Daniel Veillard | 260a68f | 1998-08-13 03:39:55 +0000 | [diff] [blame] | 1 | /* |
| 2 | * parser.c : an XML 1.0 non-verifying parser |
| 3 | * |
| 4 | * See Copyright for the status of this software. |
| 5 | * |
| 6 | * $Id$ |
| 7 | */ |
| 8 | |
| 9 | #ifdef WIN32 |
| 10 | #define HAVE_FCNTL_H |
| 11 | #include <io.h> |
| 12 | #else |
| 13 | #include <config.h> |
| 14 | #endif |
| 15 | #include <stdio.h> |
| 16 | #include <ctype.h> |
| 17 | #include <string.h> /* for memset() only */ |
| 18 | #include <malloc.h> |
| 19 | #include <sys/stat.h> |
| 20 | #ifdef HAVE_FCNTL_H |
| 21 | #include <fcntl.h> |
| 22 | #endif |
| 23 | #ifdef HAVE_UNISTD_H |
| 24 | #include <unistd.h> |
| 25 | #endif |
| 26 | #ifdef HAVE_ZLIB_H |
| 27 | #include <zlib.h> |
| 28 | #endif |
| 29 | |
| 30 | #include "tree.h" |
| 31 | #include "parser.h" |
| 32 | #include "entities.h" |
| 33 | |
| 34 | /************************************************************************ |
| 35 | * * |
| 36 | * Parser stacks related functions and macros * |
| 37 | * * |
| 38 | ************************************************************************/ |
| 39 | /* |
| 40 | * Generic function for accessing stacks in the Parser Context |
| 41 | */ |
| 42 | |
| 43 | #define PUSH_AND_POP(type, name) \ |
| 44 | int name##Push(xmlParserCtxtPtr ctxt, type value) { \ |
| 45 | if (ctxt->name##Nr >= ctxt->name##Max) { \ |
| 46 | ctxt->name##Max *= 2; \ |
| 47 | ctxt->name##Tab = (void *) realloc(ctxt->name##Tab, \ |
| 48 | ctxt->name##Max * sizeof(ctxt->name##Tab[0])); \ |
| 49 | if (ctxt->name##Tab == NULL) { \ |
| 50 | fprintf(stderr, "realloc failed !\n"); \ |
| 51 | exit(1); \ |
| 52 | } \ |
| 53 | } \ |
| 54 | ctxt->name##Tab[ctxt->name##Nr] = value; \ |
| 55 | ctxt->name = value; \ |
| 56 | return(ctxt->name##Nr++); \ |
| 57 | } \ |
| 58 | type name##Pop(xmlParserCtxtPtr ctxt) { \ |
| 59 | if (ctxt->name##Nr <= 0) return(0); \ |
| 60 | ctxt->name##Nr--; \ |
| 61 | ctxt->name = ctxt->name##Tab[ctxt->name##Nr - 1]; \ |
| 62 | return(ctxt->name); \ |
| 63 | } \ |
| 64 | |
| 65 | PUSH_AND_POP(xmlParserInputPtr, input) |
| 66 | PUSH_AND_POP(xmlNodePtr, node) |
| 67 | |
| 68 | /************* |
| 69 | #define CUR (*(ctxt->input->cur) ? *(ctxt->input->cur) : xmlPopInput(ctxt)) |
| 70 | #define NEXT (((*(ctxt->input->cur) == '\n') ? \ |
| 71 | (ctxt->input->line++, ctxt->input->col = 1) : \ |
| 72 | (ctxt->input->col++)), ctxt->input->cur++) |
| 73 | *************/ |
| 74 | |
| 75 | #define CUR (*ctxt->input->cur) |
| 76 | #define NEXT ((*ctxt->input->cur) ? \ |
| 77 | (((*(ctxt->input->cur) == '\n') ? \ |
| 78 | (ctxt->input->line++, ctxt->input->col = 1) : \ |
| 79 | (ctxt->input->col++)), ctxt->input->cur++) : \ |
| 80 | (xmlPopInput(ctxt), ctxt->input->cur)) |
| 81 | |
| 82 | #define CUR_PTR ctxt->input->cur |
| 83 | |
| 84 | #define NXT(val) ctxt->input->cur[(val)] |
| 85 | |
| 86 | #define SKIP(val) ctxt->input->cur += (val) |
| 87 | #define SKIP_BLANKS \ |
| 88 | while (IS_BLANK(*(ctxt->input->cur))) NEXT |
| 89 | |
| 90 | |
| 91 | /* |
| 92 | * xmlPopInput: the current input pointed by ctxt->input came to an end |
| 93 | * pop it and return the next char. |
| 94 | * |
| 95 | * TODO A deallocation of the popped Input structure is needed |
| 96 | */ |
| 97 | CHAR xmlPopInput(xmlParserCtxtPtr ctxt) { |
| 98 | if (ctxt->inputNr == 1) return(0); /* End of main Input */ |
| 99 | inputPop(ctxt); |
| 100 | return(CUR); |
| 101 | } |
| 102 | |
| 103 | /* |
| 104 | * xmlPushInput: switch to a new input stream which is stacked on top |
| 105 | * of the previous one(s). |
| 106 | */ |
| 107 | void xmlPushInput(xmlParserCtxtPtr ctxt, xmlParserInputPtr input) { |
| 108 | if (input == NULL) return; |
| 109 | inputPush(ctxt, input); |
| 110 | } |
| 111 | |
| 112 | /* |
| 113 | * Create a new input stream based on a memory buffer. |
| 114 | */ |
| 115 | void xmlNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { |
| 116 | xmlParserInputPtr input; |
| 117 | |
| 118 | if (entity == NULL) { |
| 119 | xmlParserError(ctxt, |
| 120 | "internal: xmlNewEntityInputStream entity = NULL\n"); |
| 121 | return; |
| 122 | } |
| 123 | if (entity->content == NULL) { |
| 124 | xmlParserError(ctxt, |
| 125 | "internal: xmlNewEntityInputStream entity->input = NULL\n"); |
| 126 | return; |
| 127 | } |
| 128 | input = (xmlParserInputPtr) malloc(sizeof(xmlParserInput)); |
| 129 | if (input == NULL) { |
| 130 | xmlParserError(ctxt, "malloc: couldn't allocate a new input stream\n"); |
| 131 | return; |
| 132 | } |
| 133 | input->filename = entity->SystemID; /* TODO !!! char <- CHAR */ |
| 134 | input->base = entity->content; |
| 135 | input->cur = entity->content; |
| 136 | input->line = 1; |
| 137 | input->col = 1; |
| 138 | xmlPushInput(ctxt, input); |
| 139 | } |
| 140 | |
| 141 | /* |
| 142 | * A few macros needed to help building the parser. |
| 143 | */ |
| 144 | |
| 145 | #ifdef UNICODE |
| 146 | /************************************************************************ |
| 147 | * * |
| 148 | * UNICODE version of the macros. * |
| 149 | * * |
| 150 | ************************************************************************/ |
| 151 | /* |
| 152 | * [2] Char ::= #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD] |
| 153 | * | [#x10000-#x10FFFF] |
| 154 | * any Unicode character, excluding the surrogate blocks, FFFE, and FFFF. |
| 155 | */ |
| 156 | #define IS_CHAR(c) \ |
| 157 | (((c) == 0x09) || ((c) == 0x0a) || ((c) == 0x0d) || \ |
| 158 | (((c) >= 0x20) && ((c) != 0xFFFE) && ((c) != 0xFFFF))) |
| 159 | |
| 160 | /* |
| 161 | * [3] S ::= (#x20 | #x9 | #xD | #xA)+ |
| 162 | */ |
| 163 | #define IS_BLANK(c) (((c) == 0x20) || ((c) == 0x09) || ((c) == 0xa) || \ |
| 164 | ((c) == 0x0D)) |
| 165 | |
| 166 | /* |
| 167 | * [85] BaseChar ::= ... long list see REC ... |
| 168 | * |
| 169 | * VI is your friend ! |
| 170 | * :1,$ s/\[#x\([0-9A-Z]*\)-#x\([0-9A-Z]*\)\]/ (((c) >= 0x\1) \&\& ((c) <= 0x\2)) ||/ |
| 171 | * and |
| 172 | * :1,$ s/#x\([0-9A-Z]*\)/ ((c) == 0x\1) ||/ |
| 173 | */ |
| 174 | #define IS_BASECHAR(c) \ |
| 175 | ((((c) >= 0x0041) && ((c) <= 0x005A)) || \ |
| 176 | (((c) >= 0x0061) && ((c) <= 0x007A)) || \ |
| 177 | (((c) >= 0x00C0) && ((c) <= 0x00D6)) || \ |
| 178 | (((c) >= 0x00D8) && ((c) <= 0x00F6)) || \ |
| 179 | (((c) >= 0x00F8) && ((c) <= 0x00FF)) || \ |
| 180 | (((c) >= 0x0100) && ((c) <= 0x0131)) || \ |
| 181 | (((c) >= 0x0134) && ((c) <= 0x013E)) || \ |
| 182 | (((c) >= 0x0141) && ((c) <= 0x0148)) || \ |
| 183 | (((c) >= 0x014A) && ((c) <= 0x017E)) || \ |
| 184 | (((c) >= 0x0180) && ((c) <= 0x01C3)) || \ |
| 185 | (((c) >= 0x01CD) && ((c) <= 0x01F0)) || \ |
| 186 | (((c) >= 0x01F4) && ((c) <= 0x01F5)) || \ |
| 187 | (((c) >= 0x01FA) && ((c) <= 0x0217)) || \ |
| 188 | (((c) >= 0x0250) && ((c) <= 0x02A8)) || \ |
| 189 | (((c) >= 0x02BB) && ((c) <= 0x02C1)) || \ |
| 190 | ((c) == 0x0386) || \ |
| 191 | (((c) >= 0x0388) && ((c) <= 0x038A)) || \ |
| 192 | ((c) == 0x038C) || \ |
| 193 | (((c) >= 0x038E) && ((c) <= 0x03A1)) || \ |
| 194 | (((c) >= 0x03A3) && ((c) <= 0x03CE)) || \ |
| 195 | (((c) >= 0x03D0) && ((c) <= 0x03D6)) || \ |
| 196 | ((c) == 0x03DA) || \ |
| 197 | ((c) == 0x03DC) || \ |
| 198 | ((c) == 0x03DE) || \ |
| 199 | ((c) == 0x03E0) || \ |
| 200 | (((c) >= 0x03E2) && ((c) <= 0x03F3)) || \ |
| 201 | (((c) >= 0x0401) && ((c) <= 0x040C)) || \ |
| 202 | (((c) >= 0x040E) && ((c) <= 0x044F)) || \ |
| 203 | (((c) >= 0x0451) && ((c) <= 0x045C)) || \ |
| 204 | (((c) >= 0x045E) && ((c) <= 0x0481)) || \ |
| 205 | (((c) >= 0x0490) && ((c) <= 0x04C4)) || \ |
| 206 | (((c) >= 0x04C7) && ((c) <= 0x04C8)) || \ |
| 207 | (((c) >= 0x04CB) && ((c) <= 0x04CC)) || \ |
| 208 | (((c) >= 0x04D0) && ((c) <= 0x04EB)) || \ |
| 209 | (((c) >= 0x04EE) && ((c) <= 0x04F5)) || \ |
| 210 | (((c) >= 0x04F8) && ((c) <= 0x04F9)) || \ |
| 211 | (((c) >= 0x0531) && ((c) <= 0x0556)) || \ |
| 212 | ((c) == 0x0559) || \ |
| 213 | (((c) >= 0x0561) && ((c) <= 0x0586)) || \ |
| 214 | (((c) >= 0x05D0) && ((c) <= 0x05EA)) || \ |
| 215 | (((c) >= 0x05F0) && ((c) <= 0x05F2)) || \ |
| 216 | (((c) >= 0x0621) && ((c) <= 0x063A)) || \ |
| 217 | (((c) >= 0x0641) && ((c) <= 0x064A)) || \ |
| 218 | (((c) >= 0x0671) && ((c) <= 0x06B7)) || \ |
| 219 | (((c) >= 0x06BA) && ((c) <= 0x06BE)) || \ |
| 220 | (((c) >= 0x06C0) && ((c) <= 0x06CE)) || \ |
| 221 | (((c) >= 0x06D0) && ((c) <= 0x06D3)) || \ |
| 222 | ((c) == 0x06D5) || \ |
| 223 | (((c) >= 0x06E5) && ((c) <= 0x06E6)) || \ |
| 224 | (((c) >= 0x0905) && ((c) <= 0x0939)) || \ |
| 225 | ((c) == 0x093D) || \ |
| 226 | (((c) >= 0x0958) && ((c) <= 0x0961)) || \ |
| 227 | (((c) >= 0x0985) && ((c) <= 0x098C)) || \ |
| 228 | (((c) >= 0x098F) && ((c) <= 0x0990)) || \ |
| 229 | (((c) >= 0x0993) && ((c) <= 0x09A8)) || \ |
| 230 | (((c) >= 0x09AA) && ((c) <= 0x09B0)) || \ |
| 231 | ((c) == 0x09B2) || \ |
| 232 | (((c) >= 0x09B6) && ((c) <= 0x09B9)) || \ |
| 233 | (((c) >= 0x09DC) && ((c) <= 0x09DD)) || \ |
| 234 | (((c) >= 0x09DF) && ((c) <= 0x09E1)) || \ |
| 235 | (((c) >= 0x09F0) && ((c) <= 0x09F1)) || \ |
| 236 | (((c) >= 0x0A05) && ((c) <= 0x0A0A)) || \ |
| 237 | (((c) >= 0x0A0F) && ((c) <= 0x0A10)) || \ |
| 238 | (((c) >= 0x0A13) && ((c) <= 0x0A28)) || \ |
| 239 | (((c) >= 0x0A2A) && ((c) <= 0x0A30)) || \ |
| 240 | (((c) >= 0x0A32) && ((c) <= 0x0A33)) || \ |
| 241 | (((c) >= 0x0A35) && ((c) <= 0x0A36)) || \ |
| 242 | (((c) >= 0x0A38) && ((c) <= 0x0A39)) || \ |
| 243 | (((c) >= 0x0A59) && ((c) <= 0x0A5C)) || \ |
| 244 | ((c) == 0x0A5E) || \ |
| 245 | (((c) >= 0x0A72) && ((c) <= 0x0A74)) || \ |
| 246 | (((c) >= 0x0A85) && ((c) <= 0x0A8B)) || \ |
| 247 | ((c) == 0x0A8D) || \ |
| 248 | (((c) >= 0x0A8F) && ((c) <= 0x0A91)) || \ |
| 249 | (((c) >= 0x0A93) && ((c) <= 0x0AA8)) || \ |
| 250 | (((c) >= 0x0AAA) && ((c) <= 0x0AB0)) || \ |
| 251 | (((c) >= 0x0AB2) && ((c) <= 0x0AB3)) || \ |
| 252 | (((c) >= 0x0AB5) && ((c) <= 0x0AB9)) || \ |
| 253 | ((c) == 0x0ABD) || \ |
| 254 | ((c) == 0x0AE0) || \ |
| 255 | (((c) >= 0x0B05) && ((c) <= 0x0B0C)) || \ |
| 256 | (((c) >= 0x0B0F) && ((c) <= 0x0B10)) || \ |
| 257 | (((c) >= 0x0B13) && ((c) <= 0x0B28)) || \ |
| 258 | (((c) >= 0x0B2A) && ((c) <= 0x0B30)) || \ |
| 259 | (((c) >= 0x0B32) && ((c) <= 0x0B33)) || \ |
| 260 | (((c) >= 0x0B36) && ((c) <= 0x0B39)) || \ |
| 261 | ((c) == 0x0B3D) || \ |
| 262 | (((c) >= 0x0B5C) && ((c) <= 0x0B5D)) || \ |
| 263 | (((c) >= 0x0B5F) && ((c) <= 0x0B61)) || \ |
| 264 | (((c) >= 0x0B85) && ((c) <= 0x0B8A)) || \ |
| 265 | (((c) >= 0x0B8E) && ((c) <= 0x0B90)) || \ |
| 266 | (((c) >= 0x0B92) && ((c) <= 0x0B95)) || \ |
| 267 | (((c) >= 0x0B99) && ((c) <= 0x0B9A)) || \ |
| 268 | ((c) == 0x0B9C) || \ |
| 269 | (((c) >= 0x0B9E) && ((c) <= 0x0B9F)) || \ |
| 270 | (((c) >= 0x0BA3) && ((c) <= 0x0BA4)) || \ |
| 271 | (((c) >= 0x0BA8) && ((c) <= 0x0BAA)) || \ |
| 272 | (((c) >= 0x0BAE) && ((c) <= 0x0BB5)) || \ |
| 273 | (((c) >= 0x0BB7) && ((c) <= 0x0BB9)) || \ |
| 274 | (((c) >= 0x0C05) && ((c) <= 0x0C0C)) || \ |
| 275 | (((c) >= 0x0C0E) && ((c) <= 0x0C10)) || \ |
| 276 | (((c) >= 0x0C12) && ((c) <= 0x0C28)) || \ |
| 277 | (((c) >= 0x0C2A) && ((c) <= 0x0C33)) || \ |
| 278 | (((c) >= 0x0C35) && ((c) <= 0x0C39)) || \ |
| 279 | (((c) >= 0x0C60) && ((c) <= 0x0C61)) || \ |
| 280 | (((c) >= 0x0C85) && ((c) <= 0x0C8C)) || \ |
| 281 | (((c) >= 0x0C8E) && ((c) <= 0x0C90)) || \ |
| 282 | (((c) >= 0x0C92) && ((c) <= 0x0CA8)) || \ |
| 283 | (((c) >= 0x0CAA) && ((c) <= 0x0CB3)) || \ |
| 284 | (((c) >= 0x0CB5) && ((c) <= 0x0CB9)) || \ |
| 285 | ((c) == 0x0CDE) || \ |
| 286 | (((c) >= 0x0CE0) && ((c) <= 0x0CE1)) || \ |
| 287 | (((c) >= 0x0D05) && ((c) <= 0x0D0C)) || \ |
| 288 | (((c) >= 0x0D0E) && ((c) <= 0x0D10)) || \ |
| 289 | (((c) >= 0x0D12) && ((c) <= 0x0D28)) || \ |
| 290 | (((c) >= 0x0D2A) && ((c) <= 0x0D39)) || \ |
| 291 | (((c) >= 0x0D60) && ((c) <= 0x0D61)) || \ |
| 292 | (((c) >= 0x0E01) && ((c) <= 0x0E2E)) || \ |
| 293 | ((c) == 0x0E30) || \ |
| 294 | (((c) >= 0x0E32) && ((c) <= 0x0E33)) || \ |
| 295 | (((c) >= 0x0E40) && ((c) <= 0x0E45)) || \ |
| 296 | (((c) >= 0x0E81) && ((c) <= 0x0E82)) || \ |
| 297 | ((c) == 0x0E84) || \ |
| 298 | (((c) >= 0x0E87) && ((c) <= 0x0E88)) || \ |
| 299 | ((c) == 0x0E8A) || \ |
| 300 | ((c) == 0x0E8D) || \ |
| 301 | (((c) >= 0x0E94) && ((c) <= 0x0E97)) || \ |
| 302 | (((c) >= 0x0E99) && ((c) <= 0x0E9F)) || \ |
| 303 | (((c) >= 0x0EA1) && ((c) <= 0x0EA3)) || \ |
| 304 | ((c) == 0x0EA5) || \ |
| 305 | ((c) == 0x0EA7) || \ |
| 306 | (((c) >= 0x0EAA) && ((c) <= 0x0EAB)) || \ |
| 307 | (((c) >= 0x0EAD) && ((c) <= 0x0EAE)) || \ |
| 308 | ((c) == 0x0EB0) || \ |
| 309 | (((c) >= 0x0EB2) && ((c) <= 0x0EB3)) || \ |
| 310 | ((c) == 0x0EBD) || \ |
| 311 | (((c) >= 0x0EC0) && ((c) <= 0x0EC4)) || \ |
| 312 | (((c) >= 0x0F40) && ((c) <= 0x0F47)) || \ |
| 313 | (((c) >= 0x0F49) && ((c) <= 0x0F69)) || \ |
| 314 | (((c) >= 0x10A0) && ((c) <= 0x10C5)) || \ |
| 315 | (((c) >= 0x10D0) && ((c) <= 0x10F6)) || \ |
| 316 | ((c) == 0x1100) || \ |
| 317 | (((c) >= 0x1102) && ((c) <= 0x1103)) || \ |
| 318 | (((c) >= 0x1105) && ((c) <= 0x1107)) || \ |
| 319 | ((c) == 0x1109) || \ |
| 320 | (((c) >= 0x110B) && ((c) <= 0x110C)) || \ |
| 321 | (((c) >= 0x110E) && ((c) <= 0x1112)) || \ |
| 322 | ((c) == 0x113C) || \ |
| 323 | ((c) == 0x113E) || \ |
| 324 | ((c) == 0x1140) || \ |
| 325 | ((c) == 0x114C) || \ |
| 326 | ((c) == 0x114E) || \ |
| 327 | ((c) == 0x1150) || \ |
| 328 | (((c) >= 0x1154) && ((c) <= 0x1155)) || \ |
| 329 | ((c) == 0x1159) || \ |
| 330 | (((c) >= 0x115F) && ((c) <= 0x1161)) || \ |
| 331 | ((c) == 0x1163) || \ |
| 332 | ((c) == 0x1165) || \ |
| 333 | ((c) == 0x1167) || \ |
| 334 | ((c) == 0x1169) || \ |
| 335 | (((c) >= 0x116D) && ((c) <= 0x116E)) || \ |
| 336 | (((c) >= 0x1172) && ((c) <= 0x1173)) || \ |
| 337 | ((c) == 0x1175) || \ |
| 338 | ((c) == 0x119E) || \ |
| 339 | ((c) == 0x11A8) || \ |
| 340 | ((c) == 0x11AB) || \ |
| 341 | (((c) >= 0x11AE) && ((c) <= 0x11AF)) || \ |
| 342 | (((c) >= 0x11B7) && ((c) <= 0x11B8)) || \ |
| 343 | ((c) == 0x11BA) || \ |
| 344 | (((c) >= 0x11BC) && ((c) <= 0x11C2)) || \ |
| 345 | ((c) == 0x11EB) || \ |
| 346 | ((c) == 0x11F0) || \ |
| 347 | ((c) == 0x11F9) || \ |
| 348 | (((c) >= 0x1E00) && ((c) <= 0x1E9B)) || \ |
| 349 | (((c) >= 0x1EA0) && ((c) <= 0x1EF9)) || \ |
| 350 | (((c) >= 0x1F00) && ((c) <= 0x1F15)) || \ |
| 351 | (((c) >= 0x1F18) && ((c) <= 0x1F1D)) || \ |
| 352 | (((c) >= 0x1F20) && ((c) <= 0x1F45)) || \ |
| 353 | (((c) >= 0x1F48) && ((c) <= 0x1F4D)) || \ |
| 354 | (((c) >= 0x1F50) && ((c) <= 0x1F57)) || \ |
| 355 | ((c) == 0x1F59) || \ |
| 356 | ((c) == 0x1F5B) || \ |
| 357 | ((c) == 0x1F5D) || \ |
| 358 | (((c) >= 0x1F5F) && ((c) <= 0x1F7D)) || \ |
| 359 | (((c) >= 0x1F80) && ((c) <= 0x1FB4)) || \ |
| 360 | (((c) >= 0x1FB6) && ((c) <= 0x1FBC)) || \ |
| 361 | ((c) == 0x1FBE) || \ |
| 362 | (((c) >= 0x1FC2) && ((c) <= 0x1FC4)) || \ |
| 363 | (((c) >= 0x1FC6) && ((c) <= 0x1FCC)) || \ |
| 364 | (((c) >= 0x1FD0) && ((c) <= 0x1FD3)) || \ |
| 365 | (((c) >= 0x1FD6) && ((c) <= 0x1FDB)) || \ |
| 366 | (((c) >= 0x1FE0) && ((c) <= 0x1FEC)) || \ |
| 367 | (((c) >= 0x1FF2) && ((c) <= 0x1FF4)) || \ |
| 368 | (((c) >= 0x1FF6) && ((c) <= 0x1FFC)) || \ |
| 369 | ((c) == 0x2126) || \ |
| 370 | (((c) >= 0x212A) && ((c) <= 0x212B)) || \ |
| 371 | ((c) == 0x212E) || \ |
| 372 | (((c) >= 0x2180) && ((c) <= 0x2182)) || \ |
| 373 | (((c) >= 0x3041) && ((c) <= 0x3094)) || \ |
| 374 | (((c) >= 0x30A1) && ((c) <= 0x30FA)) || \ |
| 375 | (((c) >= 0x3105) && ((c) <= 0x312C)) || \ |
| 376 | (((c) >= 0xAC00) && ((c) <= 0xD7A3))) |
| 377 | |
| 378 | /* |
| 379 | * [88] Digit ::= ... long list see REC ... |
| 380 | */ |
| 381 | #define IS_DIGIT(c) \ |
| 382 | ((((c) >= 0x0030) && ((c) <= 0x0039)) || \ |
| 383 | (((c) >= 0x0660) && ((c) <= 0x0669)) || \ |
| 384 | (((c) >= 0x06F0) && ((c) <= 0x06F9)) || \ |
| 385 | (((c) >= 0x0966) && ((c) <= 0x096F)) || \ |
| 386 | (((c) >= 0x09E6) && ((c) <= 0x09EF)) || \ |
| 387 | (((c) >= 0x0A66) && ((c) <= 0x0A6F)) || \ |
| 388 | (((c) >= 0x0AE6) && ((c) <= 0x0AEF)) || \ |
| 389 | (((c) >= 0x0B66) && ((c) <= 0x0B6F)) || \ |
| 390 | (((c) >= 0x0BE7) && ((c) <= 0x0BEF)) || \ |
| 391 | (((c) >= 0x0C66) && ((c) <= 0x0C6F)) || \ |
| 392 | (((c) >= 0x0CE6) && ((c) <= 0x0CEF)) || \ |
| 393 | (((c) >= 0x0D66) && ((c) <= 0x0D6F)) || \ |
| 394 | (((c) >= 0x0E50) && ((c) <= 0x0E59)) || \ |
| 395 | (((c) >= 0x0ED0) && ((c) <= 0x0ED9)) || \ |
| 396 | (((c) >= 0x0F20) && ((c) <= 0x0F29))) |
| 397 | |
| 398 | /* |
| 399 | * [87] CombiningChar ::= ... long list see REC ... |
| 400 | */ |
| 401 | #define IS_COMBINING(c) \ |
| 402 | ((((c) >= 0x0300) && ((c) <= 0x0345)) || \ |
| 403 | (((c) >= 0x0360) && ((c) <= 0x0361)) || \ |
| 404 | (((c) >= 0x0483) && ((c) <= 0x0486)) || \ |
| 405 | (((c) >= 0x0591) && ((c) <= 0x05A1)) || \ |
| 406 | (((c) >= 0x05A3) && ((c) <= 0x05B9)) || \ |
| 407 | (((c) >= 0x05BB) && ((c) <= 0x05BD)) || \ |
| 408 | ((c) == 0x05BF) || \ |
| 409 | (((c) >= 0x05C1) && ((c) <= 0x05C2)) || \ |
| 410 | ((c) == 0x05C4) || \ |
| 411 | (((c) >= 0x064B) && ((c) <= 0x0652)) || \ |
| 412 | ((c) == 0x0670) || \ |
| 413 | (((c) >= 0x06D6) && ((c) <= 0x06DC)) || \ |
| 414 | (((c) >= 0x06DD) && ((c) <= 0x06DF)) || \ |
| 415 | (((c) >= 0x06E0) && ((c) <= 0x06E4)) || \ |
| 416 | (((c) >= 0x06E7) && ((c) <= 0x06E8)) || \ |
| 417 | (((c) >= 0x06EA) && ((c) <= 0x06ED)) || \ |
| 418 | (((c) >= 0x0901) && ((c) <= 0x0903)) || \ |
| 419 | ((c) == 0x093C) || \ |
| 420 | (((c) >= 0x093E) && ((c) <= 0x094C)) || \ |
| 421 | ((c) == 0x094D) || \ |
| 422 | (((c) >= 0x0951) && ((c) <= 0x0954)) || \ |
| 423 | (((c) >= 0x0962) && ((c) <= 0x0963)) || \ |
| 424 | (((c) >= 0x0981) && ((c) <= 0x0983)) || \ |
| 425 | ((c) == 0x09BC) || \ |
| 426 | ((c) == 0x09BE) || \ |
| 427 | ((c) == 0x09BF) || \ |
| 428 | (((c) >= 0x09C0) && ((c) <= 0x09C4)) || \ |
| 429 | (((c) >= 0x09C7) && ((c) <= 0x09C8)) || \ |
| 430 | (((c) >= 0x09CB) && ((c) <= 0x09CD)) || \ |
| 431 | ((c) == 0x09D7) || \ |
| 432 | (((c) >= 0x09E2) && ((c) <= 0x09E3)) || \ |
| 433 | ((c) == 0x0A02) || \ |
| 434 | ((c) == 0x0A3C) || \ |
| 435 | ((c) == 0x0A3E) || \ |
| 436 | ((c) == 0x0A3F) || \ |
| 437 | (((c) >= 0x0A40) && ((c) <= 0x0A42)) || \ |
| 438 | (((c) >= 0x0A47) && ((c) <= 0x0A48)) || \ |
| 439 | (((c) >= 0x0A4B) && ((c) <= 0x0A4D)) || \ |
| 440 | (((c) >= 0x0A70) && ((c) <= 0x0A71)) || \ |
| 441 | (((c) >= 0x0A81) && ((c) <= 0x0A83)) || \ |
| 442 | ((c) == 0x0ABC) || \ |
| 443 | (((c) >= 0x0ABE) && ((c) <= 0x0AC5)) || \ |
| 444 | (((c) >= 0x0AC7) && ((c) <= 0x0AC9)) || \ |
| 445 | (((c) >= 0x0ACB) && ((c) <= 0x0ACD)) || \ |
| 446 | (((c) >= 0x0B01) && ((c) <= 0x0B03)) || \ |
| 447 | ((c) == 0x0B3C) || \ |
| 448 | (((c) >= 0x0B3E) && ((c) <= 0x0B43)) || \ |
| 449 | (((c) >= 0x0B47) && ((c) <= 0x0B48)) || \ |
| 450 | (((c) >= 0x0B4B) && ((c) <= 0x0B4D)) || \ |
| 451 | (((c) >= 0x0B56) && ((c) <= 0x0B57)) || \ |
| 452 | (((c) >= 0x0B82) && ((c) <= 0x0B83)) || \ |
| 453 | (((c) >= 0x0BBE) && ((c) <= 0x0BC2)) || \ |
| 454 | (((c) >= 0x0BC6) && ((c) <= 0x0BC8)) || \ |
| 455 | (((c) >= 0x0BCA) && ((c) <= 0x0BCD)) || \ |
| 456 | ((c) == 0x0BD7) || \ |
| 457 | (((c) >= 0x0C01) && ((c) <= 0x0C03)) || \ |
| 458 | (((c) >= 0x0C3E) && ((c) <= 0x0C44)) || \ |
| 459 | (((c) >= 0x0C46) && ((c) <= 0x0C48)) || \ |
| 460 | (((c) >= 0x0C4A) && ((c) <= 0x0C4D)) || \ |
| 461 | (((c) >= 0x0C55) && ((c) <= 0x0C56)) || \ |
| 462 | (((c) >= 0x0C82) && ((c) <= 0x0C83)) || \ |
| 463 | (((c) >= 0x0CBE) && ((c) <= 0x0CC4)) || \ |
| 464 | (((c) >= 0x0CC6) && ((c) <= 0x0CC8)) || \ |
| 465 | (((c) >= 0x0CCA) && ((c) <= 0x0CCD)) || \ |
| 466 | (((c) >= 0x0CD5) && ((c) <= 0x0CD6)) || \ |
| 467 | (((c) >= 0x0D02) && ((c) <= 0x0D03)) || \ |
| 468 | (((c) >= 0x0D3E) && ((c) <= 0x0D43)) || \ |
| 469 | (((c) >= 0x0D46) && ((c) <= 0x0D48)) || \ |
| 470 | (((c) >= 0x0D4A) && ((c) <= 0x0D4D)) || \ |
| 471 | ((c) == 0x0D57) || \ |
| 472 | ((c) == 0x0E31) || \ |
| 473 | (((c) >= 0x0E34) && ((c) <= 0x0E3A)) || \ |
| 474 | (((c) >= 0x0E47) && ((c) <= 0x0E4E)) || \ |
| 475 | ((c) == 0x0EB1) || \ |
| 476 | (((c) >= 0x0EB4) && ((c) <= 0x0EB9)) || \ |
| 477 | (((c) >= 0x0EBB) && ((c) <= 0x0EBC)) || \ |
| 478 | (((c) >= 0x0EC8) && ((c) <= 0x0ECD)) || \ |
| 479 | (((c) >= 0x0F18) && ((c) <= 0x0F19)) || \ |
| 480 | ((c) == 0x0F35) || \ |
| 481 | ((c) == 0x0F37) || \ |
| 482 | ((c) == 0x0F39) || \ |
| 483 | ((c) == 0x0F3E) || \ |
| 484 | ((c) == 0x0F3F) || \ |
| 485 | (((c) >= 0x0F71) && ((c) <= 0x0F84)) || \ |
| 486 | (((c) >= 0x0F86) && ((c) <= 0x0F8B)) || \ |
| 487 | (((c) >= 0x0F90) && ((c) <= 0x0F95)) || \ |
| 488 | ((c) == 0x0F97) || \ |
| 489 | (((c) >= 0x0F99) && ((c) <= 0x0FAD)) || \ |
| 490 | (((c) >= 0x0FB1) && ((c) <= 0x0FB7)) || \ |
| 491 | ((c) == 0x0FB9) || \ |
| 492 | (((c) >= 0x20D0) && ((c) <= 0x20DC)) || \ |
| 493 | ((c) == 0x20E1) || \ |
| 494 | (((c) >= 0x302A) && ((c) <= 0x302F)) || \ |
| 495 | ((c) == 0x3099) || \ |
| 496 | ((c) == 0x309A)) |
| 497 | |
| 498 | /* |
| 499 | * [89] Extender ::= #x00B7 | #x02D0 | #x02D1 | #x0387 | #x0640 | |
| 500 | * #x0E46 | #x0EC6 | #x3005 | [#x3031-#x3035] | |
| 501 | * [#x309D-#x309E] | [#x30FC-#x30FE] |
| 502 | */ |
| 503 | #define IS_EXTENDER(c) \ |
| 504 | (((c) == 0xb7) || ((c) == 0x2d0) || ((c) == 0x2d1) || \ |
| 505 | ((c) == 0x387) || ((c) == 0x640) || ((c) == 0xe46) || \ |
| 506 | ((c) == 0xec6) || ((c) == 0x3005) \ |
| 507 | (((c) >= 0x3031) && ((c) <= 0x3035)) || \ |
| 508 | (((c) >= 0x309b) && ((c) <= 0x309e)) || \ |
| 509 | (((c) >= 0x30fc) && ((c) <= 0x30fe))) |
| 510 | |
| 511 | /* |
| 512 | * [86] Ideographic ::= [#x4E00-#x9FA5] | #x3007 | [#x3021-#x3029] |
| 513 | */ |
| 514 | #define IS_IDEOGRAPHIC(c) \ |
| 515 | ((((c) >= 0x4e00) && ((c) <= 0x9fa5)) || \ |
| 516 | (((c) >= 0xf900) && ((c) <= 0xfa2d)) || \ |
| 517 | (((c) >= 0x3021) && ((c) <= 0x3029)) || \ |
| 518 | ((c) == 0x3007)) |
| 519 | |
| 520 | /* |
| 521 | * [84] Letter ::= BaseChar | Ideographic |
| 522 | */ |
| 523 | #define IS_LETTER(c) (IS_BASECHAR(c) || IS_IDEOGRAPHIC(c)) |
| 524 | |
| 525 | #else |
| 526 | /************************************************************************ |
| 527 | * * |
| 528 | * 8bits / ASCII version of the macros. * |
| 529 | * * |
| 530 | ************************************************************************/ |
| 531 | /* |
| 532 | * [2] Char ::= #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD] |
| 533 | * | [#x10000-#x10FFFF] |
| 534 | * any Unicode character, excluding the surrogate blocks, FFFE, and FFFF. |
| 535 | */ |
| 536 | #define IS_CHAR(c) \ |
| 537 | (((c) == 0x09) || ((c) == 0x0a) || ((c) == 0x0d) || ((c) >= 0x20) ||\ |
| 538 | ((c) == 0xa)) |
| 539 | |
| 540 | /* |
| 541 | * [85] BaseChar ::= ... long list see REC ... |
| 542 | */ |
| 543 | #define IS_BASECHAR(c) \ |
| 544 | ((((c) >= 0x41) && ((c) <= 0x5a)) || \ |
| 545 | (((c) >= 0x61) && ((c) <= 0x7a)) || \ |
| 546 | (((c) >= 0xaa) && ((c) <= 0x5b)) || \ |
| 547 | (((c) >= 0xc0) && ((c) <= 0xd6)) || \ |
| 548 | (((c) >= 0xd8) && ((c) <= 0xf6)) || \ |
| 549 | (((c) >= 0xf8) && ((c) <= 0xff)) || \ |
| 550 | ((c) == 0xba)) |
| 551 | |
| 552 | /* |
| 553 | * [88] Digit ::= ... long list see REC ... |
| 554 | */ |
| 555 | #define IS_DIGIT(c) (((c) >= 0x30) && ((c) <= 0x39)) |
| 556 | |
| 557 | /* |
| 558 | * [84] Letter ::= BaseChar | Ideographic |
| 559 | */ |
| 560 | #define IS_LETTER(c) IS_BASECHAR(c) |
| 561 | |
| 562 | |
| 563 | /* |
| 564 | * [87] CombiningChar ::= ... long list see REC ... |
| 565 | */ |
| 566 | #define IS_COMBINING(c) 0 |
| 567 | |
| 568 | /* |
| 569 | * [89] Extender ::= #x00B7 | #x02D0 | #x02D1 | #x0387 | #x0640 | |
| 570 | * #x0E46 | #x0EC6 | #x3005 | [#x3031-#x3035] | |
| 571 | * [#x309D-#x309E] | [#x30FC-#x30FE] |
| 572 | */ |
| 573 | #define IS_EXTENDER(c) ((c) == 0xb7) |
| 574 | |
| 575 | #endif /* !UNICODE */ |
| 576 | |
| 577 | /* |
| 578 | * Blank chars. |
| 579 | * |
| 580 | * [3] S ::= (#x20 | #x9 | #xD | #xA)+ |
| 581 | */ |
| 582 | #define IS_BLANK(c) (((c) == 0x20) || ((c) == 0x09) || ((c) == 0xa) || \ |
| 583 | ((c) == 0x0D)) |
| 584 | |
| 585 | /* |
| 586 | * [13] PubidChar ::= #x20 | #xD | #xA | [a-zA-Z0-9] | [-'()+,./:=?;!*#@$_%] |
| 587 | */ |
| 588 | #define IS_PUBIDCHAR(c) \ |
| 589 | (((c) == 0x20) || ((c) == 0x0D) || ((c) == 0x0A) || \ |
| 590 | (((c) >= 'a') && ((c) <= 'z')) || \ |
| 591 | (((c) >= 'A') && ((c) <= 'Z')) || \ |
| 592 | (((c) >= '0') && ((c) <= '9')) || \ |
| 593 | ((c) == '-') || ((c) == '\'') || ((c) == '(') || ((c) == ')') || \ |
| 594 | ((c) == '+') || ((c) == ',') || ((c) == '.') || ((c) == '/') || \ |
| 595 | ((c) == ':') || ((c) == '=') || ((c) == '?') || ((c) == ';') || \ |
| 596 | ((c) == '!') || ((c) == '*') || ((c) == '#') || ((c) == '@') || \ |
| 597 | ((c) == '$') || ((c) == '_') || ((c) == '%')) |
| 598 | |
| 599 | #define SKIP_EOL(p) \ |
| 600 | if (*(p) == 0x13) { p++ ; if (*(p) == 0x10) p++; } \ |
| 601 | if (*(p) == 0x10) { p++ ; if (*(p) == 0x13) p++; } |
| 602 | |
| 603 | #define MOVETO_ENDTAG(p) \ |
| 604 | while (IS_CHAR(*p) && (*(p) != '>')) (p)++ |
| 605 | |
| 606 | #define MOVETO_STARTTAG(p) \ |
| 607 | while (IS_CHAR(*p) && (*(p) != '<')) (p)++ |
| 608 | |
| 609 | /************************************************************************ |
| 610 | * * |
| 611 | * Commodity functions to handle CHARs * |
| 612 | * * |
| 613 | ************************************************************************/ |
| 614 | |
| 615 | /* |
| 616 | * xmlStrndup : a strndup for array of CHAR's |
| 617 | */ |
| 618 | |
| 619 | CHAR *xmlStrndup(const CHAR *cur, int len) { |
| 620 | CHAR *ret = malloc((len + 1) * sizeof(CHAR)); |
| 621 | |
| 622 | if (ret == NULL) { |
| 623 | fprintf(stderr, "malloc of %d byte failed\n", |
| 624 | (len + 1) * sizeof(CHAR)); |
| 625 | return(NULL); |
| 626 | } |
| 627 | memcpy(ret, cur, len * sizeof(CHAR)); |
| 628 | ret[len] = 0; |
| 629 | return(ret); |
| 630 | } |
| 631 | |
| 632 | /* |
| 633 | * xmlStrdup : a strdup for CHAR's |
| 634 | */ |
| 635 | |
| 636 | CHAR *xmlStrdup(const CHAR *cur) { |
| 637 | const CHAR *p = cur; |
| 638 | |
| 639 | while (IS_CHAR(*p)) p++; |
| 640 | return(xmlStrndup(cur, p - cur)); |
| 641 | } |
| 642 | |
| 643 | /* |
| 644 | * xmlCharStrndup : a strndup for char's to CHAR's |
| 645 | */ |
| 646 | |
| 647 | CHAR *xmlCharStrndup(const char *cur, int len) { |
| 648 | int i; |
| 649 | CHAR *ret = malloc((len + 1) * sizeof(CHAR)); |
| 650 | |
| 651 | if (ret == NULL) { |
| 652 | fprintf(stderr, "malloc of %d byte failed\n", |
| 653 | (len + 1) * sizeof(CHAR)); |
| 654 | return(NULL); |
| 655 | } |
| 656 | for (i = 0;i < len;i++) |
| 657 | ret[i] = (CHAR) cur[i]; |
| 658 | ret[len] = 0; |
| 659 | return(ret); |
| 660 | } |
| 661 | |
| 662 | /* |
| 663 | * xmlCharStrdup : a strdup for char's to CHAR's |
| 664 | */ |
| 665 | |
| 666 | CHAR *xmlCharStrdup(const char *cur) { |
| 667 | const char *p = cur; |
| 668 | |
| 669 | while (*p != '\0') p++; |
| 670 | return(xmlCharStrndup(cur, p - cur)); |
| 671 | } |
| 672 | |
| 673 | /* |
| 674 | * xmlStrcmp : a strcmp for CHAR's |
| 675 | */ |
| 676 | |
| 677 | int xmlStrcmp(const CHAR *str1, const CHAR *str2) { |
| 678 | register int tmp; |
| 679 | |
| 680 | do { |
| 681 | tmp = *str1++ - *str2++; |
| 682 | if (tmp != 0) return(tmp); |
| 683 | } while ((*str1 != 0) && (*str2 != 0)); |
| 684 | return (*str1 - *str2); |
| 685 | } |
| 686 | |
| 687 | /* |
| 688 | * xmlStrncmp : a strncmp for CHAR's |
| 689 | */ |
| 690 | |
| 691 | int xmlStrncmp(const CHAR *str1, const CHAR *str2, int len) { |
| 692 | register int tmp; |
| 693 | |
| 694 | if (len <= 0) return(0); |
| 695 | do { |
| 696 | tmp = *str1++ - *str2++; |
| 697 | if (tmp != 0) return(tmp); |
| 698 | len--; |
| 699 | if (len <= 0) return(0); |
| 700 | } while ((*str1 != 0) && (*str2 != 0)); |
| 701 | return (*str1 - *str2); |
| 702 | } |
| 703 | |
| 704 | /* |
| 705 | * xmlStrchr : a strchr for CHAR's |
| 706 | */ |
| 707 | |
| 708 | CHAR *xmlStrchr(const CHAR *str, CHAR val) { |
| 709 | while (*str != 0) { |
| 710 | if (*str == val) return((CHAR *) str); |
| 711 | str++; |
| 712 | } |
| 713 | return(NULL); |
| 714 | } |
| 715 | |
| 716 | /* |
| 717 | * xmlStrlen : lenght of a CHAR's string |
| 718 | */ |
| 719 | |
| 720 | int xmlStrlen(const CHAR *str) { |
| 721 | int len = 0; |
| 722 | |
| 723 | if (str == NULL) return(0); |
| 724 | while (*str != 0) { |
| 725 | str++; |
| 726 | len++; |
| 727 | } |
| 728 | return(len); |
| 729 | } |
| 730 | |
| 731 | /* |
| 732 | * xmlStrncat : a strncat for array of CHAR's |
| 733 | */ |
| 734 | |
| 735 | CHAR *xmlStrncat(CHAR *cur, const CHAR *add, int len) { |
| 736 | int size; |
| 737 | CHAR *ret; |
| 738 | |
| 739 | if ((add == NULL) || (len == 0)) |
| 740 | return(cur); |
| 741 | if (cur == NULL) |
| 742 | return(xmlStrndup(add, len)); |
| 743 | |
| 744 | size = xmlStrlen(cur); |
| 745 | ret = realloc(cur, (size + len + 1) * sizeof(CHAR)); |
| 746 | if (ret == NULL) { |
| 747 | fprintf(stderr, "xmlStrncat: realloc of %d byte failed\n", |
| 748 | (size + len + 1) * sizeof(CHAR)); |
| 749 | return(cur); |
| 750 | } |
| 751 | memcpy(&ret[size], add, len * sizeof(CHAR)); |
| 752 | ret[size + len] = 0; |
| 753 | return(ret); |
| 754 | } |
| 755 | |
| 756 | /* |
| 757 | * xmlStrcat : a strcat for CHAR's |
| 758 | */ |
| 759 | |
| 760 | CHAR *xmlStrcat(CHAR *cur, const CHAR *add) { |
| 761 | const CHAR *p = add; |
| 762 | |
| 763 | if (add == NULL) return(cur); |
| 764 | if (cur == NULL) |
| 765 | return(xmlStrdup(add)); |
| 766 | |
| 767 | while (IS_CHAR(*p)) p++; |
| 768 | return(xmlStrncat(cur, add, p - add)); |
| 769 | } |
| 770 | |
| 771 | /************************************************************************ |
| 772 | * * |
| 773 | * Commodity functions, cleanup needed ? * |
| 774 | * * |
| 775 | ************************************************************************/ |
| 776 | |
| 777 | /* |
| 778 | * Is this a sequence of blank chars that one can ignore ? |
| 779 | */ |
| 780 | |
| 781 | static int areBlanks(xmlParserCtxtPtr ctxt, const CHAR *str, int len) { |
| 782 | int i; |
| 783 | xmlNodePtr lastChild; |
| 784 | |
| 785 | for (i = 0;i < len;i++) |
| 786 | if (!(IS_BLANK(str[i]))) return(0); |
| 787 | |
| 788 | if (CUR != '<') return(0); |
| 789 | lastChild = xmlGetLastChild(ctxt->node); |
| 790 | if (lastChild == NULL) { |
| 791 | if (ctxt->node->content != NULL) return(0); |
| 792 | } else if (xmlNodeIsText(lastChild)) |
| 793 | return(0); |
| 794 | return(1); |
| 795 | } |
| 796 | |
| 797 | /* |
| 798 | * Handling of defined entities, when should we define a new input |
| 799 | * stream ? When do we just handle that as a set of chars ? |
| 800 | */ |
| 801 | |
| 802 | void xmlHandleEntity(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { |
| 803 | int len; |
| 804 | |
| 805 | if (entity->content == NULL) { |
| 806 | xmlParserError(ctxt, "xmlHandleEntity %s: content == NULL\n", |
| 807 | entity->name); |
| 808 | return; |
| 809 | } |
| 810 | len = xmlStrlen(entity->content); |
| 811 | if (len <= 2) goto handle_as_char; |
| 812 | |
| 813 | /* |
| 814 | * Redefine its content as an input stream. |
| 815 | */ |
| 816 | xmlNewEntityInputStream(ctxt, entity); |
| 817 | return; |
| 818 | |
| 819 | handle_as_char: |
| 820 | /* |
| 821 | * Just handle the content as a set of chars. |
| 822 | */ |
| 823 | if (ctxt->sax != NULL) |
| 824 | ctxt->sax->characters(ctxt, entity->content, 0, len); |
| 825 | |
| 826 | } |
| 827 | |
| 828 | /* |
| 829 | * Forward definition for recusive behaviour. |
| 830 | */ |
| 831 | xmlNodePtr xmlParseElement(xmlParserCtxtPtr ctxt); |
| 832 | CHAR *xmlParsePEReference(xmlParserCtxtPtr ctxt, int inLine); |
| 833 | CHAR *xmlParseReference(xmlParserCtxtPtr ctxt, int inLine); |
| 834 | |
| 835 | /************************************************************************ |
| 836 | * * |
| 837 | * Extra stuff for namespace support * |
| 838 | * Relates to http://www.w3.org/TR/WD-xml-names * |
| 839 | * * |
| 840 | ************************************************************************/ |
| 841 | |
| 842 | /* |
| 843 | * xmlNamespaceParseNCName : parse an XML namespace name. |
| 844 | * |
| 845 | * [NS 3] NCName ::= (Letter | '_') (NCNameChar)* |
| 846 | * |
| 847 | * [NS 4] NCNameChar ::= Letter | Digit | '.' | '-' | '_' | |
| 848 | * CombiningChar | Extender |
| 849 | */ |
| 850 | |
| 851 | CHAR *xmlNamespaceParseNCName(xmlParserCtxtPtr ctxt) { |
| 852 | const CHAR *q; |
| 853 | CHAR *ret = NULL; |
| 854 | |
| 855 | if (!IS_LETTER(CUR) && (CUR != '_')) return(NULL); |
| 856 | q = NEXT; |
| 857 | |
| 858 | while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || |
| 859 | (CUR == '.') || (CUR == '-') || |
| 860 | (CUR == '_') || |
| 861 | (IS_COMBINING(CUR)) || |
| 862 | (IS_EXTENDER(CUR))) |
| 863 | NEXT; |
| 864 | |
| 865 | ret = xmlStrndup(q, CUR_PTR - q); |
| 866 | |
| 867 | return(ret); |
| 868 | } |
| 869 | |
| 870 | /* |
| 871 | * xmlNamespaceParseQName : parse an XML qualified name |
| 872 | * |
| 873 | * [NS 5] QName ::= (Prefix ':')? LocalPart |
| 874 | * |
| 875 | * [NS 6] Prefix ::= NCName |
| 876 | * |
| 877 | * [NS 7] LocalPart ::= NCName |
| 878 | */ |
| 879 | |
| 880 | CHAR *xmlNamespaceParseQName(xmlParserCtxtPtr ctxt, CHAR **prefix) { |
| 881 | CHAR *ret = NULL; |
| 882 | |
| 883 | *prefix = NULL; |
| 884 | ret = xmlNamespaceParseNCName(ctxt); |
| 885 | if (CUR == ':') { |
| 886 | *prefix = ret; |
| 887 | NEXT; |
| 888 | ret = xmlNamespaceParseNCName(ctxt); |
| 889 | } |
| 890 | |
| 891 | return(ret); |
| 892 | } |
| 893 | |
| 894 | /* |
| 895 | * xmlNamespaceParseNSDef : parse a namespace prefix declaration |
| 896 | * |
| 897 | * [NS 1] NSDef ::= PrefixDef Eq SystemLiteral |
| 898 | * |
| 899 | * [NS 2] PrefixDef ::= 'xmlns' (':' NCName)? |
| 900 | */ |
| 901 | |
| 902 | CHAR *xmlNamespaceParseNSDef(xmlParserCtxtPtr ctxt) { |
| 903 | CHAR *name = NULL; |
| 904 | |
| 905 | if ((CUR == 'x') && (NXT(1) == 'm') && |
| 906 | (NXT(2) == 'l') && (NXT(3) == 'n') && |
| 907 | (NXT(4) == 's')) { |
| 908 | SKIP(5); |
| 909 | if (CUR == ':') { |
| 910 | NEXT; |
| 911 | name = xmlNamespaceParseNCName(ctxt); |
| 912 | } |
| 913 | } |
| 914 | return(name); |
| 915 | } |
| 916 | |
| 917 | /* |
| 918 | * [OLD] Parse and return a string between quotes or doublequotes |
| 919 | */ |
| 920 | CHAR *xmlParseQuotedString(xmlParserCtxtPtr ctxt) { |
| 921 | CHAR *ret = NULL; |
| 922 | const CHAR *q; |
| 923 | |
| 924 | if (CUR == '"') { |
| 925 | NEXT; |
| 926 | q = CUR_PTR; |
| 927 | while (IS_CHAR(CUR) && (CUR != '"')) NEXT; |
| 928 | if (CUR != '"') |
| 929 | xmlParserError(ctxt, "String not closed\"%.50s\n", q); |
| 930 | else { |
| 931 | ret = xmlStrndup(q, CUR_PTR - q); |
| 932 | NEXT; |
| 933 | } |
| 934 | } else if (CUR == '\''){ |
| 935 | NEXT; |
| 936 | q = CUR_PTR; |
| 937 | while (IS_CHAR(CUR) && (CUR != '\'')) NEXT; |
| 938 | if (CUR != '\'') |
| 939 | xmlParserError(ctxt, "String not closed\"%.50s\n", q); |
| 940 | else { |
| 941 | ret = xmlStrndup(q, CUR_PTR - q); |
| 942 | NEXT; |
| 943 | } |
| 944 | } |
| 945 | return(ret); |
| 946 | } |
| 947 | |
| 948 | /* |
| 949 | * [OLD] xmlParseNamespace: parse specific PI '<?namespace ...' constructs. |
| 950 | * |
| 951 | * This is what the older xml-name Working Draft specified, a bunch of |
| 952 | * other stuff may still rely on it, so support is still here as |
| 953 | * if ot was declared on the root of the Tree:-( |
| 954 | */ |
| 955 | |
| 956 | void xmlParseNamespace(xmlParserCtxtPtr ctxt) { |
| 957 | CHAR *href = NULL; |
| 958 | CHAR *prefix = NULL; |
| 959 | int garbage = 0; |
| 960 | |
| 961 | /* |
| 962 | * We just skipped "namespace" or "xml:namespace" |
| 963 | */ |
| 964 | SKIP_BLANKS; |
| 965 | |
| 966 | while (IS_CHAR(CUR) && (CUR != '>')) { |
| 967 | /* |
| 968 | * We can have "ns" or "prefix" attributes |
| 969 | * Old encoding as 'href' or 'AS' attributes is still supported |
| 970 | */ |
| 971 | if ((CUR == 'n') && (NXT(1) == 's')) { |
| 972 | garbage = 0; |
| 973 | SKIP(2); |
| 974 | SKIP_BLANKS; |
| 975 | |
| 976 | if (CUR != '=') continue; |
| 977 | NEXT; |
| 978 | SKIP_BLANKS; |
| 979 | |
| 980 | href = xmlParseQuotedString(ctxt); |
| 981 | SKIP_BLANKS; |
| 982 | } else if ((CUR == 'h') && (NXT(1) == 'r') && |
| 983 | (NXT(2) == 'e') && (NXT(3) == 'f')) { |
| 984 | garbage = 0; |
| 985 | SKIP(4); |
| 986 | SKIP_BLANKS; |
| 987 | |
| 988 | if (CUR != '=') continue; |
| 989 | NEXT; |
| 990 | SKIP_BLANKS; |
| 991 | |
| 992 | href = xmlParseQuotedString(ctxt); |
| 993 | SKIP_BLANKS; |
| 994 | } else if ((CUR == 'p') && (NXT(1) == 'r') && |
| 995 | (NXT(2) == 'e') && (NXT(3) == 'f') && |
| 996 | (NXT(4) == 'i') && (NXT(5) == 'x')) { |
| 997 | garbage = 0; |
| 998 | SKIP(6); |
| 999 | SKIP_BLANKS; |
| 1000 | |
| 1001 | if (CUR != '=') continue; |
| 1002 | NEXT; |
| 1003 | SKIP_BLANKS; |
| 1004 | |
| 1005 | prefix = xmlParseQuotedString(ctxt); |
| 1006 | SKIP_BLANKS; |
| 1007 | } else if ((CUR == 'A') && (NXT(1) == 'S')) { |
| 1008 | garbage = 0; |
| 1009 | SKIP(2); |
| 1010 | SKIP_BLANKS; |
| 1011 | |
| 1012 | if (CUR != '=') continue; |
| 1013 | NEXT; |
| 1014 | SKIP_BLANKS; |
| 1015 | |
| 1016 | prefix = xmlParseQuotedString(ctxt); |
| 1017 | SKIP_BLANKS; |
| 1018 | } else if ((CUR == '?') && (NXT(1) == '>')) { |
| 1019 | garbage = 0; |
| 1020 | CUR_PTR ++; |
| 1021 | } else { |
| 1022 | /* |
| 1023 | * Found garbage when parsing the namespace |
| 1024 | */ |
| 1025 | if (!garbage) |
| 1026 | xmlParserError(ctxt, "xmlParseNamespace found garbage\n"); |
| 1027 | NEXT; |
| 1028 | } |
| 1029 | } |
| 1030 | |
| 1031 | MOVETO_ENDTAG(CUR_PTR); |
| 1032 | NEXT; |
| 1033 | |
| 1034 | /* |
| 1035 | * Register the DTD. |
| 1036 | */ |
| 1037 | if (href != NULL) |
| 1038 | xmlNewGlobalNs(ctxt->doc, href, prefix); |
| 1039 | |
| 1040 | if (prefix != NULL) free(prefix); |
| 1041 | if (href != NULL) free(href); |
| 1042 | } |
| 1043 | |
| 1044 | /************************************************************************ |
| 1045 | * * |
| 1046 | * The parser itself * |
| 1047 | * Relates to http://www.w3.org/TR/REC-xml * |
| 1048 | * * |
| 1049 | ************************************************************************/ |
| 1050 | |
| 1051 | /* |
| 1052 | * xmlParseName : parse an XML name. |
| 1053 | * |
| 1054 | * [4] NameChar ::= Letter | Digit | '.' | '-' | '_' | ':' | |
| 1055 | * CombiningChar | Extender |
| 1056 | * |
| 1057 | * [5] Name ::= (Letter | '_' | ':') (NameChar)* |
| 1058 | * |
| 1059 | * [6] Names ::= Name (S Name)* |
| 1060 | */ |
| 1061 | |
| 1062 | CHAR *xmlParseName(xmlParserCtxtPtr ctxt) { |
| 1063 | const CHAR *q; |
| 1064 | CHAR *ret = NULL; |
| 1065 | |
| 1066 | if (!IS_LETTER(CUR) && (CUR != '_') && |
| 1067 | (CUR != ':')) return(NULL); |
| 1068 | q = NEXT; |
| 1069 | |
| 1070 | while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || |
| 1071 | (CUR == '.') || (CUR == '-') || |
| 1072 | (CUR == '_') || (CUR == ':') || |
| 1073 | (IS_COMBINING(CUR)) || |
| 1074 | (IS_EXTENDER(CUR))) |
| 1075 | NEXT; |
| 1076 | |
| 1077 | ret = xmlStrndup(q, CUR_PTR - q); |
| 1078 | |
| 1079 | return(ret); |
| 1080 | } |
| 1081 | |
| 1082 | /* |
| 1083 | * xmlParseNmtoken : parse an XML Nmtoken. |
| 1084 | * |
| 1085 | * [7] Nmtoken ::= (NameChar)+ |
| 1086 | * |
| 1087 | * [8] Nmtokens ::= Nmtoken (S Nmtoken)* |
| 1088 | */ |
| 1089 | |
| 1090 | CHAR *xmlParseNmtoken(xmlParserCtxtPtr ctxt) { |
| 1091 | const CHAR *q; |
| 1092 | CHAR *ret = NULL; |
| 1093 | |
| 1094 | q = NEXT; |
| 1095 | |
| 1096 | while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || |
| 1097 | (CUR == '.') || (CUR == '-') || |
| 1098 | (CUR == '_') || (CUR == ':') || |
| 1099 | (IS_COMBINING(CUR)) || |
| 1100 | (IS_EXTENDER(CUR))) |
| 1101 | NEXT; |
| 1102 | |
| 1103 | ret = xmlStrndup(q, CUR_PTR - q); |
| 1104 | |
| 1105 | return(ret); |
| 1106 | } |
| 1107 | |
| 1108 | /* |
| 1109 | * xmlParseEntityValue : parse a value for ENTITY decl. |
| 1110 | * |
| 1111 | * [9] EntityValue ::= '"' ([^%&"] | PEReference | Reference)* '"' | |
| 1112 | * "'" ([^%&'] | PEReference | Reference)* "'" |
| 1113 | */ |
| 1114 | |
| 1115 | CHAR *xmlParseEntityValue(xmlParserCtxtPtr ctxt) { |
| 1116 | CHAR *ret = NULL, *cur; |
| 1117 | const CHAR *q; |
| 1118 | |
| 1119 | if (CUR == '"') { |
| 1120 | NEXT; |
| 1121 | |
| 1122 | q = CUR_PTR; |
| 1123 | while ((IS_CHAR(CUR)) && (CUR != '"')) { |
| 1124 | if (CUR == '%') { |
| 1125 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1126 | cur = xmlParsePEReference(ctxt, 1); |
| 1127 | ret = xmlStrcat(ret, cur); |
| 1128 | q = CUR_PTR; |
| 1129 | } else if (CUR == '&') { |
| 1130 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1131 | cur = xmlParseReference(ctxt, 1); |
| 1132 | ret = xmlStrcat(ret, cur); |
| 1133 | q = CUR_PTR; |
| 1134 | } else |
| 1135 | NEXT; |
| 1136 | } |
| 1137 | if (!IS_CHAR(CUR)) { |
| 1138 | xmlParserError(ctxt, "Unfinished EntityValue\n"); |
| 1139 | } else { |
| 1140 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1141 | NEXT; |
| 1142 | } |
| 1143 | } else if (CUR == '\'') { |
| 1144 | NEXT; |
| 1145 | q = CUR_PTR; |
| 1146 | while ((IS_CHAR(CUR)) && (CUR != '\'')) { |
| 1147 | if (CUR == '%') { |
| 1148 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1149 | cur = xmlParsePEReference(ctxt, 1); |
| 1150 | ret = xmlStrcat(ret, cur); |
| 1151 | q = CUR_PTR; |
| 1152 | } else if (CUR == '&') { |
| 1153 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1154 | cur = xmlParseReference(ctxt, 1); |
| 1155 | ret = xmlStrcat(ret, cur); |
| 1156 | q = CUR_PTR; |
| 1157 | } else |
| 1158 | NEXT; |
| 1159 | } |
| 1160 | if (!IS_CHAR(CUR)) { |
| 1161 | xmlParserError(ctxt, "Unfinished EntityValue\n"); |
| 1162 | } else { |
| 1163 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1164 | NEXT; |
| 1165 | } |
| 1166 | } else { |
| 1167 | xmlParserError(ctxt, "xmlParseEntityValue \" or ' expected\n"); |
| 1168 | } |
| 1169 | |
| 1170 | return(ret); |
| 1171 | } |
| 1172 | |
| 1173 | /* |
| 1174 | * xmlParseAttValue : parse a value for an attribute |
| 1175 | * |
| 1176 | * [10] AttValue ::= '"' ([^<&"] | Reference)* '"' | |
| 1177 | * "'" ([^<&'] | Reference)* "'" |
| 1178 | */ |
| 1179 | |
| 1180 | CHAR *xmlParseAttValue(xmlParserCtxtPtr ctxt) { |
| 1181 | CHAR *ret = NULL, *cur; |
| 1182 | const CHAR *q; |
| 1183 | |
| 1184 | if (CUR == '"') { |
| 1185 | NEXT; |
| 1186 | |
| 1187 | q = CUR_PTR; |
| 1188 | while ((IS_CHAR(CUR)) && (CUR != '"')) { |
| 1189 | if (CUR == '&') { |
| 1190 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1191 | cur = xmlParseReference(ctxt, 1); |
| 1192 | ret = xmlStrcat(ret, cur); |
| 1193 | q = CUR_PTR; |
| 1194 | } else |
| 1195 | NEXT; |
| 1196 | } |
| 1197 | if (!IS_CHAR(CUR)) { |
| 1198 | xmlParserError(ctxt, "Unfinished AttValue\n"); |
| 1199 | } else { |
| 1200 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1201 | NEXT; |
| 1202 | } |
| 1203 | } else if (CUR == '\'') { |
| 1204 | NEXT; |
| 1205 | q = CUR_PTR; |
| 1206 | while ((IS_CHAR(CUR)) && (CUR != '\'')) { |
| 1207 | if (CUR == '&') { |
| 1208 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1209 | cur = xmlParseReference(ctxt, 1); |
| 1210 | ret = xmlStrcat(ret, cur); |
| 1211 | q = CUR_PTR; |
| 1212 | } else |
| 1213 | NEXT; |
| 1214 | } |
| 1215 | if (!IS_CHAR(CUR)) { |
| 1216 | xmlParserError(ctxt, "Unfinished AttValue\n"); |
| 1217 | } else { |
| 1218 | ret = xmlStrncat(ret, q, CUR_PTR - q); |
| 1219 | NEXT; |
| 1220 | } |
| 1221 | } else { |
| 1222 | xmlParserError(ctxt, "AttValue: \" or ' expected\n"); |
| 1223 | } |
| 1224 | |
| 1225 | return(ret); |
| 1226 | } |
| 1227 | |
| 1228 | /* |
| 1229 | * xmlParseSystemLiteral : parse an XML Literal |
| 1230 | * |
| 1231 | * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'") |
| 1232 | */ |
| 1233 | |
| 1234 | CHAR *xmlParseSystemLiteral(xmlParserCtxtPtr ctxt) { |
| 1235 | const CHAR *q; |
| 1236 | CHAR *ret = NULL; |
| 1237 | |
| 1238 | if (CUR == '"') { |
| 1239 | NEXT; |
| 1240 | q = CUR_PTR; |
| 1241 | while ((IS_CHAR(CUR)) && (CUR != '"')) |
| 1242 | NEXT; |
| 1243 | if (!IS_CHAR(CUR)) { |
| 1244 | xmlParserError(ctxt, "Unfinished SystemLiteral\n"); |
| 1245 | } else { |
| 1246 | ret = xmlStrndup(q, CUR_PTR - q); |
| 1247 | NEXT; |
| 1248 | } |
| 1249 | } else if (CUR == '\'') { |
| 1250 | NEXT; |
| 1251 | q = CUR_PTR; |
| 1252 | while ((IS_CHAR(CUR)) && (CUR != '\'')) |
| 1253 | NEXT; |
| 1254 | if (!IS_CHAR(CUR)) { |
| 1255 | xmlParserError(ctxt, "Unfinished SystemLiteral\n"); |
| 1256 | } else { |
| 1257 | ret = xmlStrndup(q, CUR_PTR - q); |
| 1258 | NEXT; |
| 1259 | } |
| 1260 | } else { |
| 1261 | xmlParserError(ctxt, "SystemLiteral \" or ' expected\n"); |
| 1262 | } |
| 1263 | |
| 1264 | return(ret); |
| 1265 | } |
| 1266 | |
| 1267 | /* |
| 1268 | * xmlParsePubidLiteral: parse an XML public literal |
| 1269 | * |
| 1270 | * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'" |
| 1271 | */ |
| 1272 | |
| 1273 | CHAR *xmlParsePubidLiteral(xmlParserCtxtPtr ctxt) { |
| 1274 | const CHAR *q; |
| 1275 | CHAR *ret = NULL; |
| 1276 | /* |
| 1277 | * Name ::= (Letter | '_') (NameChar)* |
| 1278 | */ |
| 1279 | if (CUR == '"') { |
| 1280 | NEXT; |
| 1281 | q = CUR_PTR; |
| 1282 | while (IS_PUBIDCHAR(CUR)) NEXT; |
| 1283 | if (CUR != '"') { |
| 1284 | xmlParserError(ctxt, "Unfinished PubidLiteral\n"); |
| 1285 | } else { |
| 1286 | ret = xmlStrndup(q, CUR_PTR - q); |
| 1287 | NEXT; |
| 1288 | } |
| 1289 | } else if (CUR == '\'') { |
| 1290 | NEXT; |
| 1291 | q = CUR_PTR; |
| 1292 | while ((IS_LETTER(CUR)) && (CUR != '\'')) |
| 1293 | NEXT; |
| 1294 | if (!IS_LETTER(CUR)) { |
| 1295 | xmlParserError(ctxt, "Unfinished PubidLiteral\n"); |
| 1296 | } else { |
| 1297 | ret = xmlStrndup(q, CUR_PTR - q); |
| 1298 | NEXT; |
| 1299 | } |
| 1300 | } else { |
| 1301 | xmlParserError(ctxt, "SystemLiteral \" or ' expected\n"); |
| 1302 | } |
| 1303 | |
| 1304 | return(ret); |
| 1305 | } |
| 1306 | |
| 1307 | /* |
| 1308 | * xmlParseCharData: parse a CharData section. |
| 1309 | * if we are within a CDATA section ']]>' marks an end of section. |
| 1310 | * |
| 1311 | * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*) |
| 1312 | */ |
| 1313 | |
| 1314 | void xmlParseCharData(xmlParserCtxtPtr ctxt, int cdata) { |
| 1315 | const CHAR *q; |
| 1316 | |
| 1317 | q = CUR_PTR; |
| 1318 | while ((IS_CHAR(CUR)) && (CUR != '<') && |
| 1319 | (CUR != '&')) { |
| 1320 | NEXT; |
| 1321 | if ((cdata) && (CUR == ']') && (NXT(1) == ']') && |
| 1322 | (NXT(2) == '>')) break; |
| 1323 | } |
| 1324 | if (q == CUR_PTR) return; |
| 1325 | |
| 1326 | /* |
| 1327 | * Ok the segment [q CUR_PTR] is to be consumed as chars. |
| 1328 | */ |
| 1329 | if (ctxt->sax != NULL) { |
| 1330 | if (areBlanks(ctxt, q, CUR_PTR - q)) |
| 1331 | ctxt->sax->ignorableWhitespace(ctxt, q, 0, CUR_PTR - q); |
| 1332 | else |
| 1333 | ctxt->sax->characters(ctxt, q, 0, CUR_PTR - q); |
| 1334 | } |
| 1335 | } |
| 1336 | |
| 1337 | /* |
| 1338 | * xmlParseExternalID: Parse an External ID |
| 1339 | * |
| 1340 | * [75] ExternalID ::= 'SYSTEM' S SystemLiteral |
| 1341 | * | 'PUBLIC' S PubidLiteral S SystemLiteral |
| 1342 | */ |
| 1343 | |
| 1344 | CHAR *xmlParseExternalID(xmlParserCtxtPtr ctxt, CHAR **publicID) { |
| 1345 | CHAR *URI = NULL; |
| 1346 | |
| 1347 | if ((CUR == 'S') && (NXT(1) == 'Y') && |
| 1348 | (NXT(2) == 'S') && (NXT(3) == 'T') && |
| 1349 | (NXT(4) == 'E') && (NXT(5) == 'M')) { |
| 1350 | SKIP(6); |
| 1351 | SKIP_BLANKS; |
| 1352 | URI = xmlParseSystemLiteral(ctxt); |
| 1353 | if (URI == NULL) |
| 1354 | xmlParserError(ctxt, |
| 1355 | "xmlParseExternalID: SYSTEM, no URI\n"); |
| 1356 | } else if ((CUR == 'P') && (NXT(1) == 'U') && |
| 1357 | (NXT(2) == 'B') && (NXT(3) == 'L') && |
| 1358 | (NXT(4) == 'I') && (NXT(5) == 'C')) { |
| 1359 | SKIP(6); |
| 1360 | SKIP_BLANKS; |
| 1361 | *publicID = xmlParsePubidLiteral(ctxt); |
| 1362 | if (*publicID == NULL) |
| 1363 | xmlParserError(ctxt, |
| 1364 | "xmlParseExternalID: PUBLIC, no Public Identifier\n"); |
| 1365 | SKIP_BLANKS; |
| 1366 | URI = xmlParseSystemLiteral(ctxt); |
| 1367 | if (URI == NULL) |
| 1368 | xmlParserError(ctxt, |
| 1369 | "xmlParseExternalID: PUBLIC, no URI\n"); |
| 1370 | } |
| 1371 | return(URI); |
| 1372 | } |
| 1373 | |
| 1374 | /* |
| 1375 | * Skip an XML (SGML) comment <!-- .... --> |
| 1376 | * This may or may not create a node (depending on the context) |
| 1377 | * The spec says that "For compatibility, the string "--" (double-hyphen) |
| 1378 | * must not occur within comments. " |
| 1379 | * |
| 1380 | * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->' |
| 1381 | */ |
| 1382 | xmlNodePtr xmlParseComment(xmlParserCtxtPtr ctxt, int create) { |
| 1383 | xmlNodePtr ret = NULL; |
| 1384 | const CHAR *q, *start; |
| 1385 | const CHAR *r; |
| 1386 | CHAR *val; |
| 1387 | |
| 1388 | /* |
| 1389 | * Check that there is a comment right here. |
| 1390 | */ |
| 1391 | if ((CUR != '<') || (NXT(1) != '!') || |
| 1392 | (NXT(2) != '-') || (NXT(3) != '-')) return(NULL); |
| 1393 | |
| 1394 | SKIP(4); |
| 1395 | start = q = CUR_PTR; |
| 1396 | NEXT; |
| 1397 | r = CUR_PTR; |
| 1398 | NEXT; |
| 1399 | while (IS_CHAR(CUR) && |
| 1400 | ((CUR == ':') || (CUR != '>') || |
| 1401 | (*r != '-') || (*q != '-'))) { |
| 1402 | if ((*r == '-') && (*q == '-')) |
| 1403 | xmlParserError(ctxt, |
| 1404 | "Comment must not contain '--' (double-hyphen)`\n"); |
| 1405 | NEXT;r++;q++; |
| 1406 | } |
| 1407 | if (!IS_CHAR(CUR)) { |
| 1408 | xmlParserError(ctxt, "Comment not terminated \n<!--%.50s\n", start); |
| 1409 | } else { |
| 1410 | NEXT; |
| 1411 | if (create) { |
| 1412 | val = xmlStrndup(start, q - start); |
| 1413 | ret = xmlNewComment(val); |
| 1414 | free(val); |
| 1415 | } |
| 1416 | } |
| 1417 | return(ret); |
| 1418 | } |
| 1419 | |
| 1420 | /* |
| 1421 | * xmlParsePITarget: parse the name of a PI |
| 1422 | * |
| 1423 | * [17] PITarget ::= Name - (('X' | 'x') ('M' | 'm') ('L' | 'l')) |
| 1424 | */ |
| 1425 | |
| 1426 | CHAR *xmlParsePITarget(xmlParserCtxtPtr ctxt) { |
| 1427 | CHAR *name; |
| 1428 | |
| 1429 | name = xmlParseName(ctxt); |
| 1430 | if ((name != NULL) && (name[3] == 0) && |
| 1431 | ((name[0] == 'x') || (name[0] == 'X')) && |
| 1432 | ((name[1] == 'm') || (name[1] == 'M')) && |
| 1433 | ((name[2] == 'l') || (name[2] == 'L'))) { |
| 1434 | xmlParserError(ctxt, "xmlParsePItarget: invalid name prefix 'xml'\n"); |
| 1435 | return(NULL); |
| 1436 | } |
| 1437 | return(name); |
| 1438 | } |
| 1439 | |
| 1440 | /* |
| 1441 | * xmlParsePI: parse an XML Processing Instruction. |
| 1442 | * |
| 1443 | * [16] PI ::= '<?' PITarget (S (Char* - (Char* '?>' Char*)))? '?>' |
| 1444 | */ |
| 1445 | |
| 1446 | void xmlParsePI(xmlParserCtxtPtr ctxt) { |
| 1447 | CHAR *target; |
| 1448 | |
| 1449 | if ((CUR == '<') && (NXT(1) == '?')) { |
| 1450 | /* |
| 1451 | * this is a Processing Instruction. |
| 1452 | */ |
| 1453 | SKIP(2); |
| 1454 | |
| 1455 | /* |
| 1456 | * Parse the target name and check for special support like |
| 1457 | * namespace. |
| 1458 | * |
| 1459 | * TODO : PI handling should be dynamically redefinable using an |
| 1460 | * API. Only namespace should be in the code IMHO ... |
| 1461 | */ |
| 1462 | target = xmlParsePITarget(ctxt); |
| 1463 | if (target != NULL) { |
| 1464 | /* |
| 1465 | * Support for the old Processing Instruction related to namespace. |
| 1466 | */ |
| 1467 | if ((target[0] == 'n') && (target[1] == 'a') && |
| 1468 | (target[2] == 'm') && (target[3] == 'e') && |
| 1469 | (target[4] == 's') && (target[5] == 'p') && |
| 1470 | (target[6] == 'a') && (target[7] == 'c') && |
| 1471 | (target[8] == 'e')) { |
| 1472 | xmlParseNamespace(ctxt); |
| 1473 | } else if ((target[0] == 'x') && (target[1] == 'm') && |
| 1474 | (target[2] == 'l') && (target[3] == ':') && |
| 1475 | (target[4] == 'n') && (target[5] == 'a') && |
| 1476 | (target[6] == 'm') && (target[7] == 'e') && |
| 1477 | (target[8] == 's') && (target[9] == 'p') && |
| 1478 | (target[10] == 'a') && (target[11] == 'c') && |
| 1479 | (target[12] == 'e')) { |
| 1480 | xmlParseNamespace(ctxt); |
| 1481 | } else { |
| 1482 | const CHAR *q = CUR_PTR; |
| 1483 | |
| 1484 | while (IS_CHAR(CUR) && |
| 1485 | ((CUR != '?') || (NXT(1) != '>'))) |
| 1486 | NEXT; |
| 1487 | if (!IS_CHAR(CUR)) { |
| 1488 | xmlParserError(ctxt, "xmlParsePI: PI %s never end ...\n", |
| 1489 | target); |
| 1490 | } else { |
| 1491 | CHAR *data; |
| 1492 | |
| 1493 | data = xmlStrndup(CUR_PTR, CUR_PTR - q); |
| 1494 | SKIP(2); |
| 1495 | |
| 1496 | /* |
| 1497 | * SAX: PI detected. |
| 1498 | */ |
| 1499 | if (ctxt->sax) |
| 1500 | ctxt->sax->processingInstruction(ctxt, target, data); |
| 1501 | /* |
| 1502 | * Unknown PI, ignore it ! |
| 1503 | */ |
| 1504 | else |
| 1505 | xmlParserWarning(ctxt, |
| 1506 | "xmlParsePI : skipping unknown PI %s\n", |
| 1507 | target); |
| 1508 | free(data); |
| 1509 | } |
| 1510 | } |
| 1511 | free(target); |
| 1512 | } else { |
| 1513 | xmlParserError(ctxt, "xmlParsePI : no target name\n"); |
| 1514 | /********* Should we try to complete parsing the PI ??? |
| 1515 | while (IS_CHAR(CUR) && |
| 1516 | (CUR != '?') && (CUR != '>')) |
| 1517 | NEXT; |
| 1518 | if (!IS_CHAR(CUR)) { |
| 1519 | fprintf(stderr, "xmlParsePI: PI %s never end ...\n", |
| 1520 | target); |
| 1521 | } |
| 1522 | ********************************************************/ |
| 1523 | } |
| 1524 | } |
| 1525 | } |
| 1526 | |
| 1527 | /* |
| 1528 | * xmlParseNotationDecl: parse a notation declaration |
| 1529 | * |
| 1530 | * [82] NotationDecl ::= '<!NOTATION' S Name S (ExternalID | PublicID) S? '>' |
| 1531 | * |
| 1532 | * [83] PublicID ::= 'PUBLIC' S PubidLiteral |
| 1533 | * |
| 1534 | * NOTE: Actually [75] and [83] interract badly since [75] can generate |
| 1535 | * 'PUBLIC' S PubidLiteral S SystemLiteral |
| 1536 | * |
| 1537 | * Hence there is actually 3 choices: |
| 1538 | * 'PUBLIC' S PubidLiteral |
| 1539 | * 'PUBLIC' S PubidLiteral S SystemLiteral |
| 1540 | * and 'SYSTEM' S SystemLiteral |
| 1541 | */ |
| 1542 | |
| 1543 | void xmlParseNotationDecl(xmlParserCtxtPtr ctxt) { |
| 1544 | CHAR *name; |
| 1545 | |
| 1546 | if ((CUR == '<') && (NXT(1) == '!') && |
| 1547 | (NXT(2) == 'N') && (NXT(3) == 'O') && |
| 1548 | (NXT(4) == 'T') && (NXT(5) == 'A') && |
| 1549 | (NXT(6) == 'T') && (NXT(7) == 'I') && |
| 1550 | (NXT(8) == 'O') && (NXT(9) == 'N') && |
| 1551 | (IS_BLANK(NXT(10)))) { |
| 1552 | SKIP(10); |
| 1553 | SKIP_BLANKS; |
| 1554 | |
| 1555 | name = xmlParseName(ctxt); |
| 1556 | if (name == NULL) { |
| 1557 | xmlParserError(ctxt, |
| 1558 | "xmlParseAttributeListDecl: no name for Element\n"); |
| 1559 | return; |
| 1560 | } |
| 1561 | SKIP_BLANKS; |
| 1562 | /* |
| 1563 | * TODO !!! |
| 1564 | */ |
| 1565 | while ((IS_CHAR(CUR)) && (CUR != '>')) |
| 1566 | NEXT; |
| 1567 | free(name); |
| 1568 | } |
| 1569 | } |
| 1570 | |
| 1571 | /* |
| 1572 | * xmlParseEntityDecl: parse <!ENTITY declarations |
| 1573 | * |
| 1574 | * [70] EntityDecl ::= GEDecl | PEDecl |
| 1575 | * |
| 1576 | * [71] GEDecl ::= '<!ENTITY' S Name S EntityDef S? '>' |
| 1577 | * |
| 1578 | * [72] PEDecl ::= '<!ENTITY' S '%' S Name S PEDef S? '>' |
| 1579 | * |
| 1580 | * [73] EntityDef ::= EntityValue | (ExternalID NDataDecl?) |
| 1581 | * |
| 1582 | * [74] PEDef ::= EntityValue | ExternalID |
| 1583 | * |
| 1584 | * [76] NDataDecl ::= S 'NDATA' S Name |
| 1585 | */ |
| 1586 | |
| 1587 | void xmlParseEntityDecl(xmlParserCtxtPtr ctxt) { |
| 1588 | CHAR *name = NULL; |
| 1589 | CHAR *value = NULL; |
| 1590 | CHAR *URI = NULL, *literal = NULL; |
| 1591 | CHAR *ndata = NULL; |
| 1592 | int isParameter = 0; |
| 1593 | |
| 1594 | if ((CUR == '<') && (NXT(1) == '!') && |
| 1595 | (NXT(2) == 'E') && (NXT(3) == 'N') && |
| 1596 | (NXT(4) == 'T') && (NXT(5) == 'I') && |
| 1597 | (NXT(6) == 'T') && (NXT(7) == 'Y') && |
| 1598 | (IS_BLANK(NXT(8)))) { |
| 1599 | SKIP(8); |
| 1600 | SKIP_BLANKS; |
| 1601 | |
| 1602 | if (CUR == '%') { |
| 1603 | NEXT; |
| 1604 | SKIP_BLANKS; |
| 1605 | isParameter = 1; |
| 1606 | } |
| 1607 | |
| 1608 | name = xmlParseName(ctxt); |
| 1609 | if (name == NULL) { |
| 1610 | xmlParserError(ctxt, "xmlParseEntityDecl: no name\n"); |
| 1611 | return; |
| 1612 | } |
| 1613 | SKIP_BLANKS; |
| 1614 | |
| 1615 | /* |
| 1616 | * TODO handle the various case of definitions... |
| 1617 | */ |
| 1618 | if (isParameter) { |
| 1619 | if ((CUR == '"') || (CUR == '\'')) |
| 1620 | value = xmlParseEntityValue(ctxt); |
| 1621 | if (value) { |
| 1622 | xmlAddDocEntity(ctxt->doc, name, |
| 1623 | XML_INTERNAL_PARAMETER_ENTITY, |
| 1624 | NULL, NULL, value); |
| 1625 | } |
| 1626 | else { |
| 1627 | URI = xmlParseExternalID(ctxt, &literal); |
| 1628 | if (URI) { |
| 1629 | xmlAddDocEntity(ctxt->doc, name, |
| 1630 | XML_EXTERNAL_PARAMETER_ENTITY, |
| 1631 | literal, URI, NULL); |
| 1632 | } |
| 1633 | } |
| 1634 | } else { |
| 1635 | if ((CUR == '"') || (CUR == '\'')) { |
| 1636 | value = xmlParseEntityValue(ctxt); |
| 1637 | xmlAddDocEntity(ctxt->doc, name, |
| 1638 | XML_INTERNAL_GENERAL_ENTITY, |
| 1639 | NULL, NULL, value); |
| 1640 | } else { |
| 1641 | URI = xmlParseExternalID(ctxt, &literal); |
| 1642 | SKIP_BLANKS; |
| 1643 | if ((CUR == 'N') && (NXT(1) == 'D') && |
| 1644 | (NXT(2) == 'A') && (NXT(3) == 'T') && |
| 1645 | (NXT(4) == 'A')) { |
| 1646 | SKIP(5); |
| 1647 | SKIP_BLANKS; |
| 1648 | ndata = xmlParseName(ctxt); |
| 1649 | xmlAddDocEntity(ctxt->doc, name, |
| 1650 | XML_EXTERNAL_GENERAL_UNPARSED_ENTITY, |
| 1651 | literal, URI, ndata); |
| 1652 | } else { |
| 1653 | xmlAddDocEntity(ctxt->doc, name, |
| 1654 | XML_EXTERNAL_GENERAL_PARSED_ENTITY, |
| 1655 | literal, URI, NULL); |
| 1656 | } |
| 1657 | } |
| 1658 | } |
| 1659 | SKIP_BLANKS; |
| 1660 | if (CUR != '>') { |
| 1661 | xmlParserError(ctxt, |
| 1662 | "xmlParseEntityDecl: entity %s not terminated\n", name); |
| 1663 | } else |
| 1664 | NEXT; |
| 1665 | if (name != NULL) free(name); |
| 1666 | if (value != NULL) free(value); |
| 1667 | if (URI != NULL) free(URI); |
| 1668 | if (literal != NULL) free(literal); |
| 1669 | if (ndata != NULL) free(ndata); |
| 1670 | } |
| 1671 | } |
| 1672 | |
| 1673 | /* |
| 1674 | * xmlParseEnumeratedType: parse and Enumerated attribute type. |
| 1675 | * |
| 1676 | * [57] EnumeratedType ::= NotationType | Enumeration |
| 1677 | * |
| 1678 | * [58] NotationType ::= 'NOTATION' S '(' S? Name (S? '|' S? Name)* S? ')' |
| 1679 | * |
| 1680 | * [59] Enumeration ::= '(' S? Nmtoken (S? '|' S? Nmtoken)* S? ')' |
| 1681 | */ |
| 1682 | |
| 1683 | void xmlParseEnumeratedType(xmlParserCtxtPtr ctxt, CHAR *name) { |
| 1684 | /* |
| 1685 | * TODO !!! |
| 1686 | */ |
| 1687 | while ((IS_CHAR(CUR)) && (CUR != '>')) |
| 1688 | NEXT; |
| 1689 | } |
| 1690 | |
| 1691 | /* |
| 1692 | * xmlParseAttributeType: parse the Attribute list def for an element |
| 1693 | * |
| 1694 | * [54] AttType ::= StringType | TokenizedType | EnumeratedType |
| 1695 | * |
| 1696 | * [55] StringType ::= 'CDATA' |
| 1697 | * |
| 1698 | * [56] TokenizedType ::= 'ID' | 'IDREF' | 'IDREFS' | 'ENTITY' | |
| 1699 | * 'ENTITIES' | 'NMTOKEN' | 'NMTOKENS' |
| 1700 | */ |
| 1701 | void xmlParseAttributeType(xmlParserCtxtPtr ctxt, CHAR *name) { |
| 1702 | /* TODO !!! */ |
| 1703 | if ((CUR == 'C') && (NXT(1) == 'D') && |
| 1704 | (NXT(2) == 'A') && (NXT(3) == 'T') && |
| 1705 | (NXT(4) == 'A')) { |
| 1706 | SKIP(5); |
| 1707 | } else if ((CUR == 'I') && (NXT(1) == 'D')) { |
| 1708 | SKIP(2); |
| 1709 | } else if ((CUR == 'I') && (NXT(1) == 'D') && |
| 1710 | (NXT(2) == 'R') && (NXT(3) == 'E') && |
| 1711 | (NXT(4) == 'F')) { |
| 1712 | SKIP(5); |
| 1713 | } else if ((CUR == 'I') && (NXT(1) == 'D') && |
| 1714 | (NXT(2) == 'R') && (NXT(3) == 'E') && |
| 1715 | (NXT(4) == 'F') && (NXT(5) == 'S')) { |
| 1716 | SKIP(6); |
| 1717 | } else if ((CUR == 'E') && (NXT(1) == 'N') && |
| 1718 | (NXT(2) == 'T') && (NXT(3) == 'I') && |
| 1719 | (NXT(4) == 'T') && (NXT(5) == 'Y')) { |
| 1720 | SKIP(6); |
| 1721 | } else if ((CUR == 'E') && (NXT(1) == 'N') && |
| 1722 | (NXT(2) == 'T') && (NXT(3) == 'I') && |
| 1723 | (NXT(4) == 'T') && (NXT(5) == 'I') && |
| 1724 | (NXT(6) == 'E') && (NXT(7) == 'S')) { |
| 1725 | SKIP(8); |
| 1726 | } else if ((CUR == 'N') && (NXT(1) == 'M') && |
| 1727 | (NXT(2) == 'T') && (NXT(3) == 'O') && |
| 1728 | (NXT(4) == 'K') && (NXT(5) == 'E') && |
| 1729 | (NXT(6) == 'N')) { |
| 1730 | SKIP(7); |
| 1731 | } else if ((CUR == 'N') && (NXT(1) == 'M') && |
| 1732 | (NXT(2) == 'T') && (NXT(3) == 'O') && |
| 1733 | (NXT(4) == 'K') && (NXT(5) == 'E') && |
| 1734 | (NXT(6) == 'N') && (NXT(7) == 'S')) { |
| 1735 | } else { |
| 1736 | xmlParseEnumeratedType(ctxt, name); |
| 1737 | } |
| 1738 | } |
| 1739 | |
| 1740 | /* |
| 1741 | * xmlParseAttributeListDecl: parse the Attribute list def for an element |
| 1742 | * |
| 1743 | * [52] AttlistDecl ::= '<!ATTLIST' S Name AttDef* S? '>' |
| 1744 | * |
| 1745 | * [53] AttDef ::= S Name S AttType S DefaultDecl |
| 1746 | */ |
| 1747 | void xmlParseAttributeListDecl(xmlParserCtxtPtr ctxt) { |
| 1748 | CHAR *name; |
| 1749 | |
| 1750 | /* TODO !!! */ |
| 1751 | if ((CUR == '<') && (NXT(1) == '!') && |
| 1752 | (NXT(2) == 'A') && (NXT(3) == 'T') && |
| 1753 | (NXT(4) == 'T') && (NXT(5) == 'L') && |
| 1754 | (NXT(6) == 'I') && (NXT(7) == 'S') && |
| 1755 | (NXT(8) == 'T') && (IS_BLANK(NXT(9)))) { |
| 1756 | SKIP(9); |
| 1757 | SKIP_BLANKS; |
| 1758 | name = xmlParseName(ctxt); |
| 1759 | if (name == NULL) { |
| 1760 | xmlParserError(ctxt, |
| 1761 | "xmlParseAttributeListDecl: no name for Element\n"); |
| 1762 | return; |
| 1763 | } |
| 1764 | SKIP_BLANKS; |
| 1765 | while (CUR != '>') { |
| 1766 | const CHAR *check = CUR_PTR; |
| 1767 | |
| 1768 | xmlParseAttributeType(ctxt, name); |
| 1769 | SKIP_BLANKS; |
| 1770 | if (check == CUR_PTR) { |
| 1771 | xmlParserError(ctxt, |
| 1772 | "xmlParseAttributeListDecl: detected error\n"); |
| 1773 | break; |
| 1774 | } |
| 1775 | } |
| 1776 | if (CUR == '>') |
| 1777 | NEXT; |
| 1778 | |
| 1779 | free(name); |
| 1780 | } |
| 1781 | } |
| 1782 | |
| 1783 | /* |
| 1784 | * xmlParseElementContentDecl: parse the declaration for an Element content |
| 1785 | * either Mixed or Children, the cases EMPTY and ANY being handled |
| 1786 | * int xmlParseElementDecl. |
| 1787 | * |
| 1788 | * [47] children ::= (choice | seq) ('?' | '*' | '+')? |
| 1789 | * |
| 1790 | * [48] cp ::= (Name | choice | seq) ('?' | '*' | '+')? |
| 1791 | * |
| 1792 | * [49] choice ::= '(' S? cp ( S? '|' S? cp )* S? ')' |
| 1793 | * |
| 1794 | * [50] seq ::= '(' S? cp ( S? ',' S? cp )* S? ')' |
| 1795 | * |
| 1796 | * or |
| 1797 | * |
| 1798 | * [51] Mixed ::= '(' S? '#PCDATA' (S? '|' S? Name)* S? ')*' | |
| 1799 | * '(' S? '#PCDATA' S? ')' |
| 1800 | */ |
| 1801 | |
| 1802 | void xmlParseElementContentDecl(xmlParserCtxtPtr ctxt, CHAR *name) { |
| 1803 | /* |
| 1804 | * TODO This has to be parsed correctly, currently we just skip until |
| 1805 | * we reach the first '>'. |
| 1806 | * !!! |
| 1807 | */ |
| 1808 | while ((IS_CHAR(CUR)) && (CUR != '>')) |
| 1809 | NEXT; |
| 1810 | } |
| 1811 | |
| 1812 | /* |
| 1813 | * xmlParseElementDecl: parse an Element declaration. |
| 1814 | * |
| 1815 | * [45] elementdecl ::= '<!ELEMENT' S Name S contentspec S? '>' |
| 1816 | * |
| 1817 | * [46] contentspec ::= 'EMPTY' | 'ANY' | Mixed | children |
| 1818 | * |
| 1819 | * TODO There is a check [ VC: Unique Element Type Declaration ] |
| 1820 | */ |
| 1821 | void xmlParseElementDecl(xmlParserCtxtPtr ctxt) { |
| 1822 | CHAR *name; |
| 1823 | |
| 1824 | if ((CUR == '<') && (NXT(1) == '!') && |
| 1825 | (NXT(2) == 'E') && (NXT(3) == 'L') && |
| 1826 | (NXT(4) == 'E') && (NXT(5) == 'M') && |
| 1827 | (NXT(6) == 'E') && (NXT(7) == 'N') && |
| 1828 | (NXT(8) == 'T') && (IS_BLANK(NXT(9)))) { |
| 1829 | SKIP(9); |
| 1830 | SKIP_BLANKS; |
| 1831 | name = xmlParseName(ctxt); |
| 1832 | if (name == NULL) { |
| 1833 | xmlParserError(ctxt, "xmlParseElementDecl: no name for Element\n"); |
| 1834 | return; |
| 1835 | } |
| 1836 | SKIP_BLANKS; |
| 1837 | if ((CUR == 'E') && (NXT(1) == 'M') && |
| 1838 | (NXT(2) == 'P') && (NXT(3) == 'T') && |
| 1839 | (NXT(4) == 'Y')) { |
| 1840 | SKIP(5); |
| 1841 | /* |
| 1842 | * Element must always be empty. |
| 1843 | */ |
| 1844 | } else if ((CUR == 'A') && (NXT(1) == 'N') && |
| 1845 | (NXT(2) == 'Y')) { |
| 1846 | SKIP(3); |
| 1847 | /* |
| 1848 | * Element is a generic container. |
| 1849 | */ |
| 1850 | } else { |
| 1851 | xmlParseElementContentDecl(ctxt, name); |
| 1852 | } |
| 1853 | SKIP_BLANKS; |
| 1854 | if (CUR != '>') { |
| 1855 | xmlParserError(ctxt, |
| 1856 | "xmlParseElementDecl: expected '>' at the end\n"); |
| 1857 | } else |
| 1858 | NEXT; |
| 1859 | } |
| 1860 | } |
| 1861 | |
| 1862 | /* |
| 1863 | * xmlParseMarkupDecl: parse Markup declarations |
| 1864 | * |
| 1865 | * [29] markupdecl ::= elementdecl | AttlistDecl | EntityDecl | |
| 1866 | * NotationDecl | PI | Comment |
| 1867 | * |
| 1868 | * TODO There is a check [ VC: Proper Declaration/PE Nesting ] |
| 1869 | */ |
| 1870 | void xmlParseMarkupDecl(xmlParserCtxtPtr ctxt) { |
| 1871 | xmlParseElementDecl(ctxt); |
| 1872 | xmlParseAttributeListDecl(ctxt); |
| 1873 | xmlParseEntityDecl(ctxt); |
| 1874 | xmlParseNotationDecl(ctxt); |
| 1875 | xmlParsePI(ctxt); |
| 1876 | xmlParseComment(ctxt, 0); |
| 1877 | } |
| 1878 | |
| 1879 | /* |
| 1880 | * xmlParseCharRef: parse Reference declarations |
| 1881 | * |
| 1882 | * [66] CharRef ::= '&#' [0-9]+ ';' | |
| 1883 | * '&#x' [0-9a-fA-F]+ ';' |
| 1884 | */ |
| 1885 | CHAR *xmlParseCharRef(xmlParserCtxtPtr ctxt, int inLine) { |
| 1886 | int val = 0; |
| 1887 | CHAR buf[2]; |
| 1888 | |
| 1889 | if ((CUR == '&') && (NXT(1) == '#') && |
| 1890 | (NXT(2) == 'x')) { |
| 1891 | SKIP(3); |
| 1892 | while (CUR != ';') { |
| 1893 | if ((CUR >= '0') && (CUR <= '9')) |
| 1894 | val = val * 16 + (CUR - '0'); |
| 1895 | else if ((CUR >= 'a') && (CUR <= 'f')) |
| 1896 | val = val * 16 + (CUR - 'a') + 10; |
| 1897 | else if ((CUR >= 'A') && (CUR <= 'F')) |
| 1898 | val = val * 16 + (CUR - 'A') + 10; |
| 1899 | else { |
| 1900 | xmlParserError(ctxt, "xmlParseCharRef: invalid value\n"); |
| 1901 | val = 0; |
| 1902 | break; |
| 1903 | } |
Daniel Veillard | 845664d | 1998-08-13 04:43:19 +0000 | [diff] [blame] | 1904 | NEXT; |
Daniel Veillard | 260a68f | 1998-08-13 03:39:55 +0000 | [diff] [blame] | 1905 | } |
| 1906 | if (CUR != ';') |
| 1907 | NEXT; |
| 1908 | } else if ((CUR == '&') && (NXT(1) == '#')) { |
| 1909 | SKIP(2); |
| 1910 | while (CUR != ';') { |
| 1911 | if ((CUR >= '0') && (CUR <= '9')) |
| 1912 | val = val * 16 + (CUR - '0'); |
| 1913 | else { |
| 1914 | xmlParserError(ctxt, "xmlParseCharRef: invalid value\n"); |
| 1915 | val = 0; |
| 1916 | break; |
| 1917 | } |
Daniel Veillard | 845664d | 1998-08-13 04:43:19 +0000 | [diff] [blame] | 1918 | NEXT; |
Daniel Veillard | 260a68f | 1998-08-13 03:39:55 +0000 | [diff] [blame] | 1919 | } |
| 1920 | if (CUR != ';') |
| 1921 | NEXT; |
| 1922 | } else { |
| 1923 | xmlParserError(ctxt, "xmlParseCharRef: invalid value\n"); |
| 1924 | } |
| 1925 | /* |
| 1926 | * Check the value IS_CHAR ... |
| 1927 | */ |
| 1928 | if (IS_CHAR(val)) { |
| 1929 | buf[0] = (CHAR) val; |
| 1930 | buf[1] = 0; |
| 1931 | if (inLine) |
| 1932 | return(xmlStrndup(buf, 1)); |
| 1933 | else if (ctxt->sax != NULL) |
| 1934 | ctxt->sax->characters(ctxt, buf, 0, 1); |
| 1935 | } else { |
| 1936 | xmlParserError(ctxt, "xmlParseCharRef: invalid value"); |
| 1937 | } |
| 1938 | return(NULL); |
| 1939 | } |
| 1940 | |
| 1941 | /* |
| 1942 | * xmlParseEntityRef: parse ENTITY references declarations |
| 1943 | * |
| 1944 | * [68] EntityRef ::= '&' Name ';' |
| 1945 | */ |
| 1946 | CHAR *xmlParseEntityRef(xmlParserCtxtPtr ctxt, int inLine) { |
| 1947 | CHAR *ret = NULL; |
| 1948 | CHAR *name; |
| 1949 | xmlEntityPtr entity; |
| 1950 | |
| 1951 | if (CUR == '&') { |
| 1952 | NEXT; |
| 1953 | name = xmlParseName(ctxt); |
| 1954 | if (name == NULL) { |
| 1955 | xmlParserError(ctxt, "xmlParseEntityRef: no name\n"); |
| 1956 | } else { |
| 1957 | if (CUR == ';') { |
| 1958 | NEXT; |
| 1959 | entity = xmlGetDocEntity(ctxt->doc, name); |
| 1960 | if (entity == NULL) { |
| 1961 | /* TODO !!! Create a reference ! */ |
| 1962 | xmlParserWarning(ctxt, |
| 1963 | "xmlParseEntityRef: &%s; not found\n", name); |
| 1964 | } |
| 1965 | /* |
| 1966 | * If we can get the content, push the entity content |
| 1967 | * as the next input stream. |
| 1968 | */ |
| 1969 | else { |
| 1970 | switch (entity->type) { |
| 1971 | case XML_INTERNAL_PARAMETER_ENTITY: |
| 1972 | case XML_EXTERNAL_PARAMETER_ENTITY: |
| 1973 | xmlParserError(ctxt, |
| 1974 | "internal: xmlGetDtdEntity returned a general entity\n"); |
| 1975 | break; |
| 1976 | case XML_INTERNAL_GENERAL_ENTITY: |
| 1977 | if (inLine) |
| 1978 | ret = entity->content; |
| 1979 | else |
| 1980 | xmlHandleEntity(ctxt, entity); |
| 1981 | break; |
| 1982 | case XML_EXTERNAL_GENERAL_PARSED_ENTITY: |
| 1983 | case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY: |
| 1984 | xmlParserWarning(ctxt, |
| 1985 | "xmlParseEntityRef: external entity &%s; not supported\n", |
| 1986 | name); |
| 1987 | break; |
| 1988 | default: |
| 1989 | xmlParserError(ctxt, |
| 1990 | "internal: xmlParseEntityRef: unknown entity type %d\n", |
| 1991 | entity->type); |
| 1992 | } |
| 1993 | } |
| 1994 | } else { |
| 1995 | char cst[2] = { '&', 0 }; |
| 1996 | |
| 1997 | xmlParserError(ctxt, "xmlParseEntityRef: expecting ';'\n"); |
| 1998 | ret = xmlStrndup(cst, 1); |
| 1999 | ret = xmlStrcat(ret, name); |
| 2000 | } |
| 2001 | free(name); |
| 2002 | } |
| 2003 | } |
| 2004 | return(ret); |
| 2005 | } |
| 2006 | |
| 2007 | /* |
| 2008 | * xmlParseReference: parse Reference declarations |
| 2009 | * |
| 2010 | * [67] Reference ::= EntityRef | CharRef |
| 2011 | */ |
| 2012 | CHAR *xmlParseReference(xmlParserCtxtPtr ctxt, int inLine) { |
| 2013 | if ((CUR == '&') && (NXT(1) == '#')) { |
| 2014 | return(xmlParseCharRef(ctxt, inLine)); |
| 2015 | } else if (CUR == '&') { |
| 2016 | return(xmlParseEntityRef(ctxt, inLine)); |
| 2017 | } |
| 2018 | return(NULL); |
| 2019 | } |
| 2020 | |
| 2021 | /* |
| 2022 | * xmlParsePEReference: parse PEReference declarations |
| 2023 | * |
| 2024 | * [69] PEReference ::= '%' Name ';' |
| 2025 | */ |
| 2026 | CHAR *xmlParsePEReference(xmlParserCtxtPtr ctxt, int inLine) { |
| 2027 | CHAR *ret = NULL; |
| 2028 | CHAR *name; |
| 2029 | xmlEntityPtr entity; |
| 2030 | |
| 2031 | if (CUR == '%') { |
| 2032 | NEXT; |
| 2033 | name = xmlParseName(ctxt); |
| 2034 | if (name == NULL) { |
| 2035 | xmlParserError(ctxt, "xmlParsePEReference: no name\n"); |
| 2036 | } else { |
| 2037 | if (CUR == ';') { |
| 2038 | NEXT; |
| 2039 | entity = xmlGetDtdEntity(ctxt->doc, name); |
| 2040 | if (entity == NULL) { |
| 2041 | xmlParserWarning(ctxt, |
| 2042 | "xmlParsePEReference: %%%s; not found\n"); |
| 2043 | } |
| 2044 | /* |
| 2045 | * If we can get the content, push the entity content |
| 2046 | * as the next input stream. |
| 2047 | */ |
| 2048 | else { |
| 2049 | switch (entity->type) { |
| 2050 | case XML_INTERNAL_PARAMETER_ENTITY: |
| 2051 | if (inLine) |
| 2052 | ret = entity->content; |
| 2053 | else |
| 2054 | xmlNewEntityInputStream(ctxt, entity); |
| 2055 | break; |
| 2056 | case XML_EXTERNAL_PARAMETER_ENTITY: |
| 2057 | xmlParserWarning(ctxt, |
| 2058 | "xmlParsePEReference: external entity %%%s; not supported\n"); |
| 2059 | break; |
| 2060 | case XML_INTERNAL_GENERAL_ENTITY: |
| 2061 | case XML_EXTERNAL_GENERAL_PARSED_ENTITY: |
| 2062 | case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY: |
| 2063 | xmlParserError(ctxt, |
| 2064 | "internal: xmlGetDtdEntity returned a general entity\n"); |
| 2065 | break; |
| 2066 | default: |
| 2067 | xmlParserError(ctxt, |
| 2068 | "internal: xmlParsePEReference: unknown entity type %d\n", |
| 2069 | entity->type); |
| 2070 | } |
| 2071 | } |
| 2072 | } else { |
| 2073 | char cst[2] = { '&', 0 }; |
| 2074 | |
| 2075 | xmlParserError(ctxt, "xmlParsePEReference: expecting ';'\n"); |
| 2076 | ret = xmlStrndup(cst, 1); |
| 2077 | ret = xmlStrcat(ret, name); |
| 2078 | } |
| 2079 | free(name); |
| 2080 | } |
| 2081 | } |
| 2082 | return(ret); |
| 2083 | } |
| 2084 | |
| 2085 | /* |
| 2086 | * xmlParseDocTypeDecl : parse a DOCTYPE declaration |
| 2087 | * |
| 2088 | * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S? |
| 2089 | * ('[' (markupdecl | PEReference | S)* ']' S?)? '>' |
| 2090 | */ |
| 2091 | |
| 2092 | void xmlParseDocTypeDecl(xmlParserCtxtPtr ctxt) { |
| 2093 | xmlDtdPtr dtd; |
| 2094 | CHAR *name; |
| 2095 | CHAR *ExternalID = NULL; |
| 2096 | CHAR *URI = NULL; |
| 2097 | |
| 2098 | /* |
| 2099 | * We know that '<!DOCTYPE' has been detected. |
| 2100 | */ |
| 2101 | SKIP(9); |
| 2102 | |
| 2103 | SKIP_BLANKS; |
| 2104 | |
| 2105 | /* |
| 2106 | * Parse the DOCTYPE name. |
| 2107 | */ |
| 2108 | name = xmlParseName(ctxt); |
| 2109 | if (name == NULL) { |
| 2110 | xmlParserError(ctxt, "xmlParseDocTypeDecl : no DOCTYPE name !\n"); |
| 2111 | } |
| 2112 | |
| 2113 | SKIP_BLANKS; |
| 2114 | |
| 2115 | /* |
| 2116 | * Check for SystemID and ExternalID |
| 2117 | */ |
| 2118 | URI = xmlParseExternalID(ctxt, &ExternalID); |
| 2119 | SKIP_BLANKS; |
| 2120 | |
| 2121 | dtd = xmlNewDtd(ctxt->doc, name, ExternalID, URI); |
| 2122 | |
| 2123 | /* |
| 2124 | * Is there any DTD definition ? |
| 2125 | */ |
| 2126 | if (CUR == '[') { |
| 2127 | NEXT; |
| 2128 | /* |
| 2129 | * Parse the succession of Markup declarations and |
| 2130 | * PEReferences. |
| 2131 | * Subsequence (markupdecl | PEReference | S)* |
| 2132 | */ |
| 2133 | while (CUR != ']') { |
| 2134 | const CHAR *check = CUR_PTR; |
| 2135 | |
| 2136 | SKIP_BLANKS; |
| 2137 | xmlParseMarkupDecl(ctxt); |
| 2138 | xmlParsePEReference(ctxt, 0); |
| 2139 | |
| 2140 | if (CUR_PTR == check) { |
| 2141 | xmlParserError(ctxt, |
| 2142 | "xmlParseDocTypeDecl: error detected in Markup declaration\n"); |
| 2143 | break; |
| 2144 | } |
| 2145 | } |
| 2146 | if (CUR == ']') NEXT; |
| 2147 | } |
| 2148 | |
| 2149 | /* |
| 2150 | * We should be at the end of the DOCTYPE declaration. |
| 2151 | */ |
| 2152 | if (CUR != '>') { |
| 2153 | xmlParserError(ctxt, "DOCTYPE unproperly terminated\n"); |
| 2154 | /* We shouldn't try to resynchronize ... */ |
| 2155 | } |
| 2156 | NEXT; |
| 2157 | |
| 2158 | /* |
| 2159 | * Cleanup, since we don't use all those identifiers |
| 2160 | * TODO : the DOCTYPE if available should be stored ! |
| 2161 | */ |
| 2162 | if (URI != NULL) free(URI); |
| 2163 | if (ExternalID != NULL) free(ExternalID); |
| 2164 | if (name != NULL) free(name); |
| 2165 | } |
| 2166 | |
| 2167 | /* |
| 2168 | * xmlParseAttribute: parse a start of tag. |
| 2169 | * |
| 2170 | * [41] Attribute ::= Name Eq AttValue |
| 2171 | * |
| 2172 | * [25] Eq ::= S? '=' S? |
| 2173 | * |
| 2174 | * With namespace: |
| 2175 | * |
| 2176 | * [NS 11] Attribute ::= QName Eq AttValue |
| 2177 | * |
| 2178 | * Also the case QName == xmlns:??? is handled independently as a namespace |
| 2179 | * definition. |
| 2180 | */ |
| 2181 | |
| 2182 | void xmlParseAttribute(xmlParserCtxtPtr ctxt, xmlNodePtr node) { |
| 2183 | CHAR *name, *value = NULL; |
| 2184 | CHAR *ns; |
| 2185 | |
| 2186 | name = xmlNamespaceParseQName(ctxt, &ns); |
| 2187 | if (name == NULL) { |
| 2188 | xmlParserError(ctxt, "error parsing attribute name\n"); |
| 2189 | return; |
| 2190 | } |
| 2191 | |
| 2192 | /* |
| 2193 | * read the value |
| 2194 | */ |
| 2195 | SKIP_BLANKS; |
| 2196 | if (CUR == '=') { |
| 2197 | NEXT; |
| 2198 | SKIP_BLANKS; |
| 2199 | value = xmlParseAttValue(ctxt); |
| 2200 | } else { |
| 2201 | xmlParserError(ctxt, "Specification mandate value for attribute %s\n", |
| 2202 | name); |
| 2203 | } |
| 2204 | |
| 2205 | /* |
| 2206 | * Check whether it's a namespace definition |
| 2207 | */ |
| 2208 | if ((ns == NULL) && |
| 2209 | (name[0] == 'x') && (name[1] == 'm') && (name[2] == 'l') && |
| 2210 | (name[3] == 'n') && (name[4] == 's') && (name[5] == 0)) { |
| 2211 | /* a default namespace definition */ |
| 2212 | xmlNewNs(node, value, NULL); |
| 2213 | if (name != NULL) |
| 2214 | free(name); |
| 2215 | if (value != NULL) |
| 2216 | free(value); |
| 2217 | return; |
| 2218 | } |
| 2219 | if ((ns != NULL) && (ns[0] == 'x') && (ns[1] == 'm') && (ns[2] == 'l') && |
| 2220 | (ns[3] == 'n') && (ns[4] == 's') && (ns[5] == 0)) { |
| 2221 | /* a standard namespace definition */ |
| 2222 | xmlNewNs(node, value, name); |
| 2223 | if (name != NULL) |
| 2224 | free(name); |
| 2225 | if (value != NULL) |
| 2226 | free(value); |
| 2227 | return; |
| 2228 | } |
| 2229 | |
| 2230 | /* |
| 2231 | * Add the attribute to the node. |
| 2232 | */ |
| 2233 | if (name != NULL) { |
| 2234 | xmlNewProp(node, name, value); |
| 2235 | free(name); |
| 2236 | } |
| 2237 | if (value != NULL) |
| 2238 | free(value); |
| 2239 | } |
| 2240 | |
| 2241 | /* |
| 2242 | * xmlParseStartTag: parse a start of tag either for rule element or |
| 2243 | * EmptyElement. In both case we don't parse the tag closing chars. |
| 2244 | * |
| 2245 | * [40] STag ::= '<' Name (S Attribute)* S? '>' |
| 2246 | * |
| 2247 | * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>' |
| 2248 | * |
| 2249 | * With namespace: |
| 2250 | * |
| 2251 | * [NS 8] STag ::= '<' QName (S Attribute)* S? '>' |
| 2252 | * |
| 2253 | * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>' |
| 2254 | */ |
| 2255 | |
| 2256 | xmlNodePtr xmlParseStartTag(xmlParserCtxtPtr ctxt) { |
| 2257 | CHAR *namespace, *name; |
| 2258 | xmlNsPtr ns = NULL; |
| 2259 | xmlNodePtr ret = NULL; |
| 2260 | |
| 2261 | if (CUR != '<') return(NULL); |
| 2262 | NEXT; |
| 2263 | |
| 2264 | name = xmlNamespaceParseQName(ctxt, &namespace); |
| 2265 | |
| 2266 | /* |
| 2267 | * Note : the namespace resolution is deferred until the end of the |
| 2268 | * attributes parsing, since local namespace can be defined as |
| 2269 | * an attribute at this level. |
| 2270 | */ |
| 2271 | ret = xmlNewNode(ns, name, NULL); |
| 2272 | |
| 2273 | /* |
| 2274 | * Now parse the attributes, it ends up with the ending |
| 2275 | * |
| 2276 | * (S Attribute)* S? |
| 2277 | */ |
| 2278 | SKIP_BLANKS; |
| 2279 | while ((IS_CHAR(CUR)) && |
| 2280 | (CUR != '>') && |
| 2281 | ((CUR != '/') || (NXT(1) != '>'))) { |
| 2282 | const CHAR *q = CUR_PTR; |
| 2283 | |
| 2284 | xmlParseAttribute(ctxt, ret); |
| 2285 | SKIP_BLANKS; |
| 2286 | |
| 2287 | if (q == CUR_PTR) { |
| 2288 | xmlParserError(ctxt, |
| 2289 | "xmlParseStartTag: problem parsing attributes\n"); |
| 2290 | break; |
| 2291 | } |
| 2292 | } |
| 2293 | |
| 2294 | /* |
| 2295 | * Search the namespace |
| 2296 | */ |
| 2297 | ns = xmlSearchNs(ctxt->doc, ret, namespace); |
| 2298 | if (ns == NULL) /* ret still doesn't have a parent yet ! */ |
| 2299 | ns = xmlSearchNs(ctxt->doc, ctxt->node, namespace); |
| 2300 | xmlSetNs(ret, ns); |
| 2301 | if (namespace != NULL) |
| 2302 | free(namespace); |
| 2303 | |
| 2304 | /* |
| 2305 | * We are parsing a new node. |
| 2306 | */ |
| 2307 | nodePush(ctxt, ret); |
| 2308 | |
| 2309 | /* |
| 2310 | * SAX: Start of Element ! |
| 2311 | */ |
| 2312 | if (ctxt->sax != NULL) |
| 2313 | ctxt->sax->startElement(ctxt, name); |
| 2314 | |
| 2315 | return(ret); |
| 2316 | } |
| 2317 | |
| 2318 | /* |
| 2319 | * xmlParseEndTag: parse an end of tag |
| 2320 | * |
| 2321 | * [42] ETag ::= '</' Name S? '>' |
| 2322 | * |
| 2323 | * With namespace |
| 2324 | * |
| 2325 | * [9] ETag ::= '</' QName S? '>' |
| 2326 | */ |
| 2327 | |
| 2328 | void xmlParseEndTag(xmlParserCtxtPtr ctxt, xmlNsPtr *nsPtr, CHAR **tagPtr) { |
| 2329 | CHAR *namespace, *name; |
| 2330 | xmlNsPtr ns = NULL; |
| 2331 | |
| 2332 | *nsPtr = NULL; |
| 2333 | *tagPtr = NULL; |
| 2334 | |
| 2335 | if ((CUR != '<') || (NXT(1) != '/')) { |
| 2336 | xmlParserError(ctxt, "xmlParseEndTag: '</' not found\n"); |
| 2337 | return; |
| 2338 | } |
| 2339 | SKIP(2); |
| 2340 | |
| 2341 | name = xmlNamespaceParseQName(ctxt, &namespace); |
| 2342 | |
| 2343 | /* |
| 2344 | * Search the namespace |
| 2345 | */ |
| 2346 | ns = xmlSearchNs(ctxt->doc, ctxt->node, namespace); |
| 2347 | if (namespace != NULL) |
| 2348 | free(namespace); |
| 2349 | |
| 2350 | *nsPtr = ns; |
| 2351 | *tagPtr = name; |
| 2352 | |
| 2353 | /* |
| 2354 | * We should definitely be at the ending "S? '>'" part |
| 2355 | */ |
| 2356 | SKIP_BLANKS; |
| 2357 | if ((!IS_CHAR(CUR)) || (CUR != '>')) { |
| 2358 | xmlParserError(ctxt, "End tag : expected '>'\n"); |
| 2359 | } else |
| 2360 | NEXT; |
| 2361 | |
| 2362 | return; |
| 2363 | } |
| 2364 | |
| 2365 | /* |
| 2366 | * xmlParseCDSect: escaped pure raw content. |
| 2367 | * |
| 2368 | * [18] CDSect ::= CDStart CData CDEnd |
| 2369 | * |
| 2370 | * [19] CDStart ::= '<![CDATA[' |
| 2371 | * |
| 2372 | * [20] Data ::= (Char* - (Char* ']]>' Char*)) |
| 2373 | * |
| 2374 | * [21] CDEnd ::= ']]>' |
| 2375 | */ |
| 2376 | void xmlParseCDSect(xmlParserCtxtPtr ctxt) { |
| 2377 | const CHAR *r, *s, *base; |
| 2378 | |
| 2379 | if ((CUR == '<') && (NXT(1) == '!') && |
| 2380 | (NXT(2) == '[') && (NXT(3) == 'C') && |
| 2381 | (NXT(4) == 'D') && (NXT(5) == 'A') && |
| 2382 | (NXT(6) == 'T') && (NXT(7) == 'A') && |
| 2383 | (NXT(8) == '[')) { |
| 2384 | SKIP(9); |
| 2385 | } else |
| 2386 | return; |
| 2387 | base = CUR_PTR; |
| 2388 | if (!IS_CHAR(CUR)) { |
| 2389 | xmlParserError(ctxt, "CData section not finished\n%.50s\n", base); |
| 2390 | return; |
| 2391 | } |
| 2392 | r = NEXT; |
| 2393 | if (!IS_CHAR(CUR)) { |
| 2394 | xmlParserError(ctxt, "CData section not finished\n%.50s\n", base); |
| 2395 | return; |
| 2396 | } |
| 2397 | s = NEXT; |
| 2398 | while (IS_CHAR(CUR) && |
| 2399 | ((*r != ']') || (*s != ']') || (CUR != '>'))) { |
| 2400 | r++;s++;NEXT; |
| 2401 | } |
| 2402 | if (!IS_CHAR(CUR)) { |
| 2403 | xmlParserError(ctxt, "CData section not finished\n%.50s\n", base); |
| 2404 | return; |
| 2405 | } |
| 2406 | |
| 2407 | /* |
| 2408 | * Ok the segment [base CUR_PTR] is to be consumed as chars. |
| 2409 | */ |
| 2410 | if (ctxt->sax != NULL) { |
| 2411 | if (areBlanks(ctxt, base, CUR_PTR - base)) |
| 2412 | ctxt->sax->ignorableWhitespace(ctxt, base, 0, CUR_PTR - base); |
| 2413 | else |
| 2414 | ctxt->sax->characters(ctxt, base, 0, CUR_PTR - base); |
| 2415 | } |
| 2416 | } |
| 2417 | |
| 2418 | /* |
| 2419 | * xmlParseContent: a content is |
| 2420 | * (element | PCData | Reference | CDSect | PI | Comment) |
| 2421 | * |
| 2422 | * [43] content ::= (element | CharData | Reference | CDSect | PI | Comment)* |
| 2423 | */ |
| 2424 | |
| 2425 | void xmlParseContent(xmlParserCtxtPtr ctxt) { |
| 2426 | xmlNodePtr ret = NULL; |
| 2427 | |
| 2428 | while ((CUR != '<') || (NXT(1) != '/')) { |
| 2429 | const CHAR *test = CUR_PTR; |
| 2430 | ret = NULL; |
| 2431 | |
| 2432 | /* |
| 2433 | * First case : a Processing Instruction. |
| 2434 | */ |
| 2435 | if ((CUR == '<') && (NXT(1) == '?')) { |
| 2436 | xmlParsePI(ctxt); |
| 2437 | } |
| 2438 | /* |
| 2439 | * Second case : a CDSection |
| 2440 | */ |
| 2441 | else if ((CUR == '<') && (NXT(1) == '!') && |
| 2442 | (NXT(2) == '[') && (NXT(3) == 'C') && |
| 2443 | (NXT(4) == 'D') && (NXT(5) == 'A') && |
| 2444 | (NXT(6) == 'T') && (NXT(7) == 'A') && |
| 2445 | (NXT(8) == '[')) { |
| 2446 | xmlParseCDSect(ctxt); |
| 2447 | } |
| 2448 | /* |
| 2449 | * Third case : a comment |
| 2450 | */ |
| 2451 | else if ((CUR == '<') && (NXT(1) == '!') && |
| 2452 | (NXT(2) == '-') && (NXT(3) == '-')) { |
| 2453 | ret = xmlParseComment(ctxt, 1); |
| 2454 | } |
| 2455 | /* |
| 2456 | * Fourth case : a sub-element. |
| 2457 | */ |
| 2458 | else if (CUR == '<') { |
| 2459 | ret = xmlParseElement(ctxt); |
| 2460 | } |
| 2461 | /* |
| 2462 | * Fifth case : a reference. |
| 2463 | */ |
| 2464 | else if (CUR == '&') { |
| 2465 | xmlParseReference(ctxt, 0); |
| 2466 | } |
| 2467 | /* |
| 2468 | * Last case, text. Note that References are handled directly. |
| 2469 | */ |
| 2470 | else { |
| 2471 | xmlParseCharData(ctxt, 0); |
| 2472 | } |
| 2473 | |
| 2474 | /* |
| 2475 | * Pop-up of finished entities. |
| 2476 | */ |
| 2477 | while ((CUR == 0) && (ctxt->inputNr > 1)) xmlPopInput(ctxt); |
| 2478 | |
| 2479 | if (test == CUR_PTR) { |
| 2480 | xmlParserError(ctxt, "detected an error in element content\n"); |
| 2481 | break; |
| 2482 | } |
| 2483 | } |
| 2484 | } |
| 2485 | |
| 2486 | /* |
| 2487 | * xmlParseElement: parse an XML element |
| 2488 | * |
| 2489 | * [39] element ::= EmptyElemTag | STag content ETag |
| 2490 | * |
| 2491 | * [41] Attribute ::= Name Eq AttValue |
| 2492 | */ |
| 2493 | |
| 2494 | |
| 2495 | xmlNodePtr xmlParseElement(xmlParserCtxtPtr ctxt) { |
| 2496 | xmlNodePtr ret; |
| 2497 | const CHAR *openTag = CUR_PTR; |
| 2498 | xmlParserNodeInfo node_info; |
| 2499 | CHAR *endTag; |
| 2500 | xmlNsPtr endNs; |
| 2501 | |
| 2502 | /* Capture start position */ |
| 2503 | node_info.begin_pos = CUR_PTR - ctxt->input->base; |
| 2504 | node_info.begin_line = ctxt->input->line; |
| 2505 | |
| 2506 | ret = xmlParseStartTag(ctxt); |
| 2507 | if (ret == NULL) { |
| 2508 | return(NULL); |
| 2509 | } |
| 2510 | |
| 2511 | /* |
| 2512 | * Check for an Empty Element. |
| 2513 | */ |
| 2514 | if ((CUR == '/') && (NXT(1) == '>')) { |
| 2515 | SKIP(2); |
| 2516 | if (ctxt->sax != NULL) |
| 2517 | ctxt->sax->endElement(ctxt, ret->name); |
| 2518 | |
| 2519 | /* |
| 2520 | * end of parsing of this node. |
| 2521 | */ |
| 2522 | nodePop(ctxt); |
| 2523 | |
| 2524 | return(ret); |
| 2525 | } |
| 2526 | if (CUR == '>') NEXT; |
| 2527 | else { |
| 2528 | xmlParserError(ctxt, "Couldn't find end of Start Tag\n%.30s\n",openTag); |
| 2529 | |
| 2530 | /* |
| 2531 | * end of parsing of this node. |
| 2532 | */ |
| 2533 | nodePop(ctxt); |
| 2534 | |
| 2535 | return(NULL); |
| 2536 | } |
| 2537 | |
| 2538 | /* |
| 2539 | * Parse the content of the element: |
| 2540 | */ |
| 2541 | xmlParseContent(ctxt); |
| 2542 | if (!IS_CHAR(CUR)) { |
| 2543 | xmlParserError(ctxt, "Premature end of data in tag %.30s\n%.30s\n", |
| 2544 | openTag); |
| 2545 | |
| 2546 | /* |
| 2547 | * end of parsing of this node. |
| 2548 | */ |
| 2549 | nodePop(ctxt); |
| 2550 | |
| 2551 | return(NULL); |
| 2552 | } |
| 2553 | |
| 2554 | /* |
| 2555 | * parse the end of tag: '</' should be here. |
| 2556 | */ |
| 2557 | xmlParseEndTag(ctxt, &endNs, &endTag); |
| 2558 | |
| 2559 | /* |
| 2560 | * Check that the Name in the ETag is the same as in the STag. |
| 2561 | */ |
| 2562 | if (endNs != ret->ns) { |
| 2563 | xmlParserError(ctxt, |
| 2564 | "Start and End tags don't use the same namespace\n%.30s\n%.30s\n", |
| 2565 | openTag, endTag); |
| 2566 | } |
| 2567 | if (endTag == NULL ) { |
| 2568 | xmlParserError(ctxt, "The End tag has no name\n%.30s\n", openTag); |
| 2569 | } else if (xmlStrcmp(ret->name, endTag)) { |
| 2570 | xmlParserError(ctxt, |
| 2571 | "Start and End tags don't use the same name\n%.30s\n%.30s\n", |
| 2572 | openTag, endTag); |
| 2573 | } |
| 2574 | /* |
| 2575 | * SAX: End of Tag |
| 2576 | */ |
| 2577 | else if (ctxt->sax != NULL) |
| 2578 | ctxt->sax->endElement(ctxt, endTag); |
| 2579 | |
| 2580 | if (endTag != NULL) |
| 2581 | free(endTag); |
| 2582 | |
| 2583 | /* Capture end position and add node */ |
| 2584 | if ( ret != NULL && ctxt->record_info ) { |
| 2585 | node_info.end_pos = CUR_PTR - ctxt->input->base; |
| 2586 | node_info.end_line = ctxt->input->line; |
| 2587 | node_info.node = ret; |
| 2588 | xmlParserAddNodeInfo(ctxt, &node_info); |
| 2589 | } |
| 2590 | |
| 2591 | /* |
| 2592 | * end of parsing of this node. |
| 2593 | */ |
| 2594 | nodePop(ctxt); |
| 2595 | |
| 2596 | return(ret); |
| 2597 | } |
| 2598 | |
| 2599 | /* |
| 2600 | * xmlParseVersionNum: parse the XML version value. |
| 2601 | * |
| 2602 | * [26] VersionNum ::= ([a-zA-Z0-9_.:] | '-')+ |
| 2603 | */ |
| 2604 | CHAR *xmlParseVersionNum(xmlParserCtxtPtr ctxt) { |
| 2605 | const CHAR *q = CUR_PTR; |
| 2606 | CHAR *ret; |
| 2607 | |
| 2608 | while (IS_CHAR(CUR) && |
| 2609 | (((CUR >= 'a') && (CUR <= 'z')) || |
| 2610 | ((CUR >= 'A') && (CUR <= 'Z')) || |
| 2611 | ((CUR >= '0') && (CUR <= '9')) || |
| 2612 | (CUR == '_') || (CUR == '.') || |
| 2613 | (CUR == ':') || (CUR == '-'))) NEXT; |
| 2614 | ret = xmlStrndup(q, CUR_PTR - q); |
| 2615 | return(ret); |
| 2616 | } |
| 2617 | |
| 2618 | /* |
| 2619 | * xmlParseVersionInfo: parse the XML version. |
| 2620 | * |
| 2621 | * [24] VersionInfo ::= S 'version' Eq (' VersionNum ' | " VersionNum ") |
| 2622 | * |
| 2623 | * [25] Eq ::= S? '=' S? |
| 2624 | */ |
| 2625 | |
| 2626 | CHAR *xmlParseVersionInfo(xmlParserCtxtPtr ctxt) { |
| 2627 | CHAR *version = NULL; |
| 2628 | const CHAR *q; |
| 2629 | |
| 2630 | if ((CUR == 'v') && (NXT(1) == 'e') && |
| 2631 | (NXT(2) == 'r') && (NXT(3) == 's') && |
| 2632 | (NXT(4) == 'i') && (NXT(5) == 'o') && |
| 2633 | (NXT(6) == 'n')) { |
| 2634 | SKIP(7); |
| 2635 | SKIP_BLANKS; |
| 2636 | if (CUR != '=') { |
| 2637 | xmlParserError(ctxt, "xmlParseVersionInfo : expected '='\n"); |
| 2638 | return(NULL); |
| 2639 | } |
| 2640 | NEXT; |
| 2641 | SKIP_BLANKS; |
| 2642 | if (CUR == '"') { |
| 2643 | NEXT; |
| 2644 | q = CUR_PTR; |
| 2645 | version = xmlParseVersionNum(ctxt); |
| 2646 | if (CUR != '"') |
| 2647 | xmlParserError(ctxt, "String not closed\n%.50s\n", q); |
| 2648 | else |
| 2649 | NEXT; |
| 2650 | } else if (CUR == '\''){ |
| 2651 | NEXT; |
| 2652 | q = CUR_PTR; |
| 2653 | version = xmlParseVersionNum(ctxt); |
| 2654 | if (CUR != '\'') |
| 2655 | xmlParserError(ctxt, "String not closed\n%.50s\n", q); |
| 2656 | else |
| 2657 | NEXT; |
| 2658 | } else { |
| 2659 | xmlParserError(ctxt, "xmlParseVersionInfo : expected ' or \"\n"); |
| 2660 | } |
| 2661 | } |
| 2662 | return(version); |
| 2663 | } |
| 2664 | |
| 2665 | /* |
| 2666 | * xmlParseEncName: parse the XML encoding name |
| 2667 | * |
| 2668 | * [81] EncName ::= [A-Za-z] ([A-Za-z0-9._] | '-')* |
| 2669 | */ |
| 2670 | CHAR *xmlParseEncName(xmlParserCtxtPtr ctxt) { |
| 2671 | const CHAR *q = CUR_PTR; |
| 2672 | CHAR *ret = NULL; |
| 2673 | |
| 2674 | if (((CUR >= 'a') && (CUR <= 'z')) || |
| 2675 | ((CUR >= 'A') && (CUR <= 'Z'))) { |
| 2676 | NEXT; |
| 2677 | while (IS_CHAR(CUR) && |
| 2678 | (((CUR >= 'a') && (CUR <= 'z')) || |
| 2679 | ((CUR >= 'A') && (CUR <= 'Z')) || |
| 2680 | ((CUR >= '0') && (CUR <= '9')) || |
| 2681 | (CUR == '-'))) NEXT; |
| 2682 | ret = xmlStrndup(q, CUR_PTR - q); |
| 2683 | } else { |
| 2684 | xmlParserError(ctxt, "Invalid XML encoding name\n"); |
| 2685 | } |
| 2686 | return(ret); |
| 2687 | } |
| 2688 | |
| 2689 | /* |
| 2690 | * xmlParseEncodingDecl: parse the XML encoding declaration |
| 2691 | * |
| 2692 | * [80] EncodingDecl ::= S 'encoding' Eq ('"' EncName '"' | "'" EncName "'") |
| 2693 | */ |
| 2694 | |
| 2695 | CHAR *xmlParseEncodingDecl(xmlParserCtxtPtr ctxt) { |
| 2696 | CHAR *encoding = NULL; |
| 2697 | const CHAR *q; |
| 2698 | |
| 2699 | SKIP_BLANKS; |
| 2700 | if ((CUR == 'e') && (NXT(1) == 'n') && |
| 2701 | (NXT(2) == 'c') && (NXT(3) == 'o') && |
| 2702 | (NXT(4) == 'd') && (NXT(5) == 'i') && |
| 2703 | (NXT(6) == 'n') && (NXT(7) == 'g')) { |
| 2704 | SKIP(8); |
| 2705 | SKIP_BLANKS; |
| 2706 | if (CUR != '=') { |
| 2707 | xmlParserError(ctxt, "xmlParseEncodingDecl : expected '='\n"); |
| 2708 | return(NULL); |
| 2709 | } |
| 2710 | NEXT; |
| 2711 | SKIP_BLANKS; |
| 2712 | if (CUR == '"') { |
| 2713 | NEXT; |
| 2714 | q = CUR_PTR; |
| 2715 | encoding = xmlParseEncName(ctxt); |
| 2716 | if (CUR != '"') |
| 2717 | xmlParserError(ctxt, "String not closed\n%.50s\n", q); |
| 2718 | else |
| 2719 | NEXT; |
| 2720 | } else if (CUR == '\''){ |
| 2721 | NEXT; |
| 2722 | q = CUR_PTR; |
| 2723 | encoding = xmlParseEncName(ctxt); |
| 2724 | if (CUR != '\'') |
| 2725 | xmlParserError(ctxt, "String not closed\n%.50s\n", q); |
| 2726 | else |
| 2727 | NEXT; |
| 2728 | } else if (CUR == '"'){ |
| 2729 | xmlParserError(ctxt, "xmlParseEncodingDecl : expected ' or \"\n"); |
| 2730 | } |
| 2731 | } |
| 2732 | return(encoding); |
| 2733 | } |
| 2734 | |
| 2735 | /* |
| 2736 | * xmlParseSDDecl: parse the XML standalone declaration |
| 2737 | * |
| 2738 | * [32] SDDecl ::= S 'standalone' Eq |
| 2739 | * (("'" ('yes' | 'no') "'") | ('"' ('yes' | 'no')'"')) |
| 2740 | */ |
| 2741 | |
| 2742 | int xmlParseSDDecl(xmlParserCtxtPtr ctxt) { |
| 2743 | int standalone = -1; |
| 2744 | |
| 2745 | SKIP_BLANKS; |
| 2746 | if ((CUR == 's') && (NXT(1) == 't') && |
| 2747 | (NXT(2) == 'a') && (NXT(3) == 'n') && |
| 2748 | (NXT(4) == 'd') && (NXT(5) == 'a') && |
| 2749 | (NXT(6) == 'l') && (NXT(7) == 'o') && |
| 2750 | (NXT(8) == 'n') && (NXT(9) == 'e')) { |
| 2751 | SKIP(10); |
| 2752 | if (CUR != '=') { |
| 2753 | xmlParserError(ctxt, "XML standalone declaration : expected '='\n"); |
| 2754 | return(standalone); |
| 2755 | } |
| 2756 | NEXT; |
| 2757 | SKIP_BLANKS; |
| 2758 | if (CUR == '\''){ |
| 2759 | NEXT; |
| 2760 | if ((CUR == 'n') && (NXT(1) == 'o')) { |
| 2761 | standalone = 0; |
| 2762 | SKIP(2); |
| 2763 | } else if ((CUR == 'y') && (NXT(1) == 'e') && |
| 2764 | (NXT(2) == 's')) { |
| 2765 | standalone = 1; |
| 2766 | SKIP(3); |
| 2767 | } else { |
| 2768 | xmlParserError(ctxt, "standalone accepts only 'yes' or 'no'\n"); |
| 2769 | } |
| 2770 | if (CUR != '\'') |
| 2771 | xmlParserError(ctxt, "String not closed\n"); |
| 2772 | else |
| 2773 | NEXT; |
| 2774 | } else if (CUR == '"'){ |
| 2775 | NEXT; |
| 2776 | if ((CUR == 'n') && (NXT(1) == 'o')) { |
| 2777 | standalone = 0; |
| 2778 | SKIP(2); |
| 2779 | } else if ((CUR == 'y') && (NXT(1) == 'e') && |
| 2780 | (NXT(2) == 's')) { |
| 2781 | standalone = 1; |
| 2782 | SKIP(3); |
| 2783 | } else { |
| 2784 | xmlParserError(ctxt, "standalone accepts only 'yes' or 'no'\n"); |
| 2785 | } |
| 2786 | if (CUR != '"') |
| 2787 | xmlParserError(ctxt, "String not closed\n"); |
| 2788 | else |
| 2789 | NEXT; |
| 2790 | } else { |
| 2791 | xmlParserError(ctxt, "Standalone value not found\n"); |
| 2792 | } |
| 2793 | } |
| 2794 | return(standalone); |
| 2795 | } |
| 2796 | |
| 2797 | /* |
| 2798 | * xmlParseXMLDecl: parse an XML declaration header |
| 2799 | * |
| 2800 | * [23] XMLDecl ::= '<?xml' VersionInfo EncodingDecl? SDDecl? S? '?>' |
| 2801 | */ |
| 2802 | |
| 2803 | void xmlParseXMLDecl(xmlParserCtxtPtr ctxt) { |
| 2804 | CHAR *version; |
| 2805 | |
| 2806 | /* |
| 2807 | * We know that '<?xml' is here. |
| 2808 | */ |
| 2809 | SKIP(5); |
| 2810 | |
| 2811 | SKIP_BLANKS; |
| 2812 | |
| 2813 | /* |
| 2814 | * We should have the VersionInfo here. |
| 2815 | */ |
| 2816 | version = xmlParseVersionInfo(ctxt); |
| 2817 | if (version == NULL) |
| 2818 | version = xmlCharStrdup(XML_DEFAULT_VERSION); |
| 2819 | ctxt->doc = xmlNewDoc(version); |
| 2820 | free(version); |
| 2821 | |
| 2822 | /* |
| 2823 | * We may have the encoding declaration |
| 2824 | */ |
| 2825 | ctxt->doc->encoding = xmlParseEncodingDecl(ctxt); |
| 2826 | |
| 2827 | /* |
| 2828 | * We may have the standalone status. |
| 2829 | */ |
| 2830 | ctxt->doc->standalone = xmlParseSDDecl(ctxt); |
| 2831 | |
| 2832 | SKIP_BLANKS; |
| 2833 | if ((CUR == '?') && (NXT(1) == '>')) { |
| 2834 | SKIP(2); |
| 2835 | } else if (CUR == '>') { |
| 2836 | /* Deprecated old WD ... */ |
| 2837 | xmlParserError(ctxt, "XML declaration must end-up with '?>'\n"); |
| 2838 | NEXT; |
| 2839 | } else { |
| 2840 | xmlParserError(ctxt, "parsing XML declaration: '?>' expected\n"); |
| 2841 | MOVETO_ENDTAG(CUR_PTR); |
| 2842 | NEXT; |
| 2843 | } |
| 2844 | } |
| 2845 | |
| 2846 | /* |
| 2847 | * xmlParseMisc: parse an XML Misc* optionnal field. |
| 2848 | * Misc* |
| 2849 | * |
| 2850 | * [27] Misc ::= Comment | PI | S |
| 2851 | */ |
| 2852 | |
| 2853 | void xmlParseMisc(xmlParserCtxtPtr ctxt) { |
| 2854 | while (((CUR == '<') && (NXT(1) == '?')) || |
| 2855 | ((CUR == '<') && (NXT(1) == '!') && |
| 2856 | (NXT(2) == '-') && (NXT(3) == '-')) || |
| 2857 | IS_BLANK(CUR)) { |
| 2858 | if ((CUR == '<') && (NXT(1) == '?')) { |
| 2859 | xmlParsePI(ctxt); |
| 2860 | } else if (IS_BLANK(CUR)) { |
| 2861 | NEXT; |
| 2862 | } else |
| 2863 | xmlParseComment(ctxt, 0); |
| 2864 | } |
| 2865 | } |
| 2866 | |
| 2867 | /* |
| 2868 | * xmlParseDocument : parse an XML document and build a tree. |
| 2869 | * |
| 2870 | * [1] document ::= prolog element Misc* |
| 2871 | * |
| 2872 | * [22] prolog ::= XMLDecl? Misc* (doctypedecl Misc*)? |
| 2873 | */ |
| 2874 | |
| 2875 | int xmlParseDocument(xmlParserCtxtPtr ctxt) { |
| 2876 | xmlDefaultSAXHandlerInit(); |
| 2877 | |
| 2878 | /* |
| 2879 | * SAX: beginning of the document processing. |
| 2880 | */ |
| 2881 | if (ctxt->sax) |
| 2882 | ctxt->sax->setDocumentLocator(ctxt, &xmlDefaultSAXLocator); |
| 2883 | if (ctxt->sax) |
| 2884 | ctxt->sax->startDocument(ctxt); |
| 2885 | |
| 2886 | /* |
| 2887 | * We should check for encoding here and plug-in some |
| 2888 | * conversion code TODO !!!! |
| 2889 | */ |
| 2890 | |
| 2891 | /* |
| 2892 | * Wipe out everything which is before the first '<' |
| 2893 | */ |
| 2894 | SKIP_BLANKS; |
| 2895 | |
| 2896 | /* |
| 2897 | * Check for the XMLDecl in the Prolog. |
| 2898 | */ |
| 2899 | if ((CUR == '<') && (NXT(1) == '?') && |
| 2900 | (NXT(2) == 'x') && (NXT(3) == 'm') && |
| 2901 | (NXT(4) == 'l')) { |
| 2902 | xmlParseXMLDecl(ctxt); |
| 2903 | /* SKIP_EOL(cur); */ |
| 2904 | SKIP_BLANKS; |
| 2905 | } else if ((CUR == '<') && (NXT(1) == '?') && |
| 2906 | (NXT(2) == 'X') && (NXT(3) == 'M') && |
| 2907 | (NXT(4) == 'L')) { |
| 2908 | /* |
| 2909 | * The first drafts were using <?XML and the final W3C REC |
| 2910 | * now use <?xml ... |
| 2911 | */ |
| 2912 | xmlParseXMLDecl(ctxt); |
| 2913 | /* SKIP_EOL(cur); */ |
| 2914 | SKIP_BLANKS; |
| 2915 | } else { |
| 2916 | CHAR *version; |
| 2917 | |
| 2918 | version = xmlCharStrdup(XML_DEFAULT_VERSION); |
| 2919 | ctxt->doc = xmlNewDoc(version); |
| 2920 | free(version); |
| 2921 | } |
| 2922 | |
| 2923 | /* |
| 2924 | * The Misc part of the Prolog |
| 2925 | */ |
| 2926 | xmlParseMisc(ctxt); |
| 2927 | |
| 2928 | /* |
| 2929 | * Then possibly doc type declaration(s) and more Misc |
| 2930 | * (doctypedecl Misc*)? |
| 2931 | */ |
| 2932 | if ((CUR == '<') && (NXT(1) == '!') && |
| 2933 | (NXT(2) == 'D') && (NXT(3) == 'O') && |
| 2934 | (NXT(4) == 'C') && (NXT(5) == 'T') && |
| 2935 | (NXT(6) == 'Y') && (NXT(7) == 'P') && |
| 2936 | (NXT(8) == 'E')) { |
| 2937 | xmlParseDocTypeDecl(ctxt); |
| 2938 | xmlParseMisc(ctxt); |
| 2939 | } |
| 2940 | |
| 2941 | /* |
| 2942 | * Time to start parsing the tree itself |
| 2943 | */ |
| 2944 | ctxt->doc->root = xmlParseElement(ctxt); |
| 2945 | |
| 2946 | /* |
| 2947 | * The Misc part at the end |
| 2948 | */ |
| 2949 | xmlParseMisc(ctxt); |
| 2950 | |
| 2951 | /* |
| 2952 | * SAX: end of the document processing. |
| 2953 | */ |
| 2954 | if (ctxt->sax) |
| 2955 | ctxt->sax->endDocument(ctxt); |
| 2956 | return(0); |
| 2957 | } |
| 2958 | |
| 2959 | /* |
| 2960 | * xmlParseDoc : parse an XML in-memory document and build a tree. |
| 2961 | */ |
| 2962 | |
| 2963 | xmlDocPtr xmlParseDoc(CHAR *cur) { |
| 2964 | xmlDocPtr ret; |
| 2965 | xmlParserCtxtPtr ctxt; |
| 2966 | xmlParserInputPtr input; |
| 2967 | |
| 2968 | if (cur == NULL) return(NULL); |
| 2969 | |
| 2970 | ctxt = (xmlParserCtxtPtr) malloc(sizeof(xmlParserCtxt)); |
| 2971 | if (ctxt == NULL) { |
| 2972 | perror("malloc"); |
| 2973 | return(NULL); |
| 2974 | } |
| 2975 | xmlInitParserCtxt(ctxt); |
| 2976 | input = (xmlParserInputPtr) malloc(sizeof(xmlParserInput)); |
| 2977 | if (input == NULL) { |
| 2978 | perror("malloc"); |
| 2979 | free(ctxt); |
| 2980 | return(NULL); |
| 2981 | } |
| 2982 | |
| 2983 | input->filename = NULL; |
| 2984 | input->line = 1; |
| 2985 | input->col = 1; |
| 2986 | input->base = cur; |
| 2987 | input->cur = cur; |
| 2988 | |
| 2989 | inputPush(ctxt, input); |
| 2990 | |
| 2991 | |
| 2992 | xmlParseDocument(ctxt); |
| 2993 | ret = ctxt->doc; |
| 2994 | free(ctxt); |
| 2995 | |
| 2996 | return(ret); |
| 2997 | } |
| 2998 | |
| 2999 | /* |
| 3000 | * xmlParseFile : parse an XML file and build a tree. |
| 3001 | */ |
| 3002 | |
| 3003 | xmlDocPtr xmlParseFile(const char *filename) { |
| 3004 | xmlDocPtr ret; |
| 3005 | #ifdef HAVE_ZLIB_H |
| 3006 | gzFile input; |
| 3007 | #else |
| 3008 | int input; |
| 3009 | #endif |
| 3010 | int res; |
| 3011 | struct stat buf; |
| 3012 | char *buffer; |
| 3013 | xmlParserCtxtPtr ctxt; |
| 3014 | xmlParserInputPtr inputStream; |
| 3015 | |
| 3016 | res = stat(filename, &buf); |
| 3017 | if (res < 0) return(NULL); |
| 3018 | |
| 3019 | #ifdef HAVE_ZLIB_H |
| 3020 | retry_bigger: |
| 3021 | buffer = malloc((buf.st_size * 20) + 100); |
| 3022 | #else |
| 3023 | buffer = malloc(buf.st_size + 100); |
| 3024 | #endif |
| 3025 | if (buffer == NULL) { |
| 3026 | perror("malloc"); |
| 3027 | return(NULL); |
| 3028 | } |
| 3029 | |
| 3030 | memset(buffer, 0, sizeof(buffer)); |
| 3031 | #ifdef HAVE_ZLIB_H |
| 3032 | input = gzopen (filename, "r"); |
| 3033 | if (input == NULL) { |
| 3034 | fprintf (stderr, "Cannot read file %s :\n", filename); |
| 3035 | perror ("gzopen failed"); |
| 3036 | return(NULL); |
| 3037 | } |
| 3038 | #else |
| 3039 | input = open (filename, O_RDONLY); |
| 3040 | if (input < 0) { |
| 3041 | fprintf (stderr, "Cannot read file %s :\n", filename); |
| 3042 | perror ("open failed"); |
| 3043 | return(NULL); |
| 3044 | } |
| 3045 | #endif |
| 3046 | #ifdef HAVE_ZLIB_H |
| 3047 | res = gzread(input, buffer, 20 * buf.st_size); |
| 3048 | #else |
| 3049 | res = read(input, buffer, buf.st_size); |
| 3050 | #endif |
| 3051 | if (res < 0) { |
| 3052 | fprintf (stderr, "Cannot read file %s :\n", filename); |
| 3053 | #ifdef HAVE_ZLIB_H |
| 3054 | perror ("gzread failed"); |
| 3055 | #else |
| 3056 | perror ("read failed"); |
| 3057 | #endif |
| 3058 | return(NULL); |
| 3059 | } |
| 3060 | #ifdef HAVE_ZLIB_H |
| 3061 | gzclose(input); |
Daniel Veillard | 70120ff | 1998-09-22 00:24:21 +0000 | [diff] [blame] | 3062 | if (res >= 20 * buf.st_size + 20) { |
Daniel Veillard | 260a68f | 1998-08-13 03:39:55 +0000 | [diff] [blame] | 3063 | free(buffer); |
| 3064 | buf.st_size *= 2; |
| 3065 | goto retry_bigger; |
| 3066 | } |
| 3067 | buf.st_size = res; |
| 3068 | #else |
| 3069 | close(input); |
| 3070 | #endif |
| 3071 | |
| 3072 | buffer[buf.st_size] = '\0'; |
| 3073 | |
| 3074 | ctxt = (xmlParserCtxtPtr) malloc(sizeof(xmlParserCtxt)); |
| 3075 | if (ctxt == NULL) { |
| 3076 | perror("malloc"); |
| 3077 | return(NULL); |
| 3078 | } |
| 3079 | xmlInitParserCtxt(ctxt); |
| 3080 | inputStream = (xmlParserInputPtr) malloc(sizeof(xmlParserInput)); |
| 3081 | if (inputStream == NULL) { |
| 3082 | perror("malloc"); |
| 3083 | free(ctxt); |
| 3084 | return(NULL); |
| 3085 | } |
| 3086 | |
| 3087 | inputStream->filename = strdup(filename); |
| 3088 | inputStream->line = 1; |
| 3089 | inputStream->col = 1; |
| 3090 | |
| 3091 | /* |
| 3092 | * TODO : plug some encoding conversion routines here. !!! |
| 3093 | */ |
| 3094 | inputStream->base = buffer; |
| 3095 | inputStream->cur = buffer; |
| 3096 | |
| 3097 | inputPush(ctxt, inputStream); |
| 3098 | |
| 3099 | xmlParseDocument(ctxt); |
| 3100 | |
| 3101 | ret = ctxt->doc; |
| 3102 | free(buffer); |
| 3103 | free(ctxt); |
| 3104 | |
| 3105 | return(ret); |
| 3106 | } |
| 3107 | |
| 3108 | |
| 3109 | /* |
| 3110 | * xmlParseMemory : parse an XML memory block and build a tree. |
| 3111 | */ |
| 3112 | xmlDocPtr xmlParseMemory(char *buffer, int size) { |
| 3113 | xmlDocPtr ret; |
| 3114 | xmlParserCtxtPtr ctxt; |
| 3115 | xmlParserInputPtr input; |
| 3116 | |
| 3117 | buffer[size - 1] = '\0'; |
| 3118 | |
| 3119 | ctxt = (xmlParserCtxtPtr) malloc(sizeof(xmlParserCtxt)); |
| 3120 | if (ctxt == NULL) { |
| 3121 | perror("malloc"); |
| 3122 | return(NULL); |
| 3123 | } |
| 3124 | xmlInitParserCtxt(ctxt); |
| 3125 | input = (xmlParserInputPtr) malloc(sizeof(xmlParserInput)); |
| 3126 | if (input == NULL) { |
| 3127 | perror("malloc"); |
| 3128 | free(ctxt); |
| 3129 | return(NULL); |
| 3130 | } |
| 3131 | |
| 3132 | input->filename = NULL; |
| 3133 | input->line = 1; |
| 3134 | input->col = 1; |
| 3135 | |
| 3136 | /* |
| 3137 | * TODO : plug some encoding conversion routines here. !!! |
| 3138 | */ |
| 3139 | input->base = buffer; |
| 3140 | input->cur = buffer; |
| 3141 | |
| 3142 | inputPush(ctxt, input); |
| 3143 | |
| 3144 | xmlParseDocument(ctxt); |
| 3145 | |
| 3146 | ret = ctxt->doc; |
| 3147 | free(ctxt); |
| 3148 | |
| 3149 | return(ret); |
| 3150 | } |
| 3151 | |
| 3152 | |
| 3153 | /* Initialize parser context */ |
| 3154 | void xmlInitParserCtxt(xmlParserCtxtPtr ctxt) |
| 3155 | { |
| 3156 | /* Allocate the Input stack */ |
| 3157 | ctxt->inputTab = (xmlParserInputPtr *) malloc(5 * sizeof(xmlParserInputPtr)); |
| 3158 | ctxt->inputNr = 0; |
| 3159 | ctxt->inputMax = 5; |
| 3160 | ctxt->input = NULL; |
| 3161 | |
| 3162 | /* Allocate the Node stack */ |
| 3163 | ctxt->nodeTab = (xmlNodePtr *) malloc(10 * sizeof(xmlNodePtr)); |
| 3164 | ctxt->nodeNr = 0; |
| 3165 | ctxt->nodeMax = 10; |
| 3166 | ctxt->node = NULL; |
| 3167 | |
| 3168 | ctxt->sax = &xmlDefaultSAXHandler; |
| 3169 | ctxt->doc = NULL; |
| 3170 | ctxt->record_info = 0; |
| 3171 | xmlInitNodeInfoSeq(&ctxt->node_seq); |
| 3172 | } |
| 3173 | |
| 3174 | |
| 3175 | /* |
| 3176 | * Clear (release owned resources) and reinitialize context |
| 3177 | */ |
| 3178 | void xmlClearParserCtxt(xmlParserCtxtPtr ctxt) |
| 3179 | { |
| 3180 | xmlClearNodeInfoSeq(&ctxt->node_seq); |
| 3181 | xmlInitParserCtxt(ctxt); |
| 3182 | } |
| 3183 | |
| 3184 | |
| 3185 | /* |
| 3186 | * Setup the parser context to parse a new buffer; Clears any prior |
| 3187 | * contents from the parser context. The buffer parameter must not be |
| 3188 | * NULL, but the filename parameter can be |
| 3189 | */ |
| 3190 | void xmlSetupParserForBuffer(xmlParserCtxtPtr ctxt, const CHAR* buffer, |
| 3191 | const char* filename) |
| 3192 | { |
| 3193 | xmlParserInputPtr input; |
| 3194 | |
| 3195 | input = (xmlParserInputPtr) malloc(sizeof(xmlParserInput)); |
| 3196 | if (input == NULL) { |
| 3197 | perror("malloc"); |
| 3198 | free(ctxt); |
| 3199 | exit(1); |
| 3200 | } |
| 3201 | |
| 3202 | xmlClearParserCtxt(ctxt); |
| 3203 | if (input->filename != NULL) |
| 3204 | input->filename = strdup(filename); |
| 3205 | else |
| 3206 | input->filename = NULL; |
| 3207 | input->line = 1; |
| 3208 | input->col = 1; |
| 3209 | input->base = buffer; |
| 3210 | input->cur = buffer; |
| 3211 | |
| 3212 | inputPush(ctxt, input); |
| 3213 | } |
| 3214 | |
| 3215 | |
| 3216 | /* |
| 3217 | * xmlParserFindNodeInfo : Find the parser node info struct for a given node |
| 3218 | */ |
| 3219 | const xmlParserNodeInfo* xmlParserFindNodeInfo(const xmlParserCtxt* ctx, |
| 3220 | const xmlNode* node) |
| 3221 | { |
| 3222 | unsigned long pos; |
| 3223 | |
| 3224 | /* Find position where node should be at */ |
| 3225 | pos = xmlParserFindNodeInfoIndex(&ctx->node_seq, node); |
| 3226 | if ( ctx->node_seq.buffer[pos].node == node ) |
| 3227 | return &ctx->node_seq.buffer[pos]; |
| 3228 | else |
| 3229 | return NULL; |
| 3230 | } |
| 3231 | |
| 3232 | |
| 3233 | /* |
| 3234 | * xmlInitNodeInfoSeq -- Initialize (set to initial state) node info sequence |
| 3235 | */ |
| 3236 | void xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq) |
| 3237 | { |
| 3238 | seq->length = 0; |
| 3239 | seq->maximum = 0; |
| 3240 | seq->buffer = NULL; |
| 3241 | } |
| 3242 | |
| 3243 | /* |
| 3244 | * xmlClearNodeInfoSeq -- Clear (release memory and reinitialize) node |
| 3245 | * info sequence |
| 3246 | */ |
| 3247 | void xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq) |
| 3248 | { |
| 3249 | if ( seq->buffer != NULL ) |
| 3250 | free(seq->buffer); |
| 3251 | xmlInitNodeInfoSeq(seq); |
| 3252 | } |
| 3253 | |
| 3254 | |
| 3255 | /* |
| 3256 | * xmlParserFindNodeInfoIndex : Find the index that the info record for |
| 3257 | * the given node is or should be at in a sorted sequence |
| 3258 | */ |
| 3259 | unsigned long xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeq* seq, |
| 3260 | const xmlNode* node) |
| 3261 | { |
| 3262 | unsigned long upper, lower, middle; |
| 3263 | int found = 0; |
| 3264 | |
| 3265 | /* Do a binary search for the key */ |
| 3266 | lower = 1; |
| 3267 | upper = seq->length; |
| 3268 | middle = 0; |
| 3269 | while ( lower <= upper && !found) { |
| 3270 | middle = lower + (upper - lower) / 2; |
| 3271 | if ( node == seq->buffer[middle - 1].node ) |
| 3272 | found = 1; |
| 3273 | else if ( node < seq->buffer[middle - 1].node ) |
| 3274 | upper = middle - 1; |
| 3275 | else |
| 3276 | lower = middle + 1; |
| 3277 | } |
| 3278 | |
| 3279 | /* Return position */ |
| 3280 | if ( middle == 0 || seq->buffer[middle - 1].node < node ) |
| 3281 | return middle; |
| 3282 | else |
| 3283 | return middle - 1; |
| 3284 | } |
| 3285 | |
| 3286 | |
| 3287 | /* |
| 3288 | * xmlParserAddNodeInfo : Insert node info record into sorted sequence |
| 3289 | */ |
| 3290 | void xmlParserAddNodeInfo(xmlParserCtxtPtr ctx, |
| 3291 | const xmlParserNodeInfo* info) |
| 3292 | { |
| 3293 | unsigned long pos; |
| 3294 | static unsigned int block_size = 5; |
| 3295 | |
| 3296 | /* Find pos and check to see if node is already in the sequence */ |
| 3297 | pos = xmlParserFindNodeInfoIndex(&ctx->node_seq, info->node); |
| 3298 | if ( pos < ctx->node_seq.length |
| 3299 | && ctx->node_seq.buffer[pos].node == info->node ) { |
| 3300 | ctx->node_seq.buffer[pos] = *info; |
| 3301 | } |
| 3302 | |
| 3303 | /* Otherwise, we need to add new node to buffer */ |
| 3304 | else { |
| 3305 | /* Expand buffer by 5 if needed */ |
| 3306 | if ( ctx->node_seq.length + 1 > ctx->node_seq.maximum ) { |
| 3307 | xmlParserNodeInfo* tmp_buffer; |
| 3308 | unsigned int byte_size = (sizeof(*ctx->node_seq.buffer) |
| 3309 | *(ctx->node_seq.maximum + block_size)); |
| 3310 | |
| 3311 | if ( ctx->node_seq.buffer == NULL ) |
| 3312 | tmp_buffer = (xmlParserNodeInfo*)malloc(byte_size); |
| 3313 | else |
| 3314 | tmp_buffer = (xmlParserNodeInfo*)realloc(ctx->node_seq.buffer, byte_size); |
| 3315 | |
| 3316 | if ( tmp_buffer == NULL ) { |
| 3317 | xmlParserError(ctx, "Out of memory"); |
| 3318 | return; |
| 3319 | } |
| 3320 | ctx->node_seq.buffer = tmp_buffer; |
| 3321 | ctx->node_seq.maximum += block_size; |
| 3322 | } |
| 3323 | |
| 3324 | /* If position is not at end, move elements out of the way */ |
| 3325 | if ( pos != ctx->node_seq.length ) { |
| 3326 | unsigned long i; |
| 3327 | |
| 3328 | for ( i = ctx->node_seq.length; i > pos; i-- ) |
| 3329 | ctx->node_seq.buffer[i] = ctx->node_seq.buffer[i - 1]; |
| 3330 | } |
| 3331 | |
| 3332 | /* Copy element and increase length */ |
| 3333 | ctx->node_seq.buffer[pos] = *info; |
| 3334 | ctx->node_seq.length++; |
| 3335 | } |
| 3336 | } |