-/* $Id: mdoc.c,v 1.159 2010/07/18 17:00:26 schwarze Exp $ */
+/* $Id: mdoc.c,v 1.177 2011/01/03 11:27:33 kristaps Exp $ */
/*
* Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
* Copyright (c) 2010 Ingo Schwarze <schwarze@openbsd.org>
#include <sys/types.h>
#include <assert.h>
-#include <ctype.h>
#include <stdarg.h>
#include <stdio.h>
#include <stdlib.h>
struct mdoc_node *);
static int mdoc_ptext(struct mdoc *, int, char *, int);
static int mdoc_pmacro(struct mdoc *, int, char *, int);
-static int macrowarn(struct mdoc *, int,
- const char *, int);
+static int mdoc_span_alloc(struct mdoc *,
+ const struct tbl_span *);
const struct mdoc_node *
mdoc_node(const struct mdoc *m)
{
- return(MDOC_HALT & m->flags ? NULL : m->first);
+ assert( ! (MDOC_HALT & m->flags));
+ return(m->first);
}
mdoc_meta(const struct mdoc *m)
{
- return(MDOC_HALT & m->flags ? NULL : &m->meta);
+ assert( ! (MDOC_HALT & m->flags));
+ return(&m->meta);
}
* Allocate volatile and non-volatile parse resources.
*/
struct mdoc *
-mdoc_alloc(struct regset *regs, void *data,
- int pflags, mandocmsg msg)
+mdoc_alloc(struct regset *regs, void *data, mandocmsg msg)
{
struct mdoc *p;
p->msg = msg;
p->data = data;
- p->pflags = pflags;
p->regs = regs;
mdoc_hash_init();
mdoc_endparse(struct mdoc *m)
{
- if (MDOC_HALT & m->flags)
- return(0);
- else if (mdoc_macroend(m))
+ assert( ! (MDOC_HALT & m->flags));
+ if (mdoc_macroend(m))
return(1);
m->flags |= MDOC_HALT;
return(0);
}
+int
+mdoc_addspan(struct mdoc *m, const struct tbl_span *sp)
+{
+
+ assert( ! (MDOC_HALT & m->flags));
+
+ /* No text before an initial macro. */
+
+ if (SEC_NONE == m->lastnamed) {
+ /* FIXME: grab from span. */
+ mdoc_pmsg(m, 0, 0, MANDOCERR_NOTEXT);
+ return(1);
+ }
+
+ return(mdoc_span_alloc(m, sp));
+}
+
/*
* Main parse routine. Parses a single line -- really just hands off to
mdoc_parseln(struct mdoc *m, int ln, char *buf, int offs)
{
- if (MDOC_HALT & m->flags)
- return(0);
+ assert( ! (MDOC_HALT & m->flags));
m->flags |= MDOC_NEWLINE;
/* If we're in the body, deny prologue calls. */
if (MDOC_PROLOGUE & mdoc_macros[tok].flags &&
- MDOC_PBODY & m->flags)
- return(mdoc_pmsg(m, line, ppos, MANDOCERR_BADBODY));
+ 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)) {
- if ( ! mdoc_pmsg(m, line, ppos, MANDOCERR_BADPROLOG))
- return(0);
+ 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)
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);
- if ( ! mdoc_action_pre(mdoc, p))
- return(0);
switch (p->type) {
case (MDOC_HEAD):
mdoc->last = p;
switch (p->type) {
+ case (MDOC_TBL):
+ /* FALLTHROUGH */
case (MDOC_TEXT):
if ( ! mdoc_valid_post(mdoc))
return(0);
- if ( ! mdoc_action_post(mdoc))
- return(0);
break;
default:
break;
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;
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);
}
+static int
+mdoc_span_alloc(struct mdoc *m, const struct tbl_span *sp)
+{
+ struct mdoc_node *n;
+
+ /* FIXME: grab from tbl_span. */
+ n = node_alloc(m, 0, 0, MDOC_MAX, MDOC_TBL);
+ n->span = sp;
+
+ if ( ! node_append(m, n))
+ return(0);
+
+ m->next = MDOC_NEXT_SIBLING;
+ return(1);
+}
+
int
mdoc_word_alloc(struct mdoc *m, int line, int pos, const char *p)
mdoc_node_free(struct mdoc_node *p)
{
- /*
- * XXX: if these end up being problematic in terms of memory
- * management and dereferencing freed blocks, then make them
- * into reference-counted double-pointers.
- */
-
- if (MDOC_Bd == p->tok && MDOC_BLOCK == p->type)
- if (p->data.Bd)
- free(p->data.Bd);
- if (MDOC_Bl == p->tok && MDOC_BLOCK == p->type)
- if (p->data.Bl)
- free(p->data.Bl);
- if (MDOC_Bf == p->tok && MDOC_HEAD == p->type)
- if (p->data.Bf)
- free(p->data.Bf);
-
+ if (MDOC_BLOCK == p->type || MDOC_ELEM == p->type)
+ free(p->norm);
if (p->string)
free(p->string);
if (p->args)
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 ('\\' == buf[offs] &&
'.' == buf[offs + 1] &&
- '"' == buf[offs + 2])
- return(mdoc_pmsg(m, line, offs, MANDOCERR_BADCOMMENT));
+ '"' == buf[offs + 2]) {
+ mdoc_pmsg(m, line, offs, MANDOCERR_BADCOMMENT);
+ return(1);
+ }
/* No text before an initial macro. */
- if (SEC_NONE == m->lastnamed)
- return(mdoc_pmsg(m, line, offs, MANDOCERR_NOTEXT));
+ if (SEC_NONE == m->lastnamed) {
+ mdoc_pmsg(m, line, offs, MANDOCERR_NOTEXT);
+ return(1);
+ }
assert(m->last);
n = m->last;
*/
if (MDOC_Bl == n->tok && MDOC_BODY == n->type &&
- LIST_column == n->data.Bl->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->data.Bl->type) {
+ 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));
*end = '\0';
if (ws)
- if ( ! mdoc_pmsg(m, line, (int)(ws-buf), MANDOCERR_EOLNSPACE))
- return(0);
+ mdoc_pmsg(m, line, (int)(ws-buf), MANDOCERR_EOLNSPACE);
if ('\0' == buf[offs] && ! (MDOC_LITERAL & m->flags)) {
- if ( ! mdoc_pmsg(m, line, (int)(c-buf), MANDOCERR_NOBLANKLN))
- return(0);
+ mdoc_pmsg(m, line, (int)(c-buf), MANDOCERR_NOBLANKLN);
/*
- * Insert a `Pp' in the case of a blank line. Technically,
+ * 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_Pp, NULL))
+ if ( ! mdoc_elem_alloc(m, line, offs, MDOC_sp, NULL))
return(0);
m->next = MDOC_NEXT_SIBLING;
}
-static int
-macrowarn(struct mdoc *m, int ln, const char *buf, int offs)
-{
- int rc;
-
- rc = mdoc_vmsg(m, MANDOCERR_MACRO, ln, offs,
- "unknown macro: %s%s",
- buf, strlen(buf) > 3 ? "..." : "");
-
- /* FIXME: logic should be in driver. */
- /* FIXME: broken, will error out and not omit a message. */
- return(MDOC_IGN_MACRO & m->pflags ? rc : 0);
-}
-
-
/*
* Parse a macro line, that is, a line beginning with the control
* character.
i = offs;
- /* Accept whitespace after the initial control char. */
+ /* Accept tabs/whitespace after the initial control char. */
- if (' ' == buf[i]) {
+ if (' ' == buf[i] || '\t' == buf[i]) {
i++;
- while (buf[i] && ' ' == buf[i])
+ while (buf[i] && (' ' == buf[i] || '\t' == buf[i]))
i++;
if ('\0' == buf[i])
return(1);
sv = i;
- /* Copy the first word into a nil-terminated buffer. */
-
- for (j = 0; j < 4; j++, i++) {
- if ('\0' == (mac[j] = buf[i]))
- break;
- else if (' ' == buf[i])
- break;
-
- /* Check for invalid characters. */
-
- if (isgraph((u_char)buf[i]))
- continue;
- if ( ! mdoc_pmsg(m, ln, i, MANDOCERR_BADCHAR))
- return(0);
- i--;
- }
+ /*
+ * Copy the first word into a nil-terminated buffer.
+ * Stop copying when a tab, space, or eoln is encountered.
+ */
+ j = 0;
+ while (j < 4 && '\0' != buf[i] && ' ' != buf[i] && '\t' != buf[i])
+ mac[j++] = buf[i++];
mac[j] = '\0';
- if (j == 4 || j < 2) {
- if ( ! macrowarn(m, ln, mac, sv))
- goto err;
- return(1);
- }
-
- if (MDOC_MAX == (tok = mdoc_hash_find(mac))) {
- if ( ! macrowarn(m, ln, mac, sv))
- goto err;
+ tok = (j > 1 || j < 4) ? mdoc_hash_find(mac) : MDOC_MAX;
+ if (MDOC_MAX == tok) {
+ mdoc_vmsg(m, MANDOCERR_MACRO, ln, sv, "%s", buf + sv - 1);
return(1);
}
- /* The macro is sane. Jump to the next word. */
+ /* Disregard the first trailing tab, if applicable. */
+
+ if ('\t' == buf[i])
+ i++;
+
+ /* Jump to the next non-whitespace word. */
while (buf[i] && ' ' == buf[i])
i++;
*/
if ('\0' == buf[i] && ' ' == buf[i - 1])
- if ( ! mdoc_pmsg(m, ln, i - 1, MANDOCERR_EOLNSPACE))
- goto err;
+ mdoc_pmsg(m, ln, i - 1, MANDOCERR_EOLNSPACE);
/*
* If an initial macro or a list invocation, divert directly
*/
if (MDOC_Bl == n->tok && MDOC_BODY == n->type &&
- LIST_column == n->data.Bl->type) {
+ LIST_column == n->norm->Bl.type) {
m->flags |= MDOC_FREECOL;
if ( ! mdoc_macro(m, MDOC_It, ln, sv, &sv, buf))
goto err;
if (MDOC_It == n->tok && MDOC_BLOCK == n->type &&
NULL != n->parent &&
MDOC_Bl == n->parent->tok &&
- LIST_column == n->parent->data.Bl->type) {
+ LIST_column == n->parent->norm->Bl.type) {
m->flags |= MDOC_FREECOL;
if ( ! mdoc_macro(m, MDOC_It, ln, sv, &sv, buf))
goto err;