a3e7fdfb5a
* Ensure we work within the array bounds when parsing command-line options; * Replace h0h0getopt with getopt(3); * Use consistent whitespace style in the function declarations. Revieweded by: dwmalone (mentor)
985 lines
23 KiB
C
985 lines
23 KiB
C
/*
|
|
* Copyright (c) 1985, 1993
|
|
* The Regents of the University of California. All rights reserved.
|
|
*
|
|
* This code is derived from software contributed to Berkeley by
|
|
* Dave Yost. Support for #if and #elif was added by Tony Finch.
|
|
*
|
|
* 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>
|
|
|
|
#ifndef lint
|
|
static const char copyright[] =
|
|
"@(#) Copyright (c) 1985, 1993\n\
|
|
The Regents of the University of California. All rights reserved.\n";
|
|
|
|
#ifdef __IDSTRING
|
|
__IDSTRING(Berkeley, "@(#)unifdef.c 8.1 (Berkeley) 6/6/93");
|
|
__IDSTRING(NetBSD, "$NetBSD: unifdef.c,v 1.8 2000/07/03 02:51:36 matt Exp $");
|
|
__IDSTRING(dotat, "$dotat: things/unifdef.c,v 1.73 2002/05/21 17:33:41 fanf Exp $");
|
|
#endif
|
|
#ifdef __FBSDID
|
|
__FBSDID("$FreeBSD$");
|
|
#endif
|
|
#endif
|
|
|
|
/*
|
|
* unifdef - remove ifdef'ed lines
|
|
*
|
|
* Warning: will not work correctly if input contains nul characters.
|
|
*
|
|
* Wishlist:
|
|
* provide an option which will append the name of the
|
|
* appropriate symbol after #else's and #endif's
|
|
* provide an option which will check symbols after
|
|
* #else's and #endif's to see that they match their
|
|
* corresponding #ifdef or #ifndef
|
|
* generate #line directives in place of deleted code
|
|
*/
|
|
|
|
#include <ctype.h>
|
|
#include <err.h>
|
|
#include <stdarg.h>
|
|
#include <stdbool.h>
|
|
#include <stdio.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <unistd.h>
|
|
|
|
/* types of input lines: */
|
|
typedef enum {
|
|
LT_PLAIN, /* ordinary line */
|
|
LT_TRUE, /* a true #if */
|
|
LT_FALSE, /* a false #if */
|
|
LT_ELTRUE, /* a true #elif */
|
|
LT_ELFALSE, /* a false #elif */
|
|
LT_IF, /* an unknown #if */
|
|
LT_ELIF, /* an unknown #elif */
|
|
LT_ELSE, /* #else */
|
|
LT_ENDIF, /* #endif */
|
|
LT_EOF /* end of file */
|
|
} Linetype;
|
|
|
|
typedef enum { /* 0 or 1: pass thru; 1 or 2: ignore comments */
|
|
REJ_NO,
|
|
REJ_IGNORE,
|
|
REJ_YES
|
|
} Reject_level;
|
|
|
|
typedef enum {
|
|
NO_COMMENT = false,
|
|
C_COMMENT,
|
|
CXX_COMMENT
|
|
} Comment_state;
|
|
|
|
typedef enum {
|
|
QUOTE_NONE = false,
|
|
QUOTE_SINGLE,
|
|
QUOTE_DOUBLE
|
|
} Quote_state;
|
|
|
|
const char *const errs[] = {
|
|
#define NO_ERR 0
|
|
"",
|
|
#define END_ERR 1
|
|
"",
|
|
#define ELIF_ERR 2
|
|
"Inappropriate elif",
|
|
#define ELSE_ERR 3
|
|
"Inappropriate else",
|
|
#define ENDIF_ERR 4
|
|
"Inappropriate endif",
|
|
#define IEOF_ERR 5
|
|
"Premature EOF in ifdef",
|
|
#define CEOF_ERR 6
|
|
"Premature EOF in comment",
|
|
#define Q1EOF_ERR 7
|
|
"Premature EOF in quoted character",
|
|
#define Q2EOF_ERR 8
|
|
"Premature EOF in quoted string"
|
|
};
|
|
|
|
/*
|
|
* These are the operators that are supported by the expression evaluator.
|
|
*/
|
|
static int op_lt(int a, int b) { return a < b; }
|
|
static int op_gt(int a, int b) { return a > b; }
|
|
static int op_le(int a, int b) { return a <= b; }
|
|
static int op_ge(int a, int b) { return a >= b; }
|
|
static int op_eq(int a, int b) { return a == b; }
|
|
static int op_ne(int a, int b) { return a != b; }
|
|
static int op_or(int a, int b) { return a || b; }
|
|
static int op_and(int a, int b) { return a && b; }
|
|
|
|
struct ops;
|
|
|
|
/*
|
|
* An evaluation function takes three arguments, as follows: (1) a pointer to
|
|
* an element of the precedence table which lists the operators at the current
|
|
* level of precedence; (2) a pointer to an integer which will receive the
|
|
* value of the expression; and (3) a pointer to a char* that points to the
|
|
* expression to be evaluated and that is updated to the end of the expression
|
|
* when evaluation is complete. The function returns LT_FALSE if the value of
|
|
* the expression is zero, LT_TRUE if it is non-zero, or LT_IF if the
|
|
* expression could not be evaluated.
|
|
*/
|
|
typedef Linetype eval_fn(struct ops *, int *, const char **);
|
|
|
|
eval_fn eval_table, eval_unary;
|
|
|
|
/*
|
|
* The precedence table. Expressions involving binary operators are evaluated
|
|
* in a table-driven way by eval_table. When it evaluates a subexpression it
|
|
* calls the inner function with its first argument pointing to the next
|
|
* element of the table. Innermost expressions have special non-table-driven
|
|
* handling.
|
|
*/
|
|
struct ops {
|
|
eval_fn *inner;
|
|
struct op {
|
|
const char *str;
|
|
int (*fn)(int, int);
|
|
} op[5];
|
|
} eval_ops[] = {
|
|
{ eval_table, { { "||", op_or } } },
|
|
{ eval_table, { { "&&", op_and } } },
|
|
{ eval_table, { { "==", op_eq },
|
|
{ "!=", op_ne } } },
|
|
{ eval_unary, { { "<=", op_le },
|
|
{ ">=", op_ge },
|
|
{ "<", op_lt },
|
|
{ ">", op_gt } } }
|
|
};
|
|
|
|
FILE *input;
|
|
const char *filename;
|
|
int linenum; /* current line number */
|
|
int stifline; /* start of current #if */
|
|
int stqcline; /* start of current coment or quote */
|
|
|
|
#define MAXLINE 1024
|
|
#define KWSIZE 8
|
|
/* tline has extra space so that it isn't overflowed when editing #elifs */
|
|
char tline[MAXLINE+KWSIZE]; /* input buffer */
|
|
char *keyword; /* used for editing #elif's */
|
|
|
|
bool complement; /* -c option in effect: do the complement */
|
|
bool debugging; /* -d option in effect: debugging reports */
|
|
bool lnblank; /* -l option in effect: blank deleted lines */
|
|
bool symlist; /* -s option in effect: output symbol list */
|
|
bool text; /* -t option in effect: this is a text file */
|
|
|
|
int exitstat; /* program exit status */
|
|
|
|
#define MAXSYMS 1000
|
|
const char *symname[MAXSYMS]; /* symbol name */
|
|
const char *value[MAXSYMS]; /* -Dsym=value */
|
|
bool ignore[MAXSYMS]; /* -iDsym or -iUsym */
|
|
|
|
int nsyms = 1; /* symbol 0 is used for tracking #ifs */
|
|
|
|
Reject_level reject; /* what kind of filtering we are doing */
|
|
Comment_state incomment; /* inside C comment */
|
|
Quote_state inquote; /* inside single or double quotes */
|
|
|
|
Linetype checkline(int *);
|
|
void debug(const char *, ...);
|
|
Linetype process(int);
|
|
void doif(int, Linetype, bool);
|
|
void elif2if(void);
|
|
void elif2endif(void);
|
|
void error(int, int);
|
|
void addsym(bool, bool, char *);
|
|
int findsym(const char *);
|
|
void flushline(bool);
|
|
int getline(char *, int, FILE *, bool);
|
|
Linetype ifeval(const char **);
|
|
int main(int, char **);
|
|
const char *skipcomment(const char *);
|
|
const char *skipquote(const char *, Quote_state);
|
|
const char *skipsym(const char *);
|
|
void usage(void);
|
|
|
|
#define endsym(c) (!isalpha((unsigned char)c) && !isdigit((unsigned char)c) && c != '_')
|
|
|
|
int
|
|
main(int argc, char *argv[])
|
|
{
|
|
int opt;
|
|
|
|
while ((opt = getopt(argc, argv, "i:D:U:I:cdlst")) != -1)
|
|
switch (opt) {
|
|
case 'i': /* treat stuff controlled by these symbols as text */
|
|
/*
|
|
* For strict backwards-compatibility the U or D
|
|
* should be immediately after the -i but it doesn't
|
|
* matter much if we relax that requirement.
|
|
*/
|
|
opt = *optarg++;
|
|
if (opt == 'D')
|
|
addsym(true, true, optarg);
|
|
else if (opt == 'U')
|
|
addsym(true, false, optarg);
|
|
else
|
|
usage();
|
|
break;
|
|
case 'D': /* define a symbol */
|
|
addsym(false, true, optarg);
|
|
break;
|
|
case 'U': /* undef a symbol */
|
|
addsym(false, false, optarg);
|
|
break;
|
|
case 'I':
|
|
/* ignore for compatibility with cpp */
|
|
break;
|
|
case 'c': /* treat -D as -U and vice versa */
|
|
complement = true;
|
|
break;
|
|
case 'd':
|
|
debugging = true;
|
|
break;
|
|
case 'l': /* blank deleted lines instead of omitting them */
|
|
lnblank = true;
|
|
break;
|
|
case 's': /* only output list of symbols that control #ifs */
|
|
symlist = true;
|
|
break;
|
|
case 't': /* don't parse C comments or strings */
|
|
text = true;
|
|
break;
|
|
default:
|
|
usage();
|
|
}
|
|
argc -= optind;
|
|
argv += optind;
|
|
if (nsyms == 1 && !symlist) {
|
|
warnx("must -D or -U at least one symbol");
|
|
usage();
|
|
}
|
|
if (argc > 1) {
|
|
errx(2, "can only do one file");
|
|
} else if (argc == 1 && strcmp(*argv, "-") != 0) {
|
|
filename = *argv;
|
|
if ((input = fopen(filename, "r")) != NULL) {
|
|
(void) process(0);
|
|
(void) fclose(input);
|
|
} else
|
|
err(2, "can't open %s", *argv);
|
|
} else {
|
|
filename = "[stdin]";
|
|
input = stdin;
|
|
(void) process(0);
|
|
}
|
|
|
|
exit(exitstat);
|
|
}
|
|
|
|
void
|
|
usage(void)
|
|
{
|
|
fprintf (stderr, "usage: %s",
|
|
"unifdef [-cdlst] [[-Dsym[=val]] [-Usym] [-iDsym[=val]] [-iUsym]] ... [file]\n");
|
|
exit (2);
|
|
}
|
|
|
|
/*
|
|
* This function processes #if lines and alters the pass-through
|
|
* state accordingly. All the complicated state transition suff is
|
|
* dealt with in this function, as well as checking that the
|
|
* #if/#elif/#else/#endif lines happen in the correct order. Lines
|
|
* between #if lines are handled by a recursive call to process().
|
|
*/
|
|
void
|
|
doif(int depth, Linetype lineval, bool ignoring)
|
|
{
|
|
Reject_level savereject;
|
|
bool active;
|
|
bool donetrue;
|
|
bool inelse;
|
|
int saveline;
|
|
|
|
debug("#if line %d code %d depth %d",
|
|
linenum, lineval, depth);
|
|
saveline = stifline;
|
|
stifline = linenum;
|
|
savereject = reject;
|
|
inelse = false;
|
|
donetrue = false;
|
|
if (lineval == LT_IF || reject != REJ_NO) {
|
|
active = false;
|
|
ignoring = false;
|
|
flushline(true);
|
|
} else if (ignoring) {
|
|
active = false;
|
|
flushline(true);
|
|
if (lineval == LT_FALSE)
|
|
reject = REJ_IGNORE;
|
|
else
|
|
donetrue = true;
|
|
} else {
|
|
active = true;
|
|
flushline(false);
|
|
if (lineval == LT_FALSE)
|
|
reject = REJ_YES;
|
|
else
|
|
donetrue = true;
|
|
}
|
|
debug("active %d ignore %d", active, ignoring);
|
|
for (;;) {
|
|
switch (lineval = process(depth)) {
|
|
case LT_ELIF:
|
|
debug("#elif start %d line %d code %d depth %d",
|
|
stifline, linenum, lineval, depth);
|
|
if (inelse)
|
|
error(ELIF_ERR, depth);
|
|
donetrue = false;
|
|
reject = savereject;
|
|
if (active) {
|
|
active = false;
|
|
elif2if();
|
|
flushline(true);
|
|
} else {
|
|
ignoring = false;
|
|
flushline(true);
|
|
}
|
|
debug("active %d ignore %d", active, ignoring);
|
|
break;
|
|
case LT_ELTRUE:
|
|
case LT_ELFALSE:
|
|
debug("#elif start %d line %d code %d depth %d",
|
|
stifline, linenum, lineval, depth);
|
|
if (inelse)
|
|
error(ELIF_ERR, depth);
|
|
if (active) {
|
|
flushline(false);
|
|
} else {
|
|
ignoring = false;
|
|
active = true;
|
|
elif2endif();
|
|
flushline(true);
|
|
}
|
|
if (lineval == LT_ELFALSE)
|
|
reject = REJ_YES;
|
|
else {
|
|
reject = REJ_NO;
|
|
donetrue = true;
|
|
}
|
|
debug("active %d ignore %d", active, ignoring);
|
|
break;
|
|
case LT_ELSE:
|
|
debug("#else start %d line %d code %d depth %d",
|
|
stifline, linenum, lineval, depth);
|
|
if (inelse)
|
|
error(ELSE_ERR, depth);
|
|
if (active) {
|
|
flushline(false);
|
|
if (reject == REJ_YES && !donetrue)
|
|
reject = REJ_NO;
|
|
else
|
|
reject = REJ_YES;
|
|
} else {
|
|
flushline(true);
|
|
if (ignoring) {
|
|
if (reject == REJ_IGNORE)
|
|
reject = REJ_NO;
|
|
else
|
|
reject = REJ_IGNORE;
|
|
}
|
|
}
|
|
inelse = true;
|
|
debug("active %d ignore %d", active, ignoring);
|
|
break;
|
|
case LT_ENDIF:
|
|
debug("#endif start %d line %d code %d depth %d",
|
|
stifline, linenum, lineval, depth);
|
|
if (active)
|
|
flushline(false);
|
|
else
|
|
flushline(true);
|
|
reject = savereject;
|
|
stifline = saveline;
|
|
return;
|
|
default:
|
|
/* bug */
|
|
abort();
|
|
}
|
|
}
|
|
}
|
|
|
|
/*
|
|
* The main file processing routine. This function deals with passing
|
|
* through normal non-#if lines, correct nesting of #if sections, and
|
|
* checking that things terminate correctly at the end of file. The
|
|
* complicated stuff is delegated to doif().
|
|
*/
|
|
Linetype
|
|
process(int depth)
|
|
{
|
|
Linetype lineval;
|
|
int cursym;
|
|
|
|
for (;;) {
|
|
linenum++;
|
|
if (getline(tline, MAXLINE, input, false) == EOF) {
|
|
if (incomment)
|
|
error(CEOF_ERR, depth);
|
|
if (inquote == QUOTE_SINGLE)
|
|
error(Q1EOF_ERR, depth);
|
|
if (inquote == QUOTE_DOUBLE)
|
|
error(Q2EOF_ERR, depth);
|
|
if (depth != 0)
|
|
error(IEOF_ERR, depth);
|
|
return LT_EOF;
|
|
}
|
|
switch (lineval = checkline(&cursym)) {
|
|
case LT_PLAIN:
|
|
flushline(true);
|
|
break;
|
|
case LT_IF:
|
|
case LT_TRUE:
|
|
case LT_FALSE:
|
|
doif(depth + 1, lineval, ignore[cursym]);
|
|
break;
|
|
case LT_ELIF:
|
|
case LT_ELTRUE:
|
|
case LT_ELFALSE:
|
|
case LT_ELSE:
|
|
case LT_ENDIF:
|
|
if (depth != 0)
|
|
return lineval;
|
|
if (lineval == LT_ENDIF)
|
|
error(ENDIF_ERR, depth);
|
|
if (lineval == LT_ELSE)
|
|
error(ELSE_ERR, depth);
|
|
error(ELIF_ERR, depth);
|
|
default:
|
|
/* bug */
|
|
abort();
|
|
}
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Parse a line and determine its type.
|
|
*/
|
|
Linetype
|
|
checkline(int *cursym)
|
|
{
|
|
const char *cp;
|
|
char *symp;
|
|
Linetype retval;
|
|
char kw[KWSIZE];
|
|
|
|
retval = LT_PLAIN;
|
|
cp = skipcomment(tline);
|
|
if (*cp != '#'
|
|
|| incomment
|
|
|| inquote == QUOTE_SINGLE
|
|
|| inquote == QUOTE_DOUBLE
|
|
)
|
|
goto eol;
|
|
|
|
cp = skipcomment(++cp);
|
|
keyword = (char *)cp;
|
|
symp = kw;
|
|
while (!endsym(*cp)) {
|
|
*symp = *cp++;
|
|
if (++symp >= &kw[KWSIZE])
|
|
goto eol;
|
|
}
|
|
*symp = '\0';
|
|
|
|
if (strcmp(kw, "ifdef") == 0) {
|
|
retval = LT_TRUE;
|
|
goto ifdef;
|
|
} else if (strcmp(kw, "ifndef") == 0) {
|
|
retval = LT_FALSE;
|
|
ifdef:
|
|
cp = skipcomment(++cp);
|
|
if (incomment) {
|
|
retval = LT_PLAIN;
|
|
goto eol;
|
|
}
|
|
if ((*cursym = findsym(cp)) == 0)
|
|
retval = LT_IF;
|
|
else if (value[*cursym] == NULL)
|
|
retval = (retval == LT_TRUE)
|
|
? LT_FALSE : LT_TRUE;
|
|
} else if (strcmp(kw, "if") == 0) {
|
|
retval = ifeval(&cp);
|
|
cp = skipcomment(cp);
|
|
if (*cp != '\n')
|
|
retval = LT_IF;
|
|
*cursym = 0;
|
|
} else if (strcmp(kw, "elif") == 0) {
|
|
retval = ifeval(&cp);
|
|
cp = skipcomment(cp);
|
|
if (retval == LT_IF)
|
|
retval = LT_ELIF;
|
|
if (retval == LT_TRUE)
|
|
retval = LT_ELTRUE;
|
|
if (retval == LT_FALSE)
|
|
retval = LT_ELFALSE;
|
|
if (*cp != '\n')
|
|
retval = LT_ELIF;
|
|
*cursym = 0;
|
|
} else if (strcmp(kw, "else") == 0)
|
|
retval = LT_ELSE;
|
|
else if (strcmp(kw, "endif") == 0)
|
|
retval = LT_ENDIF;
|
|
|
|
eol:
|
|
if (!text && reject != REJ_IGNORE)
|
|
for (; *cp;) {
|
|
if (incomment)
|
|
cp = skipcomment(cp);
|
|
else if (inquote == QUOTE_SINGLE)
|
|
cp = skipquote(cp, QUOTE_SINGLE);
|
|
else if (inquote == QUOTE_DOUBLE)
|
|
cp = skipquote(cp, QUOTE_DOUBLE);
|
|
else if (*cp == '/' && (cp[1] == '*' || cp[1] == '/'))
|
|
cp = skipcomment(cp);
|
|
else if (*cp == '\'')
|
|
cp = skipquote(cp, QUOTE_SINGLE);
|
|
else if (*cp == '"')
|
|
cp = skipquote(cp, QUOTE_DOUBLE);
|
|
else
|
|
cp++;
|
|
}
|
|
return retval;
|
|
}
|
|
|
|
/*
|
|
* Turn a #elif line into a #if. This function is used when we are
|
|
* processing a #if/#elif/#else/#endif sequence that starts off with a
|
|
* #if that we understand (and therefore it has been deleted) which is
|
|
* followed by a #elif that we don't understand and therefore must be
|
|
* kept. We turn it into a #if to keep the nesting correct.
|
|
*/
|
|
void
|
|
elif2if(void)
|
|
{
|
|
strncpy(keyword, "if ", 4);
|
|
}
|
|
|
|
/*
|
|
* Turn a #elif line into a #endif. This is used in the opposite
|
|
* situation to elif2if, i.e. a #if that we don't understand is
|
|
* followed by a #elif that we do; rather than deleting the #elif (as
|
|
* we would for a #if) we turn it into a #endif to keep the nesting
|
|
* correct.
|
|
*/
|
|
void
|
|
elif2endif(void)
|
|
{
|
|
strcpy(keyword, "endif\n");
|
|
}
|
|
|
|
/*
|
|
* Function for evaluating the innermost parts of expressions,
|
|
* viz. !expr (expr) defined(symbol) symbol number
|
|
*/
|
|
Linetype
|
|
eval_unary(struct ops *ops, int *valp, const char **cpp)
|
|
{
|
|
const char *cp;
|
|
char *ep;
|
|
int sym;
|
|
|
|
cp = skipcomment(*cpp);
|
|
if(*cp == '!') {
|
|
debug("eval%d !", ops - eval_ops);
|
|
cp++;
|
|
if (eval_unary(ops, valp, &cp) == LT_IF)
|
|
return LT_IF;
|
|
*valp = !*valp;
|
|
} else if (*cp == '(') {
|
|
cp++;
|
|
debug("eval%d (", ops - eval_ops);
|
|
if (eval_table(eval_ops, valp, &cp) == LT_IF)
|
|
return LT_IF;
|
|
cp = skipcomment(cp);
|
|
if (*cp++ != ')')
|
|
return LT_IF;
|
|
} else if (isdigit((unsigned char)*cp)) {
|
|
debug("eval%d number", ops - eval_ops);
|
|
*valp = strtol(cp, &ep, 0);
|
|
cp = skipsym(cp);
|
|
} else if (strncmp(cp, "defined", 7) == 0 && endsym(cp[7])) {
|
|
cp = skipcomment(cp+7);
|
|
debug("eval%d defined", ops - eval_ops);
|
|
if (*cp++ != '(')
|
|
return LT_IF;
|
|
cp = skipcomment(cp);
|
|
sym = findsym(cp);
|
|
if (sym == 0 && !symlist)
|
|
return LT_IF;
|
|
*valp = (value[sym] != NULL);
|
|
cp = skipsym(cp);
|
|
cp = skipcomment(cp);
|
|
if (*cp++ != ')')
|
|
return LT_IF;
|
|
} else if (!endsym(*cp)) {
|
|
debug("eval%d symbol", ops - eval_ops);
|
|
sym = findsym(cp);
|
|
if (sym == 0 && !symlist)
|
|
return LT_IF;
|
|
if (value[sym] == NULL)
|
|
*valp = 0;
|
|
else {
|
|
*valp = strtol(value[sym], &ep, 0);
|
|
if (*ep != '\0' || ep == value[sym])
|
|
return LT_IF;
|
|
}
|
|
cp = skipsym(cp);
|
|
} else
|
|
return LT_IF;
|
|
|
|
*cpp = cp;
|
|
debug("eval%d = %d", ops - eval_ops, *valp);
|
|
return *valp ? LT_TRUE : LT_FALSE;
|
|
}
|
|
|
|
/*
|
|
* Table-driven evaluation of binary operators.
|
|
*/
|
|
Linetype
|
|
eval_table(struct ops *ops, int *valp, const char **cpp)
|
|
{
|
|
const char *cp;
|
|
struct op *op;
|
|
int val;
|
|
|
|
debug("eval%d", ops - eval_ops);
|
|
cp = *cpp;
|
|
if (ops->inner(ops+1, valp, &cp) == LT_IF)
|
|
return LT_IF;
|
|
for (;;) {
|
|
cp = skipcomment(cp);
|
|
for (op = ops->op; op->str != NULL; op++)
|
|
if (strncmp(cp, op->str, strlen(op->str)) == 0)
|
|
break;
|
|
if (op->str == NULL)
|
|
break;
|
|
cp += strlen(op->str);
|
|
debug("eval%d %s", ops - eval_ops, op->str);
|
|
if (ops->inner(ops+1, &val, &cp) == LT_IF)
|
|
return LT_IF;
|
|
*valp = op->fn(*valp, val);
|
|
}
|
|
|
|
*cpp = cp;
|
|
debug("eval%d = %d", ops - eval_ops, *valp);
|
|
return *valp ? LT_TRUE : LT_FALSE;
|
|
}
|
|
|
|
/*
|
|
* Evaluate the expression on a #if or #elif line. If we can work out
|
|
* the result we return LT_TRUE or LT_FALSE accordingly, otherwise we
|
|
* return just a generic LT_IF.
|
|
*/
|
|
Linetype
|
|
ifeval(const char **cpp)
|
|
{
|
|
int val;
|
|
debug("eval %s", *cpp);
|
|
return eval_table(eval_ops, &val, cpp);
|
|
}
|
|
|
|
/*
|
|
* Skip over comments and stop at the next character position that is
|
|
* not whitespace.
|
|
*/
|
|
const char *
|
|
skipcomment(const char *cp)
|
|
{
|
|
if (incomment)
|
|
goto inside;
|
|
for (;; cp++) {
|
|
while (*cp == ' ' || *cp == '\t')
|
|
cp++;
|
|
if (text)
|
|
return cp;
|
|
if (cp[0] != '/')
|
|
return cp;
|
|
|
|
if (cp[1] == '*') {
|
|
if (!incomment) {
|
|
incomment = C_COMMENT;
|
|
stqcline = linenum;
|
|
}
|
|
} else if (cp[1] == '/') {
|
|
if (!incomment) {
|
|
incomment = CXX_COMMENT;
|
|
stqcline = linenum;
|
|
}
|
|
} else
|
|
return cp;
|
|
|
|
cp += 2;
|
|
inside:
|
|
if (incomment == C_COMMENT) {
|
|
for (;;) {
|
|
for (; *cp != '*'; cp++)
|
|
if (*cp == '\0')
|
|
return cp;
|
|
if (*++cp == '/') {
|
|
incomment = NO_COMMENT;
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
else if (incomment == CXX_COMMENT) {
|
|
for (; *cp != '\n'; cp++)
|
|
if (*cp == '\0')
|
|
return cp;
|
|
incomment = NO_COMMENT;
|
|
}
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Skip over a quoted string or character and stop at the next charaacter
|
|
* position that is not whitespace.
|
|
*/
|
|
const char *
|
|
skipquote(const char *cp, Quote_state type)
|
|
{
|
|
char qchar;
|
|
|
|
qchar = type == QUOTE_SINGLE ? '\'' : '"';
|
|
|
|
if (inquote == type)
|
|
goto inside;
|
|
for (;; cp++) {
|
|
if (*cp != qchar)
|
|
return cp;
|
|
cp++;
|
|
inquote = type;
|
|
stqcline = linenum;
|
|
inside:
|
|
for (;; cp++) {
|
|
if (*cp == qchar)
|
|
break;
|
|
if (*cp == '\0' || (*cp == '\\' && *++cp == '\0'))
|
|
return cp;
|
|
}
|
|
inquote = QUOTE_NONE;
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Skip over an identifier.
|
|
*/
|
|
const char *
|
|
skipsym(const char *cp)
|
|
{
|
|
while (!endsym(*cp))
|
|
++cp;
|
|
return cp;
|
|
}
|
|
|
|
/*
|
|
* Look for the symbol in the symbol table. If is is found, we return
|
|
* the symbol table index, else we return 0.
|
|
*/
|
|
int
|
|
findsym(const char *str)
|
|
{
|
|
const char *cp;
|
|
const char *symp;
|
|
int symind;
|
|
|
|
if (symlist) {
|
|
for (cp = str; !endsym(*cp); cp++)
|
|
continue;
|
|
printf("%.*s\n", cp-str, str);
|
|
}
|
|
for (symind = 1; symind < nsyms; ++symind) {
|
|
for (cp = str, symp = symname[symind]
|
|
; *cp && *symp && *cp == *symp
|
|
; cp++, symp++
|
|
)
|
|
continue;
|
|
if (*symp == '\0' && endsym(*cp)) {
|
|
debug("findsym %s %s", symname[symind],
|
|
value[symind] ? value[symind] : "");
|
|
return symind;
|
|
}
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* Add a symbol to the symbol table.
|
|
*/
|
|
void
|
|
addsym(bool ignorethis, bool definethis, char *sym)
|
|
{
|
|
int symind;
|
|
char *val;
|
|
|
|
symind = findsym(sym);
|
|
if (symind == 0) {
|
|
if (nsyms >= MAXSYMS)
|
|
errx(2, "too many symbols");
|
|
symind = nsyms++;
|
|
}
|
|
symname[symind] = sym;
|
|
ignore[symind] = ignorethis;
|
|
val = (char *)skipsym(sym);
|
|
if (definethis) {
|
|
if (*val == '=') {
|
|
value[symind] = val+1;
|
|
*val = '\0';
|
|
} else if (*val == '\0')
|
|
value[symind] = "";
|
|
else
|
|
usage();
|
|
} else {
|
|
if (*val != '\0')
|
|
usage();
|
|
value[symind] = NULL;
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Read a line from the input and expand tabs if requested and (if
|
|
* compiled in) treats form-feed as an end-of-line.
|
|
*/
|
|
int
|
|
getline(char *line, int maxline, FILE *inp, bool expandtabs)
|
|
{
|
|
int tmp;
|
|
int num;
|
|
int chr;
|
|
#ifdef FFSPECIAL
|
|
static bool havechar = false; /* have leftover char from last time */
|
|
static char svchar;
|
|
#endif /* FFSPECIAL */
|
|
|
|
num = 0;
|
|
#ifdef FFSPECIAL
|
|
if (havechar) {
|
|
havechar = false;
|
|
chr = svchar;
|
|
goto ent;
|
|
}
|
|
#endif /* FFSPECIAL */
|
|
while (num + 8 < maxline) { /* leave room for tab */
|
|
chr = getc(inp);
|
|
if (isprint(chr)) {
|
|
#ifdef FFSPECIAL
|
|
ent:
|
|
#endif /* FFSPECIAL */
|
|
*line++ = chr;
|
|
num++;
|
|
} else
|
|
switch (chr) {
|
|
case EOF:
|
|
return EOF;
|
|
|
|
case '\t':
|
|
if (expandtabs) {
|
|
num += tmp = 8 - (num & 7);
|
|
do
|
|
*line++ = ' ';
|
|
while (--tmp);
|
|
break;
|
|
}
|
|
default:
|
|
*line++ = chr;
|
|
num++;
|
|
break;
|
|
|
|
case '\n':
|
|
*line = '\n';
|
|
num++;
|
|
goto end;
|
|
|
|
#ifdef FFSPECIAL
|
|
case '\f':
|
|
if (++num == 1)
|
|
*line = '\f';
|
|
else {
|
|
*line = '\n';
|
|
havechar = true;
|
|
svchar = chr;
|
|
}
|
|
goto end;
|
|
#endif /* FFSPECIAL */
|
|
}
|
|
}
|
|
end:
|
|
*++line = '\0';
|
|
return num;
|
|
}
|
|
|
|
/*
|
|
* Write a line to the output or not, according to the current
|
|
* filtering state.
|
|
*/
|
|
void
|
|
flushline(bool keep)
|
|
{
|
|
if (symlist)
|
|
return;
|
|
if ((keep && reject != REJ_YES) ^ complement)
|
|
fputs(tline, stdout);
|
|
else if (lnblank)
|
|
putc('\n', stdout);
|
|
return;
|
|
}
|
|
|
|
void
|
|
debug(const char *msg, ...)
|
|
{
|
|
va_list ap;
|
|
|
|
if (debugging) {
|
|
va_start(ap, msg);
|
|
vwarnx(msg, ap);
|
|
va_end(ap);
|
|
}
|
|
}
|
|
|
|
void
|
|
error(int code, int depth)
|
|
{
|
|
if (incomment || inquote)
|
|
errx(2, "error in %s line %d: %s (#if depth %d)",
|
|
filename, stqcline, errs[code], depth);
|
|
else
|
|
errx(2, "error in %s line %d: %s"
|
|
" (#if depth %d start line %d)",
|
|
filename, linenum, errs[code], depth, stifline);
|
|
}
|