mdoc.c revision 95c635efb7c3b86efc493e0447eaec7aecca3f0f
/* $Id: mdoc.c,v 1.196 2011/09/30 00:13:28 schwarze Exp $ */
/*
* Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
* Copyright (c) 2010 Ingo Schwarze <schwarze@openbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include <assert.h>
#include <stdarg.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <time.h>
#include "mdoc.h"
#include "mandoc.h"
#include "libmdoc.h"
#include "libmandoc.h"
const char *const __mdoc_macronames[MDOC_MAX] = {
"Ap", "Dd", "Dt", "Os",
"Sh", "Ss", "Pp", "D1",
"Dl", "Bd", "Ed", "Bl",
"El", "It", "Ad", "An",
"Ar", "Cd", "Cm", "Dv",
"Er", "Ev", "Ex", "Fa",
"Fd", "Fl", "Fn", "Ft",
"Ic", "In", "Li", "Nd",
"Nm", "Op", "Ot", "Pa",
"Rv", "St", "Va", "Vt",
/* LINTED */
"Xr", "%A", "%B", "%D",
/* LINTED */
"%I", "%J", "%N", "%O",
/* LINTED */
"%P", "%R", "%T", "%V",
"Ac", "Ao", "Aq", "At",
"Bc", "Bf", "Bo", "Bq",
"Bsx", "Bx", "Db", "Dc",
"Do", "Dq", "Ec", "Ef",
"Em", "Eo", "Fx", "Ms",
"No", "Ns", "Nx", "Ox",
"Pc", "Pf", "Po", "Pq",
"Qc", "Ql", "Qo", "Qq",
"Re", "Rs", "Sc", "So",
"Sq", "Sm", "Sx", "Sy",
"Tn", "Ux", "Xc", "Xo",
"Fo", "Fc", "Oo", "Oc",
"Bk", "Ek", "Bt", "Hf",
"Fr", "Ud", "Lb", "Lp",
"Lk", "Mt", "Brq", "Bro",
/* LINTED */
"Brc", "%C", "Es", "En",
/* LINTED */
"Dx", "%Q", "br", "sp",
/* LINTED */
"%U", "Ta"
};
const char *const __mdoc_argnames[MDOC_ARG_MAX] = {
"split", "nosplit", "ragged",
"unfilled", "literal", "file",
"offset", "bullet", "dash",
"hyphen", "item", "enum",
"tag", "diag", "hang",
"ohang", "inset", "column",
"width", "compact", "std",
"filled", "words", "emphasis",
"symbolic", "nested", "centered"
};
const char * const *mdoc_macronames = __mdoc_macronames;
const char * const *mdoc_argnames = __mdoc_argnames;
static void mdoc_node_free(struct mdoc_node *);
static void mdoc_node_unlink(struct mdoc *,
struct mdoc_node *);
static void mdoc_free1(struct mdoc *);
static void mdoc_alloc1(struct mdoc *);
static int node_append(struct mdoc *,
struct mdoc_node *);
#if 0
static int mdoc_preptext(struct mdoc *, int, char *, int);
#endif
static int mdoc_ptext(struct mdoc *, int, char *, int);
static int mdoc_pmacro(struct mdoc *, int, char *, int);
const struct mdoc_node *
{
return(m->first);
}
const struct mdoc_meta *
{
return(&m->meta);
}
/*
* Frees volatile resources (parse tree, meta-data, fields).
*/
static void
{
}
/*
* Allocate all volatile resources (parse tree, meta-data, fields).
*/
static void
{
}
/*
* Free up volatile resources (see mdoc_free1()) then re-initialises the
* data with mdoc_alloc1(). After invocation, parse data has been reset
* and the parser is ready for re-invocation on a new tree; however,
* cross-parse non-volatile data is kept intact.
*/
void
{
}
/*
* Completely free up all volatile and non-volatile parse resources.
* After invocation, the pointer is no longer usable.
*/
void
{
}
/*
* Allocate volatile and non-volatile parse resources.
*/
struct mdoc *
{
struct mdoc *p;
mdoc_alloc1(p);
return(p);
}
/*
* Climb back up the parse tree, validating open scopes. Mostly calls
* through to macro_end() in macro.c.
*/
int
mdoc_endparse(struct mdoc *m)
{
if (mdoc_macroend(m))
return(1);
return(0);
}
int
{
struct mdoc_node *n;
/* No text before an initial macro. */
return(1);
}
if ( ! node_append(m, n))
return(0);
m->next = MDOC_NEXT_SIBLING;
return(1);
}
int
{
struct mdoc_node *n;
/* No text before an initial macro. */
return(1);
}
if ( ! node_append(m, n))
return(0);
m->next = MDOC_NEXT_SIBLING;
return(1);
}
/*
* Main parse routine. Parses a single line -- really just hands off to
* the macro (mdoc_pmacro()) or text parser (mdoc_ptext()).
*/
int
{
m->flags |= MDOC_NEWLINE;
/*
* Let the roff nS register switch SYNOPSIS mode early,
* such that the parser knows at all times
* whether this mode is on or off.
* Note that this mode is also switched by the Sh macro.
*/
m->flags |= MDOC_SYNOPSIS;
else
m->flags &= ~MDOC_SYNOPSIS;
}
}
int
{
/* If we're in the body, deny prologue calls. */
MDOC_PBODY & m->flags) {
return(1);
}
/* If we're in the prologue, deny "body" macros. */
! (MDOC_PBODY & m->flags)) {
m->flags |= MDOC_PBODY;
}
}
static int
{
case (MDOC_NEXT_SIBLING):
break;
case (MDOC_NEXT_CHILD):
break;
default:
abort();
/* NOTREACHED */
}
/*
* Copy over the normalised-data pointer of our parent. Not
* everybody has one, but copying a null pointer is fine.
*/
switch (p->type) {
case (MDOC_BODY):
/* FALLTHROUGH */
case (MDOC_TAIL):
/* FALLTHROUGH */
case (MDOC_HEAD):
break;
default:
break;
}
if ( ! mdoc_valid_pre(mdoc, p))
return(0);
switch (p->type) {
case (MDOC_HEAD):
break;
case (MDOC_TAIL):
break;
case (MDOC_BODY):
if (p->end)
break;
break;
default:
break;
}
switch (p->type) {
case (MDOC_TBL):
/* FALLTHROUGH */
case (MDOC_TEXT):
if ( ! mdoc_valid_post(mdoc))
return(0);
break;
default:
break;
}
return(1);
}
static struct mdoc_node *
{
struct mdoc_node *p;
/* Flag analysis. */
if (MDOC_SYNOPSIS & m->flags)
p->flags |= MDOC_SYNPRETTY;
else
p->flags &= ~MDOC_SYNPRETTY;
if (MDOC_NEWLINE & m->flags)
m->flags &= ~MDOC_NEWLINE;
return(p);
}
int
{
struct mdoc_node *p;
if ( ! node_append(m, p))
return(0);
m->next = MDOC_NEXT_CHILD;
return(1);
}
int
{
struct mdoc_node *p;
if ( ! node_append(m, p))
return(0);
m->next = MDOC_NEXT_CHILD;
return(1);
}
int
{
struct mdoc_node *p;
if ( ! node_append(m, p))
return(0);
m->next = MDOC_NEXT_CHILD;
return(1);
}
int
{
struct mdoc_node *p;
if ( ! node_append(m, p))
return(0);
m->next = MDOC_NEXT_SIBLING;
return(1);
}
int
{
struct mdoc_node *p;
if (p->args)
switch (tok) {
case (MDOC_Bd):
/* FALLTHROUGH */
case (MDOC_Bf):
/* FALLTHROUGH */
case (MDOC_Bl):
/* FALLTHROUGH */
case (MDOC_Rs):
break;
default:
break;
}
if ( ! node_append(m, p))
return(0);
m->next = MDOC_NEXT_CHILD;
return(1);
}
int
{
struct mdoc_node *p;
if (p->args)
switch (tok) {
case (MDOC_An):
break;
default:
break;
}
if ( ! node_append(m, p))
return(0);
m->next = MDOC_NEXT_CHILD;
return(1);
}
int
{
struct mdoc_node *n;
if ( ! node_append(m, n))
return(0);
m->next = MDOC_NEXT_SIBLING;
return(1);
}
static void
mdoc_node_free(struct mdoc_node *p)
{
if (p->string)
if (p->args)
mdoc_argv_free(p->args);
free(p);
}
static void
{
/* Adjust siblings. */
if (n->prev)
if (n->next)
/* Adjust parent. */
if (n->parent) {
}
/* Adjust parse point, if applicable. */
if (m && m->last == n) {
if (n->prev) {
m->next = MDOC_NEXT_SIBLING;
} else {
m->next = MDOC_NEXT_CHILD;
}
}
if (m && m->first == n)
}
void
{
while (p->child) {
mdoc_node_delete(m, p->child);
}
mdoc_node_unlink(m, p);
mdoc_node_free(p);
}
#if 0
/*
* Pre-treat a text line.
* Text lines can consist of equations, which must be handled apart from
* the regular text.
* Thus, use this function to step through a line checking if it has any
* equations embedded in it.
* This must handle multiple equations AND equations that do not end at
* the end-of-line, i.e., will re-enter in the next roff parse.
*/
static int
{
char delim;
/* Mark starting position if eqn is set. */
*start++ = '\0';
/* Parse text as normal. */
return(0);
/* Continue only if an equation exists. */
break;
/* Read past the end of the equation. */
*end++ = '\0';
while (' ' == *end)
end++;
}
/* Parse the equation itself. */
/* Process a finished equation? */
if (roff_closeeqn(m->roff))
return(0);
}
return(1);
}
#endif
/*
* Parse free-form text, that is, a line that does not begin with the
* control character.
*/
static int
{
struct mdoc_node *n;
/* No text before an initial macro. */
return(1);
}
n = m->last;
/*
* Divert directly to list processing if we're encountering a
* columnar MDOC_BLOCK with or without a prior MDOC_BLOCK entry
* (a MDOC_BODY means it's already open, in which case we should
* process within its context in the normal way).
*/
/* `Bl' is open without any children. */
m->flags |= MDOC_FREECOL;
}
/* `Bl' has block-level `It' children. */
m->flags |= MDOC_FREECOL;
}
/*
* Search for the beginning of unescaped trailing whitespace (ws)
* and for the first character not to be output (end).
*/
/* FIXME: replace with strcspn(). */
switch (*c) {
case ' ':
ws = c;
continue;
case '\t':
/*
* Always warn about trailing tabs,
* even outside literal context,
* where they should be put on the next line.
*/
ws = c;
/*
* Strip trailing tabs in literal context only;
* outside, they affect the next line.
*/
if (MDOC_LITERAL & m->flags)
continue;
break;
case '\\':
/* Skip the escaped character, too, if any. */
if (c[1])
c++;
/* FALLTHROUGH */
default:
break;
}
end = c + 1;
}
*end = '\0';
if (ws)
/*
* Insert a `sp' in the case of a blank line. Technically,
* blank lines aren't allowed, but enough manuals assume this
* behaviour that we want to work around it.
*/
return(0);
m->next = MDOC_NEXT_SIBLING;
return(1);
}
return(0);
if (MDOC_LITERAL & m->flags)
return(1);
/*
* End-of-sentence check. If the last character is an unescaped
* EOS character, then flag the node as being the end of a
* sentence. The front-end will know how to interpret this.
*/
return(1);
}
/*
* Parse a macro line, that is, a line beginning with the control
* character.
*/
static int
{
int i, sv;
char mac[5];
struct mdoc_node *n;
/* Empty post-control lines are ignored. */
return(1);
return(1);
/*
* Copy the first word into a nil-terminated buffer.
* Stop copying when a tab, space, or eoln is encountered.
*/
i = 0;
mac[i] = '\0';
return(1);
}
/* Disregard the first trailing tab, if applicable. */
offs++;
/* Jump to the next non-whitespace word. */
offs++;
/*
* Trailing whitespace. Note that tabs are allowed to be passed
* into the parser as "text", so we only warn about spaces here.
*/
/*
* If an initial macro or a list invocation, divert directly
* into macro processing.
*/
goto err;
return(1);
}
n = m->last;
/*
* If the first macro of a `Bl -column', open an `It' block
* context around the parsed macro.
*/
m->flags |= MDOC_FREECOL;
goto err;
return(1);
}
/*
* If we're following a block-level `It' within a `Bl -column'
* context (perhaps opened in the above block or in ptext()),
* then open an `It' block context around the parsed macro.
*/
m->flags |= MDOC_FREECOL;
goto err;
return(1);
}
/* Normal processing of a macro. */
goto err;
return(1);
err: /* Error out. */
return(0);
}
enum mdelim
mdoc_isdelim(const char *p)
{
if ('\0' == p[0])
return(DELIM_NONE);
if ('\0' == p[1])
switch (p[0]) {
case('('):
/* FALLTHROUGH */
case('['):
return(DELIM_OPEN);
case('|'):
return(DELIM_MIDDLE);
case('.'):
/* FALLTHROUGH */
case(','):
/* FALLTHROUGH */
case(';'):
/* FALLTHROUGH */
case(':'):
/* FALLTHROUGH */
case('?'):
/* FALLTHROUGH */
case('!'):
/* FALLTHROUGH */
case(')'):
/* FALLTHROUGH */
case(']'):
return(DELIM_CLOSE);
default:
return(DELIM_NONE);
}
if ('\\' != p[0])
return(DELIM_NONE);
return(DELIM_CLOSE);
return(DELIM_MIDDLE);
return(DELIM_NONE);
}