/usr/src/usr.bin/awk/run.c

Bug Summary

File:	src/usr.bin/awk/run.c
Warning:	line 2648, column 26 Null pointer passed as 1st argument to string length function
Annotated Source Code

Press '?' to see keyboard shortcuts
Show analyzer invocation
clang -cc1 -cc1 -triple amd64-unknown-openbsd7.4 -analyze -disable-free -clear-ast-before-backend -disable-llvm-verifier -discard-value-names -main-file-name run.c -analyzer-checker=core -analyzer-checker=apiModeling -analyzer-checker=unix -analyzer-checker=deadcode -analyzer-checker=security.insecureAPI.UncheckedReturn -analyzer-checker=security.insecureAPI.getpw -analyzer-checker=security.insecureAPI.gets -analyzer-checker=security.insecureAPI.mktemp -analyzer-checker=security.insecureAPI.mkstemp -analyzer-checker=security.insecureAPI.vfork -analyzer-checker=nullability.NullPassedToNonnull -analyzer-checker=nullability.NullReturnedFromNonnull -analyzer-output plist -w -setup-static-analyzer -mrelocation-model pic -pic-level 1 -pic-is-pie -mframe-pointer=all -relaxed-aliasing -ffp-contract=on -fno-rounding-math -mconstructor-aliases -funwind-tables=2 -target-cpu x86-64 -target-feature +retpoline-indirect-calls -target-feature +retpoline-indirect-branches -tune-cpu generic -debugger-tuning=gdb -fcoverage-compilation-dir=/usr/src/usr.bin/awk/obj -resource-dir /usr/local/llvm16/lib/clang/16 -I . -I /usr/src/usr.bin/awk -D HAS_ISBLANK -D NDEBUG -internal-isystem /usr/local/llvm16/lib/clang/16/include -internal-externc-isystem /usr/include -O2 -fdebug-compilation-dir=/usr/src/usr.bin/awk/obj -ferror-limit 19 -fwrapv -D_RET_PROTECTOR -ret-protector -fcf-protection=branch -fno-jump-tables -fgnuc-version=4.2.1 -vectorize-loops -vectorize-slp -fno-builtin-malloc -fno-builtin-calloc -fno-builtin-realloc -fno-builtin-valloc -fno-builtin-free -fno-builtin-strdup -fno-builtin-strndup -analyzer-output=html -faddrsig -D__GCC_HAVE_DWARF2_CFI_ASM=1 -o /home/ben/Projects/scan/2024-01-11-140451-98009-1 -x c /usr/src/usr.bin/awk/run.c
1/*	$OpenBSD: run.c,v 1.83 2023/11/28 20:54:38 millert Exp $	*/
2/****************************************************************
3Copyright (C) Lucent Technologies 1997
4All Rights Reserved

6Permission to use, copy, modify, and distribute this software and
7its documentation for any purpose and without fee is hereby
8granted, provided that the above copyright notice appear in all
9copies and that both that the copyright notice and this
10permission notice and warranty disclaimer appear in supporting
11documentation, and that the name Lucent Technologies or any of
12its entities not be used in advertising or publicity pertaining
13to distribution of the software without specific, written prior
14permission.

16LUCENT DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
17INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS.
18IN NO EVENT SHALL LUCENT OR ANY OF ITS ENTITIES BE LIABLE FOR ANY
19SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
20WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER
21IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,
22ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF
23THIS SOFTWARE.
24****************************************************************/

26#define DEBUG
27#include <stdio.h>
28#include <ctype.h>
29#include <errno(*__errno()).h>
30#include <wctype.h>
31#include <fcntl.h>
32#include <setjmp.h>
33#include <limits.h>
34#include <math.h>
35#include <string.h>
36#include <stdlib.h>
37#include <time.h>
38#include <sys/types.h>
39#include <sys/wait.h>
40#include "awk.h"
41#include "awkgram.tab.h"


44static void stdinit(void);
45static void flush_all(void);
46static char *wide_char_to_byte_str(int rune, size_t *outlen);

48#if 1
49#define tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0)	do { if (istemp(x)((x)->csub == 4)) tfree(x); } while (/*CONSTCOND*/0)
50#else
51void tempfree(Cell *p)do { if (((Cell *p)->csub == 4)) tfree(Cell *p); } while (
 0) {
if (p->ctype == OCELL1 && (p->csub < CUNK0 || p->csub > CFREE7)) {
WARNING("bad csub %d in Cell %d %s",
	p->csub, p->ctype, p->sval);
}
if (istemp(p)((p)->csub == 4))
tfree(p);
58}
59#endif

61/* do we really need these? */
62/* #ifdef _NFILE */
63/* #ifndef FOPEN_MAX */
64/* #define FOPEN_MAX _NFILE */
65/* #endif */
66/* #endif */
67/*  */
68/* #ifndef	FOPEN_MAX */
69/* #define	FOPEN_MAX	40 */	/* max number of open files */
70/* #endif */
71/*  */
72/* #ifndef RAND_MAX */
73/* #define RAND_MAX	32767 */	/* all that ansi guarantees */
74/* #endif */

76jmp_buf env;
77extern	int	pairstack[];
78extern	Awkfloat	srand_seed;

80Node	*winner = NULL((void *)0);	/* root of parse tree */
81Cell	*tmps;		/* free temporary cells for execution */

83static Cell	truecell	={ OBOOL2, BTRUE11, 0, 0, 1.0, NUM01, NULL((void *)0), NULL((void *)0) };
84Cell	*True	= &truecell;
85static Cell	falsecell	={ OBOOL2, BFALSE12, 0, 0, 0.0, NUM01, NULL((void *)0), NULL((void *)0) };
86Cell	*False	= &falsecell;
87static Cell	breakcell	={ OJUMP3, JBREAK23, 0, 0, 0.0, NUM01, NULL((void *)0), NULL((void *)0) };
88Cell	*jbreak	= &breakcell;
89static Cell	contcell	={ OJUMP3, JCONT24, 0, 0, 0.0, NUM01, NULL((void *)0), NULL((void *)0) };
90Cell	*jcont	= &contcell;
91static Cell	nextcell	={ OJUMP3, JNEXT22, 0, 0, 0.0, NUM01, NULL((void *)0), NULL((void *)0) };
92Cell	*jnext	= &nextcell;
93static Cell	nextfilecell	={ OJUMP3, JNEXTFILE26, 0, 0, 0.0, NUM01, NULL((void *)0), NULL((void *)0) };
94Cell	*jnextfile	= &nextfilecell;
95static Cell	exitcell	={ OJUMP3, JEXIT21, 0, 0, 0.0, NUM01, NULL((void *)0), NULL((void *)0) };
96Cell	*jexit	= &exitcell;
97static Cell	retcell		={ OJUMP3, JRET25, 0, 0, 0.0, NUM01, NULL((void *)0), NULL((void *)0) };
98Cell	*jret	= &retcell;
99static Cell	tempcell	={ OCELL1, CTEMP4, 0, EMPTY, 0.0, NUM01|STR02|DONTFREE04, NULL((void *)0), NULL((void *)0) };

101Node	*curnode = NULL((void *)0);	/* the node being executed, for debugging */

103/* buffer memory management */
104int adjbuf(char **pbuf, int *psiz, int minlen, int quantum, char **pbptr,
const char *whatrtn)
106/* 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
*/
115{
if (minlen > *psiz) {
char *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 = (char *) realloc(*pbuf, minlen);
DPRINTF("adjbuf %s: %d %d (pbuf=%p, tbuf=%p)\n", whatrtn, *psiz, minlen, (void*)*pbuf, (void*)tbuf)if (dbg) printf("adjbuf %s: %d %d (pbuf=%p, tbuf=%p)\n", whatrtn
, *psiz, minlen, (void*)*pbuf, (void*)tbuf);
if (tbuf == NULL((void *)0)) {
	if (whatrtn)
		FATAL("out of memory in %s", whatrtn);
	return 0;
}
*pbuf = tbuf;
*psiz = minlen;
if (pbptr)
	*pbptr = tbuf + boff;
}
return 1;
136}

138void run(Node *a)	/* execution of parse tree starts here */
139{

stdinit();
execute(a);
closeall();
144}

146Cell *execute(Node *u)	/* execute a node of the parse tree */
147{
Cell *(*proc)(Node **, int);
Cell *x;
Node *a;

if (u == NULL((void *)0))
return(True);
for (a = u; ; a = a->nnext) {
curnode = a;
if (isvalue(a)((a)->ntype == 1)) {
	x = (Cell *) (a->narg[0]);
	if (isfld(x)((x)->tval & 0100) && !donefld)
		fldbld();
	else if (isrec(x)((x)->tval & 0200) && !donerec)
		recbld();
	return(x);
}
if (notlegal(a->nobj)(a->nobj <= 257 || a->nobj >= 352 || proctab[a->
nobj-257] == nullproc))	/* probably a Cell* but too risky to print */
	FATAL("illegal statement");
proc = proctab[a->nobj-FIRSTTOKEN257];
x = (*proc)(a->narg, a->nobj);
if (isfld(x)((x)->tval & 0100) && !donefld)
	fldbld();
else if (isrec(x)((x)->tval & 0200) && !donerec)
	recbld();
if (isexpr(a)((a)->ntype == 3))
	return(x);
if (isjump(x)((x)->ctype == 3))
	return(x);
if (a->nnext == NULL((void *)0))
	return(x);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
180}


183Cell *program(Node **a, int n)	/* execute an awk program */
184{				/* 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)((x)->csub == 21))
	return(True);
if (isjump(x)((x)->ctype == 3))
	FATAL("illegal break, continue, next or nextfile from BEGIN");
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
if (a[1] || a[2])
while (getrec(&record, &recsize, true1) > 0) {
	x = execute(a[1]);
	if (isexit(x)((x)->csub == 21))
		break;
	tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
ex:
if (setjmp(env) != 0)	/* handles exit within END */
goto ex1;
if (a[2]) {		/* END */
x = execute(a[2]);
if (isbreak(x)((x)->csub == 23) || isnext(x)((x)->csub == 22 || (x)->csub == 26) || iscont(x)((x)->csub == 24))
	FATAL("illegal break, continue, next or nextfile from END");
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
ex1:
return(True);
215}

217struct 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 */
222};

224#define	NARGS50	50	/* max args in a call */

226struct Frame *frame = NULL((void *)0);	/* base of stack frames; dynamically allocated */
227int	nframe = 0;		/* number of frames allocated */
228struct Frame *frp = NULL((void *)0);	/* frame pointer. bottom level unused */

230Cell *call(Node **a, int n)	/* function call.  very kludgy and fragile */
231{
static const Cell newcopycell = { OCELL1, CCOPY6, 0, EMPTY, 0.0, NUM01|STR02|DONTFREE04, NULL((void *)0), NULL((void *)0) };
int i, ncall, ndef;
int freed = 0; /* handles potential double freeing when fcn & param share a tempcell */
Node *x;
Cell *args[NARGS50], *oargs[NARGS50];	/* BUG: fixed size arrays */
Cell *y, *z, *fcn;
char *s;

fcn = execute(a[0]);	/* the function itself */
s = fcn->nval;
if (!isfcn(fcn)((fcn)->tval & 040))
FATAL("calling undefined function %s", s);
if (frame == NULL((void *)0)) {
frp = frame = (struct Frame *) calloc(nframe += 100, sizeof(*frame));
if (frame == NULL((void *)0))
	FATAL("out of space for stack frames calling %s", s);
}
for (ncall = 0, x = a[1]; x != NULL((void *)0); x = x->nnext)	/* args in call */
ncall++;
ndef = (int) fcn->fval;			/* args in defn */
DPRINTF("calling %s, %d args (%d in defn), frp=%d\n", s, ncall, ndef, (int) (frp-frame))if (dbg) printf("calling %s, %d args (%d in defn), frp=%d\n",
 s, ncall, ndef, (int) (frp-frame));
if (ncall > ndef)
WARNING("function %s called with %d args, uses only %d",
	s, ncall, ndef);
if (ncall + ndef > NARGS50)
FATAL("function %s has %d arguments, limit %d", s, ncall+ndef, NARGS50);
for (i = 0, x = a[1]; x != NULL((void *)0); i++, x = x->nnext) {	/* get call args */
DPRINTF("evaluate args[%d], frp=%d:\n", i, (int) (frp-frame))if (dbg) printf("evaluate args[%d], frp=%d:\n", i, (int) (frp
-frame));
y = execute(x);
oargs[i] = y;
DPRINTF("args[%d]: %s %f <%s>, t=%o\n",if (dbg) printf("args[%d]: %s %f <%s>, t=%o\n", i, ((y->
nval) ? (y->nval) : "(null)"), y->fval, ((y)->tval &
 020) ? "(array)" : ((y->sval) ? (y->sval) : "(null)"),
 y->tval)
	i, NN(y->nval), y->fval, isarr(y) ? "(array)" : NN(y->sval), y->tval)if (dbg) printf("args[%d]: %s %f <%s>, t=%o\n", i, ((y->
nval) ? (y->nval) : "(null)"), y->fval, ((y)->tval &
 020) ? "(array)" : ((y->sval) ? (y->sval) : "(null)"),
 y->tval);
if (isfcn(y)((y)->tval & 040))
	FATAL("can't use function %s as argument in %s", y->nval, s);
if (isarr(y)((y)->tval & 020))
	args[i] = y;	/* arrays by ref */
else
	args[i] = copycell(y);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
}
for ( ; i < ndef; i++) {	/* add null args for ones not provided */
args[i] = gettemp();
*args[i] = newcopycell;
}
frp++;	/* now ok to up frame */
if (frp >= frame + nframe) {
int dfp = frp - frame;	/* old index */
frame = (struct Frame *) reallocarray(frame, (nframe += 100), sizeof(*frame));
if (frame == NULL((void *)0))
	FATAL("out of space for stack frames in %s", s);
frp = frame + dfp;
}
frp->fcncell = fcn;
frp->args = args;
frp->nargs = ndef;	/* number defined with (excess are locals) */
frp->retval = gettemp();

DPRINTF("start exec of %s, frp=%d\n", s, (int) (frp-frame))if (dbg) printf("start exec of %s, frp=%d\n", s, (int) (frp-frame
));
y = execute((Node *)(fcn->sval));	/* execute body */
DPRINTF("finished exec of %s, frp=%d\n", s, (int) (frp-frame))if (dbg) printf("finished exec of %s, frp=%d\n", s, (int) (frp
-frame));

for (i = 0; i < ndef; i++) {
Cell *t = frp->args[i];
if (isarr(t)((t)->tval & 020)) {
	if (t->csub == CCOPY6) {
		if (i >= ncall) {
			freesymtab(t);
			t->csub = CTEMP4;
			tempfree(t)do { if (((t)->csub == 4)) tfree(t); } while ( 0);
		} else {
			oargs[i]->tval = t->tval;
			oargs[i]->tval &= ~(STR02|NUM01|DONTFREE04);
			oargs[i]->sval = t->sval;
			tempfree(t)do { if (((t)->csub == 4)) tfree(t); } while ( 0);
		}
	}
} else if (t != y) {	/* kludge to prevent freeing twice */
	t->csub = CTEMP4;
	tempfree(t)do { if (((t)->csub == 4)) tfree(t); } while ( 0);
} else if (t == y && t->csub == CCOPY6) {
	t->csub = CTEMP4;
	tempfree(t)do { if (((t)->csub == 4)) tfree(t); } while ( 0);
	freed = 1;
}
}
tempfree(fcn)do { if (((fcn)->csub == 4)) tfree(fcn); } while ( 0);
if (isexit(y)((y)->csub == 21) || isnext(y)((y)->csub == 22 || (y)->csub == 26))
return y;
if (freed == 0) {
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);	/* don't free twice! */
}
z = frp->retval;			/* return value */
DPRINTF("%s returns %g |%s| %o\n", s, getfval(z), getsval(z), z->tval)if (dbg) printf("%s returns %g |%s| %o\n", s, getfval(z), getsval
(z), z->tval);
frp--;
return(z);
327}

329Cell *copycell(Cell *x)	/* make a copy of a cell in a temp */
330{
Cell *y;

/* copy is not constant or field */

y = gettemp();
y->tval = x->tval & ~(CON010|FLD0100|REC0200);
y->csub = CCOPY6;	/* prevents freeing until call is over */
y->nval = x->nval;	/* BUG? */
if (isstr(x)((x)->tval & 02) /* || x->ctype == OCELL */) {
y->sval = tostring(x->sval);
y->tval &= ~DONTFREE04;
} else
y->tval |= DONTFREE04;
y->fval = x->fval;
return y;
346}

348Cell *arg(Node **a, int n)	/* nth argument of a function */
349{

n = ptoi(a[0]);	/* argument number, counting from 0 */
DPRINTF("arg(%d), frp->nargs=%d\n", n, frp->nargs)if (dbg) printf("arg(%d), frp->nargs=%d\n", n, frp->nargs
);
if (n+1 > frp->nargs)
FATAL("argument #%d of function %s was not supplied",
	n+1, frp->fcncell->nval);
return frp->args[n];
357}

359Cell *jump(Node **a, int n)	/* break, continue, next, nextfile, return */
360{
Cell *y;

switch (n) {
case EXIT297:
if (a[0] != NULL((void *)0)) {
	y = execute(a[0]);
	errorflag = (int) getfval(y);
	tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
}
longjmp(env, 1);
case RETURN340:
if (a[0] != NULL((void *)0)) {
	y = execute(a[0]);
	if ((y->tval & (STR02|NUM01)) == (STR02|NUM01)) {
		setsval(frp->retval, getsval(y));
		frp->retval->fval = getfval(y);
		frp->retval->tval |= NUM01;
	}
	else if (y->tval & STR02)
		setsval(frp->retval, getsval(y));
	else if (y->tval & NUM01)
		setfval(frp->retval, getfval(y));
	else		/* can't happen */
		FATAL("bad type variable %d", y->tval);
	tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
}
return(jret);
case NEXT307:
return(jnext);
case NEXTFILE308:
nextfile();
return(jnextfile);
case BREAK292:
return(jbreak);
case CONTINUE294:
return(jcont);
default:	/* can't happen */
FATAL("illegal jump type %d", n);
}
return 0;	/* not reached */
401}

403Cell *awkgetline(Node **a, int n)	/* get next line from specific input */
404{		/* a[0] is variable, a[1] is operator, a[2] is filename */
Cell *r, *x;
extern Cell **fldtab;
FILE *fp;
char *buf;
int bufsize = recsize;
int mode;
bool_Bool newflag;
double result;

if ((buf = (char *) malloc(bufsize)) == NULL((void *)0))
FATAL("out of memory in getline");

fflush(stdout(&__sF[1]));	/* in case someone is waiting for a prompt */
r = gettemp();
if (a[1] != NULL((void *)0)) {		/* getline < file */
x = execute(a[2]);		/* filename */
mode = ptoi(a[1]);
if (mode == '|')		/* input pipe */
	mode = LE286;	/* arbitrary flag */
fp = openfile(mode, getsval(x), &newflag);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
if (fp == NULL((void *)0))
	n = -1;
else
	n = readrec(&buf, &bufsize, fp, newflag);
if (n <= 0) {
	;
} else if (a[0] != NULL((void *)0)) {	/* getline var <file */
	x = execute(a[0]);
	setsval(x, buf);
	if (is_number(x->sval, & result)is_valid_number(x->sval, 0, ((void *)0), & result)) {
		x->fval = result;
		x->tval |= NUM01;
	}
	tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
} else {			/* getline <file */
	setsval(fldtab[0], buf);
	if (is_number(fldtab[0]->sval, & result)is_valid_number(fldtab[0]->sval, 0, ((void *)0), & result
)) {
		fldtab[0]->fval = result;
		fldtab[0]->tval |= NUM01;
	}
}
} else {			/* bare getline; use current input */
if (a[0] == NULL((void *)0))	/* getline */
	n = getrec(&record, &recsize, true1);
else {			/* getline var */
	n = getrec(&buf, &bufsize, false0);
	if (n > 0) {
		x = execute(a[0]);
		setsval(x, buf);
		if (is_number(x->sval, & result)is_valid_number(x->sval, 0, ((void *)0), & result)) {
			x->fval = result;
			x->tval |= NUM01;
		}
		tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
	}
}
}
setfval(r, (Awkfloat) n);
free(buf);
return r;
466}

468Cell *getnf(Node **a, int n)	/* get NF */
469{
if (!donefld)
fldbld();
return (Cell *) a[0];
473}

475static char *
476makearraystring(Node *p, const char *func)
477{
char *buf;
int bufsz = recsize;
size_t blen;

if ((buf = (char *) malloc(bufsz)) == NULL((void *)0)) {
FATAL("%s: out of memory", func);
}

blen = 0;
buf[blen] = '\0';

for (; p; p = p->nnext) {
Cell *x = execute(p);	/* expr */
char *s = getsval(x);
size_t seplen = strlen(getsval(subseploc));
size_t nsub = p->nnext ? seplen : 0;
size_t slen = strlen(s);
size_t tlen = blen + slen + nsub;

if (!adjbuf(&buf, &bufsz, tlen + 1, recsize, 0, func)) {
	FATAL("%s: out of memory %s[%s...]",
	    func, x->nval, buf);
}
memcpy(buf + blen, s, slen);
if (nsub) {
	memcpy(buf + blen + slen, *SUBSEP, nsub);
}
buf[tlen] = '\0';
blen = tlen;
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
return buf;
510}

512Cell *array(Node **a, int n)	/* a[0] is symtab, a[1] is list of subscripts */
513{
Cell *x, *z;
char *buf;

x = execute(a[0]);	/* Cell* for symbol table */
buf = makearraystring(a[1], __func__);
if (!isarr(x)((x)->tval & 020)) {
DPRINTF("making %s into an array\n", NN(x->nval))if (dbg) printf("making %s into an array\n", ((x->nval) ? (
x->nval) : "(null)"));
if (freeable(x)( ((x)->tval & (02|04)) == 02 ))
	xfree(x->sval){ free((void *)(intptr_t)(x->sval)); (x->sval) = ((void
 *)0); };
x->tval &= ~(STR02|NUM01|DONTFREE04);
x->tval |= ARR020;
x->sval = (char *) makesymtab(NSYMTAB50);
}
z = setsymtab(buf, "", 0.0, STR02|NUM01, (Array *) x->sval);
z->ctype = OCELL1;
z->csub = CVAR2;
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
free(buf);
return(z);
533}

535Cell *awkdelete(Node **a, int n)	/* a[0] is symtab, a[1] is list of subscripts */
536{
Cell *x;

x = execute(a[0]);	/* Cell* for symbol table */
if (x == symtabloc) {
FATAL("cannot delete SYMTAB or its elements");
}
if (!isarr(x)((x)->tval & 020))
return True;
if (a[1] == NULL((void *)0)) {	/* delete the elements, not the table */
freesymtab(x);
x->tval &= ~STR02;
x->tval |= ARR020;
x->sval = (char *) makesymtab(NSYMTAB50);
} else {
char *buf = makearraystring(a[1], __func__);
freeelem(x, buf);
free(buf);
}
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
return True;
557}

559Cell *intest(Node **a, int n)	/* a[0] is index (list), a[1] is symtab */
560{
Cell *ap, *k;
char *buf;

ap = execute(a[1]);	/* array name */
if (!isarr(ap)((ap)->tval & 020)) {
DPRINTF("making %s into an array\n", ap->nval)if (dbg) printf("making %s into an array\n", ap->nval);
if (freeable(ap)( ((ap)->tval & (02|04)) == 02 ))
	xfree(ap->sval){ free((void *)(intptr_t)(ap->sval)); (ap->sval) = ((void
 *)0); };
ap->tval &= ~(STR02|NUM01|DONTFREE04);
ap->tval |= ARR020;
ap->sval = (char *) makesymtab(NSYMTAB50);
}
buf = makearraystring(a[0], __func__);
k = lookup(buf, (Array *) ap->sval);
tempfree(ap)do { if (((ap)->csub == 4)) tfree(ap); } while ( 0);
free(buf);
if (k == NULL((void *)0))
return(False);
else
return(True);
581}


584/* ======== utf-8 code ========== */

586/*
* Awk strings can contain ascii, random 8-bit items (eg Latin-1),
* or utf-8.  u8_isutf tests whether a string starts with a valid
* utf-8 sequence, and returns 0 if not (e.g., high bit set).
* u8_nextlen returns length of next valid sequence, which is
* 1 for ascii, 2..4 for utf-8, or 1 for high bit non-utf.
* u8_strlen returns length of string in valid utf-8 sequences
* and/or high-bit bytes.  Conversion functions go between byte
* number and character number.
*
* In theory, this behaves the same as before for non-utf8 bytes.
*
* Limited checking! This is a potential security hole.
*/

601/* is s the beginning of a valid utf-8 string? */
602/* return length 1..4 if yes, 0 if no */
603int u8_isutf(const char *s)
604{
int n, ret;
unsigned char c;

c = s[0];
if (c < 128 || awk_mb_cur_max == 1)
return 1; /* what if it's 0? */

n = strlen(s);
if (n >= 2 && ((c>>5) & 0x7) == 0x6 && (s[1] & 0xC0) == 0x80) {
ret = 2; /* 110xxxxx 10xxxxxx */
} else if (n >= 3 && ((c>>4) & 0xF) == 0xE && (s[1] & 0xC0) == 0x80
	 && (s[2] & 0xC0) == 0x80) {
ret = 3; /* 1110xxxx 10xxxxxx 10xxxxxx */
} else if (n >= 4 && ((c>>3) & 0x1F) == 0x1E && (s[1] & 0xC0) == 0x80
	 && (s[2] & 0xC0) == 0x80 && (s[3] & 0xC0) == 0x80) {
ret = 4; /* 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
} else {
ret = 0;
}
return ret;
625}

627/* Convert (prefix of) utf8 string to utf-32 rune. */
628/* Sets *rune to the value, returns the length. */
629/* No error checking: watch out. */
630int u8_rune(int *rune, const char *s)
631{
int n, ret;
unsigned char c;

c = s[0];
if (c < 128 || awk_mb_cur_max == 1) {
*rune = c;
return 1;
}

n = strlen(s);
if (n >= 2 && ((c>>5) & 0x7) == 0x6 && (s[1] & 0xC0) == 0x80) {
*rune = ((c & 0x1F) << 6) | (s[1] & 0x3F); /* 110xxxxx 10xxxxxx */
ret = 2;
} else if (n >= 3 && ((c>>4) & 0xF) == 0xE && (s[1] & 0xC0) == 0x80
	  && (s[2] & 0xC0) == 0x80) {
*rune = ((c & 0xF) << 12) | ((s[1] & 0x3F) << 6) | (s[2] & 0x3F);
	/* 1110xxxx 10xxxxxx 10xxxxxx */
ret = 3;
} else if (n >= 4 && ((c>>3) & 0x1F) == 0x1E && (s[1] & 0xC0) == 0x80
	  && (s[2] & 0xC0) == 0x80 && (s[3] & 0xC0) == 0x80) {
*rune = ((c & 0x7) << 18) | ((s[1] & 0x3F) << 12) | ((s[2] & 0x3F) << 6) | (s[3] & 0x3F);
	/* 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
ret = 4;
} else {
*rune = c;
ret = 1;
}
return ret; /* returns one byte if sequence doesn't look like utf */
660}

662/* return length of next sequence: 1 for ascii or random, 2..4 for valid utf8 */
663int u8_nextlen(const char *s)
664{
int len;

len = u8_isutf(s);
if (len == 0)
len = 1;
return len;
671}

673/* return number of utf characters or single non-utf bytes */
674int u8_strlen(const char *s)
675{
int i, len, n, totlen;
unsigned char c;

n = strlen(s);
totlen = 0;
for (i = 0; i < n; i += len) {
c = s[i];
if (c < 128 || awk_mb_cur_max == 1) {
	len = 1;
} else {
	len = u8_nextlen(&s[i]);
}
totlen++;
if (i > n)
	FATAL("bad utf count [%s] n=%d i=%d\n", s, n, i);
}
return totlen;
693}

695/* convert utf-8 char number in a string to its byte offset */
696int u8_char2byte(const char *s, int charnum)
697{
int n;
int bytenum = 0;

while (charnum > 0) {
n = u8_nextlen(s);
s += n;
bytenum += n;
charnum--;
}
return bytenum;
708}

710/* convert byte offset in s to utf-8 char number that starts there */
711int u8_byte2char(const char *s, int bytenum)
712{
int i, len, b;
int charnum = 0; /* BUG: what origin? */
/* should be 0 to match start==0 which means no match */	

b = strlen(s);
if (bytenum > b) {
return -1; /* ??? */
}
for (i = 0; i <= bytenum; i += len) {
len = u8_nextlen(s+i);
charnum++;
}
return charnum;
726}

728/* runetochar() adapted from rune.c in the Plan 9 distributione */

730enum
731{
Runeerror = 128, /* from somewhere else */
Runemax = 0x10FFFF,

Bit1    = 7,
Bitx    = 6,
Bit2    = 5,
Bit3    = 4,
Bit4    = 3,
Bit5    = 2,

T1      = ((1<<(Bit1+1))-1) ^ 0xFF,     /* 0000 0000 */
Tx      = ((1<<(Bitx+1))-1) ^ 0xFF,     /* 1000 0000 */
T2      = ((1<<(Bit2+1))-1) ^ 0xFF,     /* 1100 0000 */
T3      = ((1<<(Bit3+1))-1) ^ 0xFF,     /* 1110 0000 */
T4      = ((1<<(Bit4+1))-1) ^ 0xFF,     /* 1111 0000 */
T5      = ((1<<(Bit5+1))-1) ^ 0xFF,     /* 1111 1000 */

Rune1   = (1<<(Bit1+0*Bitx))-1,	 	/* 0000 0000 0000 0000 0111 1111 */
Rune2   = (1<<(Bit2+1*Bitx))-1,	 	/* 0000 0000 0000 0111 1111 1111 */
Rune3   = (1<<(Bit3+2*Bitx))-1,	 	/* 0000 0000 1111 1111 1111 1111 */
Rune4   = (1<<(Bit4+3*Bitx))-1,	 	/* 0011 1111 1111 1111 1111 1111 */

Maskx   = (1<<Bitx)-1,		  	/* 0011 1111 */
Testx   = Maskx ^ 0xFF,		 	/* 1100 0000 */

757};

759int runetochar(char *str, int c)
760{	
/* one character sequence 00000-0007F => 00-7F */     
if (c <= Rune1) {
str[0] = c;
return 1;
}

/* two character sequence 00080-007FF => T2 Tx */
if (c <= Rune2) {
str[0] = T2 | (c >> 1*Bitx);
str[1] = Tx | (c & Maskx);
return 2;
}

/* three character sequence 00800-0FFFF => T3 Tx Tx */
if (c > Runemax)
c = Runeerror;
if (c <= Rune3) {
str[0] = T3 |  (c >> 2*Bitx);
str[1] = Tx | ((c >> 1*Bitx) & Maskx);
str[2] = Tx |  (c & Maskx);
return 3;
}

/* four character sequence 010000-1FFFFF => T4 Tx Tx Tx */
str[0] = T4 |  (c >> 3*Bitx);
str[1] = Tx | ((c >> 2*Bitx) & Maskx);
str[2] = Tx | ((c >> 1*Bitx) & Maskx);
str[3] = Tx |  (c & Maskx);
return 4;
790}               


793/* ========== end of utf8 code =========== */



797Cell *matchop(Node **a, int n)	/* ~ and match() */
798{
Cell *x, *y;
char *s, *t;
int i;
int cstart, cpatlen, len;
fa *pfa;
int (*mf)(fa *, const char *) = match, mode = 0;

if (n == MATCHFCN306) {
mf = pmatch;
mode = 1;
}
x = execute(a[1]);	/* a[1] = target text */
s = getsval(x);
if (a[0] == NULL((void *)0))	/* a[1] == 0: already-compiled reg expr */
i = (*mf)((fa *) a[2], s);
else {
y = execute(a[2]);	/* a[2] = regular expr */
t = getsval(y);
pfa = makedfa(t, mode);
i = (*mf)(pfa, s);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
}
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
if (n == MATCHFCN306) {
int start = patbeg - s + 1; /* origin 1 */
if (patlen < 0) {
	start = 0; /* not found */
} else {
	cstart = u8_byte2char(s, start-1);
	cpatlen = 0;
	for (i = 0; i < patlen; i += len) {
		len = u8_nextlen(patbeg+i);
		cpatlen++;
	}

	start = cstart;
	patlen = cpatlen;
}

setfval(rstartloc, (Awkfloat) start);
setfval(rlengthloc, (Awkfloat) patlen);
x = gettemp();
x->tval = NUM01;
x->fval = start;
return x;
} else if ((n == MATCH265 && i == 1) || (n == NOTMATCH266 && i == 0))
return(True);
else
return(False);
848}


851Cell *boolop(Node **a, int n)	/* a[0] || a[1], a[0] && a[1], !a[0] */
852{
Cell *x, *y;
int i;

x = execute(a[0]);
i = istrue(x)((x)->csub == 11);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
switch (n) {
case BOR281:
if (i) return(True);
y = execute(a[1]);
i = istrue(y)((y)->csub == 11);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
if (i) return(True);
else return(False);
case AND280:
if ( !i ) return(False);
y = execute(a[1]);
i = istrue(y)((y)->csub == 11);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
if (i) return(True);
else return(False);
case NOT345:
if (i) return(False);
else return(True);
default:	/* can't happen */
FATAL("unknown boolean operator %d", n);
}
return 0;	/*NOTREACHED*/
881}

883Cell *relop(Node **a, int n)	/* a[0 < a[1], etc. */
884{
int i;
Cell *x, *y;
Awkfloat j;
bool_Bool x_is_nan, y_is_nan;

x = execute(a[0]);
y = execute(a[1]);
x_is_nan = isnan(x->fval)((sizeof (x->fval) == sizeof (float)) ? __isnanf(x->fval
) : (sizeof (x->fval) == sizeof (double)) ? __isnan(x->
fval) : __isnanl(x->fval));
y_is_nan = isnan(y->fval)((sizeof (y->fval) == sizeof (float)) ? __isnanf(y->fval
) : (sizeof (y->fval) == sizeof (double)) ? __isnan(y->
fval) : __isnanl(y->fval));
if (x->tval&NUM01 && y->tval&NUM01) {
if ((x_is_nan || y_is_nan) && n != NE288)
	return(False);
j = x->fval - y->fval;
i = j<0? -1: (j>0? 1: 0);
} else {
i = strcmp(getsval(x), getsval(y));
}
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
switch (n) {
case LT287:	if (i<0) return(True);
	else return(False);
case LE286:	if (i<=0) return(True);
	else return(False);
case NE288:	if (x_is_nan && y_is_nan) return(True);
	else if (i!=0) return(True);
	else return(False);
case EQ283:	if (i == 0) return(True);
	else return(False);
case GE284:	if (i>=0) return(True);
	else return(False);
case GT285:	if (i>0) return(True);
	else return(False);
default:	/* can't happen */
FATAL("unknown relational operator %d", n);
}
return 0;	/*NOTREACHED*/
922}

924void tfree(Cell *a)	/* free a tempcell */
925{
if (freeable(a)( ((a)->tval & (02|04)) == 02 )) {
DPRINTF("freeing %s %s %o\n", NN(a->nval), NN(a->sval), a->tval)if (dbg) printf("freeing %s %s %o\n", ((a->nval) ? (a->
nval) : "(null)"), ((a->sval) ? (a->sval) : "(null)"), a
->tval);
xfree(a->sval){ free((void *)(intptr_t)(a->sval)); (a->sval) = ((void
 *)0); };
}
if (a == tmps)
FATAL("tempcell list is curdled");
a->cnext = tmps;
tmps = a;
934}

936Cell *gettemp(void)	/* get a tempcell */
937{	int i;
Cell *x;

if (!tmps) {
tmps = (Cell *) calloc(100, sizeof(*tmps));
if (!tmps)
	FATAL("out of space for temporaries");
for (i = 1; i < 100; i++)
	tmps[i-1].cnext = &tmps[i];
tmps[i-1].cnext = NULL((void *)0);
}
x = tmps;
tmps = x->cnext;
*x = tempcell;
return(x);
952}

954Cell *indirect(Node **a, int n)	/* $( a[0] ) */
955{
Awkfloat val;
Cell *x;
int m;
char *s;

x = execute(a[0]);
val = getfval(x);	/* freebsd: defend against super large field numbers */
if ((Awkfloat)INT_MAX0x7fffffff < val)
FATAL("trying to access out of range field %s", x->nval);
m = (int) val;
if (m == 0 && !is_number(s = getsval(x), NULL)is_valid_number(s = getsval(x), 0, ((void *)0), ((void *)0)))	/* suspicion! */
FATAL("illegal field $(%s), name \"%s\"", s, x->nval);
/* BUG: can x->nval ever be null??? */
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = fieldadr(m);
x->ctype = OCELL1;	/* BUG?  why are these needed? */
x->csub = CFLD1;
return(x);
974}

976Cell *substr(Node **a, int nnn)		/* substr(a[0], a[1], a[2]) */
977{
int k, m, n;
int mb, nb;
char *s;
int temp;
Cell *x, *y, *z = NULL((void *)0);

x = execute(a[0]);
y = execute(a[1]);
if (a[2] != NULL((void *)0))
z = execute(a[2]);
s = getsval(x);
k = u8_strlen(s) + 1;
if (k <= 1) {
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
if (a[2] != NULL((void *)0)) {
	tempfree(z)do { if (((z)->csub == 4)) tfree(z); } while ( 0);
}
x = gettemp();
setsval(x, "");
return(x);
}
m = (int) getfval(y);
if (m <= 0)
m = 1;
else if (m > k)
m = k;
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
if (a[2] != NULL((void *)0)) {
n = (int) getfval(z);
tempfree(z)do { if (((z)->csub == 4)) tfree(z); } while ( 0);
} else
n = k - 1;
if (n < 0)
n = 0;
else if (n > k - m)
n = k - m;
/* m is start, n is length from there */
DPRINTF("substr: m=%d, n=%d, s=%s\n", m, n, s)if (dbg) printf("substr: m=%d, n=%d, s=%s\n", m, n, s);
y = gettemp();
mb = u8_char2byte(s, m-1); /* byte offset of start char in s */
nb = u8_char2byte(s, m-1+n);  /* byte offset of end+1 char in s */

temp = s[nb];	/* with thanks to John Linderman */
s[nb] = '\0';
setsval(y, s + mb);
s[nb] = temp;
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
return(y);
1027}

1029Cell *sindex(Node **a, int nnn)		/* index(a[0], a[1]) */
1030{
Cell *x, *y, *z;
char *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();
for (p1 = s1; *p1 != '\0'; p1++) {
for (q = p1, p2 = s2; *p2 != '\0' && *q == *p2; q++, p2++)
	continue;
if (*p2 == '\0') {
	/* v = (Awkfloat) (p1 - s1 + 1);	 origin 1 */

   /* should be a function: used in match() as well */
	int i, len;
	v = 0;
	for (i = 0; i < p1-s1+1; i += len) {
		len = u8_nextlen(s1+i);
		v++;
	}
	break;
}
}
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
setfval(z, v);
return(z);
1061}

1063int has_utf8(char *s)	/* return 1 if s contains any utf-8 (2 bytes or more) character */
1064{
int n;

for (n = 0; *s != 0; s += n) {
n = u8_nextlen(s);
if (n > 1)
	return 1;
}
return 0;
1073}

1075#define	MAXNUMSIZE50	50

1077int format(char **pbuf, int *pbufsize, const char *s, Node *a)	/* printf-like conversions */
1078{
char *fmt;
char *p, *t;
const char *os;
Cell *x;
int flag = 0, n;
int fmtwd; /* format width */
int fmtsz = recsize;
char *buf = *pbuf;
int bufsize = *pbufsize;
1088#define FMTSZ(a)(fmtsz - ((a) - fmt))   (fmtsz - ((a) - fmt))
1089#define BUFSZ(a)(bufsize - ((a) - buf))   (bufsize - ((a) - buf))

static bool_Bool first = true1;
static bool_Bool have_a_format = false0;

if (first) {
char xbuf[100];

snprintf(xbuf, sizeof(xbuf), "%a", 42.0);
have_a_format = (strcmp(xbuf, "0x1.5p+5") == 0);
first = false0;
}

os = s;
p = buf;
if ((fmt = (char *) malloc(fmtsz)) == NULL((void *)0))
FATAL("out of memory in format()");
while (*s) {
adjbuf(&buf, &bufsize, MAXNUMSIZE50+1+p-buf, recsize, &p, "format1");
if (*s != '%') {
	*p++ = *s++;
	continue;
}
if (*(s+1) == '%') {
	*p++ = '%';
	s += 2;
	continue;
}
fmtwd = atoi(s+1);
if (fmtwd < 0)
	fmtwd = -fmtwd;
adjbuf(&buf, &bufsize, fmtwd+1+p-buf, recsize, &p, "format2");
for (t = fmt; (*t++ = *s) != '\0'; s++) {
	if (!adjbuf(&fmt, &fmtsz, MAXNUMSIZE50+1+t-fmt, recsize, &t, "format3"))
		FATAL("format item %.30s... ran format() out of memory", os);
	/* Ignore size specifiers */
	if (strchr("hjLlqtz", *s) != NULL((void *)0)) {	/* the ansi panoply */
		t--;
		continue;
	}
	if (isalpha((uschar)*s))
		break;
	if (*s == '$') {
		FATAL("'$' not permitted in awk formats");
	}
	if (*s == '*') {
		if (a == NULL((void *)0)) {
			FATAL("not enough args in printf(%s)", os);
		}
		x = execute(a);
		a = a->nnext;
		snprintf(t - 1, FMTSZ(t - 1)(fmtsz - ((t - 1) - fmt)),
		    "%d", fmtwd=(int) getfval(x));
		if (fmtwd < 0)
			fmtwd = -fmtwd;
		adjbuf(&buf, &bufsize, fmtwd+1+p-buf, recsize, &p, "format");
		t = fmt + strlen(fmt);
		tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
	}
}
*t = '\0';
if (fmtwd < 0)
	fmtwd = -fmtwd;
adjbuf(&buf, &bufsize, fmtwd+1+p-buf, recsize, &p, "format4");
switch (*s) {
case 'a': case 'A':
	if (have_a_format)
		flag = *s;
	else
		flag = 'f';
	break;
case 'f': case 'e': case 'g': case 'E': case 'G':
	flag = 'f';
	break;
case 'd': case 'i': case 'o': case 'x': case 'X': case 'u':
	flag = (*s == 'd' || *s == 'i') ? 'd' : 'u';
	*(t-1) = 'j';
	*t = *s;
	*++t = '\0';
	break;
case 's':
	flag = 's';
	break;
case 'c':
	flag = 'c';
	break;
default:
	WARNING("weird printf conversion %s", fmt);
	flag = '?';
	break;
}
if (a == NULL((void *)0))
	FATAL("not enough args in printf(%s)", os);
x = execute(a);
a = a->nnext;
n = MAXNUMSIZE50;
if (fmtwd > n)
	n = fmtwd;
adjbuf(&buf, &bufsize, 1+n+p-buf, recsize, &p, "format5");
switch (flag) {
case '?':
	snprintf(p, BUFSZ(p)(bufsize - ((p) - buf)), "%s", fmt);	/* unknown, so dump it too */
	t = getsval(x);
	n = strlen(t);
	if (fmtwd > n)
		n = fmtwd;
	adjbuf(&buf, &bufsize, 1+strlen(p)+n+p-buf, recsize, &p, "format6");
	p += strlen(p);
	snprintf(p, BUFSZ(p)(bufsize - ((p) - buf)), "%s", t);
	break;
case 'a':
case 'A':
case 'f':	snprintf(p, BUFSZ(p)(bufsize - ((p) - buf)), fmt, getfval(x)); break;
case 'd':	snprintf(p, BUFSZ(p)(bufsize - ((p) - buf)), fmt, (intmax_t) getfval(x)); break;
case 'u':	snprintf(p, BUFSZ(p)(bufsize - ((p) - buf)), fmt, (uintmax_t) getfval(x)); break;

case 's': {
	t = getsval(x);
	n = strlen(t);
	/* if simple format or no utf-8 in the string, sprintf works */
	if (!has_utf8(t) || strcmp(fmt,"%s") == 0) {
		if (fmtwd > n)
			n = fmtwd;
		if (!adjbuf(&buf, &bufsize, 1+n+p-buf, recsize, &p, "format7"))
			FATAL("huge string/format (%d chars) in printf %.30s..." \
				" ran format() out of memory", n, t);
		snprintf(p, BUFSZ(p)(bufsize - ((p) - buf)), fmt, t);
		break;
	}

	/* get here if string has utf-8 chars and fmt is not plain %s */
	/* "%-w.ps", where -, w and .p are all optional */
	/* '0' before the w is a flag character */
	/* fmt points at % */
	int ljust = 0, wid = 0, prec = n, pad = 0;
	char *f = fmt+1;
	if (f[0] == '-') {
		ljust = 1;
		f++;
	}
	// flags '0' and '+' are recognized but skipped
	if (f[0] == '0') {
		f++;
		if (f[0] == '+')
			f++;
	}
	if (f[0] == '+') {
		f++;
		if (f[0] == '0')
			f++;
	}
	if (isdigit((uschar)f[0])) { /* there is a wid */
		wid = strtol(f, &f, 10);
	}
	if (f[0] == '.') { /* there is a .prec */
		prec = strtol(++f, &f, 10);
	}
	if (prec > u8_strlen(t))
		prec = u8_strlen(t);
	pad = wid>prec ? wid - prec : 0;  // has to be >= 0
	int i, k, n;
	
	if (ljust) { // print prec chars from t, then pad blanks
		n = u8_char2byte(t, prec);
		for (k = 0; k < n; k++) {
			//putchar(t[k]);
			*p++ = t[k];
		}
		for (i = 0; i < pad; i++) {
			//printf(" ");
			*p++ = ' ';
		}
	} else { // print pad blanks, then prec chars from t
		for (i = 0; i < pad; i++) {
			//printf(" ");
			*p++ = ' ';
		}
		n = u8_char2byte(t, prec);
		for (k = 0; k < n; k++) {
			//putchar(t[k]);
			*p++ = t[k];
		}
	}
	*p = 0;
	break;
}

             case 'c': {
	/*
	 * If a numeric value is given, awk should just turn
	 * it into a character and print it:
	 *      BEGIN { printf("%c\n", 65) }
	 * prints "A".
	 *
	 * But what if the numeric value is > 128 and
	 * represents a valid Unicode code point?!? We do
	 * our best to convert it back into UTF-8. If we
	 * can't, we output the encoding of the Unicode
	 * "invalid character", 0xFFFD.
	 */
	if (isnum(x)((x)->tval & 01)) {
		int charval = (int) getfval(x);

		if (charval != 0) {
			if (charval < 128 || awk_mb_cur_max == 1)
				snprintf(p, BUFSZ(p)(bufsize - ((p) - buf)), fmt, charval);
			else {
				// possible unicode character
				size_t count;
				char *bs = wide_char_to_byte_str(charval, &count);

				if (bs == NULL((void *)0))	{ // invalid character
					// use unicode invalid character, 0xFFFD
					bs = "\357\277\275";
					count = 3;
				}
				t = bs;
				n = count;
				goto format_percent_c;
			}
		} else {
			*p++ = '\0'; /* explicit null byte */
			*p = '\0';   /* next output will start here */
		}
		break;
	}
	t = getsval(x);
	n = u8_nextlen(t);
format_percent_c:
	if (n < 2) { /* not utf8 */
		snprintf(p, BUFSZ(p)(bufsize - ((p) - buf)), fmt, getsval(x)[0]);
		break;
	}

	// utf8 character, almost same song and dance as for %s
	int ljust = 0, wid = 0, prec = n, pad = 0;
	char *f = fmt+1;
	if (f[0] == '-') {
		ljust = 1;
		f++;
	}
	// flags '0' and '+' are recognized but skipped
	if (f[0] == '0') {
		f++;
		if (f[0] == '+')
			f++;
	}
	if (f[0] == '+') {
		f++;
		if (f[0] == '0')
			f++;
	}
	if (isdigit((uschar)f[0])) { /* there is a wid */
		wid = strtol(f, &f, 10);
	}
	if (f[0] == '.') { /* there is a .prec */
		prec = strtol(++f, &f, 10);
	}
	if (prec > 1)           // %c --> only one character
		prec = 1;
	pad = wid>prec ? wid - prec : 0;  // has to be >= 0
	int i;

	if (ljust) { // print one char from t, then pad blanks
		for (i = 0; i < n; i++)
			*p++ = t[i];
		for (i = 0; i < pad; i++) {
			//printf(" ");
			*p++ = ' ';
		}
	} else { // print pad blanks, then prec chars from t
		for (i = 0; i < pad; i++) {
			//printf(" ");
			*p++ = ' ';
		}
		for (i = 0; i < n; i++)
			*p++ = t[i];
	}
	*p = 0;
	break;
}
default:
	FATAL("can't happen: bad conversion %c in format()", flag);
}

tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
p += strlen(p);
s++;
}
*p = '\0';
free(fmt);
for ( ; a; a = a->nnext) {		/* evaluate any remaining args */
x = execute(a);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
*pbuf = buf;
*pbufsize = bufsize;
return p - buf;
1387}

1389Cell *awksprintf(Node **a, int n)		/* sprintf(a[0]) */
1390{
Cell *x;
Node *y;
char *buf;
int bufsz=3*recsize;

if ((buf = (char *) malloc(bufsz)) == NULL((void *)0))
FATAL("out of memory in awksprintf");
y = a[0]->nnext;
x = execute(a[0]);
if (format(&buf, &bufsz, getsval(x), y) == -1)
FATAL("sprintf string %.30s... too long.  can't happen.", buf);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = gettemp();
x->sval = buf;
x->tval = STR02;
return(x);
1407}

1409Cell *awkprintf(Node **a, int n)		/* printf */
1410{	/* a[0] is list of args, starting with format string */
/* a[1] is redirection operator, a[2] is redirection file */
FILE *fp;
Cell *x;
Node *y;
char *buf;
int len;
int bufsz=3*recsize;

if ((buf = (char *) malloc(bufsz)) == NULL((void *)0))
FATAL("out of memory in awkprintf");
y = a[0]->nnext;
x = execute(a[0]);
if ((len = format(&buf, &bufsz, getsval(x), y)) == -1)
FATAL("printf string %.30s... too long.  can't happen.", buf);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
if (a[1] == NULL((void *)0)) {
/* fputs(buf, stdout); */
fwrite(buf, len, 1, stdout(&__sF[1]));
if (ferror(stdout)(!__isthreaded ? ((((&__sF[1]))->_flags & 0x0040) !=
 0) : (ferror)((&__sF[1]))))
	FATAL("write error on stdout");
} else {
fp = redirect(ptoi(a[1]), a[2]);
/* fputs(buf, fp); */
fwrite(buf, len, 1, fp);
fflush(fp);
if (ferror(fp)(!__isthreaded ? (((fp)->_flags & 0x0040) != 0) : (ferror
)(fp)))
	FATAL("write error on %s", filename(fp));
}
free(buf);
return(True);
1441}

1443Cell *arith(Node **a, int n)	/* a[0] + a[1], etc.  also -a[0] */
1444{
Awkfloat i, j = 0;
double v;
Cell *x, *y, *z;

x = execute(a[0]);
i = getfval(x);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
if (n != UMINUS346 && n != UPLUS347) {
y = execute(a[1]);
j = getfval(y);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
}
z = gettemp();
switch (n) {
case ADD309:
i += j;
break;
case MINUS310:
i -= j;
break;
case MULT311:
i *= j;
break;
case DIVIDE312:
if (j == 0)
	FATAL("division by zero");
i /= j;
break;
case MOD313:
if (j == 0)
	FATAL("division by zero in mod");
modf(i/j, &v);
i = i - j * v;
break;
case UMINUS346:
i = -i;
break;
case UPLUS347: /* handled by getfval(), above */
break;
case POWER348:
if (j >= 0 && modf(j, &v) == 0.0)	/* pos integer exponent */
	i = ipow(i, (int) j);
             else {
	errno(*__errno()) = 0;
	i = errcheck(pow(i, j), "pow");
             }
break;
default:	/* can't happen */
FATAL("illegal arithmetic operator %d", n);
}
setfval(z, i);
return(z);
1497}

1499double ipow(double x, int n)	/* x**n.  ought to be done by pow, but isn't always */
1500{
double v;

if (n <= 0)
return 1;
v = ipow(x, n/2);
if (n % 2 == 0)
return v * v;
else
return x * v * v;
1510}

1512Cell *incrdecr(Node **a, int n)		/* a[0]++, etc. */
1513{
Cell *x, *z;
int k;
Awkfloat xf;

x = execute(a[0]);
xf = getfval(x);
k = (n == PREINCR329 || n == POSTINCR328) ? 1 : -1;
if (n == PREINCR329 || n == PREDECR331) {
setfval(x, xf + k);
return(x);
}
z = gettemp();
setfval(z, xf);
setfval(x, xf + k);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
return(z);
1530}

1532Cell *assign(Node **a, int n)	/* a[0] = a[1], a[0] += a[1], etc. */
1533{		/* this is subtle; don't muck with it. */
Cell *x, *y;
Awkfloat xf, yf;
double v;

y = execute(a[1]);
x = execute(a[0]);
if (n == ASSIGN314) {	/* ordinary assignment */
if (x == y && !(x->tval & (FLD0100|REC0200)) && x != nfloc)
	;	/* self-assignment: leave alone unless it's a field or NF */
else if ((y->tval & (STR02|NUM01)) == (STR02|NUM01)) {
	yf = getfval(y);
	setsval(x, getsval(y));
	x->fval = yf;
	x->tval |= NUM01;
}
else if (isstr(y)((y)->tval & 02))
	setsval(x, getsval(y));
else if (isnum(y)((y)->tval & 01))
	setfval(x, getfval(y));
else
	funnyvar(y, "read value of");
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
return(x);
}
xf = getfval(x);
yf = getfval(y);
switch (n) {
case ADDEQ316:
xf += yf;
break;
case SUBEQ317:
xf -= yf;
break;
case MULTEQ318:
xf *= yf;
break;
case DIVEQ319:
if (yf == 0)
	FATAL("division by zero in /=");
xf /= yf;
break;
case MODEQ320:
if (yf == 0)
	FATAL("division by zero in %%=");
modf(xf/yf, &v);
xf = xf - yf * v;
break;
case POWEQ321:
if (yf >= 0 && modf(yf, &v) == 0.0)	/* pos integer exponent */
	xf = ipow(xf, (int) yf);
             else {
	errno(*__errno()) = 0;
	xf = errcheck(pow(xf, yf), "pow");
             }
break;
default:
FATAL("illegal assignment operator %d", n);
break;
}
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
setfval(x, xf);
return(x);
1596}

1598Cell *cat(Node **a, int q)	/* a[0] cat a[1] */
1599{
Cell *x, *y, *z;
int n1, n2;
char *s = NULL((void *)0);
int ssz = 0;

x = execute(a[0]);
n1 = strlen(getsval(x));
adjbuf(&s, &ssz, n1 + 1, recsize, 0, "cat1");
memcpy(s, x->sval, n1);

tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);

y = execute(a[1]);
n2 = strlen(getsval(y));
adjbuf(&s, &ssz, n1 + n2 + 1, recsize, 0, "cat2");
memcpy(s + n1, y->sval, n2);
s[n1 + n2] = '\0';

tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);

z = gettemp();
z->sval = s;
z->tval = STR02;

return(z);
1625}

1627Cell *pastat(Node **a, int n)	/* a[0] { a[1] } */
1628{
Cell *x;

if (a[0] == NULL((void *)0))
x = execute(a[1]);
else {
x = execute(a[0]);
if (istrue(x)((x)->csub == 11)) {
	tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
	x = execute(a[1]);
}
}
return x;
1641}

1643Cell *dopa2(Node **a, int n)	/* a[0], a[1] { a[2] } */
1644{
Cell *x;
int pair;

pair = ptoi(a[3]);
if (pairstack[pair] == 0) {
x = execute(a[0]);
if (istrue(x)((x)->csub == 11))
	pairstack[pair] = 1;
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
if (pairstack[pair] == 1) {
x = execute(a[1]);
if (istrue(x)((x)->csub == 11))
	pairstack[pair] = 0;
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = execute(a[2]);
return(x);
}
return(False);
1664}

1666Cell *split(Node **a, int nnn)	/* split(a[0], a[1], a[2]); a[3] is type */
1667{
Cell *x = NULL((void *)0), *y, *ap;
const char *s, *origs, *t;
const char *fs = NULL((void *)0);
char *origfs = NULL((void *)0);
int sep;
char temp, num[50];
int j, n, tempstat, arg3type;
double result;

y = execute(a[0]);	/* source string */
origs = s = strdup(getsval(y));
if (s == NULL((void *)0))
FATAL("out of space in split");
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
arg3type = ptoi(a[3]);
if (a[2] == NULL((void *)0)) {		/* BUG: CSV should override implicit fs but not explicit */
fs = getsval(fsloc);
} else if (arg3type == STRING337) {	/* split(str,arr,"string") */
x = execute(a[2]);
fs = origfs = strdup(getsval(x));
if (fs == NULL((void *)0))
	FATAL("out of space in split");
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
} else if (arg3type == REGEXPR338) {
fs = "(regexpr)";	/* split(str,arr,/regexpr/) */
} else {
FATAL("illegal type of split");
}
sep = *fs;
ap = execute(a[1]);	/* array name */
/* BUG 7/26/22: this appears not to reset array: see C1/asplit */
freesymtab(ap);
DPRINTF("split: s=|%s|, a=%s, sep=|%s|\n", s, NN(ap->nval), fs)if (dbg) printf("split: s=|%s|, a=%s, sep=|%s|\n", s, ((ap->
nval) ? (ap->nval) : "(null)"), fs);
ap->tval &= ~STR02;
ap->tval |= ARR020;
ap->sval = (char *) makesymtab(NSYMTAB50);

n = 0;
      if (arg3type == REGEXPR338 && strlen((char*)((fa*)a[2])->restr) == 0) {
/* split(s, a, //); have to arrange that it looks like empty sep */
arg3type = 0;
fs = "";
sep = 0;
}
if (*s != '\0' && (strlen(fs) > 1 || arg3type == REGEXPR338)) {	/* reg expr */
fa *pfa;
if (arg3type == REGEXPR338) {	/* it's ready already */
	pfa = (fa *) a[2];
} else {
	pfa = makedfa(fs, 1);
}
if (nematch(pfa,s)) {
	tempstat = pfa->initstat;
	pfa->initstat = 2;
	do {
		n++;
		snprintf(num, sizeof(num), "%d", n);
		temp = *patbeg;
		setptr(patbeg, '\0')(*(char *)(intptr_t)(patbeg)) = ('\0');
		if (is_number(s, & result)is_valid_number(s, 0, ((void *)0), & result))
			setsymtab(num, s, result, STR02|NUM01, (Array *) ap->sval);
		else
			setsymtab(num, s, 0.0, STR02, (Array *) ap->sval);
		setptr(patbeg, temp)(*(char *)(intptr_t)(patbeg)) = (temp);
		s = patbeg + patlen;
		if (*(patbeg+patlen-1) == '\0' || *s == '\0') {
			n++;
			snprintf(num, sizeof(num), "%d", n);
			setsymtab(num, "", 0.0, STR02, (Array *) ap->sval);
			pfa->initstat = tempstat;
			goto spdone;
		}
	} while (nematch(pfa,s));
	pfa->initstat = tempstat; 	/* bwk: has to be here to reset */
					/* cf gsub and refldbld */
}
n++;
snprintf(num, sizeof(num), "%d", n);
if (is_number(s, & result)is_valid_number(s, 0, ((void *)0), & result))
	setsymtab(num, s, result, STR02|NUM01, (Array *) ap->sval);
else
	setsymtab(num, s, 0.0, STR02, (Array *) ap->sval);
spdone:
pfa = NULL((void *)0);

} else if (a[2] == NULL((void *)0) && CSV) {	/* CSV only if no explicit separator */
char *newt = (char *) malloc(strlen(s)); /* for building new string; reuse for each field */
for (;;) {
	char *fr = newt;
	n++;
	if (*s == '"' ) { /* start of "..." */
		for (s++ ; *s != '\0'; ) {
			if (*s == '"' && s[1] != '\0' && s[1] == '"') {
				s += 2; /* doubled quote */
				*fr++ = '"';
			} else if (*s == '"' && (s[1] == '\0' || s[1] == ',')) {
				s++; /* skip over closing quote */
				break;
			} else {
				*fr++ = *s++;
			}
		}
		*fr++ = 0;
	} else {	/* unquoted field */
		while (*s != ',' && *s != '\0')
			*fr++ = *s++;
		*fr++ = 0;
	}
	snprintf(num, sizeof(num), "%d", n);
	if (is_number(newt, &result)is_valid_number(newt, 0, ((void *)0), &result))
		setsymtab(num, newt, result, STR02|NUM01, (Array *) ap->sval);
	else
		setsymtab(num, newt, 0.0, STR02, (Array *) ap->sval);
	if (*s++ == '\0')
		break;
}
free(newt);

} else if (!CSV && sep == ' ') { /* usual case: split on white space */
for (n = 0; ; ) {
1788#define ISWS(c)((c) == ' ' || (c) == '\t' || (c) == '\n')	((c) == ' ' || (c) == '\t' || (c) == '\n')
	while (ISWS(*s)((*s) == ' ' || (*s) == '\t' || (*s) == '\n'))
		s++;
	if (*s == '\0')
		break;
	n++;
	t = s;
	do
		s++;
	while (*s != '\0' && !ISWS(*s)((*s) == ' ' || (*s) == '\t' || (*s) == '\n'));
	temp = *s;
	setptr(s, '\0')(*(char *)(intptr_t)(s)) = ('\0');
	snprintf(num, sizeof(num), "%d", n);
	if (is_number(t, & result)is_valid_number(t, 0, ((void *)0), & result))
		setsymtab(num, t, result, STR02|NUM01, (Array *) ap->sval);
	else
		setsymtab(num, t, 0.0, STR02, (Array *) ap->sval);
	setptr(s, temp)(*(char *)(intptr_t)(s)) = (temp);
	if (*s != '\0')
		s++;
}

} else if (sep == 0) {	/* new: split(s, a, "") => 1 char/elem */
for (n = 0; *s != '\0'; s += u8_nextlen(s)) {
	char buf[10];
	n++;
	snprintf(num, sizeof(num), "%d", n);

	for (j = 0; j < u8_nextlen(s); j++) {
		buf[j] = s[j];
	}
	buf[j] = '\0';

	if (isdigit((uschar)buf[0]))
		setsymtab(num, buf, atof(buf), STR02|NUM01, (Array *) ap->sval);
	else
		setsymtab(num, buf, 0.0, STR02, (Array *) ap->sval);
}

} else if (*s != '\0') {  /* some random single character */
for (;;) {
	n++;
	t = s;
	while (*s != sep && *s != '\n' && *s != '\0')
		s++;
	temp = *s;
	setptr(s, '\0')(*(char *)(intptr_t)(s)) = ('\0');
	snprintf(num, sizeof(num), "%d", n);
	if (is_number(t, & result)is_valid_number(t, 0, ((void *)0), & result))
		setsymtab(num, t, result, STR02|NUM01, (Array *) ap->sval);
	else
		setsymtab(num, t, 0.0, STR02, (Array *) ap->sval);
	setptr(s, temp)(*(char *)(intptr_t)(s)) = (temp);
	if (*s++ == '\0')
		break;
}
}
tempfree(ap)do { if (((ap)->csub == 4)) tfree(ap); } while ( 0);
xfree(origs){ free((void *)(intptr_t)(origs)); (origs) = ((void *)0); };
xfree(origfs){ free((void *)(intptr_t)(origfs)); (origfs) = ((void *)0); };
x = gettemp();
x->tval = NUM01;
x->fval = n;
return(x);
1852}

1854Cell *condexpr(Node **a, int n)	/* a[0] ? a[1] : a[2] */
1855{
Cell *x;

x = execute(a[0]);
if (istrue(x)((x)->csub == 11)) {
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = execute(a[1]);
} else {
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = execute(a[2]);
}
return(x);
1867}

1869Cell *ifstat(Node **a, int n)	/* if (a[0]) a[1]; else a[2] */
1870{
Cell *x;

x = execute(a[0]);
if (istrue(x)((x)->csub == 11)) {
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = execute(a[1]);
} else if (a[2] != NULL((void *)0)) {
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = execute(a[2]);
}
return(x);
1882}

1884Cell *whilestat(Node **a, int n)	/* while (a[0]) a[1] */
1885{
Cell *x;

for (;;) {
x = execute(a[0]);
if (!istrue(x)((x)->csub == 11))
	return(x);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = execute(a[1]);
if (isbreak(x)((x)->csub == 23)) {
	x = True;
	return(x);
}
if (isnext(x)((x)->csub == 22 || (x)->csub == 26) || isexit(x)((x)->csub == 21) || isret(x)((x)->csub == 25))
	return(x);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
1902}

1904Cell *dostat(Node **a, int n)	/* do a[0]; while(a[1]) */
1905{
Cell *x;

for (;;) {
x = execute(a[0]);
if (isbreak(x)((x)->csub == 23))
	return True;
if (isnext(x)((x)->csub == 22 || (x)->csub == 26) || isexit(x)((x)->csub == 21) || isret(x)((x)->csub == 25))
	return(x);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = execute(a[1]);
if (!istrue(x)((x)->csub == 11))
	return(x);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
1920}

1922Cell *forstat(Node **a, int n)	/* for (a[0]; a[1]; a[2]) a[3] */
1923{
Cell *x;

x = execute(a[0]);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
for (;;) {
if (a[1]!=NULL((void *)0)) {
	x = execute(a[1]);
	if (!istrue(x)((x)->csub == 11)) return(x);
	else tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
x = execute(a[3]);
if (isbreak(x)((x)->csub == 23))		/* turn off break */
	return True;
if (isnext(x)((x)->csub == 22 || (x)->csub == 26) || isexit(x)((x)->csub == 21) || isret(x)((x)->csub == 25))
	return(x);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = execute(a[2]);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
1943}

1945Cell *instat(Node **a, int n)	/* for (a[0] in a[1]) a[2] */
1946{
Cell *x, *vp, *arrayp, *cp, *ncp;
Array *tp;
int i;

vp = execute(a[0]);
arrayp = execute(a[1]);
if (!isarr(arrayp)((arrayp)->tval & 020)) {
return True;
}
tp = (Array *) arrayp->sval;
tempfree(arrayp)do { if (((arrayp)->csub == 4)) tfree(arrayp); } while ( 0
);
for (i = 0; i < tp->size; i++) {	/* this routine knows too much */
for (cp = tp->tab[i]; cp != NULL((void *)0); cp = ncp) {
	setsval(vp, cp->nval);
	ncp = cp->cnext;
	x = execute(a[2]);
	if (isbreak(x)((x)->csub == 23)) {
		tempfree(vp)do { if (((vp)->csub == 4)) tfree(vp); } while ( 0);
		return True;
	}
	if (isnext(x)((x)->csub == 22 || (x)->csub == 26) || isexit(x)((x)->csub == 21) || isret(x)((x)->csub == 25)) {
		tempfree(vp)do { if (((vp)->csub == 4)) tfree(vp); } while ( 0);
		return(x);
	}
	tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}
}
return True;
1975}

1977static char *nawk_convert(const char *s, int (*fun_c)(int),
  wint_t (*fun_wc)(wint_t))
1979{
char *buf      = NULL((void *)0);
char *pbuf     = NULL((void *)0);
const char *ps = NULL((void *)0);
size_t n       = 0;
wchar_t wc;
const size_t sz = awk_mb_cur_max;
int unused;

if (sz == 1) {
buf = tostring(s);

for (pbuf = buf; *pbuf; pbuf++)
	*pbuf = fun_c((uschar)*pbuf);

return buf;
} else {
/* upper/lower character may be shorter/longer */
buf = tostringN(s, strlen(s) * sz + 1);

(void) mbtowc(NULL((void *)0), NULL((void *)0), 0);	/* reset internal state */
/*
 * Reset internal state here too.
 * Assign result to avoid a compiler warning. (Casting to void
 * doesn't work.)
 * Increment said variable to avoid a different warning.
 */
unused = wctomb(NULL((void *)0), L'\0');
unused++;

ps   = s;
pbuf = buf;
while (n = mbtowc(&wc, ps, sz),
       n > 0 && n != (size_t)-1 && n != (size_t)-2)
{
	ps += n;

	n = wctomb(pbuf, fun_wc(wc));
	if (n == (size_t)-1)
		FATAL("illegal wide character %s", s);

	pbuf += n;
}

*pbuf = '\0';

if (n)
	FATAL("illegal byte sequence %s", s);

return buf;
}
2030}

2032#ifdef __DJGPP__
2033static wint_t towupper(wint_t wc)
2034{
if (wc >= 0 && wc < 256)
return toupper(wc & 0xFF);

return wc;
2039}

2041static wint_t towlower(wint_t wc)
2042{
if (wc >= 0 && wc < 256)
return tolower(wc & 0xFF);

return wc;
2047}
2048#endif

2050static char *nawk_toupper(const char *s)
2051{
return nawk_convert(s, toupper, towupper);
2053}

2055static char *nawk_tolower(const char *s)
2056{
return nawk_convert(s, tolower, towlower);
2058}

2060Cell *bltin(Node **a, int n)	/* builtin functions. a[0] is type, a[1] is arg list */
2061{
Cell *x, *y;
Awkfloat u;
int t, sz;
Awkfloat tmp;
char *buf, *fmt;
Node *nextarg;
FILE *fp;
int status = 0;
time_t tv;
struct tm *tm, tmbuf;
int estatus = 0;

t = ptoi(a[0]);
x = execute(a[1]);
nextarg = a[1]->nnext;
switch (t) {
case FLENGTH1:
if (isarr(x)((x)->tval & 020))
	u = ((Array *) x->sval)->nelem;	/* GROT.  should be function*/
else
	u = u8_strlen(getsval(x));
break;
case FLOG4:
errno(*__errno()) = 0;
u = errcheck(log(getfval(x)), "log");
break;
case FINT5:
modf(getfval(x), &u); break;
case FEXP3:
errno(*__errno()) = 0;
u = errcheck(exp(getfval(x)), "exp");
break;
case FSQRT2:
errno(*__errno()) = 0;
u = errcheck(sqrt(getfval(x)), "sqrt");
break;
case FSIN9:
u = sin(getfval(x)); break;
case FCOS10:
u = cos(getfval(x)); break;
case FATAN11:
if (nextarg == NULL((void *)0)) {
	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)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
	nextarg = nextarg->nnext;
}
break;
case FCOMPL18:
u = ~((int)getfval(x));
break;
case FAND15:
if (nextarg == 0) {
	WARNING("and requires two arguments; returning 0");
	u = 0;
	break;
}
y = execute(a[1]->nnext);
u = ((int)getfval(x)) & ((int)getfval(y));
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
nextarg = nextarg->nnext;
break;
case FFOR16:
if (nextarg == 0) {
	WARNING("or requires two arguments; returning 0");
	u = 0;
	break;
}
y = execute(a[1]->nnext);
u = ((int)getfval(x)) | ((int)getfval(y));
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
nextarg = nextarg->nnext;
break;
case FXOR17:
if (nextarg == 0) {
	WARNING("xor requires two arguments; returning 0");
	u = 0;
	break;
}
y = execute(a[1]->nnext);
u = ((int)getfval(x)) ^ ((int)getfval(y));
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
nextarg = nextarg->nnext;
break;
case FLSHIFT19:
if (nextarg == 0) {
	WARNING("lshift requires two arguments; returning 0");
	u = 0;
	break;
}
y = execute(a[1]->nnext);
u = ((int)getfval(x)) << ((int)getfval(y));
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
nextarg = nextarg->nnext;
break;
case FRSHIFT20:
if (nextarg == 0) {
	WARNING("rshift requires two arguments; returning 0");
	u = 0;
	break;
}
y = execute(a[1]->nnext);
u = ((int)getfval(x)) >> ((int)getfval(y));
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
nextarg = nextarg->nnext;
break;
case FSYSTEM6:
fflush(stdout(&__sF[1]));		/* in case something is buffered already */
estatus = status = system(getsval(x));
if (status != -1) {
	if (WIFEXITED(status)(((status) & 0177) == 0)) {
		estatus = WEXITSTATUS(status)(int)(((unsigned)(status) >> 8) & 0xff);
	} else if (WIFSIGNALED(status)(((status) & 0177) != 0177 && ((status) & 0177
) != 0)) {
		estatus = WTERMSIG(status)(((status) & 0177)) + 256;
2179#ifdef WCOREDUMP
		if (WCOREDUMP(status)((status) & 0200))
			estatus += 256;
2182#endif
	} else	/* something else?!? */
		estatus = 0;
}
/* else estatus was set to -1 */
u = estatus;
break;
case FRAND7:
/* random() returns numbers in [0..2^31-1]
 * in order to get a number in [0, 1), divide it by 2^31
 */
u = (Awkfloat) random() / (0x7fffffffL + 0x1UL);
break;
case FSRAND8:
if (isrec(x)((x)->tval & 0200)) {		/* no argument provided */
	u = time(NULL((void *)0));
	tmp = u;
	srandom((unsigned int) u);
} else {
	u = getfval(x);
	tmp = u;
	srandom_deterministic((unsigned int) u);
}
u = srand_seed;
srand_seed = tmp;
break;
case FTOUPPER12:
case FTOLOWER13:
if (t == FTOUPPER12)
	buf = nawk_toupper(getsval(x));
else
	buf = nawk_tolower(getsval(x));
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = gettemp();
setsval(x, buf);
free(buf);
return x;
case FFLUSH14:
if (isrec(x)((x)->tval & 0200) || strlen(getsval(x)) == 0) {
	flush_all();	/* fflush() or fflush("") -> all */
	u = 0;
} else if ((fp = openfile(FFLUSH14, getsval(x), NULL((void *)0))) == NULL((void *)0))
	u = EOF(-1);
else
	u = fflush(fp);
break;
case FMKTIME23:
memset(&tmbuf, 0, sizeof(tmbuf));
tm = &tmbuf;
t = sscanf(getsval(x), "%d %d %d %d %d %d %d",
    &tm->tm_year, &tm->tm_mon, &tm->tm_mday, &tm->tm_hour,
    &tm->tm_min, &tm->tm_sec, &tm->tm_isdst);
switch (t) {
case 6:
	tm->tm_isdst = -1;	/* let mktime figure it out */
	/* FALLTHROUGH */
case 7:
	tm->tm_year -= 1900;
	tm->tm_mon--;
	u = mktime(tm);
	break;
default:
	u = -1;
	break;
}
break;
case FSYSTIME21:
u = time((time_t *) 0);
break;
case FSTRFTIME22:
/* strftime([format [,timestamp]]) */
if (nextarg) {
	y = execute(nextarg);
	nextarg = nextarg->nnext;
	tv = (time_t) getfval(y);
	tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
} else
	tv = time((time_t *) 0);
tm = localtime(&tv);
if (tm == NULL((void *)0))
	FATAL("bad time %ld", (long)tv);

if (isrec(x)((x)->tval & 0200)) {
	/* format argument not provided, use default */
	fmt = tostring("%a %b %d %H:%M:%S %Z %Y");
} else
	fmt = tostring(getsval(x));

sz = 32;
buf = NULL((void *)0);
do {
	if ((buf = (char *) reallocarray(buf, 2, sz)) == NULL((void *)0))
		FATAL("out of memory in strftime");
	sz *= 2;
} while (strftime(buf, sz, fmt, tm) == 0 && fmt[0] != '\0');

y = gettemp();
setsval(y, buf);
free(fmt);
free(buf);

return y;
default:	/* can't happen */
FATAL("illegal function type %d", t);
break;
}
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = gettemp();
setfval(x, u);
if (nextarg != NULL((void *)0)) {
WARNING("warning: function has too many arguments");
for ( ; nextarg; nextarg = nextarg->nnext) {
	y = execute(nextarg);
	tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
}
}
return(x);
2299}

2301Cell *printstat(Node **a, int n)	/* print a[0] */
2302{
Node *x;
Cell *y;
FILE *fp;

if (a[1] == NULL((void *)0))	/* a[1] is redirection operator, a[2] is file */
fp = stdout(&__sF[1]);
else
fp = redirect(ptoi(a[1]), a[2]);
for (x = a[0]; x != NULL((void *)0); x = x->nnext) {
y = execute(x);
fputs(getpssval(y), fp);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
if (x->nnext == NULL((void *)0))
	fputs(getsval(orsloc), fp);
else
	fputs(getsval(ofsloc), fp);
}
if (a[1] != NULL((void *)0))
fflush(fp);
if (ferror(fp)(!__isthreaded ? (((fp)->_flags & 0x0040) != 0) : (ferror
)(fp)))
FATAL("write error on %s", filename(fp));
return(True);
2325}

2327Cell *nullproc(Node **a, int n)
2328{
return 0;
2330}


2333FILE *redirect(int a, Node *b)	/* set up all i/o redirections */
2334{
FILE *fp;
Cell *x;
char *fname;

x = execute(b);
fname = getsval(x);
fp = openfile(a, fname, NULL((void *)0));
if (fp == NULL((void *)0))
FATAL("can't open file %s", fname);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
return fp;
2346}

2348struct files {
FILE	*fp;
const char	*fname;
int	mode;	/* '|', 'a', 'w' => LE/LT, GT */
2352} *files;

2354size_t nfiles;

2356static void stdinit(void)	/* in case stdin, etc., are not constants */
2357{
nfiles = FOPEN_MAX20;
files = (struct files *) calloc(nfiles, sizeof(*files));
if (files == NULL((void *)0))
FATAL("can't allocate file memory for %zu files", nfiles);
      files[0].fp = stdin(&__sF[0]);
files[0].fname = tostring("/dev/stdin");
files[0].mode = LT287;
      files[1].fp = stdout(&__sF[1]);
files[1].fname = tostring("/dev/stdout");
files[1].mode = GT285;
      files[2].fp = stderr(&__sF[2]);
files[2].fname = tostring("/dev/stderr");
files[2].mode = GT285;
2371}

2373FILE *openfile(int a, const char *us, bool_Bool *pnewflag)
2374{
const char *s = us;
size_t i;
int m;
FILE *fp = NULL((void *)0);

if (*s == '\0')
FATAL("null file name in print or getline");
for (i = 0; i < nfiles; i++)
if (files[i].fname && strcmp(s, files[i].fname) == 0 &&
    (a == files[i].mode || (a==APPEND282 && files[i].mode==GT285) ||
     a == FFLUSH14)) {
	if (pnewflag)
		*pnewflag = false0;
	return files[i].fp;
}
if (a == FFLUSH14)	/* didn't find it, so don't create it! */
return NULL((void *)0);

for (i = 0; i < nfiles; i++)
if (files[i].fp == NULL((void *)0))
	break;
if (i >= nfiles) {
struct files *nf;
size_t nnf = nfiles + FOPEN_MAX20;
nf = (struct files *) reallocarray(files, nnf, sizeof(*nf));
if (nf == NULL((void *)0))
	FATAL("cannot grow files for %s and %zu files", s, nnf);
memset(&nf[nfiles], 0, FOPEN_MAX20 * sizeof(*nf));
nfiles = nnf;
files = nf;
}
fflush(stdout(&__sF[1]));	/* force a semblance of order */
m = a;
if (a == GT285) {
fp = fopen(s, "w");
} else if (a == APPEND282) {
fp = fopen(s, "a");
m = GT285;	/* so can mix > and >> */
} else if (a == '|') {	/* output pipe */
fp = popen(s, "w");
} else if (a == LE286) {	/* input pipe */
fp = popen(s, "r");
} else if (a == LT287) {	/* getline <file */
fp = strcmp(s, "-") == 0 ? stdin(&__sF[0]) : fopen(s, "r");	/* "-" is stdin */
} else	/* can't happen */
FATAL("illegal redirection %d", a);
if (fp != NULL((void *)0)) {
files[i].fname = tostring(s);
files[i].fp = fp;
files[i].mode = m;
if (pnewflag)
	*pnewflag = true1;
if (fp != stdin(&__sF[0]) && fp != stdout(&__sF[1]) && fp != stderr(&__sF[2]))
	(void) fcntl(fileno(fp)(!__isthreaded ? ((fp)->_file) : (fileno)(fp)), F_SETFD2, FD_CLOEXEC1);
}
return fp;
2431}

2433const char *filename(FILE *fp)
2434{
size_t i;

for (i = 0; i < nfiles; i++)
if (fp == files[i].fp)
	return files[i].fname;
return "???";
2441}

2443Cell *closefile(Node **a, int n)
2444{
Cell *x;
size_t i;
bool_Bool stat;

x = execute(a[0]);
getsval(x);
stat = true1;
for (i = 0; i < nfiles; i++) {
if (!files[i].fname || strcmp(x->sval, files[i].fname) != 0)
	continue;
if (files[i].mode == GT285 || files[i].mode == '|')
	fflush(files[i].fp);
if (ferror(files[i].fp)(!__isthreaded ? (((files[i].fp)->_flags & 0x0040) != 0
) : (ferror)(files[i].fp))) {
	if ((files[i].mode == GT285 && files[i].fp != stderr(&__sF[2]))
	  || files[i].mode == '|')
		FATAL("write error on %s", files[i].fname);
	else
		WARNING("i/o error occurred on %s", files[i].fname);
}
if (files[i].fp == stdin(&__sF[0]) || files[i].fp == stdout(&__sF[1]) ||
    files[i].fp == stderr(&__sF[2]))
	stat = freopen("/dev/null", "r+", files[i].fp) == NULL((void *)0);
else if (files[i].mode == '|' || files[i].mode == LE286)
	stat = pclose(files[i].fp) == -1;
else
	stat = fclose(files[i].fp) == EOF(-1);
if (stat)
	WARNING("i/o error occurred closing %s", files[i].fname);
xfree(files[i].fname){ free((void *)(intptr_t)(files[i].fname)); (files[i].fname) =
 ((void *)0); };
files[i].fname = NULL((void *)0);	/* watch out for ref thru this */
files[i].fp = NULL((void *)0);
break;
}
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = gettemp();
setfval(x, (Awkfloat) (stat ? -1 : 0));
return(x);
2482}

2484void closeall(void)
2485{
size_t i;
bool_Bool stat = false0;

for (i = 0; i < nfiles; i++) {
if (! files[i].fp)
	continue;
if (files[i].mode == GT285 || files[i].mode == '|')
	fflush(files[i].fp);
if (ferror(files[i].fp)(!__isthreaded ? (((files[i].fp)->_flags & 0x0040) != 0
) : (ferror)(files[i].fp))) {
	if ((files[i].mode == GT285 && files[i].fp != stderr(&__sF[2]))
	  || files[i].mode == '|')
		FATAL("write error on %s", files[i].fname);
	else
		WARNING("i/o error occurred on %s", files[i].fname);
}
if (files[i].fp == stdin(&__sF[0]) || files[i].fp == stdout(&__sF[1]) ||
    files[i].fp == stderr(&__sF[2]))
	continue;
if (files[i].mode == '|' || files[i].mode == LE286)
	stat = pclose(files[i].fp) == -1;
else
	stat = fclose(files[i].fp) == EOF(-1);
if (stat)
	WARNING("i/o error occurred while closing %s", files[i].fname);
}
2511}

2513static void flush_all(void)
2514{
size_t i;

for (i = 0; i < nfiles; i++)
if (files[i].fp)
	fflush(files[i].fp);
2520}

2522void backsub(char **pb_ptr, const char **sptr_ptr);

2524Cell *dosub(Node **a, int subop)        /* sub and gsub */
2525{
fa *pfa;
int tempstat;
char *repl;
Cell *x;

char *buf = NULL((void *)0);
char *pb = NULL((void *)0);
int bufsz = recsize;

const char *r, *s;
const char *start;
const char *noempty = NULL((void *)0);      /* empty match disallowed here */
size_t m = 0;                    /* match count */
size_t whichm;                   /* which match to select, 0 = global */
int mtype;                       /* match type */

if (a[0] == NULL((void *)0)) {	/* 0 => a[1] is already-compiled regexpr */
1
Assuming the condition is true→
2
←
Taking true branch→
pfa = (fa *) a[1];
} else {
x = execute(a[1]);
pfa = makedfa(getsval(x), 1);
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
}

x = execute(a[2]);	/* replacement string */
repl = tostring(getsval(x));
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
3
←
Assuming field 'csub' is not equal to 4→
4
←
Taking false branch→
5
←
Loop condition is false.  Exiting loop→

switch (subop) {
6
←
Control jumps to 'case 301:'  at line 2555→
case SUB301:
whichm = 1;
x = execute(a[3]);    /* source string */
break;
case GSUB302:
whichm = 0;
x = execute(a[3]);    /* source string */
break;
default:
FATAL("dosub: unrecognized subop: %d", subop);
}

start = getsval(x);
7
←
 Execution continues on line 2567→
while (pmatch(pfa, start)) {
8
←
Value assigned to 'patbeg'→
9
←
Loop condition is true.  Entering loop body→
if (buf9.1
'buf' is equal to NULL
 == NULL((void *)0)) {
10
←
Taking true branch→
	if ((pb = buf = malloc(bufsz)) == NULL((void *)0))
11
←
Assuming the condition is false→
12
←
Taking false branch→
		FATAL("out of memory in dosub");
	tempstat = pfa->initstat;
	pfa->initstat = 2;
}

/* match types */
#define	MT_IGNORE  0  /* unselected or invalid */
#define MT_INSERT  1  /* selected, empty */
#define MT_REPLACE 2  /* selected, not empty */

/* an empty match just after replacement is invalid */

if (patbeg == noempty && patlen == 0) {
13
←
Assuming 'patbeg' is equal to 'noempty'→
14
←
Assuming 'patlen' is not equal to 0→
	mtype = MT_IGNORE;    /* invalid, not counted */
} else if (whichm == ++m || whichm == 0) {
	mtype = patlen14.1
'patlen' is not equal to 0
 ? MT_REPLACE : MT_INSERT;
15
←
'?' condition is true→
} else {
	mtype = MT_IGNORE;    /* unselected, but counted */
}

/* leading text: */
if (patbeg > start) {
16
←
Assuming 'patbeg' is <= 'start'→
17
←
Taking false branch→
	adjbuf(&buf, &bufsz, (pb - buf) + (patbeg - start),
		recsize, &pb, "dosub");
	s = start;
	while (s < patbeg)
		*pb++ = *s++;
}

if (mtype17.1
'mtype' is not equal to MT_IGNORE
 == MT_IGNORE)
18
←
Taking false branch→
	goto matching_text;  /* skip replacement text */

r = repl;
while (*r != 0) {
19
←
Assuming the condition is false→
	adjbuf(&buf, &bufsz, 5+pb-buf, recsize, &pb, "dosub");
	if (*r == '\\') {
		backsub(&pb, &r);
	} else if (*r == '&') {
		r++;
		adjbuf(&buf, &bufsz, 1+patlen+pb-buf, recsize,
			&pb, "dosub");
		for (s = patbeg; s < patbeg+patlen; )
			*pb++ = *s++;
	} else {
		*pb++ = *r++;
	}
}

2619matching_text:
if (mtype19.1
'mtype' is equal to MT_REPLACE
 == MT_REPLACE || *patbeg == '\0')
	goto next_search;  /* skip matching text */
20
←
Control jumps to line 2631→

if (patlen == 0)
	patlen = u8_nextlen(patbeg);
adjbuf(&buf, &bufsz, (pb-buf) + patlen, recsize, &pb, "dosub");
s = patbeg;
while (s < patbeg + patlen)
	*pb++ = *s++;

2630next_search:
start = patbeg + patlen;
21
←
Null pointer value stored to 'start'→
if (m21.1
'm' is equal to 'whichm'
 == whichm || *patbeg == '\0')
	break;
if (mtype == MT_REPLACE)
	noempty = start;

#undef MT_IGNORE
#undef MT_INSERT
#undef MT_REPLACE
}

xfree(repl){ free((void *)(intptr_t)(repl)); (repl) = ((void *)0); };
22
←
 Execution continues on line 2642→

if (buf22.1
'buf' is not equal to NULL
 != NULL((void *)0)) {
23
←
Taking true branch→
pfa->initstat = tempstat;

/* trailing text */
adjbuf(&buf, &bufsz, 1+strlen(start)+pb-buf, 0, &pb, "dosub");
24
←
Null pointer passed as 1st argument to string length function
while ((*pb++ = *start++) != '\0')
	;

setsval(x, buf);
free(buf);
}

tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
x = gettemp();
x->tval = NUM01;
x->fval = m;
return x;
2661}

2663Cell *gensub(Node **a, int nnn)	/* global selective substitute */
/* XXX incomplete - doesn't support backreferences \0 ... \9 */
2665{
Cell *x, *y, *res, *h;
char *rptr;
const char *sptr;
char *buf, *pb;
const char *t, *q;
fa *pfa;
int mflag, tempstat, num, whichm;
int bufsz = recsize;

if ((buf = malloc(bufsz)) == NULL((void *)0))
FATAL("out of memory in gensub");
mflag = 0;	/* if mflag == 0, can replace empty string */
num = 0;
x = execute(a[4]);	/* source string */
t = getsval(x);
res = copycell(x);	/* target string - initially copy of source */
res->csub = CTEMP4;	/* result values are temporary */
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)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
}
y = execute(a[2]);	/* replacement string */
h = execute(a[3]);	/* which matches should be replaced */
sptr = getsval(h);
if (sptr[0] == 'g' || sptr[0] == 'G')
whichm = -1;
else {
/*
 * The specified number is index of replacement, starting
 * from 1. GNU awk treats index lower than 0 same as
 * 1, we do same for compatibility.
 */
whichm = (int) getfval(h) - 1;
if (whichm < 0)
	whichm = 0;
}
tempfree(h)do { if (((h)->csub == 4)) tfree(h); } while ( 0);

if (pmatch(pfa, t)) {
char *sl;

tempstat = pfa->initstat;
pfa->initstat = 2;
pb = buf;
rptr = getsval(y);
/*
 * XXX if there are any backreferences in subst string,
 * complain now.
 */
for (sl = rptr; (sl = strchr(sl, '\\')) && sl[1]; sl++) {
	if (strchr("0123456789", sl[1])) {
		FATAL("gensub doesn't support backreferences (subst \"%s\")", rptr);
	}
}

do {
	if (whichm >= 0 && whichm != num) {
		num++;
		adjbuf(&buf, &bufsz, (pb - buf) + (patbeg - t) + patlen, recsize, &pb, "gensub");

		/* copy the part of string up to and including
		 * match to output buffer */
		while (t < patbeg + patlen)
			*pb++ = *t++;
		continue;
	}

	if (patlen == 0 && *patbeg != 0) {	/* matched empty string */
		if (mflag == 0) {	/* can replace empty */
			num++;
			sptr = rptr;
			while (*sptr != 0) {
				adjbuf(&buf, &bufsz, 5+pb-buf, recsize, &pb, "gensub");
				if (*sptr == '\\') {
					backsub(&pb, &sptr);
				} else if (*sptr == '&') {
					sptr++;
					adjbuf(&buf, &bufsz, 1+patlen+pb-buf, recsize, &pb, "gensub");
					for (q = patbeg; q < patbeg+patlen; )
						*pb++ = *q++;
				} else
					*pb++ = *sptr++;
			}
		}
		if (*t == 0)	/* at end */
			goto done;
		adjbuf(&buf, &bufsz, 2+pb-buf, recsize, &pb, "gensub");
		*pb++ = *t++;
		if (pb > buf + bufsz)	/* BUG: not sure of this test */
			FATAL("gensub result0 %.30s too big; can't happen", buf);
		mflag = 0;
	}
	else {	/* matched nonempty string */
		num++;
		sptr = t;
		adjbuf(&buf, &bufsz, 1+(patbeg-sptr)+pb-buf, recsize, &pb, "gensub");
		while (sptr < patbeg)
			*pb++ = *sptr++;
		sptr = rptr;
		while (*sptr != 0) {
			adjbuf(&buf, &bufsz, 5+pb-buf, recsize, &pb, "gensub");
			if (*sptr == '\\') {
				backsub(&pb, &sptr);
			} else if (*sptr == '&') {
				sptr++;
				adjbuf(&buf, &bufsz, 1+patlen+pb-buf, recsize, &pb, "gensub");
				for (q = patbeg; q < patbeg+patlen; )
					*pb++ = *q++;
			} else
				*pb++ = *sptr++;
		}
		t = patbeg + patlen;
		if (patlen == 0 || *t == 0 || *(t-1) == 0)
			goto done;
		if (pb > buf + bufsz)
			FATAL("gensub result1 %.30s too big; can't happen", buf);
		mflag = 1;
	}
} while (pmatch(pfa,t));
sptr = t;
adjbuf(&buf, &bufsz, 1+strlen(sptr)+pb-buf, 0, &pb, "gensub");
while ((*pb++ = *sptr++) != 0)
	;
done:	if (pb > buf + bufsz)
	FATAL("gensub result2 %.30s too big; can't happen", buf);
*pb = '\0';
setsval(res, buf);
pfa->initstat = tempstat;
}
tempfree(x)do { if (((x)->csub == 4)) tfree(x); } while ( 0);
tempfree(y)do { if (((y)->csub == 4)) tfree(y); } while ( 0);
free(buf);
return(res);
2802}

2804void backsub(char **pb_ptr, const char **sptr_ptr)	/* handle \\& variations */
2805{						/* sptr[0] == '\\' */
char *pb = *pb_ptr;
const char *sptr = *sptr_ptr;

if (sptr[1] == '\\') {
if (sptr[2] == '\\' && sptr[3] == '&') { /* \\\& -> \& */
	*pb++ = '\\';
	*pb++ = '&';
	sptr += 4;
} else if (sptr[2] == '&') {	/* \\& -> \ + matched */
	*pb++ = '\\';
	sptr += 2;
} else if (do_posix) {		/* \\x -> \x */
	sptr++;
	*pb++ = *sptr++;
} 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;
2832}

2834static char *wide_char_to_byte_str(int rune, size_t *outlen)
2835{
static char buf[5];
int len;

if (rune < 0 || rune > 0x10FFFF)
return NULL((void *)0);

memset(buf, 0, sizeof(buf));

len = 0;
if (rune <= 0x0000007F) {
buf[len++] = rune;
} else if (rune <= 0x000007FF) {
// 110xxxxx 10xxxxxx
buf[len++] = 0xC0 | (rune >> 6);
buf[len++] = 0x80 | (rune & 0x3F);
} else if (rune <= 0x0000FFFF) {
// 1110xxxx 10xxxxxx 10xxxxxx
buf[len++] = 0xE0 | (rune >> 12);
buf[len++] = 0x80 | ((rune >> 6) & 0x3F);
buf[len++] = 0x80 | (rune & 0x3F);

} else {
// 0x00010000 - 0x10FFFF
// 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
buf[len++] = 0xF0 | (rune >> 18);
buf[len++] = 0x80 | ((rune >> 12) & 0x3F);
buf[len++] = 0x80 | ((rune >> 6) & 0x3F);
buf[len++] = 0x80 | (rune & 0x3F);
}

*outlen = len;
buf[len++] = '\0';

return buf;
2870}