-/* $Id: mdoc_macro.c,v 1.54 2010/05/07 05:39:35 kristaps Exp $ */
+/* $Id: mdoc_macro.c,v 1.79 2010/05/31 23:49:16 kristaps Exp $ */
/*
* Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
*
#include <string.h>
#include <time.h>
+#include "mandoc.h"
#include "libmdoc.h"
+#include "libmandoc.h"
enum rew {
REWIND_REWIND,
static int in_line_argn(MACRO_PROT_ARGS);
static int in_line(MACRO_PROT_ARGS);
static int obsolete(MACRO_PROT_ARGS);
+static int phrase_ta(MACRO_PROT_ARGS);
static int append_delims(struct mdoc *,
int, int *, char *);
-static enum mdoct lookup(int, const char *);
+static enum mdoct lookup(enum mdoct, const char *);
static enum mdoct lookup_raw(const char *);
static int phrase(struct mdoc *, int, int, char *);
static enum mdoct rew_alt(enum mdoct);
{ in_line_eoln, 0 }, /* Hf */
{ obsolete, 0 }, /* Fr */
{ in_line_eoln, 0 }, /* Ud */
- { in_line_eoln, 0 }, /* Lb */
+ { in_line, 0 }, /* Lb */
{ in_line_eoln, 0 }, /* Lp */
{ in_line, MDOC_CALLABLE | MDOC_PARSED }, /* Lk */
{ in_line, MDOC_CALLABLE | MDOC_PARSED }, /* Mt */
{ in_line_eoln, 0 }, /* br */
{ in_line_eoln, 0 }, /* sp */
{ in_line_eoln, 0 }, /* %U */
+ { phrase_ta, MDOC_CALLABLE | MDOC_PARSED }, /* Ta */
};
const struct mdoc_macro * const mdoc_macros = __mdoc_macros;
int line, int pos, const struct mdoc_node *p)
{
const char *n, *t, *tt;
+ enum mandocerr ec;
n = t = "<root>";
tt = "block";
break;
}
- if ( ! (MDOC_IGN_SCOPE & mdoc->pflags))
- return(mdoc_verr(mdoc, line, pos,
- "%s scope breaks %s scope of %s",
- tt, t, n));
- return(mdoc_vwarn(mdoc, line, pos,
- "%s scope breaks %s scope of %s",
+ ec = (MDOC_IGN_SCOPE & mdoc->pflags) ?
+ MANDOCERR_SCOPE : MANDOCERR_SYNTSCOPE;
+
+ return(mdoc_vmsg(mdoc, ec, line, pos,
+ "%s scope breaks %s of %s",
tt, t, n));
}
continue;
if ( ! (MDOC_EXPLICIT & mdoc_macros[n->tok].flags))
continue;
- return(mdoc_nerr(m, n, EOPEN));
+ mdoc_nmsg(m, n, MANDOCERR_SYNTSCOPE);
+ return(0);
}
/* Rewind to the first. */
* Look up a macro from within a subsequent context.
*/
static enum mdoct
-lookup(int from, const char *p)
+lookup(enum mdoct from, const char *p)
{
/* FIXME: make -diag lists be un-PARSED. */
static enum mdoct
lookup_raw(const char *p)
{
- int res;
+ enum mdoct res;
if (MDOC_MAX == (res = mdoc_hash_find(p)))
return(MDOC_MAX);
return(1);
if ( ! (MDOC_EXPLICIT & mdoc_macros[tok].flags))
return(1);
- return(mdoc_perr(m, line, ppos, ENOCTX));
+ /* FIXME: shouldn't raise an error */
+ mdoc_pmsg(m, line, ppos, MANDOCERR_SYNTNOSCOPE);
+ return(0);
}
if (REWIND_REWIND == c)
break;
static int
-append_delims(struct mdoc *mdoc, int line, int *pos, char *buf)
+append_delims(struct mdoc *m, int line, int *pos, char *buf)
{
- int lastarg;
- enum margerr ac;
+ int la;
+ enum margserr ac;
char *p;
- if (0 == buf[*pos])
+ if ('\0' == buf[*pos])
return(1);
for (;;) {
- lastarg = *pos;
- ac = mdoc_zargs(mdoc, line, pos, buf, ARGS_NOWARN, &p);
- assert(ARGS_PHRASE != ac);
+ la = *pos;
+ ac = mdoc_zargs(m, line, pos, buf, ARGS_NOWARN, &p);
if (ARGS_ERROR == ac)
return(0);
else if (ARGS_EOLN == ac)
break;
- assert(mdoc_isdelim(p));
- if ( ! mdoc_word_alloc(mdoc, line, lastarg, p))
+
+ assert(DELIM_NONE != mdoc_isdelim(p));
+ if ( ! mdoc_word_alloc(m, line, la, p))
return(0);
+
+ /*
+ * If we encounter end-of-sentence symbols, then trigger
+ * the double-space.
+ *
+ * XXX: it's easy to allow this to propogate outward to
+ * the last symbol, such that `. )' will cause the
+ * correct double-spacing. However, (1) groff isn't
+ * smart enough to do this and (2) it would require
+ * knowing which symbols break this behaviour, for
+ * example, `. ;' shouldn't propogate the double-space.
+ */
+ if (mandoc_eos(p, strlen(p)))
+ m->last->flags |= MDOC_EOS;
}
return(1);
static int
blk_exp_close(MACRO_PROT_ARGS)
{
- int j, lastarg, maxargs, flushed;
- enum margerr ac;
+ int j, lastarg, maxargs, flushed, nl;
+ enum margserr ac;
enum mdoct ntok;
char *p;
+ nl = MDOC_NEWLINE & m->flags;
+
switch (tok) {
case (MDOC_Ec):
maxargs = 1;
}
if ( ! (MDOC_CALLABLE & mdoc_macros[tok].flags)) {
+ /* FIXME: do this in validate */
if (buf[*pos])
- if ( ! mdoc_pwarn(m, line, ppos, ENOLINE))
+ if ( ! mdoc_pmsg(m, line, ppos, MANDOCERR_ARGSLOST))
return(0);
if ( ! rew_sub(MDOC_BODY, m, tok, line, ppos))
if ( ! flushed && ! rew_sub(MDOC_BLOCK, m, tok, line, ppos))
return(0);
- if (ppos > 1)
+ if ( ! nl)
return(1);
return(append_delims(m, line, pos, buf));
}
static int
in_line(MACRO_PROT_ARGS)
{
- int la, lastpunct, c, cnt, d, nc;
+ int la, scope, cnt, nc, nl;
+ enum margverr av;
enum mdoct ntok;
- enum margerr ac;
+ enum margserr ac;
+ enum mdelim d;
struct mdoc_arg *arg;
char *p;
+ nl = MDOC_NEWLINE & m->flags;
+
/*
* Whether we allow ignored elements (those without content,
* usually because of reserved words) to squeak by.
for (arg = NULL;; ) {
la = *pos;
- c = mdoc_argv(m, line, tok, &arg, pos, buf);
+ av = mdoc_argv(m, line, tok, &arg, pos, buf);
- if (ARGV_WORD == c) {
+ if (ARGV_WORD == av) {
*pos = la;
break;
}
- if (ARGV_EOLN == c)
+ if (ARGV_EOLN == av)
break;
- if (ARGV_ARG == c)
+ if (ARGV_ARG == av)
continue;
mdoc_argv_free(arg);
return(0);
}
- for (cnt = 0, lastpunct = 1;; ) {
+ for (cnt = scope = 0;; ) {
la = *pos;
ac = mdoc_args(m, line, pos, buf, tok, &p);
*/
if (MDOC_MAX != ntok) {
- if (0 == lastpunct && ! rew_elem(m, tok))
+ if (scope && ! rew_elem(m, tok))
return(0);
if (nc && 0 == cnt) {
if ( ! mdoc_elem_alloc(m, line, ppos, tok, arg))
return(0);
} else if ( ! nc && 0 == cnt) {
mdoc_argv_free(arg);
- if ( ! mdoc_pwarn(m, line, ppos, EIGNE))
+ if ( ! mdoc_pmsg(m, line, ppos, MANDOCERR_MACROEMPTY))
return(0);
}
if ( ! mdoc_macro(m, ntok, line, la, pos, buf))
return(0);
- if (ppos > 1)
+ if ( ! nl)
return(1);
return(append_delims(m, line, pos, buf));
}
* the word.
*/
- d = ARGS_QWORD == ac ? 0 : mdoc_isdelim(p);
-
- if (ARGS_QWORD != ac && d) {
- if (0 == lastpunct && ! rew_elem(m, tok))
+ d = ARGS_QWORD == ac ? DELIM_NONE : mdoc_isdelim(p);
+
+ if (DELIM_NONE != d) {
+ /*
+ * If we encounter closing punctuation, no word
+ * has been omitted, no scope is open, and we're
+ * allowed to have an empty element, then start
+ * a new scope. `Ar', `Fl', and `Li', only do
+ * this once per invocation. There may be more
+ * of these (all of them?).
+ */
+ if (0 == cnt && (nc || MDOC_Li == tok) &&
+ DELIM_CLOSE == d && ! scope) {
+ if ( ! mdoc_elem_alloc(m, line, ppos, tok, arg))
+ return(0);
+ if (MDOC_Ar == tok || MDOC_Li == tok ||
+ MDOC_Fl == tok)
+ cnt++;
+ scope = 1;
+ }
+ /*
+ * Close out our scope, if one is open, before
+ * any punctuation.
+ */
+ if (scope && ! rew_elem(m, tok))
return(0);
- lastpunct = 1;
- } else if (lastpunct) {
+ scope = 0;
+ } else if ( ! scope) {
if ( ! mdoc_elem_alloc(m, line, ppos, tok, arg))
return(0);
- lastpunct = 0;
+ scope = 1;
}
- if ( ! d)
+ if (DELIM_NONE == d)
cnt++;
if ( ! mdoc_word_alloc(m, line, la, p))
return(0);
* word so that the `-' can be added to each one without
* having to parse out spaces.
*/
- if (0 == lastpunct && MDOC_Fl == tok) {
+ if (scope && MDOC_Fl == tok) {
if ( ! rew_elem(m, tok))
return(0);
- lastpunct = 1;
+ scope = 0;
}
}
- if (0 == lastpunct && ! rew_elem(m, tok))
+ if (scope && ! rew_elem(m, tok))
return(0);
/*
return(0);
} else if ( ! nc && 0 == cnt) {
mdoc_argv_free(arg);
- if ( ! mdoc_pwarn(m, line, ppos, EIGNE))
+ if ( ! mdoc_pmsg(m, line, ppos, MANDOCERR_MACROEMPTY))
return(0);
}
- if (ppos > 1)
+ if ( ! nl)
return(1);
return(append_delims(m, line, pos, buf));
}
static int
blk_full(MACRO_PROT_ARGS)
{
- int c, la;
+ int la, nl;
struct mdoc_arg *arg;
struct mdoc_node *head; /* save of head macro */
struct mdoc_node *body; /* save of body macro */
#ifdef UGLY
struct mdoc_node *n;
#endif
+ enum mdoc_type mtt;
enum mdoct ntok;
- enum margerr ac;
+ enum margserr ac, lac;
+ enum margverr av;
char *p;
+ nl = MDOC_NEWLINE & m->flags;
+
/* Close out prior implicit scope. */
if ( ! (MDOC_EXPLICIT & mdoc_macros[tok].flags)) {
for (arg = NULL;; ) {
la = *pos;
- c = mdoc_argv(m, line, tok, &arg, pos, buf);
+ av = mdoc_argv(m, line, tok, &arg, pos, buf);
- if (ARGV_WORD == c) {
+ if (ARGV_WORD == av) {
*pos = la;
break;
}
- if (ARGV_EOLN == c)
+ if (ARGV_EOLN == av)
break;
- if (ARGV_ARG == c)
+ if (ARGV_ARG == av)
continue;
mdoc_argv_free(arg);
body = m->last;
}
- for (;;) {
+ ac = ARGS_ERROR;
+
+ for ( ; ; ) {
la = *pos;
+ /* Initialise last-phrase-type with ARGS_PEND. */
+ lac = ARGS_ERROR == ac ? ARGS_PEND : ac;
ac = mdoc_args(m, line, pos, buf, tok, &p);
if (ARGS_ERROR == ac)
return(0);
- if (ARGS_EOLN == ac)
+
+ if (ARGS_EOLN == ac) {
+ if (ARGS_PPHRASE != lac && ARGS_PHRASE != lac)
+ break;
+ /*
+ * This is necessary: if the last token on a
+ * line is a `Ta' or tab, then we'll get
+ * ARGS_EOLN, so we must be smart enough to
+ * reopen our scope if the last parse was a
+ * phrase or partial phrase.
+ */
+ if ( ! rew_sub(MDOC_BODY, m, tok, line, ppos))
+ return(0);
+ if ( ! mdoc_body_alloc(m, line, ppos, tok))
+ return(0);
+ body = m->last;
break;
+ }
- /* Don't emit leading punct. for phrases. */
+ /*
+ * Emit leading punctuation (i.e., punctuation before
+ * the MDOC_HEAD) for non-phrase types.
+ */
- if (NULL == head && ARGS_PHRASE != ac &&
+ if (NULL == head &&
+ ARGS_PEND != ac &&
+ ARGS_PHRASE != ac &&
+ ARGS_PPHRASE != ac &&
ARGS_QWORD != ac &&
- 1 == mdoc_isdelim(p)) {
+ DELIM_OPEN == mdoc_isdelim(p)) {
if ( ! mdoc_word_alloc(m, line, la, p))
return(0);
continue;
}
- /* Always re-open head for phrases. */
+ /* Open a head if one hasn't been opened. */
- if (NULL == head || ARGS_PHRASE == ac) {
+ if (NULL == head) {
if ( ! mdoc_head_alloc(m, line, ppos, tok))
return(0);
head = m->last;
}
- if (ARGS_PHRASE == ac) {
- if ( ! phrase(m, line, la, buf))
+ if (ARGS_PHRASE == ac ||
+ ARGS_PEND == ac ||
+ ARGS_PPHRASE == ac) {
+ /*
+ * If we haven't opened a body yet, rewind the
+ * head; if we have, rewind that instead.
+ */
+
+ mtt = body ? MDOC_BODY : MDOC_HEAD;
+ if ( ! rew_sub(mtt, m, tok, line, ppos))
return(0);
- if ( ! rew_sub(MDOC_HEAD, m, tok, line, ppos))
+
+ /* Then allocate our body context. */
+
+ if ( ! mdoc_body_alloc(m, line, ppos, tok))
+ return(0);
+ body = m->last;
+
+ /*
+ * Process phrases: set whether we're in a
+ * partial-phrase (this effects line handling)
+ * then call down into the phrase parser.
+ */
+
+ if (ARGS_PPHRASE == ac)
+ m->flags |= MDOC_PPHRASE;
+ if (ARGS_PEND == ac && ARGS_PPHRASE == lac)
+ m->flags |= MDOC_PPHRASE;
+
+ if ( ! phrase(m, line, la, buf))
return(0);
+
+ m->flags &= ~MDOC_PPHRASE;
continue;
}
head = m->last;
}
- if (1 == ppos && ! append_delims(m, line, pos, buf))
+ if (nl && ! append_delims(m, line, pos, buf))
return(0);
/* If we've already opened our body, exit now. */
if (NULL != body)
- return(1);
+ goto out;
#ifdef UGLY
/*
if ( ! mdoc_body_alloc(m, line, ppos, tok))
return(0);
+out:
+ if ( ! (MDOC_FREECOL & m->flags))
+ return(1);
+
+ if ( ! rew_sub(MDOC_BODY, m, tok, line, ppos))
+ return(0);
+ if ( ! rew_sub(MDOC_BLOCK, m, tok, line, ppos))
+ return(0);
+
+ m->flags &= ~MDOC_FREECOL;
return(1);
}
static int
blk_part_imp(MACRO_PROT_ARGS)
{
- int la;
+ int la, nl;
enum mdoct ntok;
- enum margerr ac;
+ enum margserr ac;
char *p;
struct mdoc_node *blk; /* saved block context */
struct mdoc_node *body; /* saved body context */
struct mdoc_node *n;
+ nl = MDOC_NEWLINE & m->flags;
+
/*
* A macro that spans to the end of the line. This is generally
* (but not necessarily) called as the first macro. The block
la = *pos;
ac = mdoc_args(m, line, pos, buf, tok, &p);
- assert(ARGS_PHRASE != ac);
-
if (ARGS_ERROR == ac)
return(0);
if (ARGS_EOLN == ac)
break;
if (NULL == body && ARGS_QWORD != ac &&
- 1 == mdoc_isdelim(p)) {
+ DELIM_OPEN == mdoc_isdelim(p)) {
if ( ! mdoc_word_alloc(m, line, la, p))
return(0);
continue;
body = m->last;
}
+ for (n = body->child; n && n->next; n = n->next)
+ /* Do nothing. */ ;
+
+ /*
+ * End of sentence spacing: if the last node is a text node and
+ * has a trailing period, then mark it as being end-of-sentence.
+ */
+
+ if (n && MDOC_TEXT == n->type && n->string)
+ if (mandoc_eos(n->string, strlen(n->string)))
+ n->flags |= MDOC_EOS;
+
+ /* Up-propogate the end-of-space flag. */
+
+ if (n && (MDOC_EOS & n->flags)) {
+ body->flags |= MDOC_EOS;
+ body->parent->flags |= MDOC_EOS;
+ }
+
/*
* If we can't rewind to our body, then our scope has already
* been closed by another macro (like `Oc' closing `Op'). This
* is ugly behaviour nodding its head to OpenBSD's overwhelming
* crufty use of `Op' breakage.
+ *
+ * FIXME - this should be ifdef'd OpenBSD?
*/
for (n = m->last; n; n = n->parent)
if (body == n)
break;
- if (NULL == n && ! mdoc_nwarn(m, body, EIMPBRK))
+ if (NULL == n && ! mdoc_nmsg(m, body, MANDOCERR_SCOPE))
return(0);
if (n && ! rew_last(m, body))
/* Standard appending of delimiters. */
- if (1 == ppos && ! append_delims(m, line, pos, buf))
+ if (nl && ! append_delims(m, line, pos, buf))
return(0);
/* Rewind scope, if applicable. */
static int
blk_part_exp(MACRO_PROT_ARGS)
{
- int la;
- enum margerr ac;
+ int la, nl;
+ enum margserr ac;
struct mdoc_node *head; /* keep track of head */
struct mdoc_node *body; /* keep track of body */
char *p;
enum mdoct ntok;
+ nl = MDOC_NEWLINE & m->flags;
+
/*
* The opening of an explicit macro having zero or more leading
* punctuation nodes; a head with optional single element (the
if (ARGS_EOLN == ac)
break;
- assert(ARGS_PHRASE != ac);
-
/* Flush out leading punctuation. */
if (NULL == head && ARGS_QWORD != ac &&
- 1 == mdoc_isdelim(p)) {
+ DELIM_OPEN == mdoc_isdelim(p)) {
assert(NULL == body);
if ( ! mdoc_word_alloc(m, line, la, p))
return(0);
/* Standard appending of delimiters. */
- if (ppos > 1)
+ if ( ! nl)
return(1);
-
return(append_delims(m, line, pos, buf));
}
+/* ARGSUSED */
static int
in_line_argn(MACRO_PROT_ARGS)
{
- int la, flushed, j, c, maxargs;
- enum margerr ac;
+ int la, flushed, j, maxargs, nl;
+ enum margserr ac;
+ enum margverr av;
struct mdoc_arg *arg;
char *p;
enum mdoct ntok;
+ nl = MDOC_NEWLINE & m->flags;
+
/*
* A line macro that has a fixed number of arguments (maxargs).
* Only open the scope once the first non-leading-punctuation is
for (arg = NULL; ; ) {
la = *pos;
- c = mdoc_argv(m, line, tok, &arg, pos, buf);
+ av = mdoc_argv(m, line, tok, &arg, pos, buf);
- if (ARGV_WORD == c) {
+ if (ARGV_WORD == av) {
*pos = la;
break;
}
- if (ARGV_EOLN == c)
+ if (ARGV_EOLN == av)
break;
- if (ARGV_ARG == c)
+ if (ARGV_ARG == av)
continue;
mdoc_argv_free(arg);
if ( ! (MDOC_IGNDELIM & mdoc_macros[tok].flags) &&
ARGS_QWORD != ac &&
- 0 == j && 1 == mdoc_isdelim(p)) {
+ 0 == j && DELIM_OPEN == mdoc_isdelim(p)) {
if ( ! mdoc_word_alloc(m, line, la, p))
return(0);
continue;
if ( ! (MDOC_IGNDELIM & mdoc_macros[tok].flags) &&
ARGS_QWORD != ac &&
- ! flushed && mdoc_isdelim(p)) {
+ ! flushed &&
+ DELIM_NONE != mdoc_isdelim(p)) {
if ( ! rew_elem(m, tok))
return(0);
flushed = 1;
if ( ! flushed && ! rew_elem(m, tok))
return(0);
-
- if (ppos > 1)
+ if ( ! nl)
return(1);
return(append_delims(m, line, pos, buf));
}
static int
in_line_eoln(MACRO_PROT_ARGS)
{
- int c, la;
- enum margerr ac;
+ int la;
+ enum margserr ac;
+ enum margverr av;
struct mdoc_arg *arg;
char *p;
enum mdoct ntok;
for (arg = NULL; ; ) {
la = *pos;
- c = mdoc_argv(m, line, tok, &arg, pos, buf);
+ av = mdoc_argv(m, line, tok, &arg, pos, buf);
- if (ARGV_WORD == c) {
+ if (ARGV_WORD == av) {
*pos = la;
break;
}
- if (ARGV_EOLN == c)
+ if (ARGV_EOLN == av)
break;
- if (ARGV_ARG == c)
+ if (ARGV_ARG == av)
continue;
mdoc_argv_free(arg);
static int
ctx_synopsis(MACRO_PROT_ARGS)
{
+ int nl;
+
+ nl = MDOC_NEWLINE & m->flags;
/* If we're not in the SYNOPSIS, go straight to in-line. */
if (SEC_SYNOPSIS != m->lastsec)
return(in_line(m, tok, line, ppos, pos, buf));
/* If we're a nested call, same place. */
- if (ppos > 1)
+ if ( ! nl)
return(in_line(m, tok, line, ppos, pos, buf));
/*
obsolete(MACRO_PROT_ARGS)
{
- return(mdoc_pwarn(m, line, ppos, EOBS));
+ return(mdoc_pmsg(m, line, ppos, MANDOCERR_MACROOBS));
}
phrase(struct mdoc *m, int line, int ppos, char *buf)
{
int la, pos;
- enum margerr ac;
+ enum margserr ac;
enum mdoct ntok;
char *p;
for (pos = ppos; ; ) {
la = pos;
- /* Note: no calling context! */
ac = mdoc_zargs(m, line, &pos, buf, 0, &p);
if (ARGS_ERROR == ac)
}
+/* ARGSUSED */
+static int
+phrase_ta(MACRO_PROT_ARGS)
+{
+ int la;
+ enum mdoct ntok;
+ enum margserr ac;
+ char *p;
+
+ /*
+ * FIXME: this is overly restrictive: if the `Ta' is unexpected,
+ * it should simply error out with ARGSLOST.
+ */
+
+ if ( ! rew_sub(MDOC_BODY, m, MDOC_It, line, ppos))
+ return(0);
+ if ( ! mdoc_body_alloc(m, line, ppos, MDOC_It))
+ return(0);
+
+ for (;;) {
+ la = *pos;
+ ac = mdoc_zargs(m, line, pos, buf, 0, &p);
+
+ if (ARGS_ERROR == ac)
+ return(0);
+ if (ARGS_EOLN == ac)
+ break;
+
+ ntok = ARGS_QWORD == ac ? MDOC_MAX : lookup_raw(p);
+
+ if (MDOC_MAX == ntok) {
+ if ( ! mdoc_word_alloc(m, line, la, p))
+ return(0);
+ continue;
+ }
+
+ if ( ! mdoc_macro(m, ntok, line, la, pos, buf))
+ return(0);
+ return(append_delims(m, line, pos, buf));
+ }
+
+ return(1);
+}
git.ckatri.com / mandoc.git / blobdiff