5025 import and use mandoc

Reviewed by: Hans Rosenfeld <hans.rosenfeld@nexenta.com>
Reviewed by: Igor Kozhukhov <ikozhukhov@gmail.com>
Reviewed by: Robert Mustacchi <rm@joyent.com>
Reviewed by: Albert Lee <trisk@nexenta.com>
Approved by: Dan McDonald <danmcd@omniti.com>

1 files changed, 987 insertions, 0 deletions

diff --git a/usr/src/cmd/mandoc/mdoc.c b/usr/src/cmd/mandoc/mdoc.c
new file mode 100644
index 0000000000..81a4ffc96f
--- /dev/null
+++ b/usr/src/cmd/mandoc/mdoc.c
@@ -0,0 +1,987 @@
+/*	$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 <sys/types.h>
+
+#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	struct mdoc_node *node_alloc(struct mdoc *, int, int, 
+				enum mdoct, enum mdoc_type);
+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 *
+mdoc_node(const struct mdoc *m)
+{
+
+	assert( ! (MDOC_HALT & m->flags));
+	return(m->first);
+}
+
+
+const struct mdoc_meta *
+mdoc_meta(const struct mdoc *m)
+{
+
+	assert( ! (MDOC_HALT & m->flags));
+	return(&m->meta);
+}
+
+
+/*
+ * Frees volatile resources (parse tree, meta-data, fields).
+ */
+static void
+mdoc_free1(struct mdoc *mdoc)
+{
+
+	if (mdoc->first)
+		mdoc_node_delete(mdoc, mdoc->first);
+	if (mdoc->meta.title)
+		free(mdoc->meta.title);
+	if (mdoc->meta.os)
+		free(mdoc->meta.os);
+	if (mdoc->meta.name)
+		free(mdoc->meta.name);
+	if (mdoc->meta.arch)
+		free(mdoc->meta.arch);
+	if (mdoc->meta.vol)
+		free(mdoc->meta.vol);
+	if (mdoc->meta.msec)
+		free(mdoc->meta.msec);
+	if (mdoc->meta.date)
+		free(mdoc->meta.date);
+}
+
+
+/*
+ * Allocate all volatile resources (parse tree, meta-data, fields).
+ */
+static void
+mdoc_alloc1(struct mdoc *mdoc)
+{
+
+	memset(&mdoc->meta, 0, sizeof(struct mdoc_meta));
+	mdoc->flags = 0;
+	mdoc->lastnamed = mdoc->lastsec = SEC_NONE;
+	mdoc->last = mandoc_calloc(1, sizeof(struct mdoc_node));
+	mdoc->first = mdoc->last;
+	mdoc->last->type = MDOC_ROOT;
+	mdoc->last->tok = MDOC_MAX;
+	mdoc->next = MDOC_NEXT_CHILD;
+}
+
+
+/*
+ * 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
+mdoc_reset(struct mdoc *mdoc)
+{
+
+	mdoc_free1(mdoc);
+	mdoc_alloc1(mdoc);
+}
+
+
+/*
+ * Completely free up all volatile and non-volatile parse resources.
+ * After invocation, the pointer is no longer usable.
+ */
+void
+mdoc_free(struct mdoc *mdoc)
+{
+
+	mdoc_free1(mdoc);
+	free(mdoc);
+}
+
+
+/*
+ * Allocate volatile and non-volatile parse resources.  
+ */
+struct mdoc *
+mdoc_alloc(struct roff *roff, struct mparse *parse)
+{
+	struct mdoc	*p;
+
+	p = mandoc_calloc(1, sizeof(struct mdoc));
+
+	p->parse = parse;
+	p->roff = roff;
+
+	mdoc_hash_init();
+	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)
+{
+
+	assert( ! (MDOC_HALT & m->flags));
+	if (mdoc_macroend(m))
+		return(1);
+	m->flags |= MDOC_HALT;
+	return(0);
+}
+
+int
+mdoc_addeqn(struct mdoc *m, const struct eqn *ep)
+{
+	struct mdoc_node *n;
+
+	assert( ! (MDOC_HALT & m->flags));
+
+	/* No text before an initial macro. */
+
+	if (SEC_NONE == m->lastnamed) {
+		mdoc_pmsg(m, ep->ln, ep->pos, MANDOCERR_NOTEXT);
+		return(1);
+	}
+
+	n = node_alloc(m, ep->ln, ep->pos, MDOC_MAX, MDOC_EQN);
+	n->eqn = ep;
+
+	if ( ! node_append(m, n))
+		return(0);
+
+	m->next = MDOC_NEXT_SIBLING;
+	return(1);
+}
+
+int
+mdoc_addspan(struct mdoc *m, const struct tbl_span *sp)
+{
+	struct mdoc_node *n;
+
+	assert( ! (MDOC_HALT & m->flags));
+
+	/* No text before an initial macro. */
+
+	if (SEC_NONE == m->lastnamed) {
+		mdoc_pmsg(m, sp->line, 0, MANDOCERR_NOTEXT);
+		return(1);
+	}
+
+	n = node_alloc(m, sp->line, 0, MDOC_MAX, MDOC_TBL);
+	n->span = sp;
+
+	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
+mdoc_parseln(struct mdoc *m, int ln, char *buf, int offs)
+{
+
+	assert( ! (MDOC_HALT & m->flags));
+
+	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.
+	 */
+	if (roff_regisset(m->roff, REG_nS)) {
+		if (roff_regget(m->roff, REG_nS))
+			m->flags |= MDOC_SYNOPSIS;
+		else
+			m->flags &= ~MDOC_SYNOPSIS;
+	}
+
+	return(mandoc_getcontrol(buf, &offs) ?
+			mdoc_pmacro(m, ln, buf, offs) :
+			mdoc_ptext(m, ln, buf, offs));
+}
+
+int
+mdoc_macro(MACRO_PROT_ARGS)
+{
+	assert(tok < MDOC_MAX);
+
+	/* If we're in the body, deny prologue calls. */
+
+	if (MDOC_PROLOGUE & mdoc_macros[tok].flags && 
+			MDOC_PBODY & m->flags) {
+		mdoc_pmsg(m, line, ppos, MANDOCERR_BADBODY);
+		return(1);
+	}
+
+	/* If we're in the prologue, deny "body" macros.  */
+
+	if ( ! (MDOC_PROLOGUE & mdoc_macros[tok].flags) && 
+			! (MDOC_PBODY & m->flags)) {
+		mdoc_pmsg(m, line, ppos, MANDOCERR_BADPROLOG);
+		if (NULL == m->meta.msec)
+			m->meta.msec = mandoc_strdup("1");
+		if (NULL == m->meta.title)
+			m->meta.title = mandoc_strdup("UNKNOWN");
+		if (NULL == m->meta.vol)
+			m->meta.vol = mandoc_strdup("LOCAL");
+		if (NULL == m->meta.os)
+			m->meta.os = mandoc_strdup("LOCAL");
+		if (NULL == m->meta.date)
+			m->meta.date = mandoc_normdate
+				(m->parse, NULL, line, ppos);
+		m->flags |= MDOC_PBODY;
+	}
+
+	return((*mdoc_macros[tok].fp)(m, tok, line, ppos, pos, buf));
+}
+
+
+static int
+node_append(struct mdoc *mdoc, struct mdoc_node *p)
+{
+
+	assert(mdoc->last);
+	assert(mdoc->first);
+	assert(MDOC_ROOT != p->type);
+
+	switch (mdoc->next) {
+	case (MDOC_NEXT_SIBLING):
+		mdoc->last->next = p;
+		p->prev = mdoc->last;
+		p->parent = mdoc->last->parent;
+		break;
+	case (MDOC_NEXT_CHILD):
+		mdoc->last->child = p;
+		p->parent = mdoc->last;
+		break;
+	default:
+		abort();
+		/* NOTREACHED */
+	}
+
+	p->parent->nchild++;
+
+	/*
+	 * 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):
+		p->norm = p->parent->norm;
+		break;
+	default:
+		break;
+	}
+
+	if ( ! mdoc_valid_pre(mdoc, p))
+		return(0);
+
+	switch (p->type) {
+	case (MDOC_HEAD):
+		assert(MDOC_BLOCK == p->parent->type);
+		p->parent->head = p;
+		break;
+	case (MDOC_TAIL):
+		assert(MDOC_BLOCK == p->parent->type);
+		p->parent->tail = p;
+		break;
+	case (MDOC_BODY):
+		if (p->end)
+			break;
+		assert(MDOC_BLOCK == p->parent->type);
+		p->parent->body = p;
+		break;
+	default:
+		break;
+	}
+
+	mdoc->last = p;
+
+	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 *
+node_alloc(struct mdoc *m, int line, int pos, 
+		enum mdoct tok, enum mdoc_type type)
+{
+	struct mdoc_node *p;
+
+	p = mandoc_calloc(1, sizeof(struct mdoc_node));
+	p->sec = m->lastsec;
+	p->line = line;
+	p->pos = pos;
+	p->tok = tok;
+	p->type = type;
+
+	/* Flag analysis. */
+
+	if (MDOC_SYNOPSIS & m->flags)
+		p->flags |= MDOC_SYNPRETTY;
+	else
+		p->flags &= ~MDOC_SYNPRETTY;
+	if (MDOC_NEWLINE & m->flags)
+		p->flags |= MDOC_LINE;
+	m->flags &= ~MDOC_NEWLINE;
+
+	return(p);
+}
+
+
+int
+mdoc_tail_alloc(struct mdoc *m, int line, int pos, enum mdoct tok)
+{
+	struct mdoc_node *p;
+
+	p = node_alloc(m, line, pos, tok, MDOC_TAIL);
+	if ( ! node_append(m, p))
+		return(0);
+	m->next = MDOC_NEXT_CHILD;
+	return(1);
+}
+
+
+int
+mdoc_head_alloc(struct mdoc *m, int line, int pos, enum mdoct tok)
+{
+	struct mdoc_node *p;
+
+	assert(m->first);
+	assert(m->last);
+
+	p = node_alloc(m, line, pos, tok, MDOC_HEAD);
+	if ( ! node_append(m, p))
+		return(0);
+	m->next = MDOC_NEXT_CHILD;
+	return(1);
+}
+
+
+int
+mdoc_body_alloc(struct mdoc *m, int line, int pos, enum mdoct tok)
+{
+	struct mdoc_node *p;
+
+	p = node_alloc(m, line, pos, tok, MDOC_BODY);
+	if ( ! node_append(m, p))
+		return(0);
+	m->next = MDOC_NEXT_CHILD;
+	return(1);
+}
+
+
+int
+mdoc_endbody_alloc(struct mdoc *m, int line, int pos, enum mdoct tok,
+		struct mdoc_node *body, enum mdoc_endbody end)
+{
+	struct mdoc_node *p;
+
+	p = node_alloc(m, line, pos, tok, MDOC_BODY);
+	p->pending = body;
+	p->end = end;
+	if ( ! node_append(m, p))
+		return(0);
+	m->next = MDOC_NEXT_SIBLING;
+	return(1);
+}
+
+
+int
+mdoc_block_alloc(struct mdoc *m, int line, int pos, 
+		enum mdoct tok, struct mdoc_arg *args)
+{
+	struct mdoc_node *p;
+
+	p = node_alloc(m, line, pos, tok, MDOC_BLOCK);
+	p->args = args;
+	if (p->args)
+		(args->refcnt)++;
+
+	switch (tok) {
+	case (MDOC_Bd):
+		/* FALLTHROUGH */
+	case (MDOC_Bf):
+		/* FALLTHROUGH */
+	case (MDOC_Bl):
+		/* FALLTHROUGH */
+	case (MDOC_Rs):
+		p->norm = mandoc_calloc(1, sizeof(union mdoc_data));
+		break;
+	default:
+		break;
+	}
+
+	if ( ! node_append(m, p))
+		return(0);
+	m->next = MDOC_NEXT_CHILD;
+	return(1);
+}
+
+
+int
+mdoc_elem_alloc(struct mdoc *m, int line, int pos, 
+		enum mdoct tok, struct mdoc_arg *args)
+{
+	struct mdoc_node *p;
+
+	p = node_alloc(m, line, pos, tok, MDOC_ELEM);
+	p->args = args;
+	if (p->args)
+		(args->refcnt)++;
+
+	switch (tok) {
+	case (MDOC_An):
+		p->norm = mandoc_calloc(1, sizeof(union mdoc_data));
+		break;
+	default:
+		break;
+	}
+
+	if ( ! node_append(m, p))
+		return(0);
+	m->next = MDOC_NEXT_CHILD;
+	return(1);
+}
+
+int
+mdoc_word_alloc(struct mdoc *m, int line, int pos, const char *p)
+{
+	struct mdoc_node *n;
+
+	n = node_alloc(m, line, pos, MDOC_MAX, MDOC_TEXT);
+	n->string = roff_strdup(m->roff, p);
+
+	if ( ! node_append(m, n))
+		return(0);
+
+	m->next = MDOC_NEXT_SIBLING;
+	return(1);
+}
+
+
+static void
+mdoc_node_free(struct mdoc_node *p)
+{
+
+	if (MDOC_BLOCK == p->type || MDOC_ELEM == p->type)
+		free(p->norm);
+	if (p->string)
+		free(p->string);
+	if (p->args)
+		mdoc_argv_free(p->args);
+	free(p);
+}
+
+
+static void
+mdoc_node_unlink(struct mdoc *m, struct mdoc_node *n)
+{
+
+	/* Adjust siblings. */
+
+	if (n->prev)
+		n->prev->next = n->next;
+	if (n->next)
+		n->next->prev = n->prev;
+
+	/* Adjust parent. */
+
+	if (n->parent) {
+		n->parent->nchild--;
+		if (n->parent->child == n)
+			n->parent->child = n->prev ? n->prev : n->next;
+		if (n->parent->last == n)
+			n->parent->last = n->prev ? n->prev : NULL;
+	}
+
+	/* Adjust parse point, if applicable. */
+
+	if (m && m->last == n) {
+		if (n->prev) {
+			m->last = n->prev;
+			m->next = MDOC_NEXT_SIBLING;
+		} else {
+			m->last = n->parent;
+			m->next = MDOC_NEXT_CHILD;
+		}
+	}
+
+	if (m && m->first == n)
+		m->first = NULL;
+}
+
+
+void
+mdoc_node_delete(struct mdoc *m, struct mdoc_node *p)
+{
+
+	while (p->child) {
+		assert(p->nchild);
+		mdoc_node_delete(m, p->child);
+	}
+	assert(0 == p->nchild);
+
+	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
+mdoc_preptext(struct mdoc *m, int line, char *buf, int offs)
+{
+	char		*start, *end;
+	char		 delim;
+
+	while ('\0' != buf[offs]) {
+		/* Mark starting position if eqn is set. */
+		start = NULL;
+		if ('\0' != (delim = roff_eqndelim(m->roff)))
+			if (NULL != (start = strchr(buf + offs, delim)))
+				*start++ = '\0';
+
+		/* Parse text as normal. */
+		if ( ! mdoc_ptext(m, line, buf, offs))
+			return(0);
+
+		/* Continue only if an equation exists. */
+		if (NULL == start)
+			break;
+
+		/* Read past the end of the equation. */
+		offs += start - (buf + offs);
+		assert(start == &buf[offs]);
+		if (NULL != (end = strchr(buf + offs, delim))) {
+			*end++ = '\0';
+			while (' ' == *end)
+				end++;
+		}
+
+		/* Parse the equation itself. */
+		roff_openeqn(m->roff, NULL, line, offs, buf);
+
+		/* Process a finished equation? */
+		if (roff_closeeqn(m->roff))
+			if ( ! mdoc_addeqn(m, roff_eqn(m->roff)))
+				return(0);
+		offs += (end - (buf + offs));
+	} 
+
+	return(1);
+}
+#endif
+
+/*
+ * Parse free-form text, that is, a line that does not begin with the
+ * control character.
+ */
+static int
+mdoc_ptext(struct mdoc *m, int line, char *buf, int offs)
+{
+	char		 *c, *ws, *end;
+	struct mdoc_node *n;
+
+	/* No text before an initial macro. */
+
+	if (SEC_NONE == m->lastnamed) {
+		mdoc_pmsg(m, line, offs, MANDOCERR_NOTEXT);
+		return(1);
+	}
+
+	assert(m->last);
+	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).
+	 */
+
+	if (MDOC_Bl == n->tok && MDOC_BODY == n->type &&
+			LIST_column == n->norm->Bl.type) {
+		/* `Bl' is open without any children. */
+		m->flags |= MDOC_FREECOL;
+		return(mdoc_macro(m, MDOC_It, line, offs, &offs, buf));
+	}
+
+	if (MDOC_It == n->tok && MDOC_BLOCK == n->type &&
+			NULL != n->parent &&
+			MDOC_Bl == n->parent->tok &&
+			LIST_column == n->parent->norm->Bl.type) {
+		/* `Bl' has block-level `It' children. */
+		m->flags |= MDOC_FREECOL;
+		return(mdoc_macro(m, MDOC_It, line, offs, &offs, buf));
+	}
+
+	/*
+	 * Search for the beginning of unescaped trailing whitespace (ws)
+	 * and for the first character not to be output (end).
+	 */
+
+	/* FIXME: replace with strcspn(). */
+	ws = NULL;
+	for (c = end = buf + offs; *c; c++) {
+		switch (*c) {
+		case ' ':
+			if (NULL == ws)
+				ws = c;
+			continue;
+		case '\t':
+			/*
+			 * Always warn about trailing tabs,
+			 * even outside literal context,
+			 * where they should be put on the next line.
+			 */
+			if (NULL == ws)
+				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:
+			ws = NULL;
+			break;
+		}
+		end = c + 1;
+	}
+	*end = '\0';
+
+	if (ws)
+		mdoc_pmsg(m, line, (int)(ws-buf), MANDOCERR_EOLNSPACE);
+
+	if ('\0' == buf[offs] && ! (MDOC_LITERAL & m->flags)) {
+		mdoc_pmsg(m, line, (int)(c-buf), MANDOCERR_NOBLANKLN);
+
+		/*
+		 * 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.
+		 */
+		if ( ! mdoc_elem_alloc(m, line, offs, MDOC_sp, NULL))
+			return(0);
+
+		m->next = MDOC_NEXT_SIBLING;
+		return(1);
+	}
+
+	if ( ! mdoc_word_alloc(m, line, offs, buf+offs))
+		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.
+	 */
+
+	assert(buf < end);
+
+	if (mandoc_eos(buf+offs, (size_t)(end-buf-offs), 0))
+		m->last->flags |= MDOC_EOS;
+
+	return(1);
+}
+
+
+/*
+ * Parse a macro line, that is, a line beginning with the control
+ * character.
+ */
+static int
+mdoc_pmacro(struct mdoc *m, int ln, char *buf, int offs)
+{
+	enum mdoct	  tok;
+	int		  i, sv;
+	char		  mac[5];
+	struct mdoc_node *n;
+
+	/* Empty post-control lines are ignored. */
+
+	if ('"' == buf[offs]) {
+		mdoc_pmsg(m, ln, offs, MANDOCERR_BADCOMMENT);
+		return(1);
+	} else if ('\0' == buf[offs])
+		return(1);
+
+	sv = offs;
+
+	/* 
+	 * Copy the first word into a nil-terminated buffer.
+	 * Stop copying when a tab, space, or eoln is encountered.
+	 */
+
+	i = 0;
+	while (i < 4 && '\0' != buf[offs] && 
+			' ' != buf[offs] && '\t' != buf[offs])
+		mac[i++] = buf[offs++];
+
+	mac[i] = '\0';
+
+	tok = (i > 1 || i < 4) ? mdoc_hash_find(mac) : MDOC_MAX;
+
+	if (MDOC_MAX == tok) {
+		mandoc_vmsg(MANDOCERR_MACRO, m->parse, 
+				ln, sv, "%s", buf + sv - 1);
+		return(1);
+	}
+
+	/* Disregard the first trailing tab, if applicable. */
+
+	if ('\t' == buf[offs])
+		offs++;
+
+	/* Jump to the next non-whitespace word. */
+
+	while (buf[offs] && ' ' == buf[offs])
+		offs++;
+
+	/* 
+	 * Trailing whitespace.  Note that tabs are allowed to be passed
+	 * into the parser as "text", so we only warn about spaces here.
+	 */
+
+	if ('\0' == buf[offs] && ' ' == buf[offs - 1])
+		mdoc_pmsg(m, ln, offs - 1, MANDOCERR_EOLNSPACE);
+
+	/*
+	 * If an initial macro or a list invocation, divert directly
+	 * into macro processing.
+	 */
+
+	if (NULL == m->last || MDOC_It == tok || MDOC_El == tok) {
+		if ( ! mdoc_macro(m, tok, ln, sv, &offs, buf)) 
+			goto err;
+		return(1);
+	}
+
+	n = m->last;
+	assert(m->last);
+
+	/*
+	 * If the first macro of a `Bl -column', open an `It' block
+	 * context around the parsed macro.
+	 */
+
+	if (MDOC_Bl == n->tok && MDOC_BODY == n->type &&
+			LIST_column == n->norm->Bl.type) {
+		m->flags |= MDOC_FREECOL;
+		if ( ! mdoc_macro(m, MDOC_It, ln, sv, &sv, buf))
+			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.
+	 */
+
+	if (MDOC_It == n->tok && MDOC_BLOCK == n->type &&
+			NULL != n->parent &&
+			MDOC_Bl == n->parent->tok &&
+			LIST_column == n->parent->norm->Bl.type) {
+		m->flags |= MDOC_FREECOL;
+		if ( ! mdoc_macro(m, MDOC_It, ln, sv, &sv, buf)) 
+			goto err;
+		return(1);
+	}
+
+	/* Normal processing of a macro. */
+
+	if ( ! mdoc_macro(m, tok, ln, sv, &offs, buf)) 
+		goto err;
+
+	return(1);
+
+err:	/* Error out. */
+
+	m->flags |= MDOC_HALT;
+	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);
+
+	if (0 == strcmp(p + 1, "."))
+		return(DELIM_CLOSE);
+	if (0 == strcmp(p + 1, "*(Ba"))
+		return(DELIM_MIDDLE);
+
+	return(DELIM_NONE);
+}