1 /* $NetBSD: lex.l,v 1.3 2019/12/15 22:50:47 christos Exp $ */ 2 3 %{ 4 /* 5 * Copyright (c) 1997 - 2017 Kungliga Tekniska Högskolan 6 * (Royal Institute of Technology, Stockholm, Sweden). 7 * All rights reserved. 8 * 9 * Redistribution and use in source and binary forms, with or without 10 * modification, are permitted provided that the following conditions 11 * are met: 12 * 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 20 * 3. Neither the name of the Institute nor the names of its contributors 21 * may be used to endorse or promote products derived from this software 22 * without specific prior written permission. 23 * 24 * THIS SOFTWARE IS PROVIDED BY THE INSTITUTE AND CONTRIBUTORS ``AS IS'' AND 25 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 27 * ARE DISCLAIMED. IN NO EVENT SHALL THE INSTITUTE OR CONTRIBUTORS BE LIABLE 28 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 29 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 30 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 31 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 32 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 33 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 34 * SUCH DAMAGE. 35 */ 36 37 /* Id */ 38 39 #ifdef HAVE_CONFIG_H 40 #include <config.h> 41 #endif 42 #include <stdio.h> 43 #include <stdarg.h> 44 #include <stdlib.h> 45 #include <stdint.h> 46 #include <string.h> 47 #ifdef HAVE_UNISTD_H 48 #include <unistd.h> 49 #endif 50 #undef ECHO 51 #include "symbol.h" 52 #include "asn1parse.h" 53 #include "lex.h" 54 #include "gen_locl.h" 55 56 static unsigned lineno = 1; 57 58 #undef ECHO 59 60 static void unterminated(const char *, unsigned); 61 62 %} 63 64 /* This is for broken old lexes (solaris 10 and hpux) */ 65 %e 2000 66 %p 5000 67 %a 5000 68 %n 1000 69 %o 10000 70 71 %% 72 ABSENT { return kw_ABSENT; } 73 ABSTRACT-SYNTAX { return kw_ABSTRACT_SYNTAX; } 74 ALL { return kw_ALL; } 75 APPLICATION { return kw_APPLICATION; } 76 AUTOMATIC { return kw_AUTOMATIC; } 77 BEGIN { return kw_BEGIN; } 78 BIT { return kw_BIT; } 79 BMPString { return kw_BMPString; } 80 BOOLEAN { return kw_BOOLEAN; } 81 BY { return kw_BY; } 82 CHARACTER { return kw_CHARACTER; } 83 CHOICE { return kw_CHOICE; } 84 CLASS { return kw_CLASS; } 85 COMPONENT { return kw_COMPONENT; } 86 COMPONENTS { return kw_COMPONENTS; } 87 CONSTRAINED { return kw_CONSTRAINED; } 88 CONTAINING { return kw_CONTAINING; } 89 DEFAULT { return kw_DEFAULT; } 90 DEFINITIONS { return kw_DEFINITIONS; } 91 EMBEDDED { return kw_EMBEDDED; } 92 ENCODED { return kw_ENCODED; } 93 END { return kw_END; } 94 ENUMERATED { return kw_ENUMERATED; } 95 EXCEPT { return kw_EXCEPT; } 96 EXPLICIT { return kw_EXPLICIT; } 97 EXPORTS { return kw_EXPORTS; } 98 EXTENSIBILITY { return kw_EXTENSIBILITY; } 99 EXTERNAL { return kw_EXTERNAL; } 100 FALSE { return kw_FALSE; } 101 FROM { return kw_FROM; } 102 GeneralString { return kw_GeneralString; } 103 GeneralizedTime { return kw_GeneralizedTime; } 104 GraphicString { return kw_GraphicString; } 105 IA5String { return kw_IA5String; } 106 IDENTIFIER { return kw_IDENTIFIER; } 107 IMPLICIT { return kw_IMPLICIT; } 108 IMPLIED { return kw_IMPLIED; } 109 IMPORTS { return kw_IMPORTS; } 110 INCLUDES { return kw_INCLUDES; } 111 INSTANCE { return kw_INSTANCE; } 112 INTEGER { return kw_INTEGER; } 113 INTERSECTION { return kw_INTERSECTION; } 114 ISO646String { return kw_ISO646String; } 115 MAX { return kw_MAX; } 116 MIN { return kw_MIN; } 117 MINUS-INFINITY { return kw_MINUS_INFINITY; } 118 NULL { return kw_NULL; } 119 NumericString { return kw_NumericString; } 120 OBJECT { return kw_OBJECT; } 121 OCTET { return kw_OCTET; } 122 OF { return kw_OF; } 123 OPTIONAL { return kw_OPTIONAL; } 124 ObjectDescriptor { return kw_ObjectDescriptor; } 125 PATTERN { return kw_PATTERN; } 126 PDV { return kw_PDV; } 127 PLUS-INFINITY { return kw_PLUS_INFINITY; } 128 PRESENT { return kw_PRESENT; } 129 PRIVATE { return kw_PRIVATE; } 130 PrintableString { return kw_PrintableString; } 131 REAL { return kw_REAL; } 132 RELATIVE_OID { return kw_RELATIVE_OID; } 133 SEQUENCE { return kw_SEQUENCE; } 134 SET { return kw_SET; } 135 SIZE { return kw_SIZE; } 136 STRING { return kw_STRING; } 137 SYNTAX { return kw_SYNTAX; } 138 T61String { return kw_T61String; } 139 TAGS { return kw_TAGS; } 140 TRUE { return kw_TRUE; } 141 TYPE-IDENTIFIER { return kw_TYPE_IDENTIFIER; } 142 TeletexString { return kw_TeletexString; } 143 UNION { return kw_UNION; } 144 UNIQUE { return kw_UNIQUE; } 145 UNIVERSAL { return kw_UNIVERSAL; } 146 UTCTime { return kw_UTCTime; } 147 UTF8String { return kw_UTF8String; } 148 UniversalString { return kw_UniversalString; } 149 VideotexString { return kw_VideotexString; } 150 VisibleString { return kw_VisibleString; } 151 WITH { return kw_WITH; } 152 [-,;{}()|] { return *yytext; } 153 "[" { return *yytext; } 154 "]" { return *yytext; } 155 ::= { return EEQUAL; } 156 -- { 157 int c, start_lineno = lineno; 158 int f = 0; 159 while((c = input()) != EOF) { 160 if(f && c == '-') 161 break; 162 if(c == '-') { 163 f = 1; 164 continue; 165 } 166 if(c == '\n') { 167 lineno++; 168 break; 169 } 170 f = 0; 171 } 172 if(c == EOF) 173 unterminated("comment", start_lineno); 174 } 175 \/\* { 176 int c, start_lineno = lineno; 177 int level = 1; 178 int seen_star = 0; 179 int seen_slash = 0; 180 while((c = input()) != EOF) { 181 if(c == '/') { 182 if(seen_star) { 183 if(--level == 0) 184 break; 185 seen_star = 0; 186 continue; 187 } 188 seen_slash = 1; 189 continue; 190 } 191 if(seen_star && c == '/') { 192 if(--level == 0) 193 break; 194 seen_star = 0; 195 continue; 196 } 197 if(c == '*') { 198 if(seen_slash) { 199 level++; 200 seen_star = seen_slash = 0; 201 continue; 202 } 203 seen_star = 1; 204 continue; 205 } 206 seen_star = seen_slash = 0; 207 if(c == '\n') { 208 lineno++; 209 continue; 210 } 211 } 212 if(c == EOF) 213 unterminated("comment", start_lineno); 214 } 215 "\"" { 216 int start_lineno = lineno; 217 int c; 218 char buf[1024]; 219 char *p = buf; 220 int f = 0; 221 int skip_ws = 0; 222 223 while((c = input()) != EOF) { 224 if(isspace(c) && skip_ws) { 225 if(c == '\n') 226 lineno++; 227 continue; 228 } 229 skip_ws = 0; 230 231 if(c == '"') { 232 if(f) { 233 *p++ = '"'; 234 f = 0; 235 } else 236 f = 1; 237 continue; 238 } 239 if(f == 1) { 240 unput(c); 241 break; 242 } 243 if(c == '\n') { 244 lineno++; 245 while(p > buf && isspace((unsigned char)p[-1])) 246 p--; 247 skip_ws = 1; 248 continue; 249 } 250 *p++ = c; 251 } 252 if(c == EOF) 253 unterminated("string", start_lineno); 254 *p++ = '\0'; 255 fprintf(stderr, "string -- %s\n", buf); 256 yylval.name = estrdup(buf); 257 return STRING; 258 } 259 260 -?0x[0-9A-Fa-f]+|-?[0-9]+ { char *e, *y = yytext; 261 yylval.constant = strtoll((const char *)yytext, 262 &e, 0); 263 if(e == y) 264 lex_error_message("malformed constant (%s)", yytext); 265 else 266 return NUMBER; 267 } 268 [A-Za-z][-A-Za-z0-9_]* { 269 yylval.name = estrdup ((const char *)yytext); 270 return IDENTIFIER; 271 } 272 [ \t] ; 273 \n { ++lineno; } 274 \.\.\. { return ELLIPSIS; } 275 \.\. { return RANGE; } 276 . { lex_error_message("Ignoring char(%c)\n", *yytext); } 277 %% 278 279 int 280 yywrap () 281 { 282 return 1; 283 } 284 285 void 286 lex_error_message (const char *format, ...) 287 { 288 va_list args; 289 290 va_start (args, format); 291 fprintf (stderr, "%s:%d: ", get_filename(), lineno); 292 vfprintf (stderr, format, args); 293 va_end (args); 294 error_flag++; 295 } 296 297 static void 298 unterminated(const char *type, unsigned start_lineno) 299 { 300 lex_error_message("unterminated %s, possibly started on line %d\n", type, start_lineno); 301 } 302