Print this page
3731 Update nawk to version 20121220
@@ -21,266 +21,331 @@
/*
* Copyright (c) 1989, 2010, Oracle and/or its affiliates. All rights reserved.
*/
-/* Copyright (c) 1984, 1986, 1987, 1988, 1989 AT&T */
-/* All Rights Reserved */
-
-#define tempfree(x, s) if (istemp(x)) tfree(x, s)
+/*
+ * Copyright (C) Lucent Technologies 1997
+ * All Rights Reserved
+ *
+ * Permission to use, copy, modify, and distribute this software and
+ * its documentation for any purpose and without fee is hereby
+ * granted, provided that the above copyright notice appear in all
+ * copies and that both that the copyright notice and this
+ * permission notice and warranty disclaimer appear in supporting
+ * documentation, and that the name Lucent Technologies or any of
+ * its entities not be used in advertising or publicity pertaining
+ * to distribution of the software without specific, written prior
+ * permission.
+ *
+ * LUCENT DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
+ * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS.
+ * IN NO EVENT SHALL LUCENT OR ANY OF ITS ENTITIES BE LIABLE FOR ANY
+ * SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER
+ * IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,
+ * ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF
+ * THIS SOFTWARE.
+ */
-#define execute(p) r_execute(p)
+#define tempfree(x) if (istemp(x)) tfree(x);
#define DEBUG
-#include "awk.h"
-#include <math.h>
-#include "y.tab.h"
-#include <stdio.h>
-#include <ctype.h>
#include <setjmp.h>
+#include <math.h>
#include <time.h>
+#include "awk.h"
+#include "y.tab.h"
-#ifndef FOPEN_MAX
-#define FOPEN_MAX 15 /* max number of open files, from ANSI std. */
-#endif
-
+static Cell *execute(Node *);
+static Cell *gettemp(void);
+static Cell *copycell(Cell *);
+static FILE *openfile(int, const uchar *);
+static FILE *redirect(int, Node *);
+static const char *filename(FILE *);
+static void flush_all(void);
static jmp_buf env;
+extern Awkfloat srand_seed;
-static Cell *r_execute(Node *);
-static Cell *gettemp(char *), *copycell(Cell *);
-static FILE *openfile(int, uchar *), *redirect(int, Node *);
-
-int paircnt;
-Node *winner = NULL;
-
-static Cell *tmps;
+Node *winner = NULL; /* root of parse tree */
+static Cell *tmps; /* free temporary cells for execution */
static Cell truecell = { OBOOL, BTRUE, 0, 0, 1.0, NUM };
-Cell *true = &truecell;
+Cell *True = &truecell;
static Cell falsecell = { OBOOL, BFALSE, 0, 0, 0.0, NUM };
-Cell *false = &falsecell;
+Cell *False = &falsecell;
static Cell breakcell = { OJUMP, JBREAK, 0, 0, 0.0, NUM };
Cell *jbreak = &breakcell;
static Cell contcell = { OJUMP, JCONT, 0, 0, 0.0, NUM };
Cell *jcont = &contcell;
static Cell nextcell = { OJUMP, JNEXT, 0, 0, 0.0, NUM };
Cell *jnext = &nextcell;
+static Cell nextfilecell = { OJUMP, JNEXTFILE, 0, 0, 0.0, NUM };
+Cell *jnextfile = &nextfilecell;
static Cell exitcell = { OJUMP, JEXIT, 0, 0, 0.0, NUM };
Cell *jexit = &exitcell;
static Cell retcell = { OJUMP, JRET, 0, 0, 0.0, NUM };
Cell *jret = &retcell;
-static Cell tempcell = { OCELL, CTEMP, 0, 0, 0.0, NUM };
+static Cell tempcell = { OCELL, CTEMP, 0, (uchar *)"", 0.0, NUM|
+ STR|DONTFREE};
Node *curnode = NULL; /* the node being executed, for debugging */
-static void tfree(Cell *, char *);
+static void tfree(Cell *);
static void closeall(void);
static double ipow(double, int);
+static void stdinit(void);
+
+/*
+ * buffer memory management
+ *
+ * pbuf: address of pointer to buffer being managed
+ * psiz: address of buffer size variable
+ * minlen: minimum length of buffer needed
+ * quantum: buffer size quantum
+ * pbptr: address of movable pointer into buffer, or 0 if none
+ * whatrtn: name of the calling routine if failure should cause fatal error
+ *
+ * return 0 for realloc failure, !=0 for success
+ */
+int adjbuf(uchar **pbuf, size_t *psiz, int minlen, int quantum, uchar **pbptr,
+ const char *whatrtn)
+{
+ if (minlen > *psiz) {
+ uchar *tbuf;
+ int rminlen = quantum ? minlen % quantum : 0;
+ int boff = pbptr ? *pbptr - *pbuf : 0;
+ /* round up to next multiple of quantum */
+ if (rminlen)
+ minlen += quantum - rminlen;
+ tbuf = (uchar *)realloc(*pbuf, minlen);
+ dprintf(("adjbuf %s: %d %d (pbuf=%p, tbuf=%p)\n", whatrtn,
+ *psiz, minlen, (void *)*pbuf, (void *)tbuf));
+ if (tbuf == NULL) {
+ if (whatrtn)
+ FATAL("out of memory in %s", whatrtn);
+ return (0);
+ }
+ *pbuf = (uchar *)tbuf;
+ *psiz = minlen;
+ if (pbptr)
+ *pbptr = tbuf + boff;
+ }
+ return (1);
+}
void
-run(Node *a)
+run(Node *a) /* execution of parse tree starts here */
{
+ stdinit();
(void) execute(a);
closeall();
}
static Cell *
-r_execute(Node *u)
+execute(Node *u) /* execute a node of the parse tree */
{
- register Cell *(*proc)();
- register Cell *x;
- register Node *a;
+ Cell *(*proc)(Node **, int);
+ Cell *x;
+ Node *a;
if (u == NULL)
- return (true);
+ return (True);
for (a = u; ; a = a->nnext) {
curnode = a;
if (isvalue(a)) {
- x = (Cell *) (a->narg[0]);
- if ((x->tval & FLD) && !donefld)
+ x = (Cell *)(a->narg[0]);
+ if (isfld(x) && !donefld)
fldbld();
- else if ((x->tval & REC) && !donerec)
+ else if (isrec(x) && !donerec)
recbld();
return (x);
}
/* probably a Cell* but too risky to print */
if (notlegal(a->nobj))
- ERROR "illegal statement" FATAL;
+ FATAL("illegal statement");
proc = proctab[a->nobj-FIRSTTOKEN];
x = (*proc)(a->narg, a->nobj);
- if ((x->tval & FLD) && !donefld)
+ if (isfld(x) && !donefld)
fldbld();
- else if ((x->tval & REC) && !donerec)
+ else if (isrec(x) && !donerec)
recbld();
if (isexpr(a))
return (x);
- /* a statement, goto next statement */
if (isjump(x))
return (x);
if (a->nnext == (Node *)NULL)
return (x);
- tempfree(x, "execute");
+ tempfree(x);
}
}
/*ARGSUSED*/
Cell *
-program(Node **a, int n)
-{
- register Cell *x;
+program(Node **a, int n) /* execute an awk program */
+{ /* a[0] = BEGIN, a[1] = body, a[2] = END */
+ Cell *x;
if (setjmp(env) != 0)
goto ex;
if (a[0]) { /* BEGIN */
x = execute(a[0]);
if (isexit(x))
- return (true);
+ return (True);
if (isjump(x)) {
- ERROR "illegal break, continue or next from BEGIN"
- FATAL;
+ FATAL(
+ "illegal break, continue, next or nextfile from BEGIN");
}
- tempfree(x, "");
+ tempfree(x);
}
-loop:
- if (a[1] || a[2])
- while (getrec(&record, &record_size) > 0) {
+
+ if (a[1] || a[2]) {
+ while (getrec(&record, &record_size, 1) > 0) {
x = execute(a[1]);
if (isexit(x))
break;
- tempfree(x, "");
+ tempfree(x);
+ }
}
ex:
- if (setjmp(env) != 0)
+ if (setjmp(env) != 0) /* handles exit within END */
goto ex1;
if (a[2]) { /* END */
x = execute(a[2]);
- if (iscont(x)) /* read some more */
- goto loop;
- if (isbreak(x) || isnext(x))
- ERROR "illegal break or next from END" FATAL;
- tempfree(x, "");
+ if (isbreak(x) || isnext(x) || iscont(x))
+ FATAL(
+ "illegal break, continue, next or nextfile from END");
+ tempfree(x);
}
ex1:
- return (true);
+ return (True);
}
-struct Frame {
+struct Frame { /* stack frame for awk function calls */
int nargs; /* number of arguments in this call */
Cell *fcncell; /* pointer to Cell for function */
Cell **args; /* pointer to array of arguments after execute */
Cell *retval; /* return value */
};
-#define NARGS 30
+#define NARGS 50 /* max args in a call */
struct Frame *frame = NULL; /* base of stack frames; dynamically allocated */
int nframe = 0; /* number of frames allocated */
struct Frame *fp = NULL; /* frame pointer. bottom level unused */
/*ARGSUSED*/
Cell *
-call(Node **a, int n)
+call(Node **a, int n) /* function call. very kludgy and fragile */
{
static Cell newcopycell =
{ OCELL, CCOPY, 0, (uchar *) "", 0.0, NUM|STR|DONTFREE };
- int i, ncall, ndef, freed = 0;
+ int i, ncall, ndef;
+ /* handles potential double freeing when fcn & param share a tempcell */
+ int freed = 0;
Node *x;
- Cell *args[NARGS], *oargs[NARGS], *y, *z, *fcn;
+ Cell *args[NARGS], *oargs[NARGS]; /* BUG: fixed size arrays */
+ Cell *y, *z, *fcn;
uchar *s;
fcn = execute(a[0]); /* the function itself */
s = fcn->nval;
- if (!isfunc(fcn))
- ERROR "calling undefined function %s", s FATAL;
+ if (!isfcn(fcn))
+ FATAL("calling undefined function %s", s);
if (frame == NULL) {
fp = frame = (struct Frame *)calloc(nframe += 100,
sizeof (struct Frame));
if (frame == NULL) {
- ERROR "out of space for stack frames calling %s",
- s FATAL;
+ FATAL("out of space for stack frames calling %s", s);
}
}
for (ncall = 0, x = a[1]; x != NULL; x = x->nnext) /* args in call */
ncall++;
ndef = (int)fcn->fval; /* args in defn */
dprintf(("calling %s, %d args (%d in defn), fp=%d\n",
- s, ncall, ndef, fp-frame));
+ s, ncall, ndef, (int)(fp-frame)));
+
if (ncall > ndef) {
- ERROR "function %s called with %d args, uses only %d",
- s, ncall, ndef WARNING;
+ WARNING("function %s called with %d args, uses only %d",
+ s, ncall, ndef);
}
if (ncall + ndef > NARGS) {
- ERROR "function %s has %d arguments, limit %d",
- s, ncall+ndef, NARGS FATAL;
+ FATAL("function %s has %d arguments, limit %d",
+ s, ncall + ndef, NARGS);
}
for (i = 0, x = a[1]; x != NULL; i++, x = x->nnext) {
/* get call args */
- dprintf(("evaluate args[%d], fp=%d:\n", i, fp-frame));
+ dprintf(("evaluate args[%d], fp=%d:\n", i, (int)(fp-frame)));
y = execute(x);
oargs[i] = y;
dprintf(("args[%d]: %s %f <%s>, t=%o\n",
i, y->nval, y->fval,
isarr(y) ? "(array)" : (char *)y->sval, y->tval));
- if (isfunc(y)) {
- ERROR "can't use function %s as argument in %s",
- y->nval, s FATAL;
+ if (isfcn(y)) {
+ FATAL("can't use function %s as argument in %s",
+ y->nval, s);
}
if (isarr(y))
args[i] = y; /* arrays by ref */
else
args[i] = copycell(y);
- tempfree(y, "callargs");
+ tempfree(y);
}
for (; i < ndef; i++) { /* add null args for ones not provided */
- args[i] = gettemp("nullargs");
+ args[i] = gettemp();
*args[i] = newcopycell;
}
fp++; /* now ok to up frame */
if (fp >= frame + nframe) {
int dfp = fp - frame; /* old index */
frame = (struct Frame *)
realloc(frame, (nframe += 100) * sizeof (struct Frame));
if (frame == NULL)
- ERROR "out of space for stack frames in %s", s FATAL;
+ FATAL("out of space for stack frames in %s", s);
fp = frame + dfp;
}
fp->fcncell = fcn;
fp->args = args;
fp->nargs = ndef; /* number defined with (excess are locals) */
- fp->retval = gettemp("retval");
+ fp->retval = gettemp();
- dprintf(("start exec of %s, fp=%d\n", s, fp-frame));
+ dprintf(("start exec of %s, fp=%d\n", s, (int)(fp-frame)));
/*LINTED align*/
y = execute((Node *)(fcn->sval)); /* execute body */
- dprintf(("finished exec of %s, fp=%d\n", s, fp-frame));
+ dprintf(("finished exec of %s, fp=%d\n", s, (int)(fp-frame)));
for (i = 0; i < ndef; i++) {
Cell *t = fp->args[i];
if (isarr(t)) {
if (t->csub == CCOPY) {
if (i >= ncall) {
freesymtab(t);
t->csub = CTEMP;
+ tempfree(t);
} else {
oargs[i]->tval = t->tval;
oargs[i]->tval &= ~(STR|NUM|DONTFREE);
oargs[i]->sval = t->sval;
- tempfree(t, "oargsarr");
+ tempfree(t);
}
}
- } else {
+ } else if (t != y) { /* kludge to prevent freeing twice */
t->csub = CTEMP;
- tempfree(t, "fp->args");
- if (t == y) freed = 1;
+ tempfree(t);
+ } else if (t == y && t->csub == CCOPY) {
+ t->csub = CTEMP;
+ tempfree(t);
+ freed = 1;
}
}
- tempfree(fcn, "call.fcn");
+ tempfree(fcn);
if (isexit(y) || isnext(y))
return (y);
- if (!freed)
- tempfree(y, "fcn ret"); /* this can free twice! */
+ if (freed == 0) {
+ tempfree(y); /* don't free twice! */
+ }
z = fp->retval; /* return value */
dprintf(("%s returns %g |%s| %o\n",
s, getfval(z), getsval(z), z->tval));
fp--;
return (z);
@@ -289,46 +354,45 @@
static Cell *
copycell(Cell *x) /* make a copy of a cell in a temp */
{
Cell *y;
- y = gettemp("copycell");
+ y = gettemp();
y->csub = CCOPY; /* prevents freeing until call is over */
- y->nval = x->nval;
- y->sval = x->sval ? tostring(x->sval) : NULL;
+ y->nval = x->nval; /* BUG? */
+ if (isstr(x))
+ y->sval = tostring(x->sval);
y->fval = x->fval;
/* copy is not constant or field is DONTFREE right? */
y->tval = x->tval & ~(CON|FLD|REC|DONTFREE);
return (y);
}
-/*ARGSUSED*/
Cell *
-arg(Node **a, int nnn)
+arg(Node **a, int n) /* nth argument of a function */
{
- int n;
- n = (int)a[0]; /* argument number, counting from 0 */
+ n = ptoi(a[0]); /* argument number, counting from 0 */
dprintf(("arg(%d), fp->nargs=%d\n", n, fp->nargs));
- if (n+1 > fp->nargs) {
- ERROR "argument #%d of function %s was not supplied",
- n+1, fp->fcncell->nval FATAL;
+ if (n + 1 > fp->nargs) {
+ FATAL("argument #%d of function %s was not supplied",
+ n + 1, fp->fcncell->nval);
}
return (fp->args[n]);
}
Cell *
-jump(Node **a, int n)
+jump(Node **a, int n) /* break, continue, next, nextfile, return */
{
- register Cell *y;
+ Cell *y;
switch (n) {
case EXIT:
if (a[0] != NULL) {
y = execute(a[0]);
errorflag = (int)getfval(y);
- tempfree(y, "");
+ tempfree(y);
}
longjmp(env, 1);
/*NOTREACHED*/
case RETURN:
if (a[0] != NULL) {
@@ -339,653 +403,730 @@
fp->retval->tval |= NUM;
} else if (y->tval & STR)
(void) setsval(fp->retval, getsval(y));
else if (y->tval & NUM)
(void) setfval(fp->retval, getfval(y));
- tempfree(y, "");
+ else /* can't happen */
+ FATAL("bad type variable %d", y->tval);
+ tempfree(y);
}
return (jret);
case NEXT:
return (jnext);
+ case NEXTFILE:
+ nextfile();
+ return (jnextfile);
case BREAK:
return (jbreak);
case CONTINUE:
return (jcont);
default: /* can't happen */
- ERROR "illegal jump type %d", n FATAL;
+ FATAL("illegal jump type %d", n);
}
/*NOTREACHED*/
return (NULL);
}
+/*
+ * get next line from specific input
+ *
+ * a[0] is variable, a[1] is operator, a[2] is filename
+ */
Cell *
-getaline(Node **a, int n)
+awkgetline(Node **a, int n)
{
- /* a[0] is variable, a[1] is operator, a[2] is filename */
- register Cell *r, *x;
- uchar *buf;
+ Cell *r, *x;
+ extern Cell **fldtab;
FILE *fp;
- size_t len;
+ uchar *buf;
+ size_t bufsize = record_size;
+ int mode;
+
+ if ((buf = (uchar *)malloc(bufsize)) == NULL)
+ FATAL("out of memory in getline");
(void) fflush(stdout); /* in case someone is waiting for a prompt */
- r = gettemp("");
+ r = gettemp();
if (a[1] != NULL) { /* getline < file */
x = execute(a[2]); /* filename */
- if ((int)a[1] == '|') /* input pipe */
- a[1] = (Node *)LE; /* arbitrary flag */
- fp = openfile((int)a[1], getsval(x));
- tempfree(x, "");
- buf = NULL;
+ mode = ptoi(a[1]);
+ if (mode == '|') /* input pipe */
+ mode = LE; /* arbitrary flag */
+ fp = openfile(mode, getsval(x));
+ tempfree(x);
if (fp == NULL)
n = -1;
else
- n = readrec(&buf, &len, fp);
- if (n > 0) {
+ n = readrec(&buf, &bufsize, fp);
if (a[0] != NULL) { /* getline var <file */
- (void) setsval(execute(a[0]), buf);
+ x = execute(a[0]);
+ (void) setsval(x, buf);
+ tempfree(x);
} else { /* getline <file */
- if (!(recloc->tval & DONTFREE))
- xfree(recloc->sval);
- expand_buf(&record, &record_size, len);
- (void) memcpy(record, buf, len);
- record[len] = '\0';
- recloc->sval = record;
- recloc->tval = REC | STR | DONTFREE;
- donerec = 1; donefld = 0;
+ (void) setsval(fldtab[0], buf);
+ if (is_number(fldtab[0]->sval)) {
+ fldtab[0]->fval = atof((char *)fldtab[0]->sval);
+ fldtab[0]->tval |= NUM;
}
}
- if (buf != NULL)
- free(buf);
} else { /* bare getline; use current input */
if (a[0] == NULL) /* getline */
- n = getrec(&record, &record_size);
+ n = getrec(&record, &record_size, 1);
else { /* getline var */
- init_buf(&buf, &len, LINE_INCR);
- n = getrec(&buf, &len);
- (void) setsval(execute(a[0]), buf);
- free(buf);
+ n = getrec(&buf, &bufsize, 0);
+ x = execute(a[0]);
+ (void) setsval(x, buf);
+ tempfree(x);
}
}
(void) setfval(r, (Awkfloat)n);
+ free(buf);
return (r);
}
/*ARGSUSED*/
Cell *
-getnf(Node **a, int n)
+getnf(Node **a, int n) /* get NF */
{
if (donefld == 0)
fldbld();
return ((Cell *)a[0]);
}
/*ARGSUSED*/
Cell *
-array(Node **a, int n)
+array(Node **a, int n) /* a[0] is symtab, a[1] is list of subscripts */
{
- register Cell *x, *y, *z;
- register uchar *s;
- register Node *np;
+ Cell *x, *y, *z;
+ uchar *s;
+ Node *np;
uchar *buf;
- size_t bsize, tlen, len, slen;
+ size_t bufsz = record_size;
+ size_t nsub = strlen((char *)*SUBSEP);
+
+ if ((buf = (uchar *)malloc(bufsz)) == NULL)
+ FATAL("out of memory in array");
x = execute(a[0]); /* Cell* for symbol table */
- init_buf(&buf, &bsize, LINE_INCR);
buf[0] = '\0';
- tlen = 0;
- slen = strlen((char *)*SUBSEP);
for (np = a[1]; np; np = np->nnext) {
y = execute(np); /* subscript */
s = getsval(y);
- len = strlen((char *)s);
- expand_buf(&buf, &bsize, tlen + len + slen);
- (void) memcpy(&buf[tlen], s, len);
- tlen += len;
- if (np->nnext) {
- (void) memcpy(&buf[tlen], *SUBSEP, slen);
- tlen += slen;
- }
- buf[tlen] = '\0';
- tempfree(y, "");
+ if (!adjbuf(&buf, &bufsz, strlen((char *)buf) +
+ strlen((char *)s) + nsub + 1, record_size, 0, "array"))
+ FATAL("out of memory for %s[%s...]", x->nval, buf);
+ (void) strcat((char *)buf, (char *)s);
+ if (np->nnext)
+ (void) strcat((char *)buf, (char *)*SUBSEP);
+ tempfree(y);
}
if (!isarr(x)) {
dprintf(("making %s into an array\n", x->nval));
if (freeable(x))
xfree(x->sval);
x->tval &= ~(STR|NUM|DONTFREE);
x->tval |= ARR;
- x->sval = (uchar *) makesymtab(NSYMTAB);
+ x->sval = (uchar *)makesymtab(NSYMTAB);
}
/*LINTED align*/
z = setsymtab(buf, (uchar *)"", 0.0, STR|NUM, (Array *)x->sval);
z->ctype = OCELL;
z->csub = CVAR;
- tempfree(x, "");
+ tempfree(x);
free(buf);
return (z);
}
/*ARGSUSED*/
Cell *
-delete(Node **a, int n)
+awkdelete(Node **a, int n) /* a[0] is symtab, a[1] is list of subscripts */
{
Cell *x, *y;
Node *np;
- uchar *buf, *s;
- size_t bsize, tlen, slen, len;
+ uchar *s;
+ size_t nsub = strlen((char *)*SUBSEP);
x = execute(a[0]); /* Cell* for symbol table */
if (!isarr(x))
- return (true);
- init_buf(&buf, &bsize, LINE_INCR);
+ return (True);
+ if (a[1] == 0) { /* delete the elements, not the table */
+ freesymtab(x);
+ x->tval &= ~STR;
+ x->tval |= ARR;
+ x->sval = (uchar *)makesymtab(NSYMTAB);
+ } else {
+ size_t bufsz = record_size;
+ uchar *buf;
+ if ((buf = (uchar *)malloc(bufsz)) == NULL)
+ FATAL("out of memory in delete");
buf[0] = '\0';
- tlen = 0;
- slen = strlen((char *)*SUBSEP);
for (np = a[1]; np; np = np->nnext) {
y = execute(np); /* subscript */
s = getsval(y);
- len = strlen((char *)s);
- expand_buf(&buf, &bsize, tlen + len + slen);
- (void) memcpy(&buf[tlen], s, len);
- tlen += len;
- if (np->nnext) {
- (void) memcpy(&buf[tlen], *SUBSEP, slen);
- tlen += slen;
- }
- buf[tlen] = '\0';
- tempfree(y, "");
+ if (!adjbuf(&buf, &bufsz, strlen((char *)buf) +
+ strlen((char *)s) + nsub + 1, record_size, 0,
+ "awkdelete")) {
+ FATAL("out of memory deleting %s[%s...]",
+ x->nval, buf);
+ }
+ (void) strcat((char *)buf, (char *)s);
+ if (np->nnext)
+ (void) strcat((char *)buf, (char *)*SUBSEP);
+ tempfree(y);
}
freeelem(x, buf);
- tempfree(x, "");
free(buf);
- return (true);
+ }
+ tempfree(x);
+ return (True);
}
/*ARGSUSED*/
Cell *
-intest(Node **a, int n)
+intest(Node **a, int n) /* a[0] is index (list), a[1] is symtab */
{
- register Cell *x, *ap, *k;
+ Cell *x, *ap, *k;
Node *p;
uchar *buf;
uchar *s;
- size_t bsize, tlen, slen, len;
+ size_t bufsz = record_size;
+ size_t nsub = strlen((char *)*SUBSEP);
ap = execute(a[1]); /* array name */
- if (!isarr(ap))
- ERROR "%s is not an array", ap->nval FATAL;
- init_buf(&buf, &bsize, LINE_INCR);
+ if (!isarr(ap)) {
+ dprintf(("making %s into an array\n", ap->nval));
+ if (freeable(ap))
+ xfree(ap->sval);
+ ap->tval &= ~(STR|NUM|DONTFREE);
+ ap->tval |= ARR;
+ ap->sval = (uchar *)makesymtab(NSYMTAB);
+ }
+ if ((buf = (uchar *)malloc(bufsz)) == NULL) {
+ FATAL("out of memory in intest");
+ }
buf[0] = 0;
- tlen = 0;
- slen = strlen((char *)*SUBSEP);
for (p = a[0]; p; p = p->nnext) {
x = execute(p); /* expr */
s = getsval(x);
- len = strlen((char *)s);
- expand_buf(&buf, &bsize, tlen + len + slen);
- (void) memcpy(&buf[tlen], s, len);
- tlen += len;
- tempfree(x, "");
- if (p->nnext) {
- (void) memcpy(&buf[tlen], *SUBSEP, slen);
- tlen += slen;
- }
- buf[tlen] = '\0';
+ if (!adjbuf(&buf, &bufsz, strlen((char *)buf) +
+ strlen((char *)s) + nsub + 1, record_size, 0, "intest"))
+ FATAL("out of memory deleting %s[%s...]", x->nval, buf);
+ (void) strcat((char *)buf, (char *)s);
+ tempfree(x);
+ if (p->nnext)
+ (void) strcat((char *)buf, (char *)*SUBSEP);
}
/*LINTED align*/
k = lookup(buf, (Array *)ap->sval);
- tempfree(ap, "");
+ tempfree(ap);
free(buf);
if (k == NULL)
- return (false);
+ return (False);
else
- return (true);
+ return (True);
}
-
Cell *
-matchop(Node **a, int n)
+matchop(Node **a, int n) /* ~ and match() */
{
- register Cell *x, *y;
- register uchar *s, *t;
- register int i;
+ Cell *x, *y;
+ uchar *s, *t;
+ int i;
fa *pfa;
- int (*mf)() = match, mode = 0;
+ int (*mf)(fa *, const uchar *) = match, mode = 0;
if (n == MATCHFCN) {
mf = pmatch;
mode = 1;
}
- x = execute(a[1]);
+ x = execute(a[1]); /* a[1] = target text */
s = getsval(x);
- if (a[0] == 0)
- i = (*mf)(a[2], s);
+ if (a[0] == 0) /* a[1] == 0: already-compiled reg expr */
+ i = (*mf)((fa *)a[2], s);
else {
- y = execute(a[2]);
+ y = execute(a[2]); /* a[2] = regular expr */
t = getsval(y);
pfa = makedfa(t, mode);
i = (*mf)(pfa, s);
- tempfree(y, "");
+ tempfree(y);
}
- tempfree(x, "");
+ tempfree(x);
if (n == MATCHFCN) {
int start = patbeg - s + 1;
if (patlen < 0)
start = 0;
(void) setfval(rstartloc, (Awkfloat)start);
(void) setfval(rlengthloc, (Awkfloat)patlen);
- x = gettemp("");
+ x = gettemp();
x->tval = NUM;
x->fval = start;
return (x);
- } else if (n == MATCH && i == 1 || n == NOTMATCH && i == 0)
- return (true);
+ } else if ((n == MATCH && i == 1) || (n == NOTMATCH && i == 0))
+ return (True);
else
- return (false);
+ return (False);
}
-
Cell *
-boolop(Node **a, int n)
+boolop(Node **a, int n) /* a[0] || a[1], a[0] && a[1], !a[0] */
{
- register Cell *x, *y;
- register int i;
+ Cell *x, *y;
+ int i;
x = execute(a[0]);
i = istrue(x);
- tempfree(x, "");
+ tempfree(x);
switch (n) {
case BOR:
if (i)
- return (true);
+ return (True);
y = execute(a[1]);
i = istrue(y);
- tempfree(y, "");
- return (i ? true : false);
+ tempfree(y);
+ return (i ? True : False);
case AND:
if (!i)
- return (false);
+ return (False);
y = execute(a[1]);
i = istrue(y);
- tempfree(y, "");
- return (i ? true : false);
+ tempfree(y);
+ return (i ? True : False);
case NOT:
- return (i ? false : true);
+ return (i ? False : True);
default: /* can't happen */
- ERROR "unknown boolean operator %d", n FATAL;
+ FATAL("unknown boolean operator %d", n);
}
/*NOTREACHED*/
return (NULL);
}
Cell *
-relop(Node **a, int n)
+relop(Node **a, int n) /* a[0 < a[1], etc. */
{
- register int i;
- register Cell *x, *y;
+ int i;
+ Cell *x, *y;
Awkfloat j;
x = execute(a[0]);
y = execute(a[1]);
- if (x->tval&NUM && y->tval&NUM) {
+ if (x->tval & NUM && y->tval & NUM) {
j = x->fval - y->fval;
i = j < 0 ? -1: (j > 0 ? 1: 0);
} else {
i = strcmp((char *)getsval(x), (char *)getsval(y));
}
- tempfree(x, "");
- tempfree(y, "");
+ tempfree(x);
+ tempfree(y);
switch (n) {
- case LT: return (i < 0 ? true : false);
- case LE: return (i <= 0 ? true : false);
- case NE: return (i != 0 ? true : false);
- case EQ: return (i == 0 ? true : false);
- case GE: return (i >= 0 ? true : false);
- case GT: return (i > 0 ? true : false);
+ case LT: return (i < 0 ? True : False);
+ case LE: return (i <= 0 ? True : False);
+ case NE: return (i != 0 ? True : False);
+ case EQ: return (i == 0 ? True : False);
+ case GE: return (i >= 0 ? True : False);
+ case GT: return (i > 0 ? True : False);
default: /* can't happen */
- ERROR "unknown relational operator %d", n FATAL;
+ FATAL("unknown relational operator %d", n);
}
/*NOTREACHED*/
- return (false);
+ return (False);
}
static void
-tfree(Cell *a, char *s)
+tfree(Cell *a) /* free a tempcell */
{
- if (dbg > 1) {
- (void) printf("## tfree %.8s %06lo %s\n",
- s, (ulong_t)a, a->sval ? a->sval : (uchar *)"");
- }
- if (freeable(a))
+ if (freeable(a)) {
+ dprintf(("freeing %s %s %o\n", (char *)a->nval,
+ (char *)a->sval, a->tval));
xfree(a->sval);
+ }
if (a == tmps)
- ERROR "tempcell list is curdled" FATAL;
+ FATAL("tempcell list is curdled");
a->cnext = tmps;
tmps = a;
}
static Cell *
-gettemp(char *s)
+gettemp(void) /* get a tempcell */
{
int i;
- register Cell *x;
+ Cell *x;
if (!tmps) {
tmps = (Cell *)calloc(100, sizeof (Cell));
if (!tmps)
- ERROR "no space for temporaries" FATAL;
+ FATAL("out of space for temporaries");
for (i = 1; i < 100; i++)
- tmps[i-1].cnext = &tmps[i];
- tmps[i-1].cnext = 0;
+ tmps[i - 1].cnext = &tmps[i];
+ tmps[i - 1].cnext = 0;
}
x = tmps;
tmps = x->cnext;
*x = tempcell;
- if (dbg > 1)
- (void) printf("## gtemp %.8s %06lo\n", s, (ulong_t)x);
return (x);
}
/*ARGSUSED*/
Cell *
-indirect(Node **a, int n)
+indirect(Node **a, int n) /* $( a[0] ) */
{
- register Cell *x;
- register int m;
- register uchar *s;
+ Awkfloat val;
+ Cell *x;
+ int m;
+ uchar *s;
x = execute(a[0]);
- m = (int)getfval(x);
+ /* freebsd: defend against super large field numbers */
+ val = getfval(x);
+ if ((Awkfloat)INT_MAX < val)
+ FATAL("trying to access out of range field %s", x->nval);
+ m = (int)val;
if (m == 0 && !is_number(s = getsval(x))) /* suspicion! */
- ERROR "illegal field $(%s)", s FATAL;
- tempfree(x, "");
+ FATAL("illegal field $(%s), name \"%s\"", s, x->nval);
+ /* BUG: can x->nval ever be null??? */
+ tempfree(x);
x = fieldadr(m);
- x->ctype = OCELL;
+ x->ctype = OCELL; /* BUG? why are these needed? */
x->csub = CFLD;
return (x);
}
/*ARGSUSED*/
Cell *
-substr(Node **a, int nnn)
+substr(Node **a, int nnn) /* substr(a[0], a[1], a[2]) */
{
- register int k, m, n;
- register uchar *s;
+ int k, m, n;
+ uchar *s;
int temp;
- register Cell *x, *y, *z;
+ Cell *x, *y, *z = 0;
x = execute(a[0]);
y = execute(a[1]);
if (a[2] != 0)
z = execute(a[2]);
s = getsval(x);
k = strlen((char *)s) + 1;
if (k <= 1) {
- tempfree(x, "");
- tempfree(y, "");
- if (a[2] != 0)
- tempfree(z, "");
- x = gettemp("");
+ tempfree(x);
+ tempfree(y);
+ if (a[2] != 0) {
+ tempfree(z);
+ }
+ x = gettemp();
(void) setsval(x, (uchar *)"");
return (x);
}
m = (int)getfval(y);
if (m <= 0)
m = 1;
else if (m > k)
m = k;
- tempfree(y, "");
+ tempfree(y);
if (a[2] != 0) {
n = (int)getfval(z);
- tempfree(z, "");
+ tempfree(z);
} else
n = k - 1;
if (n < 0)
n = 0;
else if (n > k - m)
n = k - m;
dprintf(("substr: m=%d, n=%d, s=%s\n", m, n, s));
- y = gettemp("");
+ y = gettemp();
temp = s[n + m - 1]; /* with thanks to John Linderman */
s[n + m - 1] = '\0';
(void) setsval(y, s + m - 1);
s[n + m - 1] = temp;
- tempfree(x, "");
+ tempfree(x);
return (y);
}
/*ARGSUSED*/
Cell *
-sindex(Node **a, int nnn)
+sindex(Node **a, int nnn) /* index(a[0], a[1]) */
{
- register Cell *x, *y, *z;
- register uchar *s1, *s2, *p1, *p2, *q;
+ Cell *x, *y, *z;
+ uchar *s1, *s2, *p1, *p2, *q;
Awkfloat v = 0.0;
x = execute(a[0]);
s1 = getsval(x);
y = execute(a[1]);
s2 = getsval(y);
- z = gettemp("");
+ z = gettemp();
for (p1 = s1; *p1 != '\0'; p1++) {
for (q = p1, p2 = s2; *p2 != '\0' && *q == *p2; q++, p2++)
;
if (*p2 == '\0') {
- v = (Awkfloat) (p1 - s1 + 1); /* origin 1 */
+ v = (Awkfloat)(p1 - s1 + 1); /* origin 1 */
break;
}
}
- tempfree(x, "");
- tempfree(y, "");
+ tempfree(x);
+ tempfree(y);
(void) setfval(z, v);
return (z);
}
-void
-format(uchar **bufp, uchar *s, Node *a)
+#define MAXNUMSIZE 50
+
+/*
+ * printf-like conversions
+ */
+int
+format(uchar **bufp, size_t *pbufsize, const uchar *s, Node *a)
{
uchar *fmt;
- register uchar *os;
- register Cell *x;
- int flag = 0, len;
- uchar_t *buf;
- size_t bufsize, fmtsize, cnt, tcnt, ret;
+ uchar *p, *t;
+ const uchar *os;
+ Cell *x;
+ int flag = 0, n;
+ int fmtwd; /* format width */
+ size_t fmtsz = record_size;
+ uchar_t *buf = *bufp;
+ size_t bufsize = *pbufsize;
- init_buf(&buf, &bufsize, LINE_INCR);
- init_buf(&fmt, &fmtsize, LINE_INCR);
os = s;
- cnt = 0;
+ p = buf;
+ if ((fmt = (uchar *)malloc(fmtsz)) == NULL)
+ FATAL("out of memory in format()");
while (*s) {
+ (void) adjbuf(&buf, &bufsize, MAXNUMSIZE + 1 + p - buf,
+ record_size, &p, "format1");
if (*s != '%') {
- expand_buf(&buf, &bufsize, cnt);
- buf[cnt++] = *s++;
+ *p++ = *s++;
continue;
}
if (*(s+1) == '%') {
- expand_buf(&buf, &bufsize, cnt);
- buf[cnt++] = '%';
+ *p++ = '%';
s += 2;
continue;
}
- for (tcnt = 0; ; s++) {
- expand_buf(&fmt, &fmtsize, tcnt);
- fmt[tcnt++] = *s;
- if (*s == '\0')
- break;
- if (isalpha(*s) && *s != 'l' && *s != 'h' && *s != 'L')
+ /*
+ * have to be real careful in case this is a huge number,
+ * eg, %100000d
+ */
+ fmtwd = atoi((char *)s + 1);
+ if (fmtwd < 0)
+ fmtwd = -fmtwd;
+ (void) adjbuf(&buf, &bufsize, fmtwd + 1 + p - buf,
+ record_size, &p, "format2");
+ for (t = fmt; (*t++ = *s) != '\0'; s++) {
+ if (!adjbuf(&fmt, &fmtsz, MAXNUMSIZE + 1 + t - fmt,
+ record_size, &t, "format3"))
+ FATAL(
+ "format item %.30s... ran format() out of memory", os);
+ if (isalpha((uchar)*s) && *s != 'l' && *s != 'h' &&
+ *s != 'L')
break; /* the ansi panoply */
if (*s == '*') {
- if (a == NULL) {
- ERROR
- "not enough args in printf(%s) or sprintf(%s)", os, os FATAL;
- }
x = execute(a);
a = a->nnext;
- tcnt--;
- expand_buf(&fmt, &fmtsize, tcnt + 12);
- ret = sprintf((char *)&fmt[tcnt], "%d",
- (int)getfval(x));
- tcnt += ret;
- tempfree(x, "");
+ (void) sprintf((char *)t - 1, "%d",
+ fmtwd = (int)getfval(x));
+ if (fmtwd < 0)
+ fmtwd = -fmtwd;
+ (void) adjbuf(&buf, &bufsize, fmtwd + 1 + p -
+ buf, record_size, &p, "format");
+ t = fmt + strlen((char *)fmt);
+ tempfree(x);
}
}
- fmt[tcnt] = '\0';
+ *t = '\0';
+ if (fmtwd < 0)
+ fmtwd = -fmtwd;
+ (void) adjbuf(&buf, &bufsize, fmtwd + 1 + p - buf,
+ record_size, &p, "format4");
switch (*s) {
case 'f': case 'e': case 'g': case 'E': case 'G':
- flag = 1;
+ flag = 'f';
break;
case 'd': case 'i':
- flag = 2;
+ flag = 'd';
if (*(s-1) == 'l')
break;
- fmt[tcnt - 1] = 'l';
- expand_buf(&fmt, &fmtsize, tcnt);
- fmt[tcnt++] = 'd';
- fmt[tcnt] = '\0';
+ *(t-1) = 'l';
+ *t = 'd';
+ *++t = '\0';
break;
case 'o': case 'x': case 'X': case 'u':
- flag = *(s-1) == 'l' ? 2 : 3;
+ flag = *(s-1) == 'l' ? 'd' : 'u';
break;
case 's':
- flag = 4;
+ flag = 's';
break;
case 'c':
- flag = 5;
+ flag = 'c';
break;
default:
- flag = 0;
+ WARNING("weird printf conversion %s", fmt);
+ flag = '?';
break;
}
- if (flag == 0) {
- len = strlen((char *)fmt);
- expand_buf(&buf, &bufsize, cnt + len);
- (void) memcpy(&buf[cnt], fmt, len);
- cnt += len;
- buf[cnt] = '\0';
- continue;
- }
if (a == NULL) {
- ERROR
- "not enough args in printf(%s) or sprintf(%s)", os, os FATAL;
+ FATAL(
+ "not enough args in printf(%s)", os);
}
x = execute(a);
a = a->nnext;
- for (;;) {
- /* make sure we have at least 1 byte space */
- expand_buf(&buf, &bufsize, cnt + 1);
- len = bufsize - cnt;
+ n = MAXNUMSIZE;
+ if (fmtwd > n)
+ n = fmtwd;
+ (void) adjbuf(&buf, &bufsize, 1 + n + p - buf, record_size,
+ &p, "format5");
+
switch (flag) {
- case 1:
- /*LINTED*/
- ret = snprintf((char *)&buf[cnt], len,
- (char *)fmt, getfval(x));
- break;
- case 2:
- /*LINTED*/
- ret = snprintf((char *)&buf[cnt], len,
- (char *)fmt, (long)getfval(x));
- break;
- case 3:
- /*LINTED*/
- ret = snprintf((char *)&buf[cnt], len,
- (char *)fmt, (int)getfval(x));
- break;
- case 4:
- /*LINTED*/
- ret = snprintf((char *)&buf[cnt], len,
- (char *)fmt, getsval(x));
+ case '?':
+ /* unknown, so dump it too */
+ /* LINTED E_SEC_SPRINTF_UNBOUNDED_COPY */
+ (void) sprintf((char *)p, "%s", (char *)fmt);
+ t = getsval(x);
+ n = strlen((char *)t);
+ if (fmtwd > n)
+ n = fmtwd;
+ (void) adjbuf(&buf, &bufsize, 1 + strlen((char *)p) +
+ n + p - buf, record_size, &p, "format6");
+ p += strlen((char *)p);
+ /* LINTED E_SEC_SPRINTF_UNBOUNDED_COPY */
+ (void) sprintf((char *)p, "%s", t);
+ break;
+ case 'f':
+ /* LINTED E_SEC_PRINTF_VAR_FMT */
+ (void) sprintf((char *)p, (char *)fmt, getfval(x));
+ break;
+ case 'd':
+ /* LINTED E_SEC_PRINTF_VAR_FMT */
+ (void) sprintf((char *)p, (char *)fmt,
+ (long) getfval(x));
+ break;
+ case 'u':
+ /* LINTED E_SEC_PRINTF_VAR_FMT */
+ (void) sprintf((char *)p, (char *)fmt,
+ (int) getfval(x));
+ break;
+ case 's':
+ t = getsval(x);
+ n = strlen((char *)t);
+ if (fmtwd > n)
+ n = fmtwd;
+ if (!adjbuf(&buf, &bufsize, 1+n+p-buf, record_size,
+ &p, "format7"))
+ FATAL(
+ "huge string/format (%d chars) in printf %.30s... "
+ "ran format() out of memory", n, t);
+ /* LINTED E_SEC_PRINTF_VAR_FMT */
+ (void) sprintf((char *)p, (char *)fmt, t);
break;
- case 5:
+ case 'c':
if (isnum(x)) {
- /*LINTED*/
- ret = snprintf((char *)&buf[cnt], len,
- (char *)fmt, (int)getfval(x));
+ if (getfval(x))
+ /* LINTED E_SEC_PRINTF_VAR_FMT */
+ (void) sprintf((char *)p, (char *)fmt,
+ (int) getfval(x));
+ else {
+ *p++ = '\0'; /* explicit null byte */
+ /* next output will start here */
+ *p = '\0';
+ }
} else {
- /*LINTED*/
- ret = snprintf((char *)&buf[cnt], len,
- (char *)fmt, getsval(x)[0]);
+ /* LINTED E_SEC_PRINTF_VAR_FMT */
+ (void) sprintf((char *)p, (char *)fmt,
+ getsval(x)[0]);
}
break;
default:
- ret = 0;
- }
- if (ret < len)
- break;
- expand_buf(&buf, &bufsize, cnt + ret);
+ FATAL("can't happen: bad conversion %c in format()",
+ flag);
}
- tempfree(x, "");
- cnt += ret;
+ tempfree(x);
+ p += strlen((char *)p);
s++;
}
- buf[cnt] = '\0';
+ *p = '\0';
+ free(fmt);
for (; a; a = a->nnext) /* evaluate any remaining args */
(void) execute(a);
- *bufp = tostring(buf);
- free(buf);
- free(fmt);
+ *bufp = buf;
+ *pbufsize = bufsize;
+ return (p - buf);
}
/*ARGSUSED*/
Cell *
-a_sprintf(Node **a, int n)
+awksprintf(Node **a, int n) /* sprintf(a[0]) */
{
- register Cell *x;
- register Node *y;
+ Cell *x;
+ Node *y;
uchar *buf;
+ size_t bufsz = 3 * record_size;
+ if ((buf = (uchar *)malloc(bufsz)) == NULL)
+ FATAL("out of memory in awksprintf");
y = a[0]->nnext;
x = execute(a[0]);
- format(&buf, getsval(x), y);
- tempfree(x, "");
- x = gettemp("");
+ if (format(&buf, &bufsz, getsval(x), y) == -1)
+ FATAL("sprintf string %.30s... too long. can't happen.", buf);
+ tempfree(x);
+ x = gettemp();
x->sval = buf;
x->tval = STR;
return (x);
}
/*ARGSUSED*/
Cell *
-aprintf(Node **a, int n)
-{
+awkprintf(Node **a, int n) /* printf */
+{ /* a[0] is list of args, starting with format string */
+ /* a[1] is redirection operator, a[2] is redirection file */
FILE *fp;
- register Cell *x;
- register Node *y;
+ Cell *x;
+ Node *y;
uchar *buf;
+ int len;
+ size_t bufsz = 3 * record_size;
+ if ((buf = (uchar *)malloc(bufsz)) == NULL)
+ FATAL("out of memory in awkprintf");
y = a[0]->nnext;
x = execute(a[0]);
- format(&buf, getsval(x), y);
- tempfree(x, "");
- if (a[1] == NULL)
- (void) fputs((char *)buf, stdout);
- else {
- fp = redirect((int)a[1], a[2]);
- (void) fputs((char *)buf, fp);
+ if ((len = format(&buf, &bufsz, getsval(x), y)) == -1)
+ FATAL("printf string %.30s... too long. can't happen.", buf);
+ tempfree(x);
+ if (a[1] == NULL) {
+ /* fputs(buf, stdout); */
+ (void) fwrite((char *)buf, len, 1, stdout);
+ if (ferror(stdout))
+ FATAL("write error on stdout");
+ } else {
+ fp = redirect(ptoi(a[1]), a[2]);
+ /* fputs(buf, fp); */
+ (void) fwrite(buf, len, 1, fp);
(void) fflush(fp);
+ if (ferror(fp))
+ FATAL("write error on %s", filename(fp));
}
free(buf);
- return (true);
+ return (True);
}
Cell *
-arith(Node **a, int n)
+arith(Node **a, int n) /* a[0] + a[1], etc. also -a[0] */
{
- Awkfloat i, j;
+ Awkfloat i, j = 0;
double v;
- register Cell *x, *y, *z;
+ Cell *x, *y, *z;
x = execute(a[0]);
i = getfval(x);
- tempfree(x, "");
+ tempfree(x);
if (n != UMINUS) {
y = execute(a[1]);
j = getfval(y);
- tempfree(y, "");
+ tempfree(y);
}
- z = gettemp("");
+ z = gettemp();
switch (n) {
case ADD:
i += j;
break;
case MINUS:
@@ -994,16 +1135,16 @@
case MULT:
i *= j;
break;
case DIVIDE:
if (j == 0)
- ERROR "division by zero" FATAL;
+ FATAL("division by zero");
i /= j;
break;
case MOD:
if (j == 0)
- ERROR "division by zero in mod" FATAL;
+ FATAL("division by zero in mod");
(void) modf(i/j, &v);
i = i - j * v;
break;
case UMINUS:
i = -i;
@@ -1013,72 +1154,77 @@
i = ipow(i, (int)j);
else
i = errcheck(pow(i, j), "pow");
break;
default: /* can't happen */
- ERROR "illegal arithmetic operator %d", n FATAL;
+ FATAL("illegal arithmetic operator %d", n);
}
(void) setfval(z, i);
return (z);
}
static double
-ipow(double x, int n)
+ipow(double x, int n) /* x**n. ought to be done by pow, but isn't always */
{
double v;
if (n <= 0)
return (1.0);
- v = ipow(x, n/2);
+ v = ipow(x, n / 2);
if (n % 2 == 0)
return (v * v);
else
return (x * v * v);
}
Cell *
-incrdecr(Node **a, int n)
+incrdecr(Node **a, int n) /* a[0]++, etc. */
{
- register Cell *x, *z;
- register int k;
+ Cell *x, *z;
+ int k;
Awkfloat xf;
x = execute(a[0]);
xf = getfval(x);
k = (n == PREINCR || n == POSTINCR) ? 1 : -1;
if (n == PREINCR || n == PREDECR) {
(void) setfval(x, xf + k);
return (x);
}
- z = gettemp("");
+ z = gettemp();
(void) setfval(z, xf);
(void) setfval(x, xf + k);
- tempfree(x, "");
+ tempfree(x);
return (z);
}
Cell *
-assign(Node **a, int n)
-{
- register Cell *x, *y;
+assign(Node **a, int n) /* a[0] = a[1], a[0] += a[1], etc. */
+{ /* this is subtle; don't muck with it. */
+ Cell *x, *y;
Awkfloat xf, yf;
double v;
y = execute(a[1]);
- x = execute(a[0]); /* order reversed from before... */
+ x = execute(a[0]);
if (n == ASSIGN) { /* ordinary assignment */
- if ((y->tval & (STR|NUM)) == (STR|NUM)) {
+ /* self-assignment: */
+ /* LINTED E_NOP_IF_STMT */
+ if (x == y && !(x->tval & (FLD|REC)))
+ /* leave alone unless it's a field */
+ ;
+ else if ((y->tval & (STR|NUM)) == (STR|NUM)) {
(void) setsval(x, getsval(y));
x->fval = getfval(y);
x->tval |= NUM;
- } else if (y->tval & STR)
+ } else if (isstr(y))
(void) setsval(x, getsval(y));
- else if (y->tval & NUM)
+ else if (isnum(y))
(void) setfval(x, getfval(y));
else
funnyvar(y, "read value of");
- tempfree(y, "");
+ tempfree(y);
return (x);
}
xf = getfval(x);
yf = getfval(y);
switch (n) {
@@ -1091,150 +1237,146 @@
case MULTEQ:
xf *= yf;
break;
case DIVEQ:
if (yf == 0)
- ERROR "division by zero in /=" FATAL;
+ FATAL("division by zero in /=");
xf /= yf;
break;
case MODEQ:
if (yf == 0)
- ERROR "division by zero in %%=" FATAL;
+ FATAL("division by zero in %%=");
(void) modf(xf/yf, &v);
xf = xf - yf * v;
break;
case POWEQ:
if (yf >= 0 && modf(yf, &v) == 0.0) /* pos integer exponent */
xf = ipow(xf, (int)yf);
else
xf = errcheck(pow(xf, yf), "pow");
break;
default:
- ERROR "illegal assignment operator %d", n FATAL;
+ FATAL("illegal assignment operator %d", n);
break;
}
- tempfree(y, "");
+ tempfree(y);
(void) setfval(x, xf);
return (x);
}
/*ARGSUSED*/
Cell *
-cat(Node **a, int q)
+cat(Node **a, int q) /* a[0] cat a[1] */
{
- register Cell *x, *y, *z;
- register int n1, n2;
- register uchar *s;
+ Cell *x, *y, *z;
+ int n1, n2;
+ uchar *s;
x = execute(a[0]);
y = execute(a[1]);
(void) getsval(x);
(void) getsval(y);
n1 = strlen((char *)x->sval);
n2 = strlen((char *)y->sval);
s = (uchar *)malloc(n1 + n2 + 1);
if (s == NULL) {
- ERROR "out of space concatenating %.15s and %.15s",
- x->sval, y->sval FATAL;
+ FATAL("out of space concatenating %.15s... and %.15s...",
+ x->sval, y->sval);
}
(void) strcpy((char *)s, (char *)x->sval);
(void) strcpy((char *)s + n1, (char *)y->sval);
- tempfree(y, "");
- z = gettemp("");
+ tempfree(x);
+ tempfree(y);
+ z = gettemp();
z->sval = s;
z->tval = STR;
- tempfree(x, "");
return (z);
}
/*ARGSUSED*/
Cell *
-pastat(Node **a, int n)
+pastat(Node **a, int n) /* a[0] { a[1] } */
{
- register Cell *x;
+ Cell *x;
if (a[0] == 0)
x = execute(a[1]);
else {
x = execute(a[0]);
if (istrue(x)) {
- tempfree(x, "");
+ tempfree(x);
x = execute(a[1]);
}
}
return (x);
}
/*ARGSUSED*/
Cell *
-dopa2(Node **a, int n)
+dopa2(Node **a, int n) /* a[0], a[1] { a[2] } */
{
Cell *x;
int pair;
- static int *pairstack = NULL;
-
- if (!pairstack) {
- /* first time */
- dprintf(("paircnt: %d\n", paircnt));
- pairstack = (int *)malloc(sizeof (int) * paircnt);
- if (!pairstack)
- ERROR "out of space in dopa2" FATAL;
- (void) memset(pairstack, 0, sizeof (int) * paircnt);
- }
- pair = (int)a[3];
+ pair = ptoi(a[3]);
if (pairstack[pair] == 0) {
x = execute(a[0]);
if (istrue(x))
pairstack[pair] = 1;
- tempfree(x, "");
+ tempfree(x);
}
if (pairstack[pair] == 1) {
x = execute(a[1]);
if (istrue(x))
pairstack[pair] = 0;
- tempfree(x, "");
+ tempfree(x);
x = execute(a[2]);
return (x);
}
- return (false);
+ return (False);
}
/*ARGSUSED*/
Cell *
-split(Node **a, int nnn)
+split(Node **a, int nnn) /* split(a[0], a[1], a[2]); a[3] is type */
{
- Cell *x, *y, *ap;
- register uchar *s;
- register int sep;
- uchar *t, temp, num[11], *fs;
- int n, tempstat;
+ Cell *x = 0, *y, *ap;
+ uchar *s, *origs;
+ int sep;
+ uchar *t, temp, num[50], *fs = 0;
+ int n, tempstat, arg3type;
y = execute(a[0]); /* source string */
- s = getsval(y);
+ origs = s = (uchar *)strdup((char *)getsval(y));
+ arg3type = ptoi(a[3]);
if (a[2] == 0) /* fs string */
fs = *FS;
- else if ((int)a[3] == STRING) { /* split(str,arr,"string") */
+ else if (arg3type == STRING) { /* split(str,arr,"string") */
x = execute(a[2]);
fs = getsval(x);
- } else if ((int)a[3] == REGEXPR)
+ } else if (arg3type == REGEXPR)
fs = (uchar *)"(regexpr)"; /* split(str,arr,/regexpr/) */
else
- ERROR "illegal type of split()" FATAL;
+ FATAL("illegal type of split");
sep = *fs;
ap = execute(a[1]); /* array name */
freesymtab(ap);
dprintf(("split: s=|%s|, a=%s, sep=|%s|\n", s, ap->nval, fs));
ap->tval &= ~STR;
ap->tval |= ARR;
ap->sval = (uchar *)makesymtab(NSYMTAB);
n = 0;
- if (*s != '\0' && strlen((char *)fs) > 1 || (int)a[3] == REGEXPR) {
+ if (arg3type == REGEXPR && strlen((char *)((fa *)a[2])->restr) == 0) {
+ arg3type = 0;
+ fs = (uchar *)"";
+ sep = 0;
+ }
+ if (*s != '\0' && (strlen((char *)fs) > 1 || arg3type == REGEXPR)) {
/* reg expr */
fa *pfa;
- if ((int)a[3] == REGEXPR) { /* it's ready already */
+ if (arg3type == REGEXPR) { /* it's ready already */
pfa = (fa *)a[2];
} else {
pfa = makedfa(fs, 1);
}
if (nematch(pfa, s)) {
@@ -1265,10 +1407,13 @@
STR, (Array *)ap->sval);
pfa->initstat = tempstat;
goto spdone;
}
} while (nematch(pfa, s));
+ /* bwk: has to be here to reset */
+ pfa->initstat = tempstat;
+ /* cf gsub and refldbld */
}
n++;
(void) sprintf((char *)num, "%d", n);
if (is_number(s)) {
(void) setsymtab(num, s, atof((char *)s),
@@ -1307,10 +1452,27 @@
}
*s = temp;
if (*s != 0)
s++;
}
+ } else if (sep == 0) { /* new: split(s, a, "") => 1 char/elem */
+ for (n = 0; *s != 0; s++) {
+ uchar buf[2];
+ n++;
+ (void) sprintf((char *)num, "%d", n);
+ buf[0] = *s;
+ buf[1] = 0;
+ if (isdigit(buf[0])) {
+ (void) setsymtab(num, buf, atof((char *)buf),
+ /* LINTED align */
+ STR | NUM, (Array *)ap->sval);
+ } else {
+ (void) setsymtab(num, buf, 0.0,
+ /* LINTED align */
+ STR, (Array *)ap->sval);
+ }
+ }
} else if (*s != 0) {
for (;;) {
n++;
t = s;
while (*s != sep && *s != '\n' && *s != '\0')
@@ -1330,173 +1492,188 @@
*s = temp;
if (*s++ == 0)
break;
}
}
- tempfree(ap, "");
- tempfree(y, "");
- if (a[2] != 0 && (int)a[3] == STRING)
- tempfree(x, "");
- x = gettemp("");
+ tempfree(ap);
+ tempfree(y);
+ free(origs);
+ if (a[2] != 0 && arg3type == STRING) {
+ tempfree(x);
+ }
+ x = gettemp();
x->tval = NUM;
x->fval = n;
return (x);
}
/*ARGSUSED*/
Cell *
-condexpr(Node **a, int n)
+condexpr(Node **a, int n) /* a[0] ? a[1] : a[2] */
{
- register Cell *x;
+ Cell *x;
x = execute(a[0]);
if (istrue(x)) {
- tempfree(x, "");
+ tempfree(x);
x = execute(a[1]);
} else {
- tempfree(x, "");
+ tempfree(x);
x = execute(a[2]);
}
return (x);
}
/*ARGSUSED*/
Cell *
-ifstat(Node **a, int n)
+ifstat(Node **a, int n) /* if (a[0]) a[1]; else a[2] */
{
- register Cell *x;
+ Cell *x;
x = execute(a[0]);
if (istrue(x)) {
- tempfree(x, "");
+ tempfree(x);
x = execute(a[1]);
} else if (a[2] != 0) {
- tempfree(x, "");
+ tempfree(x);
x = execute(a[2]);
}
return (x);
}
/*ARGSUSED*/
Cell *
-whilestat(Node **a, int n)
+whilestat(Node **a, int n) /* while (a[0]) a[1] */
{
- register Cell *x;
+ Cell *x;
for (;;) {
x = execute(a[0]);
if (!istrue(x))
return (x);
- tempfree(x, "");
+ tempfree(x);
x = execute(a[1]);
if (isbreak(x)) {
- x = true;
+ x = True;
return (x);
}
if (isnext(x) || isexit(x) || isret(x))
return (x);
- tempfree(x, "");
+ tempfree(x);
}
}
/*ARGSUSED*/
Cell *
-dostat(Node **a, int n)
+dostat(Node **a, int n) /* do a[0]; while(a[1]) */
{
- register Cell *x;
+ Cell *x;
for (;;) {
x = execute(a[0]);
if (isbreak(x))
- return (true);
+ return (True);
if (isnext(x) || isexit(x) || isret(x))
return (x);
- tempfree(x, "");
+ tempfree(x);
x = execute(a[1]);
if (!istrue(x))
return (x);
- tempfree(x, "");
+ tempfree(x);
}
}
/*ARGSUSED*/
Cell *
-forstat(Node **a, int n)
+forstat(Node **a, int n) /* for (a[0]; a[1]; a[2]) a[3] */
{
- register Cell *x;
+ Cell *x;
x = execute(a[0]);
- tempfree(x, "");
+ tempfree(x);
for (;;) {
if (a[1] != 0) {
x = execute(a[1]);
if (!istrue(x))
return (x);
else
- tempfree(x, "");
+ tempfree(x);
}
x = execute(a[3]);
if (isbreak(x)) /* turn off break */
- return (true);
+ return (True);
if (isnext(x) || isexit(x) || isret(x))
return (x);
- tempfree(x, "");
+ tempfree(x);
x = execute(a[2]);
- tempfree(x, "");
+ tempfree(x);
}
}
/*ARGSUSED*/
Cell *
-instat(Node **a, int n)
+instat(Node **a, int n) /* for (a[0] in a[1]) a[2] */
{
- register Cell *x, *vp, *arrayp, *cp, *ncp;
+ Cell *x, *vp, *arrayp, *cp, *ncp;
Array *tp;
int i;
vp = execute(a[0]);
arrayp = execute(a[1]);
- if (!isarr(arrayp))
- ERROR "%s is not an array", arrayp->nval FATAL;
+ if (!isarr(arrayp)) {
+ return (True);
+ }
/*LINTED align*/
tp = (Array *)arrayp->sval;
- tempfree(arrayp, "");
+ tempfree(arrayp);
for (i = 0; i < tp->size; i++) { /* this routine knows too much */
for (cp = tp->tab[i]; cp != NULL; cp = ncp) {
(void) setsval(vp, cp->nval);
ncp = cp->cnext;
x = execute(a[2]);
if (isbreak(x)) {
- tempfree(vp, "");
- return (true);
+ tempfree(vp);
+ return (True);
}
if (isnext(x) || isexit(x) || isret(x)) {
- tempfree(vp, "");
+ tempfree(vp);
return (x);
}
- tempfree(x, "");
+ tempfree(x);
}
}
- return (true);
+ return (True);
}
+/*
+ * builtin functions. a[0] is type, a[1] is arg list
+ */
/*ARGSUSED*/
Cell *
bltin(Node **a, int n)
{
- register Cell *x, *y;
+ Cell *x, *y;
Awkfloat u;
- register int t;
+ int t;
+ Awkfloat tmp;
uchar *p, *buf;
Node *nextarg;
+ FILE *fp;
- t = (int)a[0];
+ t = ptoi(a[0]);
x = execute(a[1]);
nextarg = a[1]->nnext;
switch (t) {
case FLENGTH:
- u = (Awkfloat)strlen((char *)getsval(x)); break;
+ if (isarr(x)) {
+ /* GROT. should be function */
+ /* LINTED align */
+ u = (Awkfloat)((Array *)x->sval)->nelem;
+ } else {
+ u = (Awkfloat)strlen((char *)getsval(x));
+ }
+ break;
case FLOG:
u = errcheck(log(getfval(x)), "log"); break;
case FINT:
(void) modf(getfval(x), &u); break;
case FEXP:
@@ -1507,35 +1684,38 @@
u = sin(getfval(x)); break;
case FCOS:
u = cos(getfval(x)); break;
case FATAN:
if (nextarg == 0) {
- ERROR "atan2 requires two arguments; returning 1.0"
- WARNING;
+ WARNING("atan2 requires two arguments; returning 1.0");
u = 1.0;
} else {
y = execute(a[1]->nnext);
u = atan2(getfval(x), getfval(y));
- tempfree(y, "");
+ tempfree(y);
nextarg = nextarg->nnext;
}
break;
case FSYSTEM:
/* in case something is buffered already */
(void) fflush(stdout);
/* 256 is unix-dep */
u = (Awkfloat)system((char *)getsval(x)) / 256;
break;
case FRAND:
- u = (Awkfloat)(rand() % 32767) / 32767.0;
+ /* in principle, rand() returns something in 0..RAND_MAX */
+ u = (Awkfloat)(rand() % RAND_MAX) / RAND_MAX;
break;
case FSRAND:
- if (x->tval & REC) /* no argument provided */
+ if (isrec(x)) /* no argument provided */
u = time((time_t *)0);
else
u = getfval(x);
- srand((int)u); u = (int)u;
+ tmp = u;
+ srand((unsigned int)u);
+ u = srand_seed;
+ srand_seed = tmp;
break;
case FTOUPPER:
case FTOLOWER:
buf = tostring(getsval(x));
if (t == FTOUPPER) {
@@ -1545,108 +1725,154 @@
} else {
for (p = buf; *p; p++)
if (isupper(*p))
*p = tolower(*p);
}
- tempfree(x, "");
- x = gettemp("");
+ tempfree(x);
+ x = gettemp();
(void) setsval(x, buf);
free(buf);
return (x);
+ case FFLUSH:
+ if (isrec(x) || strlen((char *)getsval(x)) == 0) {
+ /* fflush() or fflush("") -> all */
+ flush_all();
+ u = 0;
+ } else if ((fp = openfile(FFLUSH, getsval(x))) == NULL)
+ u = EOF;
+ else
+ u = fflush(fp);
+ break;
default: /* can't happen */
- ERROR "illegal function type %d", t FATAL;
+ FATAL("illegal function type %d", t);
break;
}
- tempfree(x, "");
- x = gettemp("");
+ tempfree(x);
+ x = gettemp();
(void) setfval(x, u);
if (nextarg != 0) {
- ERROR "warning: function has too many arguments" WARNING;
+ WARNING("warning: function has too many arguments");
for (; nextarg; nextarg = nextarg->nnext)
(void) execute(nextarg);
}
return (x);
}
/*ARGSUSED*/
Cell *
-print(Node **a, int n)
+printstat(Node **a, int n) /* print a[0] */
{
- register Node *x;
- register Cell *y;
+ Node *x;
+ Cell *y;
FILE *fp;
- if (a[1] == 0)
+ if (a[1] == 0) /* a[1] is redirection operator, a[2] is file */
fp = stdout;
else
- fp = redirect((int)a[1], a[2]);
+ fp = redirect(ptoi(a[1]), a[2]);
for (x = a[0]; x != NULL; x = x->nnext) {
y = execute(x);
(void) fputs((char *)getsval(y), fp);
- tempfree(y, "");
+ tempfree(y);
if (x->nnext == NULL)
(void) fputs((char *)*ORS, fp);
else
(void) fputs((char *)*OFS, fp);
}
if (a[1] != 0)
(void) fflush(fp);
- return (true);
+ if (ferror(fp))
+ FATAL("write error on %s", filename(fp));
+ return (True);
}
/*ARGSUSED*/
Cell *
nullproc(Node **a, int n)
{
return (0);
}
-struct {
- FILE *fp;
- uchar *fname;
- int mode; /* '|', 'a', 'w' */
-} files[FOPEN_MAX];
-
static FILE *
-redirect(int a, Node *b)
+redirect(int a, Node *b) /* set up all i/o redirections */
{
FILE *fp;
Cell *x;
uchar *fname;
x = execute(b);
fname = getsval(x);
fp = openfile(a, fname);
if (fp == NULL)
- ERROR "can't open file %s", fname FATAL;
- tempfree(x, "");
+ FATAL("can't open file %s", fname);
+ tempfree(x);
return (fp);
}
+struct files {
+ FILE *fp;
+ const uchar *fname;
+ int mode; /* '|', 'a', 'w' => LE/LT, GT */
+} *files;
+
+int nfiles;
+
+static void
+stdinit(void) /* in case stdin, etc., are not constants */
+{
+ nfiles = FOPEN_MAX;
+ files = calloc(nfiles, sizeof (*files));
+ if (files == NULL)
+ FATAL("can't allocate file memory for %u files", nfiles);
+ files[0].fp = stdin;
+ files[0].fname = (uchar *)"/dev/stdin";
+ files[0].mode = LT;
+ files[1].fp = stdout;
+ files[1].fname = (uchar *)"/dev/stdout";
+ files[1].mode = GT;
+ files[2].fp = stderr;
+ files[2].fname = (uchar *)"/dev/stderr";
+ files[2].mode = GT;
+}
+
static FILE *
-openfile(int a, uchar *s)
+openfile(int a, const uchar *us)
{
- register int i, m;
- register FILE *fp;
+ const uchar *s = us;
+ int i, m;
+ FILE *fp = 0;
if (*s == '\0')
- ERROR "null file name in print or getline" FATAL;
- for (i = 0; i < FOPEN_MAX; i++) {
+ FATAL("null file name in print or getline");
+ for (i = 0; i < nfiles; i++) {
if (files[i].fname &&
- strcmp((char *)s, (char *)files[i].fname) == 0) {
+ (strcmp((char *)s, (char *)files[i].fname) == 0)) {
if (a == files[i].mode ||
- a == APPEND && files[i].mode == GT) {
+ (a == APPEND && files[i].mode == GT)) {
return (files[i].fp);
}
+ if (a == FFLUSH)
+ return (files[i].fp);
}
}
- for (i = 0; i < FOPEN_MAX; i++) {
+ if (a == FFLUSH) /* didn't find it, so don't create it! */
+ return (NULL);
+
+ for (i = 0; i < nfiles; i++) {
if (files[i].fp == 0)
break;
}
- if (i >= FOPEN_MAX)
- ERROR "%s makes too many open files", s FATAL;
+ if (i >= nfiles) {
+ struct files *nf;
+ int nnf = nfiles + FOPEN_MAX;
+ nf = realloc(files, nnf * sizeof (*nf));
+ if (nf == NULL)
+ FATAL("cannot grow files for %s and %d files", s, nnf);
+ (void) memset(&nf[nfiles], 0, FOPEN_MAX * sizeof (*nf));
+ nfiles = nnf;
+ files = nf;
+ }
(void) fflush(stdout); /* force a semblance of order */
m = a;
if (a == GT) {
fp = fopen((char *)s, "w");
} else if (a == APPEND) {
@@ -1658,247 +1884,314 @@
fp = popen((char *)s, "r");
} else if (a == LT) { /* getline <file */
fp = strcmp((char *)s, "-") == 0 ?
stdin : fopen((char *)s, "r"); /* "-" is stdin */
} else /* can't happen */
- ERROR "illegal redirection" FATAL;
+ FATAL("illegal redirection %d", a);
if (fp != NULL) {
files[i].fname = tostring(s);
files[i].fp = fp;
files[i].mode = m;
}
return (fp);
}
+static const char *
+filename(FILE *fp)
+{
+ int i;
+
+ for (i = 0; i < nfiles; i++) {
+ if (fp == files[i].fp)
+ return ((char *)files[i].fname);
+ }
+
+ return ("???");
+}
+
/*ARGSUSED*/
Cell *
closefile(Node **a, int n)
{
- register Cell *x;
+ Cell *x;
int i, stat;
x = execute(a[0]);
(void) getsval(x);
- for (i = 0; i < FOPEN_MAX; i++) {
+ stat = -1;
+ for (i = 0; i < nfiles; i++) {
if (files[i].fname &&
strcmp((char *)x->sval, (char *)files[i].fname) == 0) {
if (ferror(files[i].fp)) {
- ERROR "i/o error occurred on %s",
- files[i].fname WARNING;
+ WARNING("i/o error occurred on %s",
+ files[i].fname);
}
if (files[i].mode == '|' || files[i].mode == LE)
stat = pclose(files[i].fp);
else
stat = fclose(files[i].fp);
if (stat == EOF) {
- ERROR "i/o error occurred closing %s",
- files[i].fname WARNING;
+ WARNING("i/o error occurred closing %s",
+ files[i].fname);
}
+ if (i > 2) /* don't do /dev/std... */
xfree(files[i].fname);
/* watch out for ref thru this */
files[i].fname = NULL;
files[i].fp = NULL;
}
}
- tempfree(x, "close");
- return (true);
+ tempfree(x);
+ x = gettemp();
+ (void) setfval(x, (Awkfloat)stat);
+ return (x);
}
static void
closeall(void)
{
int i, stat;
for (i = 0; i < FOPEN_MAX; i++) {
if (files[i].fp) {
if (ferror(files[i].fp)) {
- ERROR "i/o error occurred on %s",
- files[i].fname WARNING;
+ WARNING("i/o error occurred on %s",
+ files[i].fname);
}
if (files[i].mode == '|' || files[i].mode == LE)
stat = pclose(files[i].fp);
else
stat = fclose(files[i].fp);
if (stat == EOF) {
- ERROR "i/o error occurred while closing %s",
- files[i].fname WARNING;
+ WARNING("i/o error occurred while closing %s",
+ files[i].fname);
}
}
}
}
+static void
+flush_all(void)
+{
+ int i;
+
+ for (i = 0; i < nfiles; i++)
+ if (files[i].fp)
+ (void) fflush(files[i].fp);
+}
+
+void backsub(uchar **pb_ptr, uchar **sptr_ptr);
+
/*ARGSUSED*/
Cell *
-sub(Node **a, int nnn)
+sub(Node **a, int nnn) /* substitute command */
{
- register uchar *sptr;
- register Cell *x, *y, *result;
- uchar *buf, *t;
+ uchar *sptr, *pb, *q;
+ Cell *x, *y, *result;
+ uchar *t, *buf;
fa *pfa;
- size_t bsize, cnt, len;
+ size_t bufsz = record_size;
+ if ((buf = (uchar *)malloc(bufsz)) == NULL)
+ FATAL("out of memory in sub");
x = execute(a[3]); /* target string */
t = getsval(x);
- if (a[0] == 0)
+ if (a[0] == 0) /* 0 => a[1] is already-compiled regexpr */
pfa = (fa *)a[1]; /* regular expression */
else {
y = execute(a[1]);
pfa = makedfa(getsval(y), 1);
- tempfree(y, "");
+ tempfree(y);
}
y = execute(a[2]); /* replacement string */
- result = false;
+ result = False;
if (pmatch(pfa, t)) {
- init_buf(&buf, &bsize, LINE_INCR);
- cnt = 0;
sptr = t;
- len = patbeg - sptr;
- if (len > 0) {
- expand_buf(&buf, &bsize, cnt + len);
- (void) memcpy(buf, sptr, len);
- cnt += len;
- }
+ (void) adjbuf(&buf, &bufsz, 1+patbeg-sptr, record_size,
+ 0, "sub");
+ pb = buf;
+ while (sptr < patbeg)
+ *pb++ = *sptr++;
sptr = getsval(y);
while (*sptr != 0) {
- expand_buf(&buf, &bsize, cnt);
- if (*sptr == '\\' &&
- (*(sptr+1) == '&' || *(sptr+1) == '\\')) {
- sptr++; /* skip \, */
- buf[cnt++] = *sptr++; /* add & or \ */
+ (void) adjbuf(&buf, &bufsz, 5 + pb - buf, record_size,
+ &pb, "sub");
+ if (*sptr == '\\') {
+ backsub(&pb, &sptr);
} else if (*sptr == '&') {
- expand_buf(&buf, &bsize, cnt + patlen);
sptr++;
- (void) memcpy(&buf[cnt], patbeg, patlen);
- cnt += patlen;
+ (void) adjbuf(&buf, &bufsz, 1+patlen+pb-buf,
+ record_size, &pb, "sub");
+ for (q = patbeg; q < patbeg + patlen; )
+ *pb++ = *q++;
} else {
- buf[cnt++] = *sptr++;
+ *pb++ = *sptr++;
}
}
+ *pb = '\0';
+ if (pb > buf + bufsz)
+ FATAL("sub result1 %.30s too big; can't happen", buf);
sptr = patbeg + patlen;
if ((patlen == 0 && *patbeg) || (patlen && *(sptr-1))) {
- len = strlen((char *)sptr);
- expand_buf(&buf, &bsize, cnt + len);
- (void) memcpy(&buf[cnt], sptr, len);
- cnt += len;
+ (void) adjbuf(&buf, &bufsz, 1 + strlen((char *)sptr) +
+ pb - buf, 0, &pb, "sub");
+ while ((*pb++ = *sptr++) != 0)
+ ;
}
- buf[cnt] = '\0';
+ if (pb > buf + bufsz)
+ FATAL("sub result2 %.30s too big; can't happen", buf);
+ /* BUG: should be able to avoid copy */
(void) setsval(x, buf);
- free(buf);
- result = true;
+ result = True;
}
- tempfree(x, "");
- tempfree(y, "");
+ tempfree(x);
+ tempfree(y);
+ free(buf);
return (result);
}
/*ARGSUSED*/
Cell *
-gsub(Node **a, int nnn)
+gsub(Node **a, int nnn) /* global substitute */
{
- register Cell *x, *y;
- register uchar *rptr, *sptr, *t;
+ Cell *x, *y;
+ uchar *rptr, *sptr, *t, *pb, *q;
uchar *buf;
- register fa *pfa;
+ fa *pfa;
int mflag, tempstat, num;
- size_t bsize, cnt, len;
+ size_t bufsz = record_size;
+ if ((buf = (uchar *)malloc(bufsz)) == NULL)
+ FATAL("out of memory in gsub");
mflag = 0; /* if mflag == 0, can replace empty string */
num = 0;
x = execute(a[3]); /* target string */
t = getsval(x);
- if (a[0] == 0)
- pfa = (fa *) a[1]; /* regular expression */
+ if (a[0] == 0) /* 0 => a[1] is already-compiled regexpr */
+ pfa = (fa *)a[1]; /* regular expression */
else {
y = execute(a[1]);
pfa = makedfa(getsval(y), 1);
- tempfree(y, "");
+ tempfree(y);
}
y = execute(a[2]); /* replacement string */
if (pmatch(pfa, t)) {
tempstat = pfa->initstat;
pfa->initstat = 2;
- init_buf(&buf, &bsize, LINE_INCR);
+ pb = buf;
rptr = getsval(y);
- cnt = 0;
do {
if (patlen == 0 && *patbeg != 0) {
/* matched empty string */
if (mflag == 0) { /* can replace empty */
num++;
sptr = rptr;
while (*sptr != 0) {
- expand_buf(&buf, &bsize, cnt);
- if (*sptr == '\\' &&
- (*(sptr+1) == '&' ||
- *(sptr+1) == '\\')) {
- sptr++;
- buf[cnt++] = *sptr++;
+ (void) adjbuf(&buf, &bufsz, 5+pb-buf,
+ record_size, &pb, "gsub");
+ if (*sptr == '\\') {
+ backsub(&pb, &sptr);
} else if (*sptr == '&') {
- expand_buf(&buf,
- &bsize,
- cnt + patlen);
sptr++;
- (void) memcpy(&buf[cnt],
- patbeg, patlen);
- cnt += patlen;
+ (void) adjbuf(&buf, &bufsz, 1 +
+ patlen+pb-buf, record_size,
+ &pb, "gsub");
+ for (q = patbeg; q < patbeg+patlen; )
+ *pb++ = *q++;
} else {
- buf[cnt++] = *sptr++;
+ *pb++ = *sptr++;
}
}
}
if (*t == 0) /* at end */
goto done;
- expand_buf(&buf, &bsize, cnt);
- buf[cnt++] = *t++;
+ (void) adjbuf(&buf, &bufsz, 2 + pb - buf,
+ record_size, &pb, "gsub");
+ *pb++ = *t++;
+ /* BUG: not sure of this test */
+ if (pb > buf + bufsz)
+ FATAL(
+ "gsub result0 %.30s too big; can't happen", buf);
mflag = 0;
} else { /* matched nonempty string */
num++;
sptr = t;
- len = patbeg - sptr;
- if (len > 0) {
- expand_buf(&buf, &bsize, cnt + len);
- (void) memcpy(&buf[cnt], sptr, len);
- cnt += len;
- }
+ (void) adjbuf(&buf, &bufsz, 1 +
+ (patbeg - sptr) + pb - buf, record_size,
+ &pb, "gsub");
+ while (sptr < patbeg)
+ *pb++ = *sptr++;
sptr = rptr;
while (*sptr != 0) {
- expand_buf(&buf, &bsize, cnt);
- if (*sptr == '\\' &&
- (*(sptr+1) == '&' ||
- *(sptr+1) == '\\')) {
- sptr++;
- buf[cnt++] = *sptr++;
+ (void) adjbuf(&buf, &bufsz, 5 + pb -
+ buf, record_size, &pb, "gsub");
+ if (*sptr == '\\') {
+ backsub(&pb, &sptr);
} else if (*sptr == '&') {
- expand_buf(&buf, &bsize,
- cnt + patlen);
sptr++;
- (void) memcpy(&buf[cnt],
- patbeg, patlen);
- cnt += patlen;
+ (void) adjbuf(&buf, &bufsz, 1 +
+ patlen + pb - buf, record_size,
+ &pb, "gsub");
+ for (q = patbeg; q < patbeg + patlen; )
+ *pb++ = *q++;
} else {
- buf[cnt++] = *sptr++;
+ *pb++ = *sptr++;
}
}
t = patbeg + patlen;
- if ((*(t-1) == 0) || (*t == 0))
+ if (patlen == 0 || *t == 0 || *(t - 1) == 0)
goto done;
+ if (pb > buf + bufsz)
+ FATAL(
+ "gsub result1 %.30s too big; can't happen", buf);
mflag = 1;
}
} while (pmatch(pfa, t));
sptr = t;
- len = strlen((char *)sptr);
- expand_buf(&buf, &bsize, len + cnt);
- (void) memcpy(&buf[cnt], sptr, len);
- cnt += len;
- done:
- buf[cnt] = '\0';
+ (void) adjbuf(&buf, &bufsz, 1 + strlen((char *)sptr) + pb -
+ buf, 0, &pb, "gsub");
+ while ((*pb++ = *sptr++) != 0)
+ ;
+ done: if (pb < buf + bufsz)
+ *pb = '\0';
+ else if (*(pb-1) != '\0')
+ FATAL("gsub result2 %.30s truncated; can't happen",
+ buf);
+ /* BUG: should be able to avoid copy + free */
(void) setsval(x, buf);
- free(buf);
pfa->initstat = tempstat;
}
- tempfree(x, "");
- tempfree(y, "");
- x = gettemp("");
+ tempfree(x);
+ tempfree(y);
+ x = gettemp();
x->tval = NUM;
x->fval = num;
+ free(buf);
return (x);
}
+
+void
+backsub(uchar **pb_ptr, uchar **sptr_ptr) /* handle \\& variations */
+{ /* sptr[0] == '\\' */
+ uchar *pb = *pb_ptr, *sptr = *sptr_ptr;
+
+ if (sptr[1] == '\\') {
+ if (sptr[2] == '\\' && sptr[3] == '&') { /* \\\& -> \& */
+ *pb++ = '\\';
+ *pb++ = '&';
+ sptr += 4;
+ } else if (sptr[2] == '&') { /* \\& -> \ + matched */
+ *pb++ = '\\';
+ sptr += 2;
+ } else { /* \\x -> \\x */
+ *pb++ = *sptr++;
+ *pb++ = *sptr++;
+ }
+ } else if (sptr[1] == '&') { /* literal & */
+ sptr++;
+ *pb++ = *sptr++;
+ } else /* literal \ */
+ *pb++ = *sptr++;
+
+ *pb_ptr = pb;
+ *sptr_ptr = sptr;
+}