-/* $Id: mandoc.c,v 1.101 2017/06/11 19:37:01 schwarze Exp $ */
+/* $Id: mandoc.c,v 1.117 2020/01/19 16:44:50 schwarze Exp $ */
/*
* Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
- * Copyright (c) 2011-2015, 2017 Ingo Schwarze <schwarze@openbsd.org>
+ * Copyright (c) 2011-2015, 2017-2020 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
#include "mandoc.h"
#include "roff.h"
#include "libmandoc.h"
+#include "roff_int.h"
static int a2time(time_t *, const char *, const char *);
static char *time2a(time_t);
+enum mandoc_esc
+mandoc_font(const char *cp, int sz)
+{
+ switch (sz) {
+ case 0:
+ return ESCAPE_FONTPREV;
+ case 1:
+ switch (cp[0]) {
+ case 'B':
+ case '3':
+ return ESCAPE_FONTBOLD;
+ case 'I':
+ case '2':
+ return ESCAPE_FONTITALIC;
+ case 'P':
+ return ESCAPE_FONTPREV;
+ case 'R':
+ case '1':
+ return ESCAPE_FONTROMAN;
+ case '4':
+ return ESCAPE_FONTBI;
+ default:
+ return ESCAPE_ERROR;
+ }
+ case 2:
+ switch (cp[0]) {
+ case 'B':
+ switch (cp[1]) {
+ case 'I':
+ return ESCAPE_FONTBI;
+ default:
+ return ESCAPE_ERROR;
+ }
+ case 'C':
+ switch (cp[1]) {
+ case 'B':
+ return ESCAPE_FONTBOLD;
+ case 'I':
+ return ESCAPE_FONTITALIC;
+ case 'R':
+ case 'W':
+ return ESCAPE_FONTCW;
+ default:
+ return ESCAPE_ERROR;
+ }
+ default:
+ return ESCAPE_ERROR;
+ }
+ default:
+ return ESCAPE_ERROR;
+ }
+}
+
enum mandoc_esc
mandoc_escape(const char **end, const char **start, int *sz)
{
const char *local_start;
- int local_sz;
+ int local_sz, c, i;
char term;
enum mandoc_esc gly;
if (NULL == sz)
sz = &local_sz;
+ /*
+ * Treat "\E" just like "\";
+ * it only makes a difference in copy mode.
+ */
+
+ if (**end == 'E')
+ ++*end;
+
/*
* Beyond the backslash, at least one input character
* is part of the escape sequence. With one exception
*sz = 2;
break;
case '[':
+ if (**start == ' ') {
+ ++*end;
+ return ESCAPE_ERROR;
+ }
gly = ESCAPE_SPECIAL;
term = ']';
break;
/*
* Escapes taking no arguments at all.
*/
- case 'd':
- case 'u':
+ case '!':
+ case '?':
+ return ESCAPE_UNSUPP;
+ case '%':
+ case '&':
+ case ')':
case ',':
case '/':
+ case '^':
+ case 'a':
+ case 'd':
+ case 'r':
+ case 't':
+ case 'u':
+ case '{':
+ case '|':
+ case '}':
return ESCAPE_IGNORE;
+ case 'c':
+ return ESCAPE_NOSPACE;
+ case 'p':
+ return ESCAPE_BREAK;
/*
* The \z escape is supposed to output the following
* 'X' is the trigger. These have opaque sub-strings.
*/
case 'F':
+ case 'f':
case 'g':
case 'k':
case 'M':
case 'm':
case 'n':
+ case 'O':
case 'V':
case 'Y':
- gly = ESCAPE_IGNORE;
- /* FALLTHROUGH */
- case 'f':
- if (ESCAPE_ERROR == gly)
- gly = ESCAPE_FONT;
+ gly = (*start)[-1] == 'f' ? ESCAPE_FONT : ESCAPE_IGNORE;
switch (**start) {
case '(':
+ if ((*start)[-1] == 'O')
+ gly = ESCAPE_ERROR;
*start = ++*end;
*sz = 2;
break;
case '[':
+ if ((*start)[-1] == 'O')
+ gly = (*start)[1] == '5' ?
+ ESCAPE_UNSUPP : ESCAPE_ERROR;
*start = ++*end;
term = ']';
break;
default:
+ if ((*start)[-1] == 'O') {
+ switch (**start) {
+ case '0':
+ gly = ESCAPE_UNSUPP;
+ break;
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ break;
+ default:
+ gly = ESCAPE_ERROR;
+ break;
+ }
+ }
*sz = 1;
break;
}
break;
+ case '*':
+ if (strncmp(*start, "(.T", 3) != 0)
+ abort();
+ gly = ESCAPE_DEVICE;
+ *start = ++*end;
+ *sz = 2;
+ break;
/*
* These escapes are of the form \X'Y', where 'X' is the trigger
break;
/*
- * Anything else is assumed to be a glyph.
- * In this case, pass back the character after the backslash.
+ * Several special characters can be encoded as
+ * one-byte escape sequences without using \[].
*/
- default:
+ case ' ':
+ case '\'':
+ case '-':
+ case '.':
+ case '0':
+ case ':':
+ case '_':
+ case '`':
+ case 'e':
+ case '~':
gly = ESCAPE_SPECIAL;
+ /* FALLTHROUGH */
+ default:
+ if (gly == ESCAPE_ERROR)
+ gly = ESCAPE_UNDEF;
*start = --*end;
*sz = 1;
break;
}
- assert(ESCAPE_ERROR != gly);
-
/*
* Read up to the terminating character,
* paying attention to nested escapes.
}
}
*sz = (*end)++ - *start;
+
+ /*
+ * The file chars.c only provides one common list
+ * of character names, but \[-] == \- is the only
+ * one of the characters with one-byte names that
+ * allows enclosing the name in brackets.
+ */
+ if (gly == ESCAPE_SPECIAL && *sz == 1 && **start != '-')
+ return ESCAPE_ERROR;
} else {
assert(*sz > 0);
if ((size_t)*sz > strlen(*start))
switch (gly) {
case ESCAPE_FONT:
- if (2 == *sz) {
- if ('C' == **start) {
- /*
- * Treat constant-width font modes
- * just like regular font modes.
- */
- (*start)++;
- (*sz)--;
- } else {
- if ('B' == (*start)[0] && 'I' == (*start)[1])
- gly = ESCAPE_FONTBI;
+ gly = mandoc_font(*start, *sz);
+ break;
+ case ESCAPE_SPECIAL:
+ if (**start == 'c') {
+ if (*sz < 6 || *sz > 7 ||
+ strncmp(*start, "char", 4) != 0 ||
+ (int)strspn(*start + 4, "0123456789") + 4 < *sz)
break;
- }
- } else if (1 != *sz)
- break;
-
- switch (**start) {
- case '3':
- case 'B':
- gly = ESCAPE_FONTBOLD;
- break;
- case '2':
- case 'I':
- gly = ESCAPE_FONTITALIC;
- break;
- case 'P':
- gly = ESCAPE_FONTPREV;
- break;
- case '1':
- case 'R':
- gly = ESCAPE_FONTROMAN;
+ c = 0;
+ for (i = 4; i < *sz; i++)
+ c = 10 * c + ((*start)[i] - '0');
+ if (c < 0x21 || (c > 0x7e && c < 0xa0) || c > 0xff)
+ break;
+ *start += 4;
+ *sz -= 4;
+ gly = ESCAPE_NUMBERED;
break;
}
- break;
- case ESCAPE_SPECIAL:
- if (1 == *sz && 'c' == **start)
- gly = ESCAPE_NOSPACE;
+
/*
* Unicode escapes are defined in groff as \[u0000]
* to \[u10FFFF], where the contained value must be
return gly;
}
-/*
- * Parse a quoted or unquoted roff-style request or macro argument.
- * Return a pointer to the parsed argument, which is either the original
- * pointer or advanced by one byte in case the argument is quoted.
- * NUL-terminate the argument in place.
- * Collapse pairs of quotes inside quoted arguments.
- * Advance the argument pointer to the next argument,
- * or to the NUL byte terminating the argument line.
- */
-char *
-mandoc_getarg(struct mparse *parse, char **cpp, int ln, int *pos)
-{
- char *start, *cp;
- int quoted, pairs, white;
-
- /* Quoting can only start with a new word. */
- start = *cpp;
- quoted = 0;
- if ('"' == *start) {
- quoted = 1;
- start++;
- }
-
- pairs = 0;
- white = 0;
- for (cp = start; '\0' != *cp; cp++) {
-
- /*
- * Move the following text left
- * after quoted quotes and after "\\" and "\t".
- */
- if (pairs)
- cp[-pairs] = cp[0];
-
- if ('\\' == cp[0]) {
- /*
- * In copy mode, translate double to single
- * backslashes and backslash-t to literal tabs.
- */
- switch (cp[1]) {
- case 't':
- cp[0] = '\t';
- /* FALLTHROUGH */
- case '\\':
- pairs++;
- cp++;
- break;
- case ' ':
- /* Skip escaped blanks. */
- if (0 == quoted)
- cp++;
- break;
- default:
- break;
- }
- } else if (0 == quoted) {
- if (' ' == cp[0]) {
- /* Unescaped blanks end unquoted args. */
- white = 1;
- break;
- }
- } else if ('"' == cp[0]) {
- if ('"' == cp[1]) {
- /* Quoted quotes collapse. */
- pairs++;
- cp++;
- } else {
- /* Unquoted quotes end quoted args. */
- quoted = 2;
- break;
- }
- }
- }
-
- /* Quoted argument without a closing quote. */
- if (1 == quoted)
- mandoc_msg(MANDOCERR_ARG_QUOTE, parse, ln, *pos, NULL);
-
- /* NUL-terminate this argument and move to the next one. */
- if (pairs)
- cp[-pairs] = '\0';
- if ('\0' != *cp) {
- *cp++ = '\0';
- while (' ' == *cp)
- cp++;
- }
- *pos += (int)(cp - start) + (quoted ? 1 : 0);
- *cpp = cp;
-
- if ('\0' == *cp && (white || ' ' == cp[-1]))
- mandoc_msg(MANDOCERR_SPACE_EOL, parse, ln, *pos, NULL);
-
- return start;
-}
-
static int
a2time(time_t *t, const char *fmt, const char *p)
{
size_t ssz;
int isz;
+ buf = NULL;
tm = localtime(&t);
if (tm == NULL)
- return NULL;
+ goto fail;
/*
* Reserve space:
* of looking at LC_TIME.
*/
- if ((isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday)) == -1)
+ isz = snprintf(p, 4 + 1, "%d, ", tm->tm_mday);
+ if (isz < 0 || isz > 4)
goto fail;
p += isz;
fail:
free(buf);
- return NULL;
+ return mandoc_strdup("");
}
char *
-mandoc_normdate(struct roff_man *man, char *in, int ln, int pos)
+mandoc_normdate(struct roff_node *nch, struct roff_node *nbl)
{
+ char *cp;
time_t t;
- /* No date specified: use today's date. */
+ /* No date specified. */
- if (in == NULL || *in == '\0' || strcmp(in, "$" "Mdocdate$") == 0) {
- mandoc_msg(MANDOCERR_DATE_MISSING, man->parse, ln, pos, NULL);
- return time2a(time(NULL));
+ if (nch == NULL) {
+ if (nbl == NULL)
+ mandoc_msg(MANDOCERR_DATE_MISSING, 0, 0, NULL);
+ else
+ mandoc_msg(MANDOCERR_DATE_MISSING, nbl->line,
+ nbl->pos, "%s", roff_name[nbl->tok]);
+ return mandoc_strdup("");
}
+ if (*nch->string == '\0') {
+ mandoc_msg(MANDOCERR_DATE_MISSING, nch->line,
+ nch->pos, "%s", roff_name[nbl->tok]);
+ return mandoc_strdup("");
+ }
+ if (strcmp(nch->string, "$" "Mdocdate$") == 0)
+ return time2a(time(NULL));
/* Valid mdoc(7) date format. */
- if (a2time(&t, "$" "Mdocdate: %b %d %Y $", in) ||
- a2time(&t, "%b %d, %Y", in))
- return time2a(t);
+ if (a2time(&t, "$" "Mdocdate: %b %d %Y $", nch->string) ||
+ a2time(&t, "%b %d, %Y", nch->string)) {
+ cp = time2a(t);
+ if (t > time(NULL) + 86400)
+ mandoc_msg(MANDOCERR_DATE_FUTURE, nch->line,
+ nch->pos, "%s %s", roff_name[nbl->tok], cp);
+ else if (*nch->string != '$' &&
+ strcmp(nch->string, cp) != 0)
+ mandoc_msg(MANDOCERR_DATE_NORM, nch->line,
+ nch->pos, "%s %s", roff_name[nbl->tok], cp);
+ return cp;
+ }
/* In man(7), do not warn about the legacy format. */
- if (a2time(&t, "%Y-%m-%d", in) == 0)
- mandoc_msg(MANDOCERR_DATE_BAD, man->parse, ln, pos, in);
- else if (man->macroset == MACROSET_MDOC)
- mandoc_vmsg(MANDOCERR_DATE_LEGACY, man->parse,
- ln, pos, "Dd %s", in);
+ if (a2time(&t, "%Y-%m-%d", nch->string) == 0)
+ mandoc_msg(MANDOCERR_DATE_BAD, nch->line, nch->pos,
+ "%s %s", roff_name[nbl->tok], nch->string);
+ else if (t > time(NULL) + 86400)
+ mandoc_msg(MANDOCERR_DATE_FUTURE, nch->line, nch->pos,
+ "%s %s", roff_name[nbl->tok], nch->string);
+ else if (nbl->tok == MDOC_Dd)
+ mandoc_msg(MANDOCERR_DATE_LEGACY, nch->line, nch->pos,
+ "Dd %s", nch->string);
/* Use any non-mdoc(7) date verbatim. */
- return mandoc_strdup(in);
+ return mandoc_strdup(nch->string);
}
int
git.ckatri.com / mandoc.git / blobdiff