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 <stddef.h>
27#include <stdint.h>
28
29#ifndef bool
30#define bool int
31#define true 1
32#define false 0
33#endif
34
35#if __STDC_VERSION__ <= 199901L
36#define noreturn
37#else
38#include <stdnoreturn.h>
39#endif
40
41typedef double Awkfloat;
42
43/* unsigned char is more trouble than it's worth */
44
45typedef unsigned char uschar;
46
47#define xfree(a) \
48 { \
49 free((void *)(intptr_t)(a)); \
50 (a) = NULL; \
51 }
52/*
53 * We sometimes cheat writing read-only pointers to NUL-terminate them
54 * and then put back the original value
55 */
56#define setptr(ptr, a) (*(char *)(intptr_t)(ptr)) = (a)
57
58#define NN(p) \
59 ((p) ? (p) : "(null)") /* guaranteed non-null for DPRINTF \
60 */
61#define DEBUG
62#ifdef DEBUG
63#define DPRINTF(...) \
64 if (dbg) \
65 printf(__VA_ARGS__)
66#else
67#define DPRINTF(...)
68#endif
69
70extern enum compile_states { RUNNING, COMPILING, ERROR_PRINTING } compile_time;
71
72extern bool safe; /* false => unsafe, true => safe */
73
74#define RECSIZE (8 * 1024) /* sets limit on records, fields, etc., etc. */
75extern int recsize; /* size of current record, orig RECSIZE */
76
77extern size_t awk_mb_cur_max; /* max size of a multi-byte character */
78
79extern char EMPTY[]; /* this avoid -Wwritable-strings issues */
80extern char **FS;
81extern char **RS;
82extern char **ORS;
83extern char **OFS;
84extern char **OFMT;
85extern Awkfloat *NR;
86extern Awkfloat *FNR;
87extern Awkfloat *NF;
88extern char **FILENAME;
89extern char **SUBSEP;
90extern Awkfloat *RSTART;
91extern Awkfloat *RLENGTH;
92
93extern bool CSV; /* true for csv input */
94
95extern char *record; /* points to $0 */
96extern int lineno; /* line number in awk program */
97extern int errorflag; /* 1 if error has occurred */
98extern bool donefld; /* true if record broken into fields */
99extern bool donerec; /* true if record is valid (no fld has changed */
100extern int dbg;
101
102extern const char *patbeg; /* beginning of pattern matched */
103extern int patlen; /* length of pattern matched. set in b.c */
104
105/* Cell: all information about a variable or constant */
106
107typedef struct Cell {
108 uschar ctype; /* OCELL, OBOOL, OJUMP, etc. */
109 uschar csub; /* CCON, CTEMP, CFLD, etc. */
110 char *nval; /* name, for variables only */
111 char *sval; /* string value */
112 Awkfloat fval; /* value as number */
113 int tval; /* type info: STR|NUM|ARR|FCN|FLD|CON|DONTFREE|CONVC|CONVO */
114 char *fmt; /* CONVFMT/OFMT value used to convert from number */
115 struct Cell *cnext; /* ptr to next if chained */
116} Cell;
117
118typedef struct Array { /* symbol table array */
119 int nelem; /* elements in table right now */
120 int size; /* size of tab */
121 Cell **tab; /* hash table pointers */
122} Array;
123
124#define NSYMTAB 50 /* initial size of a symbol table */
125extern Array *symtab;
126
127extern Cell *nrloc; /* NR */
128extern Cell *fnrloc; /* FNR */
129extern Cell *fsloc; /* FS */
130extern Cell *nfloc; /* NF */
131extern Cell *ofsloc; /* OFS */
132extern Cell *orsloc; /* ORS */
133extern Cell *rsloc; /* RS */
134extern Cell *rstartloc; /* RSTART */
135extern Cell *rlengthloc; /* RLENGTH */
136extern Cell *subseploc; /* SUBSEP */
137extern Cell *symtabloc; /* SYMTAB */
138
139/* Cell.tval values: */
140#define NUM 01 /* number value is valid */
141#define STR 02 /* string value is valid */
142#define DONTFREE 04 /* string space is not freeable */
143#define CON 010 /* this is a constant */
144#define ARR 020 /* this is an array */
145#define FCN 040 /* this is a function name */
146#define FLD 0100 /* this is a field $1, $2, ... */
147#define REC 0200 /* this is $0 */
148#define CONVC 0400 /* string was converted from number via CONVFMT */
149#define CONVO 01000 /* string was converted from number via OFMT */
150
151/* function types */
152#define FLENGTH 1
153#define FSQRT 2
154#define FEXP 3
155#define FLOG 4
156#define FINT 5
157#define FSYSTEM 6
158#define FRAND 7
159#define FSRAND 8
160#define FSIN 9
161#define FCOS 10
162#define FATAN 11
163#define FTOUPPER 12
164#define FTOLOWER 13
165#define FFLUSH 14
166
167/* Node: parse tree is made of nodes, with Cell's at bottom */
168
169typedef struct Node {
170 int ntype;
171 struct Node *nnext;
172 int lineno;
173 int nobj;
174 struct Node *narg[1]; /* variable: actual size set by calling malloc */
175} Node;
176
177#define NIL ((Node *)0)
178
179extern Node *winner;
180extern Node *nullnode;
181
182/* ctypes */
183#define OCELL 1
184#define OBOOL 2
185#define OJUMP 3
186
187/* Cell subtypes: csub */
188#define CFREE 7
189#define CCOPY 6
190#define CCON 5
191#define CTEMP 4
192#define CNAME 3
193#define CVAR 2
194#define CFLD 1
195#define CUNK 0
196
197/* bool subtypes */
198#define BTRUE 11
199#define BFALSE 12
200
201/* jump subtypes */
202#define JEXIT 21
203#define JNEXT 22
204#define JBREAK 23
205#define JCONT 24
206#define JRET 25
207#define JNEXTFILE 26
208
209/* node types */
210#define NVALUE 1
211#define NSTAT 2
212#define NEXPR 3
213
214extern int pairstack[], paircnt;
215
216#define notlegal(n) (n <= FIRSTTOKEN || n >= LASTTOKEN || proctab[n - FIRSTTOKEN] == nullproc)
217#define isvalue(n) ((n)->ntype == NVALUE)
218#define isexpr(n) ((n)->ntype == NEXPR)
219#define isjump(n) ((n)->ctype == OJUMP)
220#define isexit(n) ((n)->csub == JEXIT)
221#define isbreak(n) ((n)->csub == JBREAK)
222#define iscont(n) ((n)->csub == JCONT)
223#define isnext(n) ((n)->csub == JNEXT || (n)->csub == JNEXTFILE)
224#define isret(n) ((n)->csub == JRET)
225#define isrec(n) ((n)->tval & REC)
226#define isfld(n) ((n)->tval & FLD)
227#define isstr(n) ((n)->tval & STR)
228#define isnum(n) ((n)->tval & NUM)
229#define isarr(n) ((n)->tval & ARR)
230#define isfcn(n) ((n)->tval & FCN)
231#define istrue(n) ((n)->csub == BTRUE)
232#define istemp(n) ((n)->csub == CTEMP)
233#define isargument(n) ((n)->nobj == ARG)
234/* #define freeable(p) (!((p)->tval & DONTFREE)) */
235#define freeable(p) (((p)->tval & (STR | DONTFREE)) == STR)
236
237/* structures used by regular expression matching machinery, mostly b.c: */
238
239#define NCHARS (1256 + 3) /* 256 handles 8-bit chars; 128 does 7-bit */
240 /* BUG: some overflows (caught) if we use 256 */
241 /* watch out in match(), etc. */
242#define HAT (NCHARS + 2) /* matches ^ in regular expr */
243#define NSTATES 32
244
245typedef struct rrow {
246 long ltype; /* long avoids pointer warnings on 64-bit */
247 union {
248 int i;
249 Node *np;
250 uschar *up;
251 int *rp; /* rune representation of char class */
252 } lval; /* because Al stores a pointer in it! */
253 int *lfollow;
254} rrow;
255
256typedef struct gtte { /* gototab entry */
257 unsigned int ch;
258 unsigned int state;
259} gtte;
260
261#define GOTO_DIRECT 128 /* size of ASCII direct-index fast path */
262
263typedef struct gtt { /* gototab */
264 size_t allocated;
265 size_t inuse;
266 gtte *entries;
267 int direct[GOTO_DIRECT]; /* fast path for ASCII; 0 == miss */
268} gtt;
269
270typedef struct fa {
271 gtt *gototab;
272 uschar *out;
273 uschar *restr;
274 int **posns;
275 int state_count;
276 bool anchor;
277 int use;
278 int initstat;
279 int curstat;
280 int accept;
281 struct rrow re[1]; /* variable: actual size set by calling malloc */
282} fa;
283
284#include "proto.h"
