1 .\" $Id: mandoc.3,v 1.23 2014/01/05 20:26:36 schwarze Exp $
3 .\" Copyright (c) 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
4 .\" Copyright (c) 2010 Ingo Schwarze <schwarze@openbsd.org>
6 .\" Permission to use, copy, modify, and distribute this software for any
7 .\" purpose with or without fee is hereby granted, provided that the above
8 .\" copyright notice and this permission notice appear in all copies.
10 .\" THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
11 .\" WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
12 .\" MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
13 .\" ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
14 .\" WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
15 .\" ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
16 .\" OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
18 .Dd $Mdocdate: January 5 2014 $
44 .Nd mandoc macro compiler library
53 .Fa "const char **end"
54 .Fa "const char **start"
57 .Ft "const struct man_meta *"
59 .Fa "const struct man *man"
61 .Ft "const struct mparse *"
63 .Fa "const struct man *man"
65 .Ft "const struct man_node *"
67 .Fa "const struct man *man"
70 .Fn mchars_alloc "void"
72 .Fn mchars_free "struct mchars *p"
74 .Fn mchars_num2char "const char *cp" "size_t sz"
76 .Fn mchars_num2uc "const char *cp" "size_t sz"
79 .Fa "const struct mchars *p"
86 .Fa "const struct mchars *p"
90 .Ft "const struct mdoc_meta *"
92 .Fa "const struct mdoc *mdoc"
94 .Ft "const struct mdoc_node *"
96 .Fa "const struct mdoc *mdoc"
100 .Fa "enum mparset inttype"
101 .Fa "enum mandoclevel wlevel"
108 .Fa "struct mparse *parse"
112 .Fa "const struct mparse *parse"
116 .Fa "struct mparse *parse"
118 .Ft "enum mandoclevel"
120 .Fa "struct mparse *parse"
122 .Fa "const char *fname"
126 .Fa "struct mparse *parse"
130 .Fa "struct mparse *parse"
131 .Fa "struct mdoc **mdoc"
132 .Fa "struct man **man"
140 .Fa "enum mandoclevel"
142 .Vt extern const char * const * man_macronames;
143 .Vt extern const char * const * mdoc_argnames;
144 .Vt extern const char * const * mdoc_macronames;
145 .Fd "#define ASCII_NBRSP"
146 .Fd "#define ASCII_HYPH"
152 manual into an abstract syntax tree (AST).
154 manuals are composed of
158 and may be mixed with
165 The following describes a general parse sequence:
168 initiate a parsing sequence with
171 parse files or file descriptors with
174 retrieve a parsed syntax tree, if the parse was successful, with
177 iterate over parse nodes with
182 free all allocated memory with
191 library also contains routines for translating character strings into glyphs
192 .Pq see Fn mchars_alloc
193 and parsing escape sequences from strings
194 .Pq see Fn mandoc_escape .
196 This section documents the functions, types, and variables available
201 .It Vt "enum mandoc_esc"
202 An escape sequence classification.
203 .It Vt "enum mandocerr"
204 A fatal error, error, or warning message during parsing.
205 .It Vt "enum mandoclevel"
206 A classification of an
208 as regards system operation.
209 .It Vt "struct mchars"
210 An opaque pointer to an object allowing for translation between
211 character strings and glyphs.
214 .It Vt "enum mparset"
215 The type of parser when reading input.
216 This should usually be
219 .It Vt "struct mparse"
220 An opaque pointer to a running parse sequence.
225 This may be used across parsed input if
227 is called between parses.
229 A prototype for a function to handle fatal error, error, and warning
230 messages emitted by the parser.
235 Scan an escape sequence, i.e., a character string beginning with
237 Pass a pointer to the character after the
241 it will be set to the supremum of the parsed escape sequence unless
244 in which case the string is bogus and should be
251 is set to the first relevant character of the substring (font, glyph,
265 Obtain the meta-data of a successful parse.
266 This may only be used on a pointer returned by
273 Get the parser used for the current output.
279 Obtain the root node of a successful parse.
280 This may only be used on a pointer returned by
288 .Vt "struct mchars *"
289 object for translating special characters into glyphs.
292 for an overview of special characters.
293 The object must be freed with
300 Free an object created with
306 .It Fn mchars_num2char
307 Convert a character index (e.g., the \eN\(aq\(aq escape) into a
308 printable ASCII character.
309 Returns \e0 (the nil character) if the input sequence is malformed.
315 Convert a hexadecimal character index (e.g., the \e[uNNNN] escape) into
317 Returns \e0 (the nil character) if the input sequence is malformed.
322 .It Fn mchars_spec2cp
323 Convert a special character into a valid Unicode codepoint.
324 Returns \-1 on failure or a non-zero Unicode codepoint on success.
329 .It Fn mchars_spec2str
330 Convert a special character into an ASCII string.
339 Obtain the meta-data of a successful parse.
340 This may only be used on a pointer returned by
347 Obtain the root node of a successful parse.
348 This may only be used on a pointer returned by
356 The arguments have the following effect:
357 .Bl -tag -offset 5n -width inttype
363 only that parser will be used.
366 the document type will be automatically detected.
369 .Dv MANDOCLEVEL_FATAL ,
370 .Dv MANDOCLEVEL_ERROR ,
372 .Dv MANDOCLEVEL_WARNING .
373 Messages below the selected level will be suppressed.
375 A callback function to handle errors and warnings.
380 A default string for the
383 macro, overriding the
385 preprocessor definition and the results of
388 When set, parsing is aborted after the NAME section.
389 This is for example useful to quickly build minimal databases.
392 The same parser may be used for multiple files so long as
394 is called between parses.
396 must be called to free the memory allocated by this function.
402 Free all memory allocated by
408 .It Fn mparse_getkeep
409 Acquire the keep buffer.
410 Must follow a call of
417 Instruct the parser to retain a copy of its parsed input.
418 This can be acquired with subsequent
426 Parse a file or file descriptor.
431 is opened for reading.
434 is assumed to be the name associated with
436 This may be called multiple times with different parameters; however,
438 should be invoked between parses.
444 Reset a parser so that
452 Obtain the result of a parse.
453 Only successful parses
457 returned less than MANDOCLEVEL_FATAL
459 should invoke this function, in which case one of the two pointers will
465 .It Fn mparse_strerror
466 Return a statically-allocated string representation of an error code.
471 .It Fn mparse_strlevel
472 Return a statically-allocated string representation of a level code.
480 .It Va man_macronames
481 The string representation of a man macro as indexed by
484 The string representation of a mdoc macro argument as indexed by
485 .Vt "enum mdocargt" .
486 .It Va mdoc_macronames
487 The string representation of a mdoc macro as indexed by
490 .Sh IMPLEMENTATION NOTES
491 This section consists of structural documentation for
495 syntax trees and strings.
496 .Ss Man and Mdoc Strings
497 Strings may be extracted from mdoc and man meta-data, or from text
498 nodes (MDOC_TEXT and MAN_TEXT, respectively).
499 These strings have special non-printing formatting cues embedded in the
500 text itself, as well as
502 escapes preserved from input.
503 Implementing systems will need to handle both situations to produce
505 In general, strings may be assumed to consist of 7-bit ASCII characters.
507 The following non-printing characters may be embedded in text strings:
510 A non-breaking space character.
515 Escape characters are also passed verbatim into text strings.
516 An escape character is a sequence of characters beginning with the
519 To construct human-readable text, these should be intercepted with
521 and converted with one of
522 .Fn mchars_num2char ,
523 .Fn mchars_spec2str ,
525 .Ss Man Abstract Syntax Tree
526 This AST is governed by the ontological rules dictated in
528 and derives its terminology accordingly.
530 The AST is composed of
532 nodes with element, root and text types as declared by the
535 Each node also provides its parse point (the
540 fields), its position in the tree (the
546 fields) and some type-specific data.
548 The tree itself is arranged according to the following normal form,
549 where capitalised non-terminals represent nodes.
551 .Bl -tag -width "ELEMENTXX" -compact
555 \(<- ELEMENT | TEXT | BLOCK
568 The only elements capable of nesting other elements are those with
569 next-lint scope as documented in
571 .Ss Mdoc Abstract Syntax Tree
572 This AST is governed by the ontological
575 and derives its terminology accordingly.
577 elements described in
579 are described simply as
582 The AST is composed of
584 nodes with block, head, body, element, root and text types as declared
588 Each node also provides its parse point (the
593 fields), its position in the tree (the
600 fields) and some type-specific data, in particular, for nodes generated
601 from macros, the generating macro in the
605 The tree itself is arranged according to the following normal form,
606 where capitalised non-terminals represent nodes.
608 .Bl -tag -width "ELEMENTXX" -compact
612 \(<- BLOCK | ELEMENT | TEXT
614 \(<- HEAD [TEXT] (BODY [TEXT])+ [TAIL [TEXT]]
620 \(<- mnode* [ENDBODY mnode*]
627 Of note are the TEXT nodes following the HEAD, BODY and TAIL nodes of
628 the BLOCK production: these refer to punctuation marks.
629 Furthermore, although a TEXT node will generally have a non-zero-length
630 string, in the specific case of
631 .Sq \&.Bd \-literal ,
632 an empty line will produce a zero-length string.
633 Multiple body parts are only found in invocations of
635 where a new body introduces a new phrase.
639 syntax tree accommodates for broken block structures as well.
640 The ENDBODY node is available to end the formatting associated
641 with a given block before the physical end of that block.
644 field, is of the BODY
648 as the BLOCK it is ending, and has a
650 field pointing to that BLOCK's BODY node.
651 It is an indirect child of that BODY node
652 and has no children of its own.
654 An ENDBODY node is generated when a block ends while one of its child
655 blocks is still open, like in the following example:
656 .Bd -literal -offset indent
663 This example results in the following block structure:
664 .Bd -literal -offset indent
669 BLOCK Bo, pending -> Ao
674 ENDBODY Ao, pending -> Ao
679 Here, the formatting of the
681 block extends from TEXT ao to TEXT ac,
682 while the formatting of the
684 block extends from TEXT bo to TEXT bc.
685 It renders as follows in
689 .Dl <ao [bo ac> bc] end
691 Support for badly-nested blocks is only provided for backward
692 compatibility with some older
695 Using badly-nested blocks is
696 .Em strongly discouraged ;
703 are unable to render them in any meaningful way.
704 Furthermore, behaviour when encountering badly-nested blocks is not
705 consistent across troff implementations, especially when using multiple
706 levels of badly-nested blocks.
718 library was written by
719 .An Kristaps Dzonsons Aq Mt kristaps@bsd.lv .