freebsd-nq/usr.bin/sed/compile.c

875 lines
20 KiB
C
Raw Normal View History

1994-05-27 12:33:43 +00:00
/*-
* Copyright (c) 1992 Diomidis Spinellis.
* Copyright (c) 1992, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* Diomidis Spinellis of Imperial College, University of London.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the University of
* California, Berkeley and its contributors.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
1994-05-27 12:33:43 +00:00
#ifndef lint
static const char sccsid[] = "@(#)compile.c 8.1 (Berkeley) 6/6/93";
#endif
1994-05-27 12:33:43 +00:00
#include <sys/types.h>
#include <sys/stat.h>
#include <ctype.h>
#include <err.h>
1994-05-27 12:33:43 +00:00
#include <fcntl.h>
#include <limits.h>
#include <regex.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "defs.h"
#include "extern.h"
#define LHSZ 128
#define LHMASK (LHSZ - 1)
static struct labhash {
struct labhash *lh_next;
u_int lh_hash;
struct s_command *lh_cmd;
int lh_ref;
} *labels[LHSZ];
2002-03-22 01:42:45 +00:00
static char *compile_addr(char *, struct s_addr *);
static char *compile_ccl(char **, char *);
static char *compile_delimited(char *, char *);
static char *compile_flags(char *, struct s_subst *);
static char *compile_re(char *, regex_t **);
static char *compile_subst(char *, struct s_subst *);
static char *compile_text(void);
static char *compile_tr(char *, char **);
1994-05-27 12:33:43 +00:00
static struct s_command
2002-03-22 01:42:45 +00:00
**compile_stream(struct s_command **);
static char *duptoeol(char *, const char *);
static void enterlabel(struct s_command *);
1994-05-27 12:33:43 +00:00
static struct s_command
2002-03-22 01:42:45 +00:00
*findlabel(char *);
static void fixuplabel(struct s_command *, struct s_command *);
static void uselabel(void);
1994-05-27 12:33:43 +00:00
/*
* Command specification. This is used to drive the command parser.
*/
struct s_format {
char code; /* Command code */
int naddr; /* Number of address args */
enum e_args args; /* Argument type */
};
static struct s_format cmd_fmts[] = {
{'{', 2, GROUP},
{'}', 0, ENDGROUP},
1994-05-27 12:33:43 +00:00
{'a', 1, TEXT},
{'b', 2, BRANCH},
{'c', 2, TEXT},
{'d', 2, EMPTY},
{'D', 2, EMPTY},
{'g', 2, EMPTY},
{'G', 2, EMPTY},
{'h', 2, EMPTY},
{'H', 2, EMPTY},
{'i', 1, TEXT},
{'l', 2, EMPTY},
{'n', 2, EMPTY},
{'N', 2, EMPTY},
{'p', 2, EMPTY},
{'P', 2, EMPTY},
{'q', 1, EMPTY},
{'r', 1, RFILE},
{'s', 2, SUBST},
{'t', 2, BRANCH},
{'w', 2, WFILE},
{'x', 2, EMPTY},
{'y', 2, TR},
{'!', 2, NONSEL},
{':', 0, LABEL},
{'#', 0, COMMENT},
{'=', 1, EMPTY},
{'\0', 0, COMMENT},
};
/* The compiled program. */
struct s_command *prog;
/*
* Compile the program into prog.
* Initialise appends.
*/
void
compile()
{
*compile_stream(&prog) = NULL;
1994-05-27 12:33:43 +00:00
fixuplabel(prog, NULL);
uselabel();
if (appendnum == 0)
appends = NULL;
else if ((appends = malloc(sizeof(struct s_appends) * appendnum)) ==
NULL)
err(1, "malloc");
if ((match = malloc((maxnsub + 1) * sizeof(regmatch_t))) == NULL)
err(1, "malloc");
1994-05-27 12:33:43 +00:00
}
#define EATSPACE() do { \
if (p) \
1996-08-11 17:46:35 +00:00
while (*p && isspace((unsigned char)*p)) \
1994-05-27 12:33:43 +00:00
p++; \
} while (0)
static struct s_command **
compile_stream(link)
1994-05-27 12:33:43 +00:00
struct s_command **link;
{
char *p;
1994-05-27 12:33:43 +00:00
static char lbuf[_POSIX2_LINE_MAX + 1]; /* To save stack */
struct s_command *cmd, *cmd2, *stack;
1994-05-27 12:33:43 +00:00
struct s_format *fp;
int naddr; /* Number of addresses */
stack = 0;
1994-05-27 12:33:43 +00:00
for (;;) {
if ((p = cu_fgets(lbuf, sizeof(lbuf), NULL)) == NULL) {
if (stack != 0)
errx(1, "%lu: %s: unexpected EOF (pending }'s)",
linenum, fname);
1994-05-27 12:33:43 +00:00
return (link);
}
semicolon: EATSPACE();
if (p) {
if (*p == '#' || *p == '\0')
continue;
else if (*p == ';') {
p++;
goto semicolon;
}
}
if ((*link = cmd = malloc(sizeof(struct s_command))) == NULL)
err(1, "malloc");
1994-05-27 12:33:43 +00:00
link = &cmd->next;
cmd->nonsel = cmd->inrange = 0;
/* First parse the addresses */
naddr = 0;
/* Valid characters to start an address */
#define addrchar(c) (strchr("0123456789/\\$", (c)))
if (addrchar(*p)) {
naddr++;
if ((cmd->a1 = malloc(sizeof(struct s_addr))) == NULL)
err(1, "malloc");
1994-05-27 12:33:43 +00:00
p = compile_addr(p, cmd->a1);
EATSPACE(); /* EXTENSION */
if (*p == ',') {
p++;
EATSPACE(); /* EXTENSION */
naddr++;
if ((cmd->a2 = malloc(sizeof(struct s_addr)))
== NULL)
err(1, "malloc");
1994-05-27 12:33:43 +00:00
p = compile_addr(p, cmd->a2);
EATSPACE();
} else
cmd->a2 = 0;
} else
cmd->a1 = cmd->a2 = 0;
1994-05-27 12:33:43 +00:00
nonsel: /* Now parse the command */
if (!*p)
errx(1, "%lu: %s: command expected", linenum, fname);
1994-05-27 12:33:43 +00:00
cmd->code = *p;
for (fp = cmd_fmts; fp->code; fp++)
if (fp->code == *p)
break;
if (!fp->code)
errx(1, "%lu: %s: invalid command code %c", linenum, fname, *p);
1994-05-27 12:33:43 +00:00
if (naddr > fp->naddr)
errx(1,
"%lu: %s: command %c expects up to %d address(es), found %d",
linenum, fname, *p, fp->naddr, naddr);
1994-05-27 12:33:43 +00:00
switch (fp->args) {
case NONSEL: /* ! */
p++;
EATSPACE();
cmd->nonsel = ! cmd->nonsel;
1994-05-27 12:33:43 +00:00
goto nonsel;
case GROUP: /* { */
p++;
EATSPACE();
cmd->next = stack;
stack = cmd;
link = &cmd->u.c;
if (*p)
goto semicolon;
1994-05-27 12:33:43 +00:00
break;
case ENDGROUP:
/*
* Short-circuit command processing, since end of
* group is really just a noop.
*/
cmd->nonsel = 1;
if (stack == 0)
errx(1, "%lu: %s: unexpected }", linenum, fname);
cmd2 = stack;
stack = cmd2->next;
cmd2->next = cmd;
/*FALLTHROUGH*/
1994-05-27 12:33:43 +00:00
case EMPTY: /* d D g G h H l n N p P q x = \0 */
p++;
EATSPACE();
if (*p == ';') {
p++;
link = &cmd->next;
goto semicolon;
}
if (*p)
errx(1, "%lu: %s: extra characters at the end of %c command",
linenum, fname, cmd->code);
1994-05-27 12:33:43 +00:00
break;
case TEXT: /* a c i */
p++;
EATSPACE();
if (*p != '\\')
errx(1,
"%lu: %s: command %c expects \\ followed by text", linenum, fname, cmd->code);
1994-05-27 12:33:43 +00:00
p++;
EATSPACE();
if (*p)
errx(1,
"%lu: %s: extra characters after \\ at the end of %c command",
linenum, fname, cmd->code);
1994-05-27 12:33:43 +00:00
cmd->t = compile_text();
break;
case COMMENT: /* \0 # */
break;
case WFILE: /* w */
p++;
EATSPACE();
if (*p == '\0')
errx(1, "%lu: %s: filename expected", linenum, fname);
1994-05-27 12:33:43 +00:00
cmd->t = duptoeol(p, "w command");
if (aflag)
cmd->u.fd = -1;
else if ((cmd->u.fd = open(p,
1994-05-27 12:33:43 +00:00
O_WRONLY|O_APPEND|O_CREAT|O_TRUNC,
DEFFILEMODE)) == -1)
err(1, "%s", p);
1994-05-27 12:33:43 +00:00
break;
case RFILE: /* r */
p++;
EATSPACE();
if (*p == '\0')
errx(1, "%lu: %s: filename expected", linenum, fname);
1994-05-27 12:33:43 +00:00
else
cmd->t = duptoeol(p, "read command");
break;
case BRANCH: /* b t */
p++;
EATSPACE();
if (*p == '\0')
cmd->t = NULL;
else
cmd->t = duptoeol(p, "branch");
break;
case LABEL: /* : */
p++;
EATSPACE();
cmd->t = duptoeol(p, "label");
if (strlen(p) == 0)
errx(1, "%lu: %s: empty label", linenum, fname);
1994-05-27 12:33:43 +00:00
enterlabel(cmd);
break;
case SUBST: /* s */
p++;
if (*p == '\0' || *p == '\\')
errx(1,
"%lu: %s: substitute pattern can not be delimited by newline or backslash",
linenum, fname);
if ((cmd->u.s = malloc(sizeof(struct s_subst))) == NULL)
err(1, "malloc");
1994-05-27 12:33:43 +00:00
p = compile_re(p, &cmd->u.s->re);
if (p == NULL)
errx(1,
"%lu: %s: unterminated substitute pattern", linenum, fname);
1994-05-27 12:33:43 +00:00
--p;
p = compile_subst(p, cmd->u.s);
p = compile_flags(p, cmd->u.s);
EATSPACE();
if (*p == ';') {
p++;
link = &cmd->next;
goto semicolon;
}
break;
case TR: /* y */
p++;
p = compile_tr(p, (char **)&cmd->u.y);
EATSPACE();
if (*p == ';') {
p++;
link = &cmd->next;
goto semicolon;
}
if (*p)
errx(1,
"%lu: %s: extra text at the end of a transform command", linenum, fname);
1994-05-27 12:33:43 +00:00
break;
}
}
}
/*
* Get a delimited string. P points to the delimeter of the string; d points
* to a buffer area. Newline and delimiter escapes are processed; other
* escapes are ignored.
*
* Returns a pointer to the first character after the final delimiter or NULL
* in the case of a non-terminated string. The character array d is filled
* with the processed string.
*/
static char *
compile_delimited(p, d)
char *p, *d;
{
char c;
c = *p++;
if (c == '\0')
return (NULL);
else if (c == '\\')
errx(1, "%lu: %s: \\ can not be used as a string delimiter",
linenum, fname);
1994-05-27 12:33:43 +00:00
else if (c == '\n')
errx(1, "%lu: %s: newline can not be used as a string delimiter",
linenum, fname);
1994-05-27 12:33:43 +00:00
while (*p) {
if (*p == '[') {
if ((d = compile_ccl(&p, d)) == NULL)
errx(1, "%lu: %s: unbalanced brackets ([])", linenum, fname);
continue;
} else if (*p == '\\' && p[1] == '[') {
*d++ = *p++;
} else if (*p == '\\' && p[1] == c)
1994-05-27 12:33:43 +00:00
p++;
else if (*p == '\\' && p[1] == 'n') {
*d++ = '\n';
p += 2;
continue;
} else if (*p == '\\' && p[1] == '\\')
*d++ = *p++;
else if (*p == c) {
*d = '\0';
return (p + 1);
}
*d++ = *p++;
}
return (NULL);
}
/* compile_ccl: expand a POSIX character class */
static char *
compile_ccl(sp, t)
char **sp;
char *t;
{
int c, d;
char *s = *sp;
*t++ = *s++;
if (*s == '^')
*t++ = *s++;
if (*s == ']')
*t++ = *s++;
for (; *s && (*t = *s) != ']'; s++, t++)
if (*s == '[' && ((d = *(s+1)) == '.' || d == ':' || d == '=')) {
*++t = *++s, t++, s++;
for (c = *s; (*t = *s) != ']' || c != d; s++, t++)
if ((c = *s) == '\0')
return NULL;
} else if (*s == '\\' && s[1] == 'n')
*t = '\n', s++;
return (*s == ']') ? *sp = ++s, ++t : NULL;
}
1994-05-27 12:33:43 +00:00
/*
* Get a regular expression. P points to the delimiter of the regular
* expression; repp points to the address of a regexp pointer. Newline
* and delimiter escapes are processed; other escapes are ignored.
* Returns a pointer to the first character after the final delimiter
* or NULL in the case of a non terminated regular expression. The regexp
* pointer is set to the compiled regular expression.
* Cflags are passed to regcomp.
*/
static char *
compile_re(p, repp)
char *p;
regex_t **repp;
{
int eval;
char re[_POSIX2_LINE_MAX + 1];
p = compile_delimited(p, re);
if (p && strlen(re) == 0) {
*repp = NULL;
return (p);
}
if ((*repp = malloc(sizeof(regex_t))) == NULL)
err(1, "malloc");
if (p && (eval = regcomp(*repp, re, rflags)) != 0)
errx(1, "%lu: %s: RE error: %s",
linenum, fname, strregerror(eval, *repp));
1994-05-27 12:33:43 +00:00
if (maxnsub < (*repp)->re_nsub)
maxnsub = (*repp)->re_nsub;
return (p);
}
/*
* Compile the substitution string of a regular expression and set res to
* point to a saved copy of it. Nsub is the number of parenthesized regular
* expressions.
*/
static char *
compile_subst(p, s)
char *p;
struct s_subst *s;
{
static char lbuf[_POSIX2_LINE_MAX + 1];
int asize, size;
u_char ref;
1994-05-27 12:33:43 +00:00
char c, *text, *op, *sp;
int more = 1, sawesc = 0;
1994-05-27 12:33:43 +00:00
c = *p++; /* Terminator character */
if (c == '\0')
return (NULL);
s->maxbref = 0;
s->linenum = linenum;
asize = 2 * _POSIX2_LINE_MAX + 1;
if ((text = malloc(asize)) == NULL)
err(1, "malloc");
1994-05-27 12:33:43 +00:00
size = 0;
do {
op = sp = text + size;
for (; *p; p++) {
if (*p == '\\' || sawesc) {
/*
* If this is a continuation from the last
* buffer, we won't have a character to
* skip over.
*/
if (sawesc)
sawesc = 0;
else
p++;
if (*p == '\0') {
/*
* This escaped character is continued
* in the next part of the line. Note
* this fact, then cause the loop to
* exit w/ normal EOL case and reenter
* above with the new buffer.
*/
sawesc = 1;
p--;
continue;
} else if (strchr("123456789", *p) != NULL) {
1994-05-27 12:33:43 +00:00
*sp++ = '\\';
ref = *p - '0';
if (s->re != NULL &&
ref > s->re->re_nsub)
errx(1, "%lu: %s: \\%c not defined in the RE",
linenum, fname, *p);
1994-05-27 12:33:43 +00:00
if (s->maxbref < ref)
s->maxbref = ref;
} else if (*p == '&' || *p == '\\')
*sp++ = '\\';
} else if (*p == c) {
if (*++p == '\0' && more) {
if (cu_fgets(lbuf, sizeof(lbuf), &more))
p = lbuf;
}
1994-05-27 12:33:43 +00:00
*sp++ = '\0';
size += sp - op;
if ((s->new = realloc(text, size)) == NULL)
err(1, "realloc");
1994-05-27 12:33:43 +00:00
return (p);
} else if (*p == '\n') {
errx(1,
"%lu: %s: unescaped newline inside substitute pattern", linenum, fname);
1994-05-27 12:33:43 +00:00
/* NOTREACHED */
}
*sp++ = *p;
}
size += sp - op;
if (asize - size < _POSIX2_LINE_MAX + 1) {
asize *= 2;
if ((text = realloc(text, asize)) == NULL)
err(1, "realloc");
1994-05-27 12:33:43 +00:00
}
} while (cu_fgets(p = lbuf, sizeof(lbuf), &more));
errx(1, "%lu: %s: unterminated substitute in regular expression",
linenum, fname);
1994-05-27 12:33:43 +00:00
/* NOTREACHED */
}
/*
* Compile the flags of the s command
*/
static char *
compile_flags(p, s)
char *p;
struct s_subst *s;
{
int gn; /* True if we have seen g or n */
char wfile[_POSIX2_LINE_MAX + 1], *q;
s->n = 1; /* Default */
s->p = 0;
s->wfile = NULL;
s->wfd = -1;
for (gn = 0;;) {
EATSPACE(); /* EXTENSION */
switch (*p) {
case 'g':
if (gn)
errx(1,
"%lu: %s: more than one number or 'g' in substitute flags", linenum, fname);
1994-05-27 12:33:43 +00:00
gn = 1;
s->n = 0;
break;
case '\0':
case '\n':
case ';':
return (p);
case 'p':
s->p = 1;
break;
case '1': case '2': case '3':
case '4': case '5': case '6':
case '7': case '8': case '9':
if (gn)
errx(1,
"%lu: %s: more than one number or 'g' in substitute flags", linenum, fname);
1994-05-27 12:33:43 +00:00
gn = 1;
/* XXX Check for overflow */
s->n = (int)strtol(p, &p, 10);
break;
case 'w':
p++;
#ifdef HISTORIC_PRACTICE
if (*p != ' ') {
warnx("%lu: %s: space missing before w wfile", linenum, fname);
1994-05-27 12:33:43 +00:00
return (p);
}
#endif
EATSPACE();
q = wfile;
while (*p) {
if (*p == '\n')
break;
*q++ = *p++;
}
*q = '\0';
if (q == wfile)
errx(1, "%lu: %s: no wfile specified", linenum, fname);
1994-05-27 12:33:43 +00:00
s->wfile = strdup(wfile);
if (!aflag && (s->wfd = open(wfile,
O_WRONLY|O_APPEND|O_CREAT|O_TRUNC,
DEFFILEMODE)) == -1)
err(1, "%s", wfile);
1994-05-27 12:33:43 +00:00
return (p);
default:
errx(1, "%lu: %s: bad flag in substitute command: '%c'",
linenum, fname, *p);
1994-05-27 12:33:43 +00:00
break;
}
p++;
}
}
/*
* Compile a translation set of strings into a lookup table.
*/
static char *
compile_tr(p, transtab)
char *p;
char **transtab;
{
int i;
char *lt, *op, *np;
char old[_POSIX2_LINE_MAX + 1];
char new[_POSIX2_LINE_MAX + 1];
if (*p == '\0' || *p == '\\')
errx(1,
"%lu: %s: transform pattern can not be delimited by newline or backslash",
linenum, fname);
1994-05-27 12:33:43 +00:00
p = compile_delimited(p, old);
if (p == NULL)
errx(1, "%lu: %s: unterminated transform source string",
linenum, fname);
1994-05-27 12:33:43 +00:00
p = compile_delimited(--p, new);
if (p == NULL)
errx(1, "%lu: %s: unterminated transform target string",
linenum, fname);
1994-05-27 12:33:43 +00:00
EATSPACE();
if (strlen(new) != strlen(old))
errx(1, "%lu: %s: transform strings are not the same length",
linenum, fname);
1994-05-27 12:33:43 +00:00
/* We assume characters are 8 bits */
if ((lt = malloc(UCHAR_MAX)) == NULL)
err(1, "malloc");
1994-05-27 12:33:43 +00:00
for (i = 0; i <= UCHAR_MAX; i++)
lt[i] = (char)i;
for (op = old, np = new; *op; op++, np++)
lt[(u_char)*op] = *np;
*transtab = lt;
return (p);
}
/*
* Compile the text following an a or i command.
*/
static char *
compile_text()
{
int asize, esc_nl, size;
1994-05-27 12:33:43 +00:00
char *text, *p, *op, *s;
char lbuf[_POSIX2_LINE_MAX + 1];
asize = 2 * _POSIX2_LINE_MAX + 1;
if ((text = malloc(asize)) == NULL)
err(1, "malloc");
1994-05-27 12:33:43 +00:00
size = 0;
while (cu_fgets(lbuf, sizeof(lbuf), NULL)) {
1994-05-27 12:33:43 +00:00
op = s = text + size;
p = lbuf;
EATSPACE();
for (esc_nl = 0; *p != '\0'; p++) {
if (*p == '\\' && p[1] != '\0' && *++p == '\n')
esc_nl = 1;
1994-05-27 12:33:43 +00:00
*s++ = *p;
}
size += s - op;
if (!esc_nl) {
1994-05-27 12:33:43 +00:00
*s = '\0';
break;
}
if (asize - size < _POSIX2_LINE_MAX + 1) {
asize *= 2;
if ((text = realloc(text, asize)) == NULL)
err(1, "realloc");
1994-05-27 12:33:43 +00:00
}
}
text[size] = '\0';
if ((p = realloc(text, size + 1)) == NULL)
err(1, "realloc");
return (p);
1994-05-27 12:33:43 +00:00
}
/*
* Get an address and return a pointer to the first character after
* it. Fill the structure pointed to according to the address.
*/
static char *
compile_addr(p, a)
char *p;
struct s_addr *a;
{
char *end;
switch (*p) {
case '\\': /* Context address */
++p;
/* FALLTHROUGH */
case '/': /* Context address */
p = compile_re(p, &a->u.r);
if (p == NULL)
errx(1, "%lu: %s: unterminated regular expression", linenum, fname);
1994-05-27 12:33:43 +00:00
a->type = AT_RE;
return (p);
case '$': /* Last line */
a->type = AT_LAST;
return (p + 1);
/* Line number */
1995-05-30 06:41:30 +00:00
case '0': case '1': case '2': case '3': case '4':
1994-05-27 12:33:43 +00:00
case '5': case '6': case '7': case '8': case '9':
a->type = AT_LINE;
a->u.l = strtol(p, &end, 10);
return (end);
default:
errx(1, "%lu: %s: expected context address", linenum, fname);
1994-05-27 12:33:43 +00:00
return (NULL);
}
}
/*
* duptoeol --
* Return a copy of all the characters up to \n or \0.
*/
static char *
duptoeol(s, ctype)
char *s;
const char *ctype;
1994-05-27 12:33:43 +00:00
{
size_t len;
int ws;
char *p, *start;
1994-05-27 12:33:43 +00:00
ws = 0;
for (start = s; *s != '\0' && *s != '\n'; ++s)
1996-08-11 17:46:35 +00:00
ws = isspace((unsigned char)*s);
1994-05-27 12:33:43 +00:00
*s = '\0';
if (ws)
warnx("%lu: %s: whitespace after %s", linenum, fname, ctype);
1994-05-27 12:33:43 +00:00
len = s - start + 1;
if ((p = malloc(len)) == NULL)
err(1, "malloc");
return (memmove(p, start, len));
1994-05-27 12:33:43 +00:00
}
/*
* Convert goto label names to addresses, and count a and r commands, in
* the given subset of the script. Free the memory used by labels in b
* and t commands (but not by :).
*
* TODO: Remove } nodes
*/
static void
fixuplabel(cp, end)
struct s_command *cp, *end;
{
for (; cp != end; cp = cp->next)
switch (cp->code) {
case 'a':
case 'r':
appendnum++;
break;
case 'b':
case 't':
/* Resolve branch target. */
if (cp->t == NULL) {
cp->u.c = NULL;
break;
}
if ((cp->u.c = findlabel(cp->t)) == NULL)
errx(1, "%lu: %s: undefined label '%s'", linenum, fname, cp->t);
1994-05-27 12:33:43 +00:00
free(cp->t);
break;
case '{':
/* Do interior commands. */
fixuplabel(cp->u.c, cp->next);
break;
}
}
/*
* Associate the given command label for later lookup.
*/
static void
enterlabel(cp)
struct s_command *cp;
{
struct labhash **lhp, *lh;
u_char *p;
u_int h, c;
1994-05-27 12:33:43 +00:00
for (h = 0, p = (u_char *)cp->t; (c = *p) != 0; p++)
h = (h << 5) + h + c;
lhp = &labels[h & LHMASK];
for (lh = *lhp; lh != NULL; lh = lh->lh_next)
if (lh->lh_hash == h && strcmp(cp->t, lh->lh_cmd->t) == 0)
errx(1, "%lu: %s: duplicate label '%s'", linenum, fname, cp->t);
if ((lh = malloc(sizeof *lh)) == NULL)
err(1, "malloc");
1994-05-27 12:33:43 +00:00
lh->lh_next = *lhp;
lh->lh_hash = h;
lh->lh_cmd = cp;
lh->lh_ref = 0;
*lhp = lh;
}
/*
* Find the label contained in the command l in the command linked
* list cp. L is excluded from the search. Return NULL if not found.
*/
static struct s_command *
findlabel(name)
char *name;
{
struct labhash *lh;
u_char *p;
u_int h, c;
1994-05-27 12:33:43 +00:00
for (h = 0, p = (u_char *)name; (c = *p) != 0; p++)
h = (h << 5) + h + c;
for (lh = labels[h & LHMASK]; lh != NULL; lh = lh->lh_next) {
if (lh->lh_hash == h && strcmp(name, lh->lh_cmd->t) == 0) {
lh->lh_ref = 1;
return (lh->lh_cmd);
}
}
return (NULL);
}
1995-05-30 06:41:30 +00:00
/*
1994-05-27 12:33:43 +00:00
* Warn about any unused labels. As a side effect, release the label hash
* table space.
*/
static void
uselabel()
{
struct labhash *lh, *next;
int i;
1994-05-27 12:33:43 +00:00
for (i = 0; i < LHSZ; i++) {
for (lh = labels[i]; lh != NULL; lh = next) {
next = lh->lh_next;
if (!lh->lh_ref)
warnx("%lu: %s: unused label '%s'",
linenum, fname, lh->lh_cmd->t);
1994-05-27 12:33:43 +00:00
free(lh);
}
}
}