version 1.136, 2010/05/17 22:11:42 |
version 1.197, 2012/05/27 17:48:57 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se> |
* 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 |
* Permission to use, copy, modify, and distribute this software for any |
* purpose with or without fee is hereby granted, provided that the above |
* purpose with or without fee is hereby granted, provided that the above |
|
|
#include <sys/types.h> |
#include <sys/types.h> |
|
|
#include <assert.h> |
#include <assert.h> |
#include <ctype.h> |
|
#include <stdarg.h> |
#include <stdarg.h> |
#include <stdio.h> |
#include <stdio.h> |
#include <stdlib.h> |
#include <stdlib.h> |
#include <string.h> |
#include <string.h> |
#include <time.h> |
#include <time.h> |
|
|
|
#include "mdoc.h" |
#include "mandoc.h" |
#include "mandoc.h" |
#include "libmdoc.h" |
#include "libmdoc.h" |
#include "libmandoc.h" |
#include "libmandoc.h" |
Line 69 const char *const __mdoc_macronames[MDOC_MAX] = { |
|
Line 70 const char *const __mdoc_macronames[MDOC_MAX] = { |
|
/* LINTED */ |
/* LINTED */ |
"Dx", "%Q", "br", "sp", |
"Dx", "%Q", "br", "sp", |
/* LINTED */ |
/* LINTED */ |
"%U" |
"%U", "Ta" |
}; |
}; |
|
|
const char *const __mdoc_argnames[MDOC_ARG_MAX] = { |
const char *const __mdoc_argnames[MDOC_ARG_MAX] = { |
Line 96 static struct mdoc_node *node_alloc(struct mdoc *, int |
|
Line 97 static struct mdoc_node *node_alloc(struct mdoc *, int |
|
enum mdoct, enum mdoc_type); |
enum mdoct, enum mdoc_type); |
static int node_append(struct mdoc *, |
static int node_append(struct mdoc *, |
struct mdoc_node *); |
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_ptext(struct mdoc *, int, char *, int); |
static int mdoc_pmacro(struct mdoc *, int, char *, int); |
static int mdoc_pmacro(struct mdoc *, int, char *, int); |
static int macrowarn(struct mdoc *, int, |
|
const char *, int); |
|
|
|
|
|
const struct mdoc_node * |
const struct mdoc_node * |
mdoc_node(const struct mdoc *m) |
mdoc_node(const struct mdoc *m) |
{ |
{ |
|
|
return(MDOC_HALT & m->flags ? NULL : m->first); |
assert( ! (MDOC_HALT & m->flags)); |
|
return(m->first); |
} |
} |
|
|
|
|
Line 114 const struct mdoc_meta * |
|
Line 116 const struct mdoc_meta * |
|
mdoc_meta(const struct mdoc *m) |
mdoc_meta(const struct mdoc *m) |
{ |
{ |
|
|
return(MDOC_HALT & m->flags ? NULL : &m->meta); |
assert( ! (MDOC_HALT & m->flags)); |
|
return(&m->meta); |
} |
} |
|
|
|
|
Line 139 mdoc_free1(struct mdoc *mdoc) |
|
Line 142 mdoc_free1(struct mdoc *mdoc) |
|
free(mdoc->meta.vol); |
free(mdoc->meta.vol); |
if (mdoc->meta.msec) |
if (mdoc->meta.msec) |
free(mdoc->meta.msec); |
free(mdoc->meta.msec); |
|
if (mdoc->meta.date) |
|
free(mdoc->meta.date); |
} |
} |
|
|
|
|
Line 155 mdoc_alloc1(struct mdoc *mdoc) |
|
Line 160 mdoc_alloc1(struct mdoc *mdoc) |
|
mdoc->last = mandoc_calloc(1, sizeof(struct mdoc_node)); |
mdoc->last = mandoc_calloc(1, sizeof(struct mdoc_node)); |
mdoc->first = mdoc->last; |
mdoc->first = mdoc->last; |
mdoc->last->type = MDOC_ROOT; |
mdoc->last->type = MDOC_ROOT; |
|
mdoc->last->tok = MDOC_MAX; |
mdoc->next = MDOC_NEXT_CHILD; |
mdoc->next = MDOC_NEXT_CHILD; |
} |
} |
|
|
Line 191 mdoc_free(struct mdoc *mdoc) |
|
Line 197 mdoc_free(struct mdoc *mdoc) |
|
* Allocate volatile and non-volatile parse resources. |
* Allocate volatile and non-volatile parse resources. |
*/ |
*/ |
struct mdoc * |
struct mdoc * |
mdoc_alloc(void *data, int pflags, mandocmsg msg) |
mdoc_alloc(struct roff *roff, struct mparse *parse, char *defos) |
{ |
{ |
struct mdoc *p; |
struct mdoc *p; |
|
|
p = mandoc_calloc(1, sizeof(struct mdoc)); |
p = mandoc_calloc(1, sizeof(struct mdoc)); |
|
|
p->msg = msg; |
p->parse = parse; |
p->data = data; |
p->defos = defos; |
p->pflags = pflags; |
p->roff = roff; |
|
|
mdoc_hash_init(); |
mdoc_hash_init(); |
mdoc_alloc1(p); |
mdoc_alloc1(p); |
|
|
mdoc_endparse(struct mdoc *m) |
mdoc_endparse(struct mdoc *m) |
{ |
{ |
|
|
if (MDOC_HALT & m->flags) |
assert( ! (MDOC_HALT & m->flags)); |
return(0); |
if (mdoc_macroend(m)) |
else if (mdoc_macroend(m)) |
|
return(1); |
return(1); |
m->flags |= MDOC_HALT; |
m->flags |= MDOC_HALT; |
return(0); |
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 |
* Main parse routine. Parses a single line -- really just hands off to |
* the macro (mdoc_pmacro()) or text parser (mdoc_ptext()). |
* the macro (mdoc_pmacro()) or text parser (mdoc_ptext()). |
|
|
mdoc_parseln(struct mdoc *m, int ln, char *buf, int offs) |
mdoc_parseln(struct mdoc *m, int ln, char *buf, int offs) |
{ |
{ |
|
|
if (MDOC_HALT & m->flags) |
assert( ! (MDOC_HALT & m->flags)); |
return(0); |
|
|
|
m->flags |= MDOC_NEWLINE; |
m->flags |= MDOC_NEWLINE; |
return(('.' == buf[offs] || '\'' == buf[offs]) ? |
|
|
/* |
|
* 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_pmacro(m, ln, buf, offs) : |
mdoc_ptext(m, ln, buf, offs)); |
mdoc_ptext(m, ln, buf, offs)); |
} |
} |
|
|
|
|
int |
int |
mdoc_vmsg(struct mdoc *mdoc, enum mandocerr t, |
mdoc_macro(MACRO_PROT_ARGS) |
int ln, int pos, const char *fmt, ...) |
|
{ |
{ |
char buf[256]; |
|
va_list ap; |
|
|
|
va_start(ap, fmt); |
|
vsnprintf(buf, sizeof(buf) - 1, fmt, ap); |
|
va_end(ap); |
|
|
|
return((*mdoc->msg)(t, mdoc->data, ln, pos, buf)); |
|
} |
|
|
|
|
|
int |
|
mdoc_macro(struct mdoc *m, enum mdoct tok, |
|
int ln, int pp, int *pos, char *buf) |
|
{ |
|
assert(tok < MDOC_MAX); |
assert(tok < MDOC_MAX); |
|
|
/* If we're in the body, deny prologue calls. */ |
/* If we're in the body, deny prologue calls. */ |
|
|
if (MDOC_PROLOGUE & mdoc_macros[tok].flags && |
if (MDOC_PROLOGUE & mdoc_macros[tok].flags && |
MDOC_PBODY & m->flags) |
MDOC_PBODY & m->flags) { |
return(mdoc_pmsg(m, ln, pp, MANDOCERR_BADBODY)); |
mdoc_pmsg(m, line, ppos, MANDOCERR_BADBODY); |
|
return(1); |
|
} |
|
|
/* If we're in the prologue, deny "body" macros. */ |
/* If we're in the prologue, deny "body" macros. */ |
|
|
if ( ! (MDOC_PROLOGUE & mdoc_macros[tok].flags) && |
if ( ! (MDOC_PROLOGUE & mdoc_macros[tok].flags) && |
! (MDOC_PBODY & m->flags)) { |
! (MDOC_PBODY & m->flags)) { |
if ( ! mdoc_pmsg(m, ln, pp, MANDOCERR_BADPROLOG)) |
mdoc_pmsg(m, line, ppos, MANDOCERR_BADPROLOG); |
return(0); |
if (NULL == m->meta.msec) |
|
m->meta.msec = mandoc_strdup("1"); |
if (NULL == m->meta.title) |
if (NULL == m->meta.title) |
m->meta.title = mandoc_strdup("unknown"); |
m->meta.title = mandoc_strdup("UNKNOWN"); |
if (NULL == m->meta.vol) |
if (NULL == m->meta.vol) |
m->meta.vol = mandoc_strdup("local"); |
m->meta.vol = mandoc_strdup("LOCAL"); |
if (NULL == m->meta.os) |
if (NULL == m->meta.os) |
m->meta.os = mandoc_strdup("local"); |
m->meta.os = mandoc_strdup("LOCAL"); |
if (0 == m->meta.date) |
if (NULL == m->meta.date) |
m->meta.date = time(NULL); |
m->meta.date = mandoc_normdate |
|
(m->parse, NULL, line, ppos); |
m->flags |= MDOC_PBODY; |
m->flags |= MDOC_PBODY; |
} |
} |
|
|
return((*mdoc_macros[tok].fp)(m, tok, ln, pp, pos, buf)); |
return((*mdoc_macros[tok].fp)(m, tok, line, ppos, pos, buf)); |
} |
} |
|
|
|
|
Line 315 node_append(struct mdoc *mdoc, struct mdoc_node *p) |
|
Line 368 node_append(struct mdoc *mdoc, struct mdoc_node *p) |
|
|
|
p->parent->nchild++; |
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)) |
if ( ! mdoc_valid_pre(mdoc, p)) |
return(0); |
return(0); |
if ( ! mdoc_action_pre(mdoc, p)) |
|
return(0); |
|
|
|
switch (p->type) { |
switch (p->type) { |
case (MDOC_HEAD): |
case (MDOC_HEAD): |
Line 330 node_append(struct mdoc *mdoc, struct mdoc_node *p) |
|
Line 398 node_append(struct mdoc *mdoc, struct mdoc_node *p) |
|
p->parent->tail = p; |
p->parent->tail = p; |
break; |
break; |
case (MDOC_BODY): |
case (MDOC_BODY): |
|
if (p->end) |
|
break; |
assert(MDOC_BLOCK == p->parent->type); |
assert(MDOC_BLOCK == p->parent->type); |
p->parent->body = p; |
p->parent->body = p; |
break; |
break; |
Line 340 node_append(struct mdoc *mdoc, struct mdoc_node *p) |
|
Line 410 node_append(struct mdoc *mdoc, struct mdoc_node *p) |
|
mdoc->last = p; |
mdoc->last = p; |
|
|
switch (p->type) { |
switch (p->type) { |
|
case (MDOC_TBL): |
|
/* FALLTHROUGH */ |
case (MDOC_TEXT): |
case (MDOC_TEXT): |
if ( ! mdoc_valid_post(mdoc)) |
if ( ! mdoc_valid_post(mdoc)) |
return(0); |
return(0); |
if ( ! mdoc_action_post(mdoc)) |
|
return(0); |
|
break; |
break; |
default: |
default: |
break; |
break; |
Line 366 node_alloc(struct mdoc *m, int line, int pos, |
|
Line 436 node_alloc(struct mdoc *m, int line, int pos, |
|
p->pos = pos; |
p->pos = pos; |
p->tok = tok; |
p->tok = tok; |
p->type = type; |
p->type = type; |
|
|
|
/* Flag analysis. */ |
|
|
|
if (MDOC_SYNOPSIS & m->flags) |
|
p->flags |= MDOC_SYNPRETTY; |
|
else |
|
p->flags &= ~MDOC_SYNPRETTY; |
if (MDOC_NEWLINE & m->flags) |
if (MDOC_NEWLINE & m->flags) |
p->flags |= MDOC_LINE; |
p->flags |= MDOC_LINE; |
m->flags &= ~MDOC_NEWLINE; |
m->flags &= ~MDOC_NEWLINE; |
|
|
return(p); |
return(p); |
} |
} |
|
|
Line 416 mdoc_body_alloc(struct mdoc *m, int line, int pos, enu |
|
Line 494 mdoc_body_alloc(struct mdoc *m, int line, int pos, enu |
|
|
|
|
|
int |
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, |
mdoc_block_alloc(struct mdoc *m, int line, int pos, |
enum mdoct tok, struct mdoc_arg *args) |
enum mdoct tok, struct mdoc_arg *args) |
{ |
{ |
Line 425 mdoc_block_alloc(struct mdoc *m, int line, int pos, |
|
Line 519 mdoc_block_alloc(struct mdoc *m, int line, int pos, |
|
p->args = args; |
p->args = args; |
if (p->args) |
if (p->args) |
(args->refcnt)++; |
(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)) |
if ( ! node_append(m, p)) |
return(0); |
return(0); |
m->next = MDOC_NEXT_CHILD; |
m->next = MDOC_NEXT_CHILD; |
Line 442 mdoc_elem_alloc(struct mdoc *m, int line, int pos, |
|
Line 551 mdoc_elem_alloc(struct mdoc *m, int line, int pos, |
|
p->args = args; |
p->args = args; |
if (p->args) |
if (p->args) |
(args->refcnt)++; |
(args->refcnt)++; |
|
|
|
switch (tok) { |
|
case (MDOC_An): |
|
p->norm = mandoc_calloc(1, sizeof(union mdoc_data)); |
|
break; |
|
default: |
|
break; |
|
} |
|
|
if ( ! node_append(m, p)) |
if ( ! node_append(m, p)) |
return(0); |
return(0); |
m->next = MDOC_NEXT_CHILD; |
m->next = MDOC_NEXT_CHILD; |
return(1); |
return(1); |
} |
} |
|
|
|
|
int |
int |
mdoc_word_alloc(struct mdoc *m, int line, int pos, const char *p) |
mdoc_word_alloc(struct mdoc *m, int line, int pos, const char *p) |
{ |
{ |
struct mdoc_node *n; |
struct mdoc_node *n; |
size_t sv, len; |
|
|
|
len = strlen(p); |
|
|
|
n = node_alloc(m, line, pos, MDOC_MAX, MDOC_TEXT); |
n = node_alloc(m, line, pos, MDOC_MAX, MDOC_TEXT); |
n->string = mandoc_malloc(len + 1); |
n->string = roff_strdup(m->roff, p); |
sv = strlcpy(n->string, p, len + 1); |
|
|
|
/* Prohibit truncation. */ |
|
assert(sv < len + 1); |
|
|
|
if ( ! node_append(m, n)) |
if ( ! node_append(m, n)) |
return(0); |
return(0); |
|
|
Line 472 mdoc_word_alloc(struct mdoc *m, int line, int pos, con |
|
Line 582 mdoc_word_alloc(struct mdoc *m, int line, int pos, con |
|
} |
} |
|
|
|
|
void |
static void |
mdoc_node_free(struct mdoc_node *p) |
mdoc_node_free(struct mdoc_node *p) |
{ |
{ |
|
|
|
if (MDOC_BLOCK == p->type || MDOC_ELEM == p->type) |
|
free(p->norm); |
if (p->string) |
if (p->string) |
free(p->string); |
free(p->string); |
if (p->args) |
if (p->args) |
Line 501 mdoc_node_unlink(struct mdoc *m, struct mdoc_node *n) |
|
Line 613 mdoc_node_unlink(struct mdoc *m, struct mdoc_node *n) |
|
n->parent->nchild--; |
n->parent->nchild--; |
if (n->parent->child == n) |
if (n->parent->child == n) |
n->parent->child = n->prev ? n->prev : n->next; |
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. */ |
/* Adjust parse point, if applicable. */ |
Line 534 mdoc_node_delete(struct mdoc *m, struct mdoc_node *p) |
|
Line 648 mdoc_node_delete(struct mdoc *m, struct mdoc_node *p) |
|
mdoc_node_free(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 |
* Parse free-form text, that is, a line that does not begin with the |
* control character. |
* control character. |
Line 542 mdoc_node_delete(struct mdoc *m, struct mdoc_node *p) |
|
Line 709 mdoc_node_delete(struct mdoc *m, struct mdoc_node *p) |
|
static int |
static int |
mdoc_ptext(struct mdoc *m, int line, char *buf, int offs) |
mdoc_ptext(struct mdoc *m, int line, char *buf, int offs) |
{ |
{ |
int i; |
char *c, *ws, *end; |
|
struct mdoc_node *n; |
|
|
/* Ignore bogus comments. */ |
/* No text before an initial macro. */ |
|
|
if ('\\' == buf[offs] && |
if (SEC_NONE == m->lastnamed) { |
'.' == buf[offs + 1] && |
mdoc_pmsg(m, line, offs, MANDOCERR_NOTEXT); |
'"' == buf[offs + 2]) |
return(1); |
return(mdoc_pmsg(m, line, offs, MANDOCERR_BADCOMMENT)); |
} |
|
|
/* No text before an initial macro. */ |
assert(m->last); |
|
n = m->last; |
|
|
if (SEC_NONE == m->lastnamed) |
/* |
return(mdoc_pmsg(m, line, offs, MANDOCERR_NOTEXT)); |
* 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). |
|
*/ |
|
|
/* Literal just gets pulled in as-is. */ |
if (MDOC_Bl == n->tok && MDOC_BODY == n->type && |
|
LIST_column == n->norm->Bl.type) { |
if (MDOC_LITERAL & m->flags) |
/* `Bl' is open without any children. */ |
return(mdoc_word_alloc(m, line, offs, buf + offs)); |
m->flags |= MDOC_FREECOL; |
|
return(mdoc_macro(m, MDOC_It, line, offs, &offs, buf)); |
|
} |
|
|
/* Check for a blank line, which may also consist of spaces. */ |
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)); |
|
} |
|
|
for (i = offs; ' ' == buf[i]; i++) |
/* |
/* Skip to first non-space. */ ; |
* Search for the beginning of unescaped trailing whitespace (ws) |
|
* and for the first character not to be output (end). |
|
*/ |
|
|
if ('\0' == buf[i]) { |
/* FIXME: replace with strcspn(). */ |
if ( ! mdoc_pmsg(m, line, offs, MANDOCERR_NOBLANKLN)) |
ws = NULL; |
return(0); |
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 `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 |
* blank lines aren't allowed, but enough manuals assume this |
* behaviour that we want to work around it. |
* 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); |
return(0); |
|
|
m->next = MDOC_NEXT_SIBLING; |
m->next = MDOC_NEXT_SIBLING; |
return(1); |
return(1); |
} |
} |
|
|
/* |
if ( ! mdoc_word_alloc(m, line, offs, buf+offs)) |
* Warn if the last un-escaped character is whitespace. Then |
|
* strip away the remaining spaces (tabs stay!). |
|
*/ |
|
|
|
i = (int)strlen(buf); |
|
assert(i); |
|
|
|
if (' ' == buf[i - 1] || '\t' == buf[i - 1]) { |
|
if (i > 1 && '\\' != buf[i - 2]) |
|
if ( ! mdoc_pmsg(m, line, i - 1, MANDOCERR_EOLNSPACE)) |
|
return(0); |
|
|
|
for (--i; i && ' ' == buf[i]; i--) |
|
/* Spin back to non-space. */ ; |
|
|
|
/* Jump ahead of escaped whitespace. */ |
|
i += '\\' == buf[i] ? 2 : 1; |
|
|
|
buf[i] = '\0'; |
|
} |
|
|
|
/* Allocate the whole word. */ |
|
|
|
if ( ! mdoc_word_alloc(m, line, offs, buf + offs)) |
|
return(0); |
return(0); |
|
|
|
if (MDOC_LITERAL & m->flags) |
|
return(1); |
|
|
/* |
/* |
* End-of-sentence check. If the last character is an unescaped |
* End-of-sentence check. If the last character is an unescaped |
* EOS character, then flag the node as being the end of a |
* EOS character, then flag the node as being the end of a |
* sentence. The front-end will know how to interpret this. |
* sentence. The front-end will know how to interpret this. |
*/ |
*/ |
|
|
assert(i); |
assert(buf < end); |
if (mandoc_eos(buf, (size_t)i)) |
|
|
if (mandoc_eos(buf+offs, (size_t)(end-buf-offs), 0)) |
m->last->flags |= MDOC_EOS; |
m->last->flags |= MDOC_EOS; |
|
|
return(1); |
return(1); |
} |
} |
|
|
|
|
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. */ |
|
return(MDOC_IGN_MACRO & m->pflags ? rc : 0); |
|
} |
|
|
|
|
|
/* |
/* |
* Parse a macro line, that is, a line beginning with the control |
* Parse a macro line, that is, a line beginning with the control |
* character. |
* character. |
*/ |
*/ |
int |
static int |
mdoc_pmacro(struct mdoc *m, int ln, char *buf, int offs) |
mdoc_pmacro(struct mdoc *m, int ln, char *buf, int offs) |
{ |
{ |
enum mdoct tok; |
enum mdoct tok; |
int i, j, sv; |
int i, sv; |
char mac[5]; |
char mac[5]; |
|
struct mdoc_node *n; |
|
|
/* Empty lines are ignored. */ |
/* Empty post-control lines are ignored. */ |
|
|
offs++; |
if ('"' == buf[offs]) { |
|
mdoc_pmsg(m, ln, offs, MANDOCERR_BADCOMMENT); |
if ('\0' == buf[offs]) |
|
return(1); |
return(1); |
|
} else if ('\0' == buf[offs]) |
|
return(1); |
|
|
i = offs; |
sv = offs; |
|
|
/* Accept whitespace after the initial control char. */ |
/* |
|
* Copy the first word into a nil-terminated buffer. |
|
* Stop copying when a tab, space, or eoln is encountered. |
|
*/ |
|
|
if (' ' == buf[i]) { |
i = 0; |
i++; |
while (i < 4 && '\0' != buf[offs] && |
while (buf[i] && ' ' == buf[i]) |
' ' != buf[offs] && '\t' != buf[offs]) |
i++; |
mac[i++] = buf[offs++]; |
if ('\0' == buf[i]) |
|
return(1); |
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); |
} |
} |
|
|
sv = i; |
/* Disregard the first trailing tab, if applicable. */ |
|
|
/* Copy the first word into a nil-terminated buffer. */ |
if ('\t' == buf[offs]) |
|
offs++; |
|
|
for (j = 0; j < 4; j++, i++) { |
/* Jump to the next non-whitespace word. */ |
if ('\0' == (mac[j] = buf[i])) |
|
break; |
|
else if (' ' == buf[i]) |
|
break; |
|
|
|
/* Check for invalid characters. */ |
while (buf[offs] && ' ' == buf[offs]) |
|
offs++; |
|
|
if (isgraph((u_char)buf[i])) |
/* |
continue; |
* Trailing whitespace. Note that tabs are allowed to be passed |
if ( ! mdoc_pmsg(m, ln, i, MANDOCERR_BADCHAR)) |
* into the parser as "text", so we only warn about spaces here. |
return(0); |
*/ |
i--; |
|
} |
|
|
|
mac[j] = '\0'; |
if ('\0' == buf[offs] && ' ' == buf[offs - 1]) |
|
mdoc_pmsg(m, ln, offs - 1, MANDOCERR_EOLNSPACE); |
|
|
if (j == 4 || j < 2) { |
/* |
if ( ! macrowarn(m, ln, mac, sv)) |
* 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; |
goto err; |
return(1); |
return(1); |
} |
|
|
|
if (MDOC_MAX == (tok = mdoc_hash_find(mac))) { |
|
if ( ! macrowarn(m, ln, mac, sv)) |
|
goto err; |
|
return(1); |
|
} |
} |
|
|
/* The macro is sane. Jump to the next word. */ |
n = m->last; |
|
assert(m->last); |
|
|
while (buf[i] && ' ' == buf[i]) |
/* |
i++; |
* If the first macro of a `Bl -column', open an `It' block |
|
* context around the parsed macro. |
/* |
|
* 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[i] && ' ' == buf[i - 1]) |
if (MDOC_Bl == n->tok && MDOC_BODY == n->type && |
if ( ! mdoc_pmsg(m, ln, i - 1, MANDOCERR_EOLNSPACE)) |
LIST_column == n->norm->Bl.type) { |
|
m->flags |= MDOC_FREECOL; |
|
if ( ! mdoc_macro(m, MDOC_It, ln, sv, &sv, buf)) |
goto err; |
goto err; |
|
return(1); |
|
} |
|
|
/* |
/* |
* Begin recursive parse sequence. Since we're at the start of |
* If we're following a block-level `It' within a `Bl -column' |
* the line, we don't need to do callable/parseable checks. |
* context (perhaps opened in the above block or in ptext()), |
|
* then open an `It' block context around the parsed macro. |
*/ |
*/ |
if ( ! mdoc_macro(m, tok, ln, sv, &i, 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) { |
|
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; |
goto err; |
|
|
return(1); |
return(1); |
Line 729 err: /* Error out. */ |
|
Line 941 err: /* Error out. */ |
|
return(0); |
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); |
|
} |