Commit Graph

Select branches

Hide Pull Requests

master

Mono Color

• f603d86175 Fixes https://github.com/kristapsdz/kcgi/issues/57 -- thanks! kristapsdz 2018-12-26 17:42:21 +00:00
• 2c002d897f Note linking with linux. kristapsdz 2018-12-26 17:37:16 +00:00
• 74818bf9b7 Add some informative bits. kristapsdz 2018-12-26 17:36:36 +00:00
• e9f843ac5b Note FreeBSD's -lmd requirement. kristapsdz 2018-12-26 17:32:08 +00:00
• f9eaf15e25 Finally fix https://github.com/kristapsdz/kcgi/issues/51 kristapsdz 2018-12-26 17:28:16 +00:00
• 91f8e6b0d7 Mark targets as phony. kristapsdz 2018-12-26 17:25:09 +00:00
• 95decd553b Simplify this rule. kristapsdz 2018-12-26 16:53:46 +00:00
• 8d25b35a0d The culprit to the regression test problem. kristapsdz 2018-12-26 16:49:54 +00:00
• 9afe6599b7 Fix compilation by using .for loops. This binds us to BSD make, but that's ok because either way we need extensions from the usual grammar. Largely for the regression tests. kristapsdz 2018-12-26 16:07:42 +00:00
• 5160d806da Remove GNU makefile in favour of BSD makefile. kristapsdz 2018-12-24 23:06:08 +00:00
• e7e1dfe69a Using BSD makefile (again) for simplicity. kristapsdz 2018-12-24 23:04:21 +00:00
• 3eecfd0d8b Upgrade oconfigure. kristapsdz 2018-12-24 22:40:46 +00:00
• 5c451370eb Version. kristapsdz 2018-12-18 13:28:51 +00:00
• fc2fb7a98a Remove dead class. kristapsdz 2018-12-18 13:28:45 +00:00
• ccdd2922f3 Note musl/glibc. kristapsdz 2018-12-18 13:26:43 +00:00
• ffd90e08b8 Make sure we're not considering the value as a signed char. Great find! kristapsdz 2018-12-18 10:42:18 +00:00
• cbc8d652d5 Add test for JSON parser mangling control characters. kristapsdz 2018-12-18 10:40:52 +00:00
• f084064afe Compilation warning under musl. kristapsdz 2018-12-18 10:19:39 +00:00
• f8ab80418c Move HAVE_ZLIB macros around to prevent compiler warnings about unused functions and variables. kristapsdz 2018-12-18 09:21:11 +00:00
• 1695d9f70b sys/poll.h -> poll.h kristapsdz 2018-12-18 09:03:51 +00:00
• 613aa8ef8c Bring up to date with newest oconfigure. kristapsdz 2018-12-18 09:03:36 +00:00
• 9726880e6f Fix minor scan-build nits. kristapsdz 2018-09-28 14:28:36 +00:00
• 69f2c319b8 No more systrace. kristapsdz 2018-09-28 14:24:30 +00:00
• e5389dab78 Get rid of all allocation/free of sandbox resources, which was only used by systrace. kristapsdz 2018-09-28 14:21:08 +00:00
• 6746ead36f Remove the argument to sandboxers, as that was only used by systrace. kristapsdz 2018-09-28 14:16:06 +00:00
• fc16574dee Remove parent configuration of sandbox, as only systrace used it and our systrace support is no more. kristapsdz 2018-09-28 14:08:07 +00:00
• 49005ddbcb Start process of removing systrace support. kristapsdz 2018-09-28 14:04:56 +00:00
• dc2d864e7a As noted by Valentin Pistol, Mojave works fine now with the regression suite. So reenable it documentation-wise. Thank you! kristapsdz 2018-09-28 14:03:48 +00:00
• 25b5ede081 Tweak language about third-party ports. Thank you! kristapsdz 2018-09-28 13:45:16 +00:00
• a50b737967 Typo fix, thanks! kristapsdz 2018-09-28 13:42:10 +00:00
• f19e4eaa73 Explicitly note whether an argument is a pointer to a struct kreq. kristapsdz 2018-09-28 13:37:03 +00:00
• 337609eabf Note that -lm is required for Linux. kristapsdz 2018-06-19 09:45:23 +00:00
• b91ef48ad7 Document -lm and -lbsd on Linux. kristapsdz 2018-06-19 09:43:56 +00:00
• 4c1647223a Move seccomp debug option into linux-only area. kristapsdz 2018-06-19 09:41:09 +00:00
• 1f829bc6a3 Remove reference to libbsd: it's no longer needed. kristapsdz 2018-06-19 09:39:10 +00:00
• c481ace0fa Linux needs -lm for fpclassify in kcgijson. Also, we don't need libbsd any more. kristapsdz 2018-06-19 09:33:00 +00:00
• 656fad65c4 Remove arc4random depends. (Just use random(), otherwise.) kristapsdz 2018-06-19 09:29:28 +00:00
• d9af66533b Version up. kristapsdz 2018-06-19 09:20:30 +00:00
• 5031a1a63e Make Mac OS X be able to at least run make regress. kristapsdz 2018-06-19 09:16:45 +00:00
• 0688b4e4ba Add application/xml MIME type. We still use TEXT_XML as the default map from .xml suffixes, but this is a start for getting a fix in for https://github.com/kristapsdz/kcaldav/issues/88. Thank you! kristapsdz 2018-06-18 14:12:11 +00:00
• 29d0d295dd Make sure we rebuild regress objects on kcgi.h change. kristapsdz 2018-06-18 14:09:27 +00:00
• 472d00d3fe Add success and failure cases for https://github.com/kristapsdz/kcgi/pull/42, thanks! kristapsdz 2018-06-18 12:48:44 +00:00
• d650be179f Accept hex nc's, inspired by https://github.com/kristapsdz/kcgi/pull/42. This should fix some earlier-noted issues with kcaldav, which uses a lot of the digest authentication. Thank you! kristapsdz 2018-06-18 12:46:46 +00:00
• 44fffcdb25 Add PSA regarding Mac OS X and kcgi, and while here note that systrace has long since been deprecated. kristapsdz 2018-06-14 20:36:24 +00:00
• b7fc5249e7 Protect system crash for FastCGI case. And remove some snark... kristapsdz 2018-06-12 17:09:41 +00:00
• b701ab7d2c How to crash Mac OS X: run make regress. Known for several years, reported again by https://github.com/kristapsdz/kcgi/issues/39. Thanks! kristapsdz 2018-06-12 08:26:55 +00:00
• 35d100137a Version up. kristapsdz 2018-06-09 09:42:49 +00:00
• 2f1aedd1a9 Merge https://github.com/kristapsdz/kcgi/pull/37, thanks! kristapsdz 2018-06-09 09:38:37 +00:00
• fb8724f032 Fix from Ross Richardson to prevent high-bits from being printed as signed. Thanks! kristapsdz 2018-05-30 13:53:39 +00:00
• f7e5134072 Oops: forgot to install manpage. Hot-fix version. kristapsdz 2018-04-19 19:14:13 +00:00
• fa2be1a75e Use INSTALL_ macro instead of install or cp. kristapsdz 2018-04-19 19:06:45 +00:00
• 6a357b28aa Bring up to date with oconfigure. kristapsdz 2018-04-19 18:56:10 +00:00
• 20295fe275 Version up. kristapsdz 2018-04-19 18:55:13 +00:00
• bb12b8185a Clean up kutil_invalidate() primarily as suggested by schwarze@, thanks! kristapsdz 2018-04-19 18:39:36 +00:00
• be16f5fdf9 Clarify operation. kristapsdz 2018-04-19 18:39:11 +00:00
• 5e5b2faf6a Add test of kutil_invalidate. kristapsdz 2018-04-19 18:29:58 +00:00
• 31550ac7d7 Split kutil_invalidate manpage into its own. kristapsdz 2018-04-19 18:29:34 +00:00
• d77b0e20f3 Use proper template function. kristapsdz 2018-04-19 14:29:01 +00:00
• ffd28f213a Clarify. kristapsdz 2018-04-19 14:27:25 +00:00
• 00959e99ba Add a fast-track for printing non-delimeter content. kristapsdz 2018-04-19 14:25:45 +00:00
• 3d85f3632a Allow for escaping the template delimiters. kristapsdz 2018-04-19 13:36:50 +00:00
• 9f4ef20dcf style(9) kristapsdz 2018-04-19 01:13:52 +00:00
• e4b4f052a3 Much more extensive testing of templates. kristapsdz 2018-04-18 23:39:59 +00:00
• edbcf8ea65 Let @@@@ be properly handled, handle trailing @@, and properly handle suppressing delimiters in the case of fallthrough. Analysis inspired by an e-mail from schwarze@, who suggested alternative (but functionally similarly) fixes. Thanks! kristapsdz 2018-04-18 23:39:51 +00:00
• 3adcb1f1b1 Clarify behaviour of when key sequence is printed or not. Inspired by a series of mails from schwarze@, thanks! kristapsdz 2018-04-18 23:38:05 +00:00
• dc1cb55513 style(9) kristapsdz 2018-04-18 21:08:22 +00:00
• c926852dd4 Add kcgi_strerror(3). kristapsdz 2018-04-17 18:07:56 +00:00
• 533058cda1 Use kcgi_buf. kristapsdz 2018-04-10 20:37:46 +00:00
• 0c3332492f Version notes. kristapsdz 2018-04-10 20:31:39 +00:00
• 1de34a8f50 Versioning up... kristapsdz 2018-04-10 20:31:28 +00:00
• e9248b6d66 Clean up version notes. kristapsdz 2018-04-10 18:43:11 +00:00
• 53db040cbe Const-ify and fix a check for "buf" when it should have been "b". kristapsdz 2018-04-10 18:29:00 +00:00
• 2342b0c4b2 Don't use "ssz" for this temporary value---we need it! kristapsdz 2018-04-10 18:19:35 +00:00
• d9b6552ed0 From a follow-up in https://github.com/kristapsdz/kcgi/issues/27: fix sandbox running on arm. Thanks again! kristapsdz 2018-04-10 18:18:11 +00:00
• 96bce5ee08 Don't use EXIT_xxxx macros, apply style(9), make sure r.mime is checked. kristapsdz 2018-04-10 15:51:05 +00:00
• 55f2dacb5a Keep up to date. kristapsdz 2018-04-10 15:50:32 +00:00
• 91f8e6ac06 Check MIME and run style(9). kristapsdz 2018-04-10 15:45:45 +00:00
• 0082c48332 Run through style(9) and check for r.mime. kristapsdz 2018-04-10 15:44:20 +00:00
• 6568bd1dd4 Flip logic for r.mime. kristapsdz 2018-04-10 15:43:03 +00:00
• c5f177da26 Run through style(9) and don't let a bogus MIME type slip through. kristapsdz 2018-04-10 15:40:59 +00:00
• 48852037e1 Add in FastCGI pledging. kristapsdz 2018-04-10 15:30:49 +00:00
• 7e870a0b02 No need to do error checking for HUP and don't include stdlib. kristapsdz 2018-04-10 15:30:36 +00:00
• 42a6a7420c Fix promises required by FastCGI. kristapsdz 2018-04-10 15:27:06 +00:00
• 53421602a4 Adding new tutorial. kristapsdz 2018-04-10 15:09:24 +00:00
• 38f9dee3f8 Lots of polish. kristapsdz 2018-04-10 15:08:03 +00:00
• 3a18570767 Error checking should be done. kristapsdz 2018-04-10 15:07:39 +00:00
• bc54ae4b51 Adding prettify locally. kristapsdz 2018-04-10 15:07:00 +00:00
• a4827dfe4f Use local prettify, simplify example, fix copyrights, don't use permlink for articles, use github author page. kristapsdz 2018-04-10 15:06:39 +00:00
• b03a176921 Fix also from https://github.com/kristapsdz/kcgi/pull/33 --- thanks again! kristapsdz 2018-04-10 01:22:54 +00:00
• e0645ba9dd Merge https://github.com/kristapsdz/kcgi/pull/33/files --- thanks! kristapsdz 2018-04-10 01:20:21 +00:00
• d93db8213f Reference khttp_write(3), which documents the inherited return values. kristapsdz 2018-04-09 22:58:21 +00:00
• 363843e868 Continue working on release notes. kristapsdz 2018-04-09 22:53:09 +00:00
• 7a5342ef53 Required for fcgi.c now. kristapsdz 2018-04-09 22:53:01 +00:00
• 593f2a7196 Clarify free status. kristapsdz 2018-04-09 22:52:37 +00:00
• 525566876d Clean up presentation of this important phrase. kristapsdz 2018-04-09 22:50:56 +00:00
• ba55cbd6f0 Clean up kreq on failure. kristapsdz 2018-04-09 22:50:26 +00:00
• ed1cff9f97 Fix code KERR -> KCGI. kristapsdz 2018-04-09 22:36:36 +00:00
• 4d0a328f9c Clarify the role of KCGI_HUP and FastCGI connection closing. kristapsdz 2018-04-09 22:35:49 +00:00
• e961f4f6e6 Specify write errors for FastCGI connections. kristapsdz 2018-04-09 22:27:04 +00:00
• 3b65c179f1 Don't make khttp_write functions assert on KSTATE_BODY. It's too easy for this condition to be reached if callers don't check the return code of khttp_body. kristapsdz 2018-04-09 22:14:12 +00:00