mandoc - UNIX manpage compiler toolset

	Commit message (Collapse)	Author	Age	Files	Lines
*	Rename mandoc-db to makewhatis. On the suggestion of schwarze@; I agree.	Kristaps Dzonsons	2011-05-13	1	-962/+0
\| \| \| \|	Add initial version notes.
*	Correctly catch `Vt' semicolon in mandoc-db.	Kristaps Dzonsons	2011-05-12	1	-2/+3
\|
*	Make sure that we check for the \(en, \(em, \-, and - separators for	Kristaps Dzonsons	2011-05-12	1	-3/+8
\| \| \| \|	the heuristic dance to grab manual descriptions from man documents.
*	Grok manual architecture in mandoc-db.c.	Kristaps Dzonsons	2011-05-04	1	-0/+5
\|
*	Add configurations (`Cd') to mandoc-db mining.	Kristaps Dzonsons	2011-05-04	1	-2/+21
\| \| \| \| \|	Also put some notes into index.sgml to the effect that mandoc-db exists, but is not linked to the build.
*	Pick up authors in mandoc-db.	Kristaps Dzonsons	2011-05-03	1	-2/+21
\|
*	Have mandoc-db grok `Dt'/`TH' manual title as well.	Kristaps Dzonsons	2011-05-03	1	-0/+6
\|
*	Back out stripping of non-predef and non-special escape sequences from	Kristaps Dzonsons	2011-05-02	1	-4/+3
\| \| \| \| \|	input (this is not yet possible with mandoc_escape(), which depends on nil-terminated strings).
*	Have mandoc-db strip out escapes that aren't special or predefined	Kristaps Dzonsons	2011-05-01	1	-15/+54
\| \| \| \|	characters. This is a work-in-progress that has some issues.
*	Let mandoc-db grok `St' tokens.	Kristaps Dzonsons	2011-04-28	1	-2/+20
\|
*	Let mandoc-db also collect -man descriptions.	Kristaps Dzonsons	2011-04-12	1	-5/+25
\|
*	Have mandoc-db accumulate manual page descriptions (`Nd' in -mdoc parlance)	Kristaps Dzonsons	2011-04-11	1	-29/+79
\| \| \| \| \|	in the index. This allows, with both the btree and index, full emulation of apropos(1) and other goodies.
*	Use dbt_xxxx functions to stash both filename and manual section in the	Kristaps Dzonsons	2011-04-11	1	-16/+18
\| \| \| \| \| \|	value part of the index. This is the actual manual section---before, mandoc.cgi was relying on the file suffix, but this can be (e.g.) .man or whatnot. This is The Correct Way (tm).
*	Lint-checks over mandoc-db.c.	Kristaps Dzonsons	2011-04-05	1	-18/+14
\|
*	Use a little more horsepower in parsing out NAME sections from -man	Kristaps Dzonsons	2011-04-05	1	-16/+56
\| \| \| \|	manuals. This covers the majority case.
*	Fix type- and NULL-check to be correct node (last, not child). Prevents	Kristaps Dzonsons	2011-04-05	1	-1/+1
\| \| \| \|	segfault in NetBSD.
*	Add mandoc-db.c checking for utility-name in NAME section of -man	Kristaps Dzonsons	2011-04-05	1	-3/+77
\| \| \| \|	documents.
*	Have `Fd' detection in mandoc-db also look for local includes (i.e.,	Kristaps Dzonsons	2011-04-04	1	-2/+2
\| \| \| \|	those that are quoted and not angular-bracketed).
*	Add manual page for mandoc-db (mostly to document the file format of	Kristaps Dzonsons	2011-04-04	1	-16/+17
\| \| \| \| \|	the generated index and keyword databases). Add some documentation within mandoc-db.c.
*	Have mandoc-db create an recno-addressed index of files alongside the	Kristaps Dzonsons	2011-04-03	1	-59/+83
\| \| \| \|	keyword database, which references the index of hard-coding files.
*	Have mandoc-db use config.h for strlcat(). Then create the btree	Kristaps Dzonsons	2011-04-03	1	-29/+37
\| \| \| \| \| \|	database from a directory, instead of a path. This is because it'll also output an index of files to that same directory. Add documentation to the local variable names, too.
*	Add an initial keyword-indexer as an alternative front-end to libmandoc.a.	Kristaps Dzonsons	2011-04-02	1	-0/+636
	This will be used during BSDCan-2011, paired with a web front-end, as a demonstrandum of semantic annotations (mdoc) versus those of man. This isn't linked to the build in any way but for cleaning created files. This code has been heavily tested under valgrind and is known to work on GNU/Linux (needs -ldb library), all BSDs, and Mac OSX. Please don't repost this, for the time being, as I don't want to spoil the fun for the conference. This utility is still constantly under development (e.g., it will also generate a recno database of filenames and `Nd' output so that paths needn't be hard-coded) but is mature enough to warrant being checked in.