1/****************************************************************
  2Copyright (C) Lucent Technologies 1997
  3All Rights Reserved
  4
  5Permission to use, copy, modify, and distribute this software and
  6its documentation for any purpose and without fee is hereby
  7granted, provided that the above copyright notice appear in all
  8copies and that both that the copyright notice and this
  9permission notice and warranty disclaimer appear in supporting
 10documentation, and that the name Lucent Technologies or any of
 11its entities not be used in advertising or publicity pertaining
 12to distribution of the software without specific, written prior
 13permission.
 14
 15LUCENT DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
 16INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS.
 17IN NO EVENT SHALL LUCENT OR ANY OF ITS ENTITIES BE LIABLE FOR ANY
 18SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
 19WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER
 20IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,
 21ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF
 22THIS SOFTWARE.
 23****************************************************************/
 24
 25#include <assert.h>
 26#include <stdint.h>
 27
 28#ifndef bool
 29#define bool int
 30#define true 1
 31#define false 0
 32#endif
 33
 34#if __STDC_VERSION__ <= 199901L
 35#define noreturn
 36#else
 37#include <stdnoreturn.h>
 38#endif
 39
 40typedef double	Awkfloat;
 41
 42/* unsigned char is more trouble than it's worth */
 43
 44typedef	unsigned char uschar;
 45
 46#define	xfree(a)	{ free((void *)(intptr_t)(a)); (a) = NULL; }
 47/*
 48 * We sometimes cheat writing read-only pointers to NUL-terminate them
 49 * and then put back the original value
 50 */
 51#define setptr(ptr, a)	(*(char *)(intptr_t)(ptr)) = (a)
 52
 53#define	NN(p)	((p) ? (p) : "(null)")	/* guaranteed non-null for DPRINTF
 54*/
 55#define	DEBUG
 56#ifdef	DEBUG
 57#	define	DPRINTF(...)	if (dbg) printf(__VA_ARGS__)
 58#else
 59#	define	DPRINTF(...)
 60#endif
 61
 62extern enum compile_states {
 63	RUNNING,
 64	COMPILING,
 65	ERROR_PRINTING
 66} compile_time;
 67
 68extern bool	safe;		/* false => unsafe, true => safe */
 69
 70#define	RECSIZE	(8 * 1024)	/* sets limit on records, fields, etc., etc. */
 71extern int	recsize;	/* size of current record, orig RECSIZE */
 72
 73extern size_t	awk_mb_cur_max;	/* max size of a multi-byte character */
 74
 75extern char	EMPTY[];	/* this avoid -Wwritable-strings issues */
 76extern char	**FS;
 77extern char	**RS;
 78extern char	**ORS;
 79extern char	**OFS;
 80extern char	**OFMT;
 81extern Awkfloat *NR;
 82extern Awkfloat *FNR;
 83extern Awkfloat *NF;
 84extern char	**FILENAME;
 85extern char	**SUBSEP;
 86extern Awkfloat *RSTART;
 87extern Awkfloat *RLENGTH;
 88
 89extern bool	CSV;		/* true for csv input */
 90
 91extern char	*record;	/* points to $0 */
 92extern int	lineno;		/* line number in awk program */
 93extern int	errorflag;	/* 1 if error has occurred */
 94extern bool	donefld;	/* true if record broken into fields */
 95extern bool	donerec;	/* true if record is valid (no fld has changed */
 96extern int	dbg;
 97
 98extern const char *patbeg;	/* beginning of pattern matched */
 99extern	int	patlen;		/* length of pattern matched.  set in b.c */
100
101/* Cell:  all information about a variable or constant */
102
103typedef struct Cell {
104	uschar	ctype;		/* OCELL, OBOOL, OJUMP, etc. */
105	uschar	csub;		/* CCON, CTEMP, CFLD, etc. */
106	char	*nval;		/* name, for variables only */
107	char	*sval;		/* string value */
108	Awkfloat fval;		/* value as number */
109	int	 tval;		/* type info: STR|NUM|ARR|FCN|FLD|CON|DONTFREE|CONVC|CONVO */
110	char	*fmt;		/* CONVFMT/OFMT value used to convert from number */
111	struct Cell *cnext;	/* ptr to next if chained */
112} Cell;
113
114typedef struct Array {		/* symbol table array */
115	int	nelem;		/* elements in table right now */
116	int	size;		/* size of tab */
117	Cell	**tab;		/* hash table pointers */
118} Array;
119
120#define	NSYMTAB	50	/* initial size of a symbol table */
121extern Array	*symtab;
122
123extern Cell	*nrloc;		/* NR */
124extern Cell	*fnrloc;	/* FNR */
125extern Cell	*fsloc;		/* FS */
126extern Cell	*nfloc;		/* NF */
127extern Cell	*ofsloc;	/* OFS */
128extern Cell	*orsloc;	/* ORS */
129extern Cell	*rsloc;		/* RS */
130extern Cell	*rstartloc;	/* RSTART */
131extern Cell	*rlengthloc;	/* RLENGTH */
132extern Cell	*subseploc;	/* SUBSEP */
133extern Cell	*symtabloc;	/* SYMTAB */
134
135/* Cell.tval values: */
136#define	NUM	01	/* number value is valid */
137#define	STR	02	/* string value is valid */
138#define DONTFREE 04	/* string space is not freeable */
139#define	CON	010	/* this is a constant */
140#define	ARR	020	/* this is an array */
141#define	FCN	040	/* this is a function name */
142#define FLD	0100	/* this is a field $1, $2, ... */
143#define	REC	0200	/* this is $0 */
144#define CONVC	0400	/* string was converted from number via CONVFMT */
145#define CONVO	01000	/* string was converted from number via OFMT */
146
147
148/* function types */
149#define	FLENGTH	1
150#define	FSQRT	2
151#define	FEXP	3
152#define	FLOG	4
153#define	FINT	5
154#define	FSYSTEM	6
155#define	FRAND	7
156#define	FSRAND	8
157#define	FSIN	9
158#define	FCOS	10
159#define	FATAN	11
160#define	FTOUPPER 12
161#define	FTOLOWER 13
162#define	FFLUSH	14
163
164/* Node:  parse tree is made of nodes, with Cell's at bottom */
165
166typedef struct Node {
167	int	ntype;
168	struct	Node *nnext;
169	int	lineno;
170	int	nobj;
171	struct	Node *narg[1];	/* variable: actual size set by calling malloc */
172} Node;
173
174#define	NIL	((Node *) 0)
175
176extern Node	*winner;
177extern Node	*nullnode;
178
179/* ctypes */
180#define OCELL	1
181#define OBOOL	2
182#define OJUMP	3
183
184/* Cell subtypes: csub */
185#define	CFREE	7
186#define CCOPY	6
187#define CCON	5
188#define CTEMP	4
189#define CNAME	3
190#define CVAR	2
191#define CFLD	1
192#define	CUNK	0
193
194/* bool subtypes */
195#define BTRUE	11
196#define BFALSE	12
197
198/* jump subtypes */
199#define JEXIT	21
200#define JNEXT	22
201#define	JBREAK	23
202#define	JCONT	24
203#define	JRET	25
204#define	JNEXTFILE	26
205
206/* node types */
207#define NVALUE	1
208#define NSTAT	2
209#define NEXPR	3
210
211
212extern	int	pairstack[], paircnt;
213
214#define notlegal(n)	(n <= FIRSTTOKEN || n >= LASTTOKEN || proctab[n-FIRSTTOKEN] == nullproc)
215#define isvalue(n)	((n)->ntype == NVALUE)
216#define isexpr(n)	((n)->ntype == NEXPR)
217#define isjump(n)	((n)->ctype == OJUMP)
218#define isexit(n)	((n)->csub == JEXIT)
219#define	isbreak(n)	((n)->csub == JBREAK)
220#define	iscont(n)	((n)->csub == JCONT)
221#define	isnext(n)	((n)->csub == JNEXT || (n)->csub == JNEXTFILE)
222#define	isret(n)	((n)->csub == JRET)
223#define isrec(n)	((n)->tval & REC)
224#define isfld(n)	((n)->tval & FLD)
225#define isstr(n)	((n)->tval & STR)
226#define isnum(n)	((n)->tval & NUM)
227#define isarr(n)	((n)->tval & ARR)
228#define isfcn(n)	((n)->tval & FCN)
229#define istrue(n)	((n)->csub == BTRUE)
230#define istemp(n)	((n)->csub == CTEMP)
231#define	isargument(n)	((n)->nobj == ARG)
232/* #define freeable(p)	(!((p)->tval & DONTFREE)) */
233#define freeable(p)	( ((p)->tval & (STR|DONTFREE)) == STR )
234
235/* structures used by regular expression matching machinery, mostly b.c: */
236
237#define NCHARS	(1256+3)		/* 256 handles 8-bit chars; 128 does 7-bit */
238				/* BUG: some overflows (caught) if we use 256 */
239				/* watch out in match(), etc. */
240#define	HAT	(NCHARS+2)	/* matches ^ in regular expr */
241#define NSTATES	32
242
243typedef struct rrow {
244	long	ltype;	/* long avoids pointer warnings on 64-bit */
245	union {
246		int i;
247		Node *np;
248		uschar *up;
249		int *rp; /* rune representation of char class */
250	} lval;		/* because Al stores a pointer in it! */
251	int	*lfollow;
252} rrow;
253
254typedef struct gtte { /* gototab entry */
255	unsigned int ch;
256	unsigned int state;
257} gtte;
258
259#define GOTO_DIRECT	128	/* size of ASCII direct-index fast path */
260
261typedef struct gtt {	/* gototab */
262	size_t	allocated;
263	size_t	inuse;
264	gtte	*entries;
265	int	direct[GOTO_DIRECT];	/* fast path for ASCII; 0 == miss */
266} gtt;
267
268typedef struct fa {
269	gtt	*gototab;
270	uschar	*out;
271	uschar	*restr;
272	int	**posns;
273	int	state_count;
274	bool	anchor;
275	int	use;
276	int	initstat;
277	int	curstat;
278	int	accept;
279	struct	rrow re[1];	/* variable: actual size set by calling malloc */
280} fa;
281
282
283#include "proto.h"