk-requestable addresses, talk requests don't cross NAT or firewalls safely, talk doesn't support ipv6, and nobody runs talkd anyway. It doesn't make sense to try to replace it with some other kind of notify scheme either; there might still be a multiuser machine somewhere with enough simultaneous users that broadcasting something akin to talk requests locally might make sense, and where a nontrivial number of the users actually play hunt, but probably not, and if so those users are likely to be better off using wall(1) anyway. Nowadays games will be set up by looking for people in chat or by emailing or texting friends. Theoretically someone could set up an internet hunt metaserver for finding hunt games, but that would be something entirely different anyway and I doubt there's demand.
* Remove empty header file.dholland2014-03-296-50/+9
|
* Use setproctitle() instead of handrolled version.dholland2014-03-292-26/+7
|
* Make the code for issuing talk requests to find players actually build.dholland2014-03-296-45/+55
|
* don't declare own boolean typedholland2014-03-299-128/+120
|
* use random(), not a handrolled 16-bit generatordholland2014-03-291-12/+4
|
* Use getaddrinfo() instead of gethostbyname().dholland2014-03-291-7/+23
|
* typo in error messagedholland2014-03-291-3/+3
|
* constdholland2014-03-291-4/+4
|
* simplify silly logicdholland2014-03-291-14/+11
|
* temp is a pointer, and tempmax, not sizeof(temp), is the length ofapb2014-03-281-3/+3
| | | the buffer that it points to. Adjust a strlcat() call to suit.
* If you are going to use memcpy instead of struct assignment, then atapb2014-03-281-3/+3
| | | | | least use sizeof(the correct variable). The use of memcmp to test two pointers for equality, one line earlier, is also weird but may be correct.
* Add note cautioning against bothering with WARNS=6 until gcc improvesdholland2014-03-231-1/+4
| | | | | | | | (or -Wconversion is removed from WARNS=6) as it produces loads of false positives. The most entertaining of these that I've seen this afternoon: games/hack/hack.apply.c:143:22: error: conversion to 'unsigned char:1' from 'int' may alter its value [-Werror=conversion] flags.move = multi = 0; ^
* fix signedness issuedholland2014-03-231-4/+4
|
* gratuitous (but not harmful) changes to pass gcc48's -Wconversiondholland2014-03-231-10/+12
|
* Use ssize_t for read() results. Don't use int for lseek() results.dholland2014-03-221-6/+5
| | | Call srandom() with time(), not getpid().
* Use ssize_t for read() results. Don't use int for lseek() results.dholland2014-03-221-6/+6
|
* Use uid_t to hold getuid() results, and don't check getuid for failuredholland2014-03-221-5/+3
| | | as it isn't allowed to fail.
* use size_t instead of int for a lengthdholland2014-03-221-3/+3
|
* Use random(), not rand().dholland2014-03-222-5/+5
|
* avoid unnecessary floating point opsdholland2014-03-221-4/+7
|
* avoid signed left-shiftdholland2014-03-221-5/+5
|
* Use CHAR_BIT, not 8.dholland2014-03-221-2/+3
|
* don't declare ospeed; it's in <termcap.h>dholland2014-03-221-3/+2
|
* Make the list of place types an enumeration. Use unsigned values fordholland2014-03-226-29/+30
| | | place numbers.
* use a more suitable typedholland2014-03-221-3/+4
|
* use more suitable typesdholland2014-03-223-11/+13
|
* woops, more of previous.dholland2014-03-222-7/+7
|
* Make some things unsigned that should be (why would we need e.g. adholland2014-03-224-45/+52
| | | negative number of airports?) and remove some related bogus casts.
* Don't cast the return value of malloc.dholland2014-03-222-13/+13
|
* Minor tidyup.dholland2014-03-221-6/+6
|
* tag decl() __noinline, as inlining it would defeat its purposedholland2014-03-221-3/+3
| | | | (which is to not have interesting string literals appear in the compiled binary)
* Revert part of previous -r1.21 and fix the errno-handling bug instead.dholland2014-03-221-1/+4
|
* move extern decls to .h filesdholland2014-03-222-6/+6
|
* Merge riastradh-drm2 to HEAD.riastradh2014-03-181-1/+1
|
* fix typo, from Henning Petersen in PR misc/48613dholland2014-02-221-1/+1
|
* remove duplicatedholland2013-12-111-4/+0
|
* another minor fixdholland2013-12-111-1/+1
|
* Reinstate typo since kre claims it's by purpose.wiz2013-12-111-1/+1
| | | Discussion in PR 48442.
* Error and typo fixes from OpenBSD and FreeBSD via Henning Petersenwiz2013-12-111-19/+17
| | | | | | in PR 48443. While here, remove a duplication, fix another typo, and don't claim that Mr. Sheperd was the first man in space.
* Error and typo fixes from OpenBSD and FreeBSD via Henning Petersenwiz2013-12-111-16/+16
| | | in PR 48442.
* WARNS=5dholland2013-12-071-2/+1
|
* hals_end(6) outputs the last words of the supercomputer HAL 9000 aboardmbalmer2013-11-124-2/+228
| | | | | | the spaceship in Stanley Kubrick's famous film "2001 - A Space Odissey". The source code and output of this program were used to illustrate an article in the book "Total Interaction". How this looks in print can be at http://www.netbsd.org/~mbalmer/hals_end.jpg.
* fix unused variable warningschristos2013-10-196-27/+16
|
* Untabify the header definition to avoid misaligned comment onuwe2013-09-191-11/+11
| | | str_delim in PostScript output.
* Use .Fa instead of .Ar to refer to header fields.uwe2013-09-191-4/+4
|
generated by cgit v1.2.3-56-ge451 (git 2.39.0) at 2024-06-30 03:54:06 +0000