Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1 | /* |
| 2 | * HTMLparser.c : an HTML 4.0 non-verifying parser |
| 3 | * |
| 4 | * See Copyright for the status of this software. |
| 5 | * |
| 6 | * Daniel.Veillard@w3.org |
| 7 | */ |
| 8 | |
Bjorn Reese | 70a9da5 | 2001-04-21 16:57:29 +0000 | [diff] [blame] | 9 | #include "libxml.h" |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 10 | #ifdef LIBXML_HTML_ENABLED |
Bjorn Reese | 70a9da5 | 2001-04-21 16:57:29 +0000 | [diff] [blame] | 11 | |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 12 | #include <string.h> |
| 13 | #ifdef HAVE_CTYPE_H |
| 14 | #include <ctype.h> |
| 15 | #endif |
| 16 | #ifdef HAVE_STDLIB_H |
| 17 | #include <stdlib.h> |
| 18 | #endif |
| 19 | #ifdef HAVE_SYS_STAT_H |
| 20 | #include <sys/stat.h> |
| 21 | #endif |
| 22 | #ifdef HAVE_FCNTL_H |
| 23 | #include <fcntl.h> |
| 24 | #endif |
| 25 | #ifdef HAVE_UNISTD_H |
| 26 | #include <unistd.h> |
| 27 | #endif |
| 28 | #ifdef HAVE_ZLIB_H |
| 29 | #include <zlib.h> |
| 30 | #endif |
| 31 | |
| 32 | #include <libxml/xmlmemory.h> |
| 33 | #include <libxml/tree.h> |
| 34 | #include <libxml/parser.h> |
| 35 | #include <libxml/parserInternals.h> |
| 36 | #include <libxml/xmlerror.h> |
| 37 | #include <libxml/HTMLparser.h> |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 38 | #include <libxml/HTMLtree.h> |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 39 | #include <libxml/entities.h> |
| 40 | #include <libxml/encoding.h> |
| 41 | #include <libxml/valid.h> |
| 42 | #include <libxml/xmlIO.h> |
| 43 | |
| 44 | #define HTML_MAX_NAMELEN 1000 |
| 45 | #define HTML_PARSER_BIG_BUFFER_SIZE 1000 |
| 46 | #define HTML_PARSER_BUFFER_SIZE 100 |
| 47 | |
| 48 | /* #define DEBUG */ |
| 49 | /* #define DEBUG_PUSH */ |
| 50 | |
| 51 | int htmlOmittedDefaultValue = 1; |
| 52 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 53 | xmlChar * htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len, |
| 54 | xmlChar end, xmlChar end2, xmlChar end3); |
| 55 | |
| 56 | /************************************************************************ |
| 57 | * * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 58 | * Parser stacks related functions and macros * |
| 59 | * * |
| 60 | ************************************************************************/ |
| 61 | |
| 62 | /* |
| 63 | * Generic function for accessing stacks in the Parser Context |
| 64 | */ |
| 65 | |
| 66 | #define PUSH_AND_POP(scope, type, name) \ |
| 67 | scope int html##name##Push(htmlParserCtxtPtr ctxt, type value) { \ |
| 68 | if (ctxt->name##Nr >= ctxt->name##Max) { \ |
| 69 | ctxt->name##Max *= 2; \ |
| 70 | ctxt->name##Tab = (type *) xmlRealloc(ctxt->name##Tab, \ |
| 71 | ctxt->name##Max * sizeof(ctxt->name##Tab[0])); \ |
| 72 | if (ctxt->name##Tab == NULL) { \ |
| 73 | xmlGenericError(xmlGenericErrorContext, \ |
| 74 | "realloc failed !\n"); \ |
| 75 | return(0); \ |
| 76 | } \ |
| 77 | } \ |
| 78 | ctxt->name##Tab[ctxt->name##Nr] = value; \ |
| 79 | ctxt->name = value; \ |
| 80 | return(ctxt->name##Nr++); \ |
| 81 | } \ |
| 82 | scope type html##name##Pop(htmlParserCtxtPtr ctxt) { \ |
| 83 | type ret; \ |
| 84 | if (ctxt->name##Nr < 0) return(0); \ |
| 85 | ctxt->name##Nr--; \ |
| 86 | if (ctxt->name##Nr < 0) return(0); \ |
| 87 | if (ctxt->name##Nr > 0) \ |
| 88 | ctxt->name = ctxt->name##Tab[ctxt->name##Nr - 1]; \ |
| 89 | else \ |
| 90 | ctxt->name = NULL; \ |
| 91 | ret = ctxt->name##Tab[ctxt->name##Nr]; \ |
| 92 | ctxt->name##Tab[ctxt->name##Nr] = 0; \ |
| 93 | return(ret); \ |
| 94 | } \ |
| 95 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 96 | /* PUSH_AND_POP(static, xmlNodePtr, node) */ |
| 97 | PUSH_AND_POP(static, xmlChar*, name) |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 98 | |
| 99 | /* |
| 100 | * Macros for accessing the content. Those should be used only by the parser, |
| 101 | * and not exported. |
| 102 | * |
| 103 | * Dirty macros, i.e. one need to make assumption on the context to use them |
| 104 | * |
| 105 | * CUR_PTR return the current pointer to the xmlChar to be parsed. |
| 106 | * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled |
| 107 | * in ISO-Latin or UTF-8, and the current 16 bit value if compiled |
| 108 | * in UNICODE mode. This should be used internally by the parser |
| 109 | * only to compare to ASCII values otherwise it would break when |
| 110 | * running with UTF-8 encoding. |
| 111 | * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only |
| 112 | * to compare on ASCII based substring. |
| 113 | * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR |
| 114 | * it should be used only to compare on ASCII based substring. |
| 115 | * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined |
| 116 | * strings within the parser. |
| 117 | * |
| 118 | * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding |
| 119 | * |
| 120 | * CURRENT Returns the current char value, with the full decoding of |
| 121 | * UTF-8 if we are using this mode. It returns an int. |
| 122 | * NEXT Skip to the next character, this does the proper decoding |
| 123 | * in UTF-8 mode. It also pop-up unfinished entities on the fly. |
| 124 | * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly |
| 125 | */ |
| 126 | |
| 127 | #define UPPER (toupper(*ctxt->input->cur)) |
| 128 | |
| 129 | #define SKIP(val) ctxt->nbChars += (val),ctxt->input->cur += (val) |
| 130 | |
| 131 | #define NXT(val) ctxt->input->cur[(val)] |
| 132 | |
| 133 | #define UPP(val) (toupper(ctxt->input->cur[(val)])) |
| 134 | |
| 135 | #define CUR_PTR ctxt->input->cur |
| 136 | |
| 137 | #define SHRINK xmlParserInputShrink(ctxt->input) |
| 138 | |
| 139 | #define GROW xmlParserInputGrow(ctxt->input, INPUT_CHUNK) |
| 140 | |
| 141 | #define CURRENT ((int) (*ctxt->input->cur)) |
| 142 | |
| 143 | #define SKIP_BLANKS htmlSkipBlankChars(ctxt) |
| 144 | |
| 145 | /* Inported from XML */ |
| 146 | |
| 147 | /* #define CUR (ctxt->token ? ctxt->token : (int) (*ctxt->input->cur)) */ |
| 148 | #define CUR ((int) (*ctxt->input->cur)) |
| 149 | #define NEXT xmlNextChar(ctxt),ctxt->nbChars++ |
| 150 | |
| 151 | #define RAW (ctxt->token ? -1 : (*ctxt->input->cur)) |
| 152 | #define NXT(val) ctxt->input->cur[(val)] |
| 153 | #define CUR_PTR ctxt->input->cur |
| 154 | |
| 155 | |
| 156 | #define NEXTL(l) do { \ |
| 157 | if (*(ctxt->input->cur) == '\n') { \ |
| 158 | ctxt->input->line++; ctxt->input->col = 1; \ |
| 159 | } else ctxt->input->col++; \ |
| 160 | ctxt->token = 0; ctxt->input->cur += l; ctxt->nbChars++; \ |
| 161 | } while (0) |
| 162 | |
| 163 | /************ |
| 164 | \ |
| 165 | if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \ |
| 166 | if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt); |
| 167 | ************/ |
| 168 | |
| 169 | #define CUR_CHAR(l) htmlCurrentChar(ctxt, &l) |
| 170 | #define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l) |
| 171 | |
| 172 | #define COPY_BUF(l,b,i,v) \ |
| 173 | if (l == 1) b[i++] = (xmlChar) v; \ |
| 174 | else i += xmlCopyChar(l,&b[i],v) |
| 175 | |
| 176 | /** |
| 177 | * htmlCurrentChar: |
| 178 | * @ctxt: the HTML parser context |
| 179 | * @len: pointer to the length of the char read |
| 180 | * |
| 181 | * The current char value, if using UTF-8 this may actaully span multiple |
| 182 | * bytes in the input buffer. Implement the end of line normalization: |
| 183 | * 2.11 End-of-Line Handling |
| 184 | * If the encoding is unspecified, in the case we find an ISO-Latin-1 |
| 185 | * char, then the encoding converter is plugged in automatically. |
| 186 | * |
| 187 | * Returns the current char value and its lenght |
| 188 | */ |
| 189 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 190 | static int |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 191 | htmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) { |
| 192 | if (ctxt->instate == XML_PARSER_EOF) |
| 193 | return(0); |
| 194 | |
| 195 | if (ctxt->token != 0) { |
| 196 | *len = 0; |
| 197 | return(ctxt->token); |
| 198 | } |
| 199 | if (ctxt->charset == XML_CHAR_ENCODING_UTF8) { |
| 200 | /* |
| 201 | * We are supposed to handle UTF8, check it's valid |
| 202 | * From rfc2044: encoding of the Unicode values on UTF-8: |
| 203 | * |
| 204 | * UCS-4 range (hex.) UTF-8 octet sequence (binary) |
| 205 | * 0000 0000-0000 007F 0xxxxxxx |
| 206 | * 0000 0080-0000 07FF 110xxxxx 10xxxxxx |
| 207 | * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx |
| 208 | * |
| 209 | * Check for the 0x110000 limit too |
| 210 | */ |
| 211 | const unsigned char *cur = ctxt->input->cur; |
| 212 | unsigned char c; |
| 213 | unsigned int val; |
| 214 | |
| 215 | c = *cur; |
| 216 | if (c & 0x80) { |
| 217 | if (cur[1] == 0) |
| 218 | xmlParserInputGrow(ctxt->input, INPUT_CHUNK); |
| 219 | if ((cur[1] & 0xc0) != 0x80) |
| 220 | goto encoding_error; |
| 221 | if ((c & 0xe0) == 0xe0) { |
| 222 | |
| 223 | if (cur[2] == 0) |
| 224 | xmlParserInputGrow(ctxt->input, INPUT_CHUNK); |
| 225 | if ((cur[2] & 0xc0) != 0x80) |
| 226 | goto encoding_error; |
| 227 | if ((c & 0xf0) == 0xf0) { |
| 228 | if (cur[3] == 0) |
| 229 | xmlParserInputGrow(ctxt->input, INPUT_CHUNK); |
| 230 | if (((c & 0xf8) != 0xf0) || |
| 231 | ((cur[3] & 0xc0) != 0x80)) |
| 232 | goto encoding_error; |
| 233 | /* 4-byte code */ |
| 234 | *len = 4; |
| 235 | val = (cur[0] & 0x7) << 18; |
| 236 | val |= (cur[1] & 0x3f) << 12; |
| 237 | val |= (cur[2] & 0x3f) << 6; |
| 238 | val |= cur[3] & 0x3f; |
| 239 | } else { |
| 240 | /* 3-byte code */ |
| 241 | *len = 3; |
| 242 | val = (cur[0] & 0xf) << 12; |
| 243 | val |= (cur[1] & 0x3f) << 6; |
| 244 | val |= cur[2] & 0x3f; |
| 245 | } |
| 246 | } else { |
| 247 | /* 2-byte code */ |
| 248 | *len = 2; |
| 249 | val = (cur[0] & 0x1f) << 6; |
| 250 | val |= cur[1] & 0x3f; |
| 251 | } |
| 252 | if (!IS_CHAR(val)) { |
| 253 | ctxt->errNo = XML_ERR_INVALID_ENCODING; |
| 254 | if ((ctxt->sax != NULL) && |
| 255 | (ctxt->sax->error != NULL)) |
| 256 | ctxt->sax->error(ctxt->userData, |
| 257 | "Char 0x%X out of allowed range\n", val); |
| 258 | ctxt->wellFormed = 0; |
| 259 | ctxt->disableSAX = 1; |
| 260 | } |
| 261 | return(val); |
| 262 | } else { |
| 263 | /* 1-byte code */ |
| 264 | *len = 1; |
| 265 | return((int) *ctxt->input->cur); |
| 266 | } |
| 267 | } |
| 268 | /* |
| 269 | * Assume it's a fixed lenght encoding (1) with |
| 270 | * a compatibke encoding for the ASCII set, since |
| 271 | * XML constructs only use < 128 chars |
| 272 | */ |
| 273 | *len = 1; |
| 274 | if ((int) *ctxt->input->cur < 0x80) |
| 275 | return((int) *ctxt->input->cur); |
| 276 | |
| 277 | /* |
| 278 | * Humm this is bad, do an automatic flow conversion |
| 279 | */ |
| 280 | xmlSwitchEncoding(ctxt, XML_CHAR_ENCODING_8859_1); |
| 281 | ctxt->charset = XML_CHAR_ENCODING_UTF8; |
| 282 | return(xmlCurrentChar(ctxt, len)); |
| 283 | |
| 284 | encoding_error: |
| 285 | /* |
| 286 | * If we detect an UTF8 error that probably mean that the |
| 287 | * input encoding didn't get properly advertized in the |
| 288 | * declaration header. Report the error and switch the encoding |
| 289 | * to ISO-Latin-1 (if you don't like this policy, just declare the |
| 290 | * encoding !) |
| 291 | */ |
| 292 | ctxt->errNo = XML_ERR_INVALID_ENCODING; |
| 293 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) { |
| 294 | ctxt->sax->error(ctxt->userData, |
| 295 | "Input is not proper UTF-8, indicate encoding !\n"); |
| 296 | ctxt->sax->error(ctxt->userData, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n", |
| 297 | ctxt->input->cur[0], ctxt->input->cur[1], |
| 298 | ctxt->input->cur[2], ctxt->input->cur[3]); |
| 299 | } |
| 300 | |
| 301 | ctxt->charset = XML_CHAR_ENCODING_8859_1; |
| 302 | *len = 1; |
| 303 | return((int) *ctxt->input->cur); |
| 304 | } |
| 305 | |
| 306 | /** |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 307 | * htmlSkipBlankChars: |
| 308 | * @ctxt: the HTML parser context |
| 309 | * |
| 310 | * skip all blanks character found at that point in the input streams. |
| 311 | * |
| 312 | * Returns the number of space chars skipped |
| 313 | */ |
| 314 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 315 | static int |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 316 | htmlSkipBlankChars(xmlParserCtxtPtr ctxt) { |
| 317 | int res = 0; |
| 318 | |
| 319 | while (IS_BLANK(*(ctxt->input->cur))) { |
| 320 | if ((*ctxt->input->cur == 0) && |
| 321 | (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) { |
| 322 | xmlPopInput(ctxt); |
| 323 | } else { |
| 324 | if (*(ctxt->input->cur) == '\n') { |
| 325 | ctxt->input->line++; ctxt->input->col = 1; |
| 326 | } else ctxt->input->col++; |
| 327 | ctxt->input->cur++; |
| 328 | ctxt->nbChars++; |
| 329 | if (*ctxt->input->cur == 0) |
| 330 | xmlParserInputGrow(ctxt->input, INPUT_CHUNK); |
| 331 | } |
| 332 | res++; |
| 333 | } |
| 334 | return(res); |
| 335 | } |
| 336 | |
| 337 | |
| 338 | |
| 339 | /************************************************************************ |
| 340 | * * |
| 341 | * The list of HTML elements and their properties * |
| 342 | * * |
| 343 | ************************************************************************/ |
| 344 | |
| 345 | /* |
| 346 | * Start Tag: 1 means the start tag can be ommited |
| 347 | * End Tag: 1 means the end tag can be ommited |
| 348 | * 2 means it's forbidden (empty elements) |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 349 | * 3 means the tag is stylistic and should be closed easilly |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 350 | * Depr: this element is deprecated |
| 351 | * DTD: 1 means that this element is valid only in the Loose DTD |
| 352 | * 2 means that this element is valid only in the Frameset DTD |
| 353 | * |
| 354 | * Name,Start Tag,End Tag,Save End, Empty, Depr., DTD, Description |
| 355 | */ |
| 356 | htmlElemDesc html40ElementTable[] = { |
| 357 | { "a", 0, 0, 0, 0, 0, 0, "anchor " }, |
| 358 | { "abbr", 0, 0, 0, 0, 0, 0, "abbreviated form" }, |
| 359 | { "acronym", 0, 0, 0, 0, 0, 0, "" }, |
| 360 | { "address", 0, 0, 0, 0, 0, 0, "information on author " }, |
| 361 | { "applet", 0, 0, 0, 0, 1, 1, "java applet " }, |
| 362 | { "area", 0, 2, 2, 1, 0, 0, "client-side image map area " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 363 | { "b", 0, 3, 0, 0, 0, 0, "bold text style" }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 364 | { "base", 0, 2, 2, 1, 0, 0, "document base uri " }, |
| 365 | { "basefont", 0, 2, 2, 1, 1, 1, "base font size " }, |
| 366 | { "bdo", 0, 0, 0, 0, 0, 0, "i18n bidi over-ride " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 367 | { "big", 0, 3, 0, 0, 0, 0, "large text style" }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 368 | { "blockquote", 0, 0, 0, 0, 0, 0, "long quotation " }, |
| 369 | { "body", 1, 1, 0, 0, 0, 0, "document body " }, |
| 370 | { "br", 0, 2, 2, 1, 0, 0, "forced line break " }, |
| 371 | { "button", 0, 0, 0, 0, 0, 0, "push button " }, |
| 372 | { "caption", 0, 0, 0, 0, 0, 0, "table caption " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 373 | { "center", 0, 3, 0, 0, 1, 1, "shorthand for div align=center " }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 374 | { "cite", 0, 0, 0, 0, 0, 0, "citation" }, |
| 375 | { "code", 0, 0, 0, 0, 0, 0, "computer code fragment" }, |
| 376 | { "col", 0, 2, 2, 1, 0, 0, "table column " }, |
| 377 | { "colgroup", 0, 1, 0, 0, 0, 0, "table column group " }, |
| 378 | { "dd", 0, 1, 0, 0, 0, 0, "definition description " }, |
| 379 | { "del", 0, 0, 0, 0, 0, 0, "deleted text " }, |
| 380 | { "dfn", 0, 0, 0, 0, 0, 0, "instance definition" }, |
| 381 | { "dir", 0, 0, 0, 0, 1, 1, "directory list" }, |
| 382 | { "div", 0, 0, 0, 0, 0, 0, "generic language/style container"}, |
| 383 | { "dl", 0, 0, 0, 0, 0, 0, "definition list " }, |
| 384 | { "dt", 0, 1, 0, 0, 0, 0, "definition term " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 385 | { "em", 0, 3, 0, 0, 0, 0, "emphasis" }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 386 | { "fieldset", 0, 0, 0, 0, 0, 0, "form control group " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 387 | { "font", 0, 3, 0, 0, 1, 1, "local change to font " }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 388 | { "form", 0, 0, 0, 0, 0, 0, "interactive form " }, |
| 389 | { "frame", 0, 2, 2, 1, 0, 2, "subwindow " }, |
| 390 | { "frameset", 0, 0, 0, 0, 0, 2, "window subdivision" }, |
| 391 | { "h1", 0, 0, 0, 0, 0, 0, "heading " }, |
| 392 | { "h2", 0, 0, 0, 0, 0, 0, "heading " }, |
| 393 | { "h3", 0, 0, 0, 0, 0, 0, "heading " }, |
| 394 | { "h4", 0, 0, 0, 0, 0, 0, "heading " }, |
| 395 | { "h5", 0, 0, 0, 0, 0, 0, "heading " }, |
| 396 | { "h6", 0, 0, 0, 0, 0, 0, "heading " }, |
| 397 | { "head", 1, 1, 0, 0, 0, 0, "document head " }, |
| 398 | { "hr", 0, 2, 2, 1, 0, 0, "horizontal rule " }, |
| 399 | { "html", 1, 1, 0, 0, 0, 0, "document root element " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 400 | { "i", 0, 3, 0, 0, 0, 0, "italic text style" }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 401 | { "iframe", 0, 0, 0, 0, 0, 1, "inline subwindow " }, |
| 402 | { "img", 0, 2, 2, 1, 0, 0, "embedded image " }, |
| 403 | { "input", 0, 2, 2, 1, 0, 0, "form control " }, |
| 404 | { "ins", 0, 0, 0, 0, 0, 0, "inserted text" }, |
| 405 | { "isindex", 0, 2, 2, 1, 1, 1, "single line prompt " }, |
| 406 | { "kbd", 0, 0, 0, 0, 0, 0, "text to be entered by the user" }, |
| 407 | { "label", 0, 0, 0, 0, 0, 0, "form field label text " }, |
| 408 | { "legend", 0, 0, 0, 0, 0, 0, "fieldset legend " }, |
| 409 | { "li", 0, 1, 1, 0, 0, 0, "list item " }, |
| 410 | { "link", 0, 2, 2, 1, 0, 0, "a media-independent link " }, |
| 411 | { "map", 0, 0, 0, 0, 0, 0, "client-side image map " }, |
| 412 | { "menu", 0, 0, 0, 0, 1, 1, "menu list " }, |
| 413 | { "meta", 0, 2, 2, 1, 0, 0, "generic metainformation " }, |
| 414 | { "noframes", 0, 0, 0, 0, 0, 2, "alternate content container for non frame-based rendering " }, |
| 415 | { "noscript", 0, 0, 0, 0, 0, 0, "alternate content container for non script-based rendering " }, |
| 416 | { "object", 0, 0, 0, 0, 0, 0, "generic embedded object " }, |
| 417 | { "ol", 0, 0, 0, 0, 0, 0, "ordered list " }, |
| 418 | { "optgroup", 0, 0, 0, 0, 0, 0, "option group " }, |
| 419 | { "option", 0, 1, 0, 0, 0, 0, "selectable choice " }, |
| 420 | { "p", 0, 1, 1, 0, 0, 0, "paragraph " }, |
| 421 | { "param", 0, 2, 2, 1, 0, 0, "named property value " }, |
| 422 | { "pre", 0, 0, 0, 0, 0, 0, "preformatted text " }, |
| 423 | { "q", 0, 0, 0, 0, 0, 0, "short inline quotation " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 424 | { "s", 0, 3, 0, 0, 1, 1, "strike-through text style" }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 425 | { "samp", 0, 0, 0, 0, 0, 0, "sample program output, scripts, etc." }, |
| 426 | { "script", 0, 0, 0, 0, 0, 0, "script statements " }, |
| 427 | { "select", 0, 0, 0, 0, 0, 0, "option selector " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 428 | { "small", 0, 3, 0, 0, 0, 0, "small text style" }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 429 | { "span", 0, 0, 0, 0, 0, 0, "generic language/style container " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 430 | { "strike", 0, 3, 0, 0, 1, 1, "strike-through text" }, |
| 431 | { "strong", 0, 3, 0, 0, 0, 0, "strong emphasis" }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 432 | { "style", 0, 0, 0, 0, 0, 0, "style info " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 433 | { "sub", 0, 3, 0, 0, 0, 0, "subscript" }, |
| 434 | { "sup", 0, 3, 0, 0, 0, 0, "superscript " }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 435 | { "table", 0, 0, 0, 0, 0, 0, " " }, |
| 436 | { "tbody", 1, 0, 0, 0, 0, 0, "table body " }, |
| 437 | { "td", 0, 0, 0, 0, 0, 0, "table data cell" }, |
| 438 | { "textarea", 0, 0, 0, 0, 0, 0, "multi-line text field " }, |
| 439 | { "tfoot", 0, 1, 0, 0, 0, 0, "table footer " }, |
| 440 | { "th", 0, 1, 0, 0, 0, 0, "table header cell" }, |
| 441 | { "thead", 0, 1, 0, 0, 0, 0, "table header " }, |
| 442 | { "title", 0, 0, 0, 0, 0, 0, "document title " }, |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 443 | { "tr", 0, 0, 0, 0, 0, 0, "table row " }, |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 444 | { "tt", 0, 3, 0, 0, 0, 0, "teletype or monospaced text style" }, |
| 445 | { "u", 0, 3, 0, 0, 1, 1, "underlined text style" }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 446 | { "ul", 0, 0, 0, 0, 0, 0, "unordered list " }, |
| 447 | { "var", 0, 0, 0, 0, 0, 0, "instance of a variable or program argument" }, |
| 448 | }; |
| 449 | |
| 450 | /* |
| 451 | * start tags that imply the end of a current element |
| 452 | * any tag of each line implies the end of the current element if the type of |
| 453 | * that element is in the same line |
| 454 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 455 | const char *htmlEquEnd[] = { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 456 | "dt", "dd", "li", "option", NULL, |
| 457 | "h1", "h2", "h3", "h4", "h5", "h6", NULL, |
| 458 | "ol", "menu", "dir", "address", "pre", "listing", "xmp", NULL, |
| 459 | NULL |
| 460 | }; |
| 461 | /* |
| 462 | * acording the HTML DTD, HR should be added to the 2nd line above, as it |
| 463 | * is not allowed within a H1, H2, H3, etc. But we should tolerate that case |
| 464 | * because many documents contain rules in headings... |
| 465 | */ |
| 466 | |
| 467 | /* |
| 468 | * start tags that imply the end of current element |
| 469 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 470 | const char *htmlStartClose[] = { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 471 | "form", "form", "p", "hr", "h1", "h2", "h3", "h4", "h5", "h6", |
| 472 | "dl", "ul", "ol", "menu", "dir", "address", "pre", |
| 473 | "listing", "xmp", "head", NULL, |
| 474 | "head", "p", NULL, |
| 475 | "title", "p", NULL, |
| 476 | "body", "head", "style", "link", "title", "p", NULL, |
| 477 | "li", "p", "h1", "h2", "h3", "h4", "h5", "h6", "dl", "address", |
| 478 | "pre", "listing", "xmp", "head", "li", NULL, |
| 479 | "hr", "p", "head", NULL, |
| 480 | "h1", "p", "head", NULL, |
| 481 | "h2", "p", "head", NULL, |
| 482 | "h3", "p", "head", NULL, |
| 483 | "h4", "p", "head", NULL, |
| 484 | "h5", "p", "head", NULL, |
| 485 | "h6", "p", "head", NULL, |
| 486 | "dir", "p", "head", NULL, |
| 487 | "address", "p", "head", "ul", NULL, |
| 488 | "pre", "p", "head", "ul", NULL, |
| 489 | "listing", "p", "head", NULL, |
| 490 | "xmp", "p", "head", NULL, |
| 491 | "blockquote", "p", "head", NULL, |
| 492 | "dl", "p", "dt", "menu", "dir", "address", "pre", "listing", |
| 493 | "xmp", "head", NULL, |
| 494 | "dt", "p", "menu", "dir", "address", "pre", "listing", "xmp", |
| 495 | "head", "dd", NULL, |
| 496 | "dd", "p", "menu", "dir", "address", "pre", "listing", "xmp", |
| 497 | "head", "dt", NULL, |
| 498 | "ul", "p", "head", "ol", "menu", "dir", "address", "pre", |
| 499 | "listing", "xmp", NULL, |
| 500 | "ol", "p", "head", "ul", NULL, |
| 501 | "menu", "p", "head", "ul", NULL, |
| 502 | "p", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", NULL, |
| 503 | "div", "p", "head", NULL, |
| 504 | "noscript", "p", "head", NULL, |
| 505 | "center", "font", "b", "i", "p", "head", NULL, |
| 506 | "a", "a", NULL, |
| 507 | "caption", "p", NULL, |
| 508 | "colgroup", "caption", "colgroup", "col", "p", NULL, |
| 509 | "col", "caption", "col", "p", NULL, |
| 510 | "table", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", "pre", |
| 511 | "listing", "xmp", "a", NULL, |
Daniel Veillard | 43dadeb | 2001-04-24 11:23:35 +0000 | [diff] [blame] | 512 | "th", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL, |
| 513 | "td", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 514 | "tr", "th", "td", "tr", "caption", "col", "colgroup", "p", NULL, |
| 515 | "thead", "caption", "col", "colgroup", NULL, |
| 516 | "tfoot", "th", "td", "tr", "caption", "col", "colgroup", "thead", |
| 517 | "tbody", "p", NULL, |
| 518 | "tbody", "th", "td", "tr", "caption", "col", "colgroup", "thead", |
| 519 | "tfoot", "tbody", "p", NULL, |
| 520 | "optgroup", "option", NULL, |
| 521 | "option", "option", NULL, |
| 522 | "fieldset", "legend", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", |
| 523 | "pre", "listing", "xmp", "a", NULL, |
| 524 | NULL |
| 525 | }; |
| 526 | |
| 527 | /* |
| 528 | * The list of HTML elements which are supposed not to have |
| 529 | * CDATA content and where a p element will be implied |
| 530 | * |
| 531 | * TODO: extend that list by reading the HTML SGML DtD on |
| 532 | * implied paragraph |
| 533 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 534 | static const char *htmlNoContentElements[] = { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 535 | "html", |
| 536 | "head", |
| 537 | "body", |
| 538 | NULL |
| 539 | }; |
| 540 | |
| 541 | /* |
| 542 | * The list of HTML attributes which are of content %Script; |
| 543 | * NOTE: when adding ones, check htmlIsScriptAttribute() since |
| 544 | * it assumes the name starts with 'on' |
| 545 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 546 | static const char *htmlScriptAttributes[] = { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 547 | "onclick", |
| 548 | "ondblclick", |
| 549 | "onmousedown", |
| 550 | "onmouseup", |
| 551 | "onmouseover", |
| 552 | "onmousemove", |
| 553 | "onmouseout", |
| 554 | "onkeypress", |
| 555 | "onkeydown", |
| 556 | "onkeyup", |
| 557 | "onload", |
| 558 | "onunload", |
| 559 | "onfocus", |
| 560 | "onblur", |
| 561 | "onsubmit", |
| 562 | "onrest", |
| 563 | "onchange", |
| 564 | "onselect" |
| 565 | }; |
| 566 | |
| 567 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 568 | static const char** htmlStartCloseIndex[100]; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 569 | static int htmlStartCloseIndexinitialized = 0; |
| 570 | |
| 571 | /************************************************************************ |
| 572 | * * |
| 573 | * functions to handle HTML specific data * |
| 574 | * * |
| 575 | ************************************************************************/ |
| 576 | |
| 577 | /** |
| 578 | * htmlInitAutoClose: |
| 579 | * |
| 580 | * Initialize the htmlStartCloseIndex for fast lookup of closing tags names. |
| 581 | * This is not reentrant. Call xmlInitParser() once before processing in |
| 582 | * case of use in multithreaded programs. |
| 583 | */ |
| 584 | void |
| 585 | htmlInitAutoClose(void) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 586 | int indx, i = 0; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 587 | |
| 588 | if (htmlStartCloseIndexinitialized) return; |
| 589 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 590 | for (indx = 0;indx < 100;indx ++) htmlStartCloseIndex[indx] = NULL; |
| 591 | indx = 0; |
| 592 | while ((htmlStartClose[i] != NULL) && (indx < 100 - 1)) { |
| 593 | htmlStartCloseIndex[indx++] = &htmlStartClose[i]; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 594 | while (htmlStartClose[i] != NULL) i++; |
| 595 | i++; |
| 596 | } |
| 597 | htmlStartCloseIndexinitialized = 1; |
| 598 | } |
| 599 | |
| 600 | /** |
| 601 | * htmlTagLookup: |
| 602 | * @tag: The tag name in lowercase |
| 603 | * |
| 604 | * Lookup the HTML tag in the ElementTable |
| 605 | * |
| 606 | * Returns the related htmlElemDescPtr or NULL if not found. |
| 607 | */ |
| 608 | htmlElemDescPtr |
| 609 | htmlTagLookup(const xmlChar *tag) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 610 | unsigned int i; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 611 | |
| 612 | for (i = 0; i < (sizeof(html40ElementTable) / |
| 613 | sizeof(html40ElementTable[0]));i++) { |
Daniel Veillard | 1ed3f88 | 2001-04-18 09:45:35 +0000 | [diff] [blame] | 614 | if (!xmlStrcasecmp(tag, BAD_CAST html40ElementTable[i].name)) |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 615 | return(&html40ElementTable[i]); |
| 616 | } |
| 617 | return(NULL); |
| 618 | } |
| 619 | |
| 620 | /** |
| 621 | * htmlCheckAutoClose: |
| 622 | * @newtag: The new tag name |
| 623 | * @oldtag: The old tag name |
| 624 | * |
| 625 | * Checks wether the new tag is one of the registered valid tags for closing old. |
| 626 | * Initialize the htmlStartCloseIndex for fast lookup of closing tags names. |
| 627 | * |
| 628 | * Returns 0 if no, 1 if yes. |
| 629 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 630 | static int |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 631 | htmlCheckAutoClose(const xmlChar *newtag, const xmlChar *oldtag) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 632 | int i, indx; |
| 633 | const char **closed = NULL; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 634 | |
| 635 | if (htmlStartCloseIndexinitialized == 0) htmlInitAutoClose(); |
| 636 | |
| 637 | /* inefficient, but not a big deal */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 638 | for (indx = 0; indx < 100;indx++) { |
| 639 | closed = htmlStartCloseIndex[indx]; |
| 640 | if (closed == NULL) return(0); |
| 641 | if (xmlStrEqual(BAD_CAST *closed, newtag)) break; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 642 | } |
| 643 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 644 | i = closed - htmlStartClose; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 645 | i++; |
| 646 | while (htmlStartClose[i] != NULL) { |
| 647 | if (xmlStrEqual(BAD_CAST htmlStartClose[i], oldtag)) { |
| 648 | return(1); |
| 649 | } |
| 650 | i++; |
| 651 | } |
| 652 | return(0); |
| 653 | } |
| 654 | |
| 655 | /** |
| 656 | * htmlAutoCloseOnClose: |
| 657 | * @ctxt: an HTML parser context |
| 658 | * @newtag: The new tag name |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 659 | * @force: force the tag closure |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 660 | * |
| 661 | * The HTmL DtD allows an ending tag to implicitely close other tags. |
| 662 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 663 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 664 | htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar *newtag) { |
| 665 | htmlElemDescPtr info; |
| 666 | xmlChar *oldname; |
| 667 | int i; |
| 668 | |
| 669 | #ifdef DEBUG |
| 670 | xmlGenericError(xmlGenericErrorContext,"Close of %s stack: %d elements\n", newtag, ctxt->nameNr); |
| 671 | for (i = 0;i < ctxt->nameNr;i++) |
| 672 | xmlGenericError(xmlGenericErrorContext,"%d : %s\n", i, ctxt->nameTab[i]); |
| 673 | #endif |
| 674 | |
| 675 | for (i = (ctxt->nameNr - 1);i >= 0;i--) { |
| 676 | if (xmlStrEqual(newtag, ctxt->nameTab[i])) break; |
| 677 | } |
| 678 | if (i < 0) return; |
| 679 | |
| 680 | while (!xmlStrEqual(newtag, ctxt->name)) { |
| 681 | info = htmlTagLookup(ctxt->name); |
| 682 | if ((info == NULL) || (info->endTag == 1)) { |
| 683 | #ifdef DEBUG |
| 684 | xmlGenericError(xmlGenericErrorContext,"htmlAutoCloseOnClose: %s closes %s\n", newtag, ctxt->name); |
| 685 | #endif |
Daniel Veillard | 56098d4 | 2001-04-24 12:51:09 +0000 | [diff] [blame] | 686 | } else if (info->endTag == 3) { |
| 687 | #ifdef DEBUG |
| 688 | xmlGenericError(xmlGenericErrorContext,"End of tag %s: expecting %s\n", name, ctxt->name); |
| 689 | #endif |
| 690 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 691 | ctxt->sax->error(ctxt->userData, |
| 692 | "Opening and ending tag mismatch: %s and %s\n", |
| 693 | newtag, ctxt->name); |
| 694 | ctxt->wellFormed = 0; |
| 695 | } else { |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 696 | return; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 697 | } |
| 698 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL)) |
| 699 | ctxt->sax->endElement(ctxt->userData, ctxt->name); |
| 700 | oldname = htmlnamePop(ctxt); |
| 701 | if (oldname != NULL) { |
| 702 | #ifdef DEBUG |
| 703 | xmlGenericError(xmlGenericErrorContext,"htmlAutoCloseOnClose: popped %s\n", oldname); |
| 704 | #endif |
| 705 | xmlFree(oldname); |
| 706 | } |
| 707 | } |
| 708 | } |
| 709 | |
| 710 | /** |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 711 | * htmlAutoCloseOnEnd: |
| 712 | * @ctxt: an HTML parser context |
| 713 | * |
| 714 | * Close all remaining tags at the end of the stream |
| 715 | */ |
| 716 | static void |
| 717 | htmlAutoCloseOnEnd(htmlParserCtxtPtr ctxt) { |
| 718 | xmlChar *oldname; |
| 719 | int i; |
| 720 | |
| 721 | if (ctxt->nameNr == 0) |
| 722 | return; |
| 723 | #ifdef DEBUG |
| 724 | xmlGenericError(xmlGenericErrorContext,"Close of stack: %d elements\n", ctxt->nameNr); |
| 725 | #endif |
| 726 | |
| 727 | for (i = (ctxt->nameNr - 1);i >= 0;i--) { |
| 728 | #ifdef DEBUG |
| 729 | xmlGenericError(xmlGenericErrorContext,"%d : %s\n", i, ctxt->nameTab[i]); |
| 730 | #endif |
| 731 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL)) |
| 732 | ctxt->sax->endElement(ctxt->userData, ctxt->name); |
| 733 | oldname = htmlnamePop(ctxt); |
| 734 | if (oldname != NULL) { |
| 735 | #ifdef DEBUG |
| 736 | xmlGenericError(xmlGenericErrorContext,"htmlAutoCloseOnEnd: popped %s\n", oldname); |
| 737 | #endif |
| 738 | xmlFree(oldname); |
| 739 | } |
| 740 | } |
| 741 | } |
| 742 | |
| 743 | /** |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 744 | * htmlAutoClose: |
| 745 | * @ctxt: an HTML parser context |
| 746 | * @newtag: The new tag name or NULL |
| 747 | * |
| 748 | * The HTmL DtD allows a tag to implicitely close other tags. |
| 749 | * The list is kept in htmlStartClose array. This function is |
| 750 | * called when a new tag has been detected and generates the |
| 751 | * appropriates closes if possible/needed. |
| 752 | * If newtag is NULL this mean we are at the end of the resource |
| 753 | * and we should check |
| 754 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 755 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 756 | htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar *newtag) { |
| 757 | xmlChar *oldname; |
| 758 | while ((newtag != NULL) && (ctxt->name != NULL) && |
| 759 | (htmlCheckAutoClose(newtag, ctxt->name))) { |
| 760 | #ifdef DEBUG |
| 761 | xmlGenericError(xmlGenericErrorContext,"htmlAutoClose: %s closes %s\n", newtag, ctxt->name); |
| 762 | #endif |
| 763 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL)) |
| 764 | ctxt->sax->endElement(ctxt->userData, ctxt->name); |
| 765 | oldname = htmlnamePop(ctxt); |
| 766 | if (oldname != NULL) { |
| 767 | #ifdef DEBUG |
| 768 | xmlGenericError(xmlGenericErrorContext,"htmlAutoClose: popped %s\n", oldname); |
| 769 | #endif |
| 770 | xmlFree(oldname); |
| 771 | } |
| 772 | } |
| 773 | if (newtag == NULL) { |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 774 | htmlAutoCloseOnEnd(ctxt); |
| 775 | return; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 776 | } |
| 777 | while ((newtag == NULL) && (ctxt->name != NULL) && |
| 778 | ((xmlStrEqual(ctxt->name, BAD_CAST"head")) || |
| 779 | (xmlStrEqual(ctxt->name, BAD_CAST"body")) || |
| 780 | (xmlStrEqual(ctxt->name, BAD_CAST"html")))) { |
| 781 | #ifdef DEBUG |
| 782 | xmlGenericError(xmlGenericErrorContext,"htmlAutoClose: EOF closes %s\n", ctxt->name); |
| 783 | #endif |
| 784 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL)) |
| 785 | ctxt->sax->endElement(ctxt->userData, ctxt->name); |
| 786 | oldname = htmlnamePop(ctxt); |
| 787 | if (oldname != NULL) { |
| 788 | #ifdef DEBUG |
| 789 | xmlGenericError(xmlGenericErrorContext,"htmlAutoClose: popped %s\n", oldname); |
| 790 | #endif |
| 791 | xmlFree(oldname); |
| 792 | } |
| 793 | } |
| 794 | |
| 795 | } |
| 796 | |
| 797 | /** |
| 798 | * htmlAutoCloseTag: |
| 799 | * @doc: the HTML document |
| 800 | * @name: The tag name |
| 801 | * @elem: the HTML element |
| 802 | * |
| 803 | * The HTmL DtD allows a tag to implicitely close other tags. |
| 804 | * The list is kept in htmlStartClose array. This function checks |
| 805 | * if the element or one of it's children would autoclose the |
| 806 | * given tag. |
| 807 | * |
| 808 | * Returns 1 if autoclose, 0 otherwise |
| 809 | */ |
| 810 | int |
| 811 | htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) { |
| 812 | htmlNodePtr child; |
| 813 | |
| 814 | if (elem == NULL) return(1); |
| 815 | if (xmlStrEqual(name, elem->name)) return(0); |
| 816 | if (htmlCheckAutoClose(elem->name, name)) return(1); |
| 817 | child = elem->children; |
| 818 | while (child != NULL) { |
| 819 | if (htmlAutoCloseTag(doc, name, child)) return(1); |
| 820 | child = child->next; |
| 821 | } |
| 822 | return(0); |
| 823 | } |
| 824 | |
| 825 | /** |
| 826 | * htmlIsAutoClosed: |
| 827 | * @doc: the HTML document |
| 828 | * @elem: the HTML element |
| 829 | * |
| 830 | * The HTmL DtD allows a tag to implicitely close other tags. |
| 831 | * The list is kept in htmlStartClose array. This function checks |
| 832 | * if a tag is autoclosed by one of it's child |
| 833 | * |
| 834 | * Returns 1 if autoclosed, 0 otherwise |
| 835 | */ |
| 836 | int |
| 837 | htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) { |
| 838 | htmlNodePtr child; |
| 839 | |
| 840 | if (elem == NULL) return(1); |
| 841 | child = elem->children; |
| 842 | while (child != NULL) { |
| 843 | if (htmlAutoCloseTag(doc, elem->name, child)) return(1); |
| 844 | child = child->next; |
| 845 | } |
| 846 | return(0); |
| 847 | } |
| 848 | |
| 849 | /** |
| 850 | * htmlCheckImplied: |
| 851 | * @ctxt: an HTML parser context |
| 852 | * @newtag: The new tag name |
| 853 | * |
| 854 | * The HTML DtD allows a tag to exists only implicitely |
| 855 | * called when a new tag has been detected and generates the |
| 856 | * appropriates implicit tags if missing |
| 857 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 858 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 859 | htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) { |
| 860 | if (!htmlOmittedDefaultValue) |
| 861 | return; |
| 862 | if (xmlStrEqual(newtag, BAD_CAST"html")) |
| 863 | return; |
| 864 | if (ctxt->nameNr <= 0) { |
| 865 | #ifdef DEBUG |
| 866 | xmlGenericError(xmlGenericErrorContext,"Implied element html: pushed html\n"); |
| 867 | #endif |
| 868 | htmlnamePush(ctxt, xmlStrdup(BAD_CAST"html")); |
| 869 | if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) |
| 870 | ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL); |
| 871 | } |
| 872 | if ((xmlStrEqual(newtag, BAD_CAST"body")) || (xmlStrEqual(newtag, BAD_CAST"head"))) |
| 873 | return; |
| 874 | if ((ctxt->nameNr <= 1) && |
| 875 | ((xmlStrEqual(newtag, BAD_CAST"script")) || |
| 876 | (xmlStrEqual(newtag, BAD_CAST"style")) || |
| 877 | (xmlStrEqual(newtag, BAD_CAST"meta")) || |
| 878 | (xmlStrEqual(newtag, BAD_CAST"link")) || |
| 879 | (xmlStrEqual(newtag, BAD_CAST"title")) || |
| 880 | (xmlStrEqual(newtag, BAD_CAST"base")))) { |
| 881 | /* |
| 882 | * dropped OBJECT ... i you put it first BODY will be |
| 883 | * assumed ! |
| 884 | */ |
| 885 | #ifdef DEBUG |
| 886 | xmlGenericError(xmlGenericErrorContext,"Implied element head: pushed head\n"); |
| 887 | #endif |
| 888 | htmlnamePush(ctxt, xmlStrdup(BAD_CAST"head")); |
| 889 | if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) |
| 890 | ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL); |
| 891 | } else if ((!xmlStrEqual(newtag, BAD_CAST"noframes")) && |
| 892 | (!xmlStrEqual(newtag, BAD_CAST"frame")) && |
| 893 | (!xmlStrEqual(newtag, BAD_CAST"frameset"))) { |
| 894 | int i; |
| 895 | for (i = 0;i < ctxt->nameNr;i++) { |
| 896 | if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"body")) { |
| 897 | return; |
| 898 | } |
| 899 | if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"head")) { |
| 900 | return; |
| 901 | } |
| 902 | } |
| 903 | |
| 904 | #ifdef DEBUG |
| 905 | xmlGenericError(xmlGenericErrorContext,"Implied element body: pushed body\n"); |
| 906 | #endif |
| 907 | htmlnamePush(ctxt, xmlStrdup(BAD_CAST"body")); |
| 908 | if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) |
| 909 | ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL); |
| 910 | } |
| 911 | } |
| 912 | |
| 913 | /** |
| 914 | * htmlCheckParagraph |
| 915 | * @ctxt: an HTML parser context |
| 916 | * |
| 917 | * Check whether a p element need to be implied before inserting |
| 918 | * characters in the current element. |
| 919 | * |
| 920 | * Returns 1 if a paragraph has been inserted, 0 if not and -1 |
| 921 | * in case of error. |
| 922 | */ |
| 923 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 924 | static int |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 925 | htmlCheckParagraph(htmlParserCtxtPtr ctxt) { |
| 926 | const xmlChar *tag; |
| 927 | int i; |
| 928 | |
| 929 | if (ctxt == NULL) |
| 930 | return(-1); |
| 931 | tag = ctxt->name; |
| 932 | if (tag == NULL) { |
| 933 | htmlAutoClose(ctxt, BAD_CAST"p"); |
| 934 | htmlCheckImplied(ctxt, BAD_CAST"p"); |
| 935 | htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p")); |
| 936 | if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) |
| 937 | ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL); |
| 938 | return(1); |
| 939 | } |
| 940 | if (!htmlOmittedDefaultValue) |
| 941 | return(0); |
| 942 | for (i = 0; htmlNoContentElements[i] != NULL; i++) { |
| 943 | if (xmlStrEqual(tag, BAD_CAST htmlNoContentElements[i])) { |
| 944 | #ifdef DEBUG |
| 945 | xmlGenericError(xmlGenericErrorContext,"Implied element paragraph\n"); |
| 946 | #endif |
| 947 | htmlAutoClose(ctxt, BAD_CAST"p"); |
| 948 | htmlCheckImplied(ctxt, BAD_CAST"p"); |
| 949 | htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p")); |
| 950 | if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) |
| 951 | ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL); |
| 952 | return(1); |
| 953 | } |
| 954 | } |
| 955 | return(0); |
| 956 | } |
| 957 | |
| 958 | /** |
| 959 | * htmlIsScriptAttribute: |
| 960 | * @name: an attribute name |
| 961 | * |
| 962 | * Check if an attribute is of content type Script |
| 963 | * |
| 964 | * Returns 1 is the attribute is a script 0 otherwise |
| 965 | */ |
| 966 | int |
| 967 | htmlIsScriptAttribute(const xmlChar *name) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 968 | unsigned int i; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 969 | |
| 970 | if (name == NULL) |
| 971 | return(0); |
| 972 | /* |
| 973 | * all script attributes start with 'on' |
| 974 | */ |
| 975 | if ((name[0] != 'o') || (name[1] != 'n')) |
| 976 | return(0); |
| 977 | for (i = 0; |
| 978 | i < sizeof(htmlScriptAttributes)/sizeof(htmlScriptAttributes[0]); |
| 979 | i++) { |
| 980 | if (xmlStrEqual(name, (const xmlChar *) htmlScriptAttributes[i])) |
| 981 | return(1); |
| 982 | } |
| 983 | return(0); |
| 984 | } |
| 985 | |
| 986 | /************************************************************************ |
| 987 | * * |
| 988 | * The list of HTML predefined entities * |
| 989 | * * |
| 990 | ************************************************************************/ |
| 991 | |
| 992 | |
| 993 | htmlEntityDesc html40EntitiesTable[] = { |
| 994 | /* |
| 995 | * the 4 absolute ones, plus apostrophe. |
| 996 | */ |
| 997 | { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" }, |
| 998 | { 38, "amp", "ampersand, U+0026 ISOnum" }, |
| 999 | { 39, "apos", "single quote" }, |
| 1000 | { 60, "lt", "less-than sign, U+003C ISOnum" }, |
| 1001 | { 62, "gt", "greater-than sign, U+003E ISOnum" }, |
| 1002 | |
| 1003 | /* |
| 1004 | * A bunch still in the 128-255 range |
| 1005 | * Replacing them depend really on the charset used. |
| 1006 | */ |
| 1007 | { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" }, |
| 1008 | { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" }, |
| 1009 | { 162, "cent", "cent sign, U+00A2 ISOnum" }, |
| 1010 | { 163, "pound","pound sign, U+00A3 ISOnum" }, |
| 1011 | { 164, "curren","currency sign, U+00A4 ISOnum" }, |
| 1012 | { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" }, |
| 1013 | { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" }, |
| 1014 | { 167, "sect", "section sign, U+00A7 ISOnum" }, |
| 1015 | { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" }, |
| 1016 | { 169, "copy", "copyright sign, U+00A9 ISOnum" }, |
| 1017 | { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" }, |
| 1018 | { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" }, |
| 1019 | { 172, "not", "not sign, U+00AC ISOnum" }, |
| 1020 | { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" }, |
| 1021 | { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" }, |
| 1022 | { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" }, |
| 1023 | { 176, "deg", "degree sign, U+00B0 ISOnum" }, |
| 1024 | { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" }, |
| 1025 | { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" }, |
| 1026 | { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" }, |
| 1027 | { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" }, |
| 1028 | { 181, "micro","micro sign, U+00B5 ISOnum" }, |
| 1029 | { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" }, |
| 1030 | { 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" }, |
| 1031 | { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" }, |
| 1032 | { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" }, |
| 1033 | { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" }, |
| 1034 | { 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" }, |
| 1035 | { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" }, |
| 1036 | { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" }, |
| 1037 | { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" }, |
| 1038 | { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" }, |
| 1039 | { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" }, |
| 1040 | { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" }, |
| 1041 | { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" }, |
| 1042 | { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" }, |
| 1043 | { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" }, |
| 1044 | { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" }, |
| 1045 | { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" }, |
| 1046 | { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" }, |
| 1047 | { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" }, |
| 1048 | { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" }, |
| 1049 | { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" }, |
| 1050 | { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" }, |
| 1051 | { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" }, |
| 1052 | { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" }, |
| 1053 | { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" }, |
| 1054 | { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" }, |
| 1055 | { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" }, |
| 1056 | { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" }, |
| 1057 | { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" }, |
| 1058 | { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" }, |
| 1059 | { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" }, |
| 1060 | { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" }, |
| 1061 | { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" }, |
| 1062 | { 215, "times","multiplication sign, U+00D7 ISOnum" }, |
| 1063 | { 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" }, |
| 1064 | { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" }, |
| 1065 | { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" }, |
| 1066 | { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" }, |
| 1067 | { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" }, |
| 1068 | { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" }, |
| 1069 | { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" }, |
| 1070 | { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" }, |
| 1071 | { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" }, |
| 1072 | { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" }, |
| 1073 | { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" }, |
| 1074 | { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" }, |
| 1075 | { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" }, |
| 1076 | { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" }, |
| 1077 | { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" }, |
| 1078 | { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" }, |
| 1079 | { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" }, |
| 1080 | { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" }, |
| 1081 | { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" }, |
| 1082 | { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" }, |
| 1083 | { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" }, |
| 1084 | { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" }, |
| 1085 | { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" }, |
| 1086 | { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" }, |
| 1087 | { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" }, |
| 1088 | { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" }, |
| 1089 | { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" }, |
| 1090 | { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" }, |
| 1091 | { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" }, |
| 1092 | { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" }, |
| 1093 | { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" }, |
| 1094 | { 247, "divide","division sign, U+00F7 ISOnum" }, |
| 1095 | { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" }, |
| 1096 | { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" }, |
| 1097 | { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" }, |
| 1098 | { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" }, |
| 1099 | { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" }, |
| 1100 | { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" }, |
| 1101 | { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" }, |
| 1102 | { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" }, |
| 1103 | |
| 1104 | { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" }, |
| 1105 | { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" }, |
| 1106 | { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" }, |
| 1107 | { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" }, |
| 1108 | { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" }, |
| 1109 | |
| 1110 | /* |
| 1111 | * Anything below should really be kept as entities references |
| 1112 | */ |
| 1113 | { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" }, |
| 1114 | |
| 1115 | { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" }, |
| 1116 | { 732, "tilde","small tilde, U+02DC ISOdia" }, |
| 1117 | |
| 1118 | { 913, "Alpha","greek capital letter alpha, U+0391" }, |
| 1119 | { 914, "Beta", "greek capital letter beta, U+0392" }, |
| 1120 | { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" }, |
| 1121 | { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" }, |
| 1122 | { 917, "Epsilon","greek capital letter epsilon, U+0395" }, |
| 1123 | { 918, "Zeta", "greek capital letter zeta, U+0396" }, |
| 1124 | { 919, "Eta", "greek capital letter eta, U+0397" }, |
| 1125 | { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" }, |
| 1126 | { 921, "Iota", "greek capital letter iota, U+0399" }, |
| 1127 | { 922, "Kappa","greek capital letter kappa, U+039A" }, |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1128 | { 923, "Lambda", "greek capital letter lambda, U+039B ISOgrk3" }, |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1129 | { 924, "Mu", "greek capital letter mu, U+039C" }, |
| 1130 | { 925, "Nu", "greek capital letter nu, U+039D" }, |
| 1131 | { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" }, |
| 1132 | { 927, "Omicron","greek capital letter omicron, U+039F" }, |
| 1133 | { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" }, |
| 1134 | { 929, "Rho", "greek capital letter rho, U+03A1" }, |
| 1135 | { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" }, |
| 1136 | { 932, "Tau", "greek capital letter tau, U+03A4" }, |
| 1137 | { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" }, |
| 1138 | { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" }, |
| 1139 | { 935, "Chi", "greek capital letter chi, U+03A7" }, |
| 1140 | { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" }, |
| 1141 | { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" }, |
| 1142 | |
| 1143 | { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" }, |
| 1144 | { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" }, |
| 1145 | { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" }, |
| 1146 | { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" }, |
| 1147 | { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" }, |
| 1148 | { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" }, |
| 1149 | { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" }, |
| 1150 | { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" }, |
| 1151 | { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" }, |
| 1152 | { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" }, |
| 1153 | { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" }, |
| 1154 | { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" }, |
| 1155 | { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" }, |
| 1156 | { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" }, |
| 1157 | { 959, "omicron","greek small letter omicron, U+03BF NEW" }, |
| 1158 | { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" }, |
| 1159 | { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" }, |
| 1160 | { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" }, |
| 1161 | { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" }, |
| 1162 | { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" }, |
| 1163 | { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" }, |
| 1164 | { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" }, |
| 1165 | { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" }, |
| 1166 | { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" }, |
| 1167 | { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" }, |
| 1168 | { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" }, |
| 1169 | { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" }, |
| 1170 | { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" }, |
| 1171 | |
| 1172 | { 8194, "ensp", "en space, U+2002 ISOpub" }, |
| 1173 | { 8195, "emsp", "em space, U+2003 ISOpub" }, |
| 1174 | { 8201, "thinsp","thin space, U+2009 ISOpub" }, |
| 1175 | { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" }, |
| 1176 | { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" }, |
| 1177 | { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" }, |
| 1178 | { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" }, |
| 1179 | { 8211, "ndash","en dash, U+2013 ISOpub" }, |
| 1180 | { 8212, "mdash","em dash, U+2014 ISOpub" }, |
| 1181 | { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" }, |
| 1182 | { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" }, |
| 1183 | { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" }, |
| 1184 | { 8220, "ldquo","left double quotation mark, U+201C ISOnum" }, |
| 1185 | { 8221, "rdquo","right double quotation mark, U+201D ISOnum" }, |
| 1186 | { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" }, |
| 1187 | { 8224, "dagger","dagger, U+2020 ISOpub" }, |
| 1188 | { 8225, "Dagger","double dagger, U+2021 ISOpub" }, |
| 1189 | |
| 1190 | { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" }, |
| 1191 | { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" }, |
| 1192 | |
| 1193 | { 8240, "permil","per mille sign, U+2030 ISOtech" }, |
| 1194 | |
| 1195 | { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" }, |
| 1196 | { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" }, |
| 1197 | |
| 1198 | { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" }, |
| 1199 | { 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" }, |
| 1200 | |
| 1201 | { 8254, "oline","overline = spacing overscore, U+203E NEW" }, |
| 1202 | { 8260, "frasl","fraction slash, U+2044 NEW" }, |
| 1203 | |
| 1204 | { 8364, "euro", "euro sign, U+20AC NEW" }, |
| 1205 | |
| 1206 | { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" }, |
| 1207 | { 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" }, |
| 1208 | { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" }, |
| 1209 | { 8482, "trade","trade mark sign, U+2122 ISOnum" }, |
| 1210 | { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" }, |
| 1211 | { 8592, "larr", "leftwards arrow, U+2190 ISOnum" }, |
| 1212 | { 8593, "uarr", "upwards arrow, U+2191 ISOnum" }, |
| 1213 | { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" }, |
| 1214 | { 8595, "darr", "downwards arrow, U+2193 ISOnum" }, |
| 1215 | { 8596, "harr", "left right arrow, U+2194 ISOamsa" }, |
| 1216 | { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" }, |
| 1217 | { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" }, |
| 1218 | { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" }, |
| 1219 | { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" }, |
| 1220 | { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" }, |
| 1221 | { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" }, |
| 1222 | |
| 1223 | { 8704, "forall","for all, U+2200 ISOtech" }, |
| 1224 | { 8706, "part", "partial differential, U+2202 ISOtech" }, |
| 1225 | { 8707, "exist","there exists, U+2203 ISOtech" }, |
| 1226 | { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" }, |
| 1227 | { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" }, |
| 1228 | { 8712, "isin", "element of, U+2208 ISOtech" }, |
| 1229 | { 8713, "notin","not an element of, U+2209 ISOtech" }, |
| 1230 | { 8715, "ni", "contains as member, U+220B ISOtech" }, |
| 1231 | { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" }, |
| 1232 | { 8721, "sum", "n-ary sumation, U+2211 ISOamsb" }, |
| 1233 | { 8722, "minus","minus sign, U+2212 ISOtech" }, |
| 1234 | { 8727, "lowast","asterisk operator, U+2217 ISOtech" }, |
| 1235 | { 8730, "radic","square root = radical sign, U+221A ISOtech" }, |
| 1236 | { 8733, "prop", "proportional to, U+221D ISOtech" }, |
| 1237 | { 8734, "infin","infinity, U+221E ISOtech" }, |
| 1238 | { 8736, "ang", "angle, U+2220 ISOamso" }, |
| 1239 | { 8743, "and", "logical and = wedge, U+2227 ISOtech" }, |
| 1240 | { 8744, "or", "logical or = vee, U+2228 ISOtech" }, |
| 1241 | { 8745, "cap", "intersection = cap, U+2229 ISOtech" }, |
| 1242 | { 8746, "cup", "union = cup, U+222A ISOtech" }, |
| 1243 | { 8747, "int", "integral, U+222B ISOtech" }, |
| 1244 | { 8756, "there4","therefore, U+2234 ISOtech" }, |
| 1245 | { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" }, |
| 1246 | { 8773, "cong", "approximately equal to, U+2245 ISOtech" }, |
| 1247 | { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" }, |
| 1248 | { 8800, "ne", "not equal to, U+2260 ISOtech" }, |
| 1249 | { 8801, "equiv","identical to, U+2261 ISOtech" }, |
| 1250 | { 8804, "le", "less-than or equal to, U+2264 ISOtech" }, |
| 1251 | { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" }, |
| 1252 | { 8834, "sub", "subset of, U+2282 ISOtech" }, |
| 1253 | { 8835, "sup", "superset of, U+2283 ISOtech" }, |
| 1254 | { 8836, "nsub", "not a subset of, U+2284 ISOamsn" }, |
| 1255 | { 8838, "sube", "subset of or equal to, U+2286 ISOtech" }, |
| 1256 | { 8839, "supe", "superset of or equal to, U+2287 ISOtech" }, |
| 1257 | { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" }, |
| 1258 | { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" }, |
| 1259 | { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" }, |
| 1260 | { 8901, "sdot", "dot operator, U+22C5 ISOamsb" }, |
| 1261 | { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" }, |
| 1262 | { 8969, "rceil","right ceiling, U+2309 ISOamsc" }, |
| 1263 | { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" }, |
| 1264 | { 8971, "rfloor","right floor, U+230B ISOamsc" }, |
| 1265 | { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" }, |
| 1266 | { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" }, |
| 1267 | { 9674, "loz", "lozenge, U+25CA ISOpub" }, |
| 1268 | |
| 1269 | { 9824, "spades","black spade suit, U+2660 ISOpub" }, |
| 1270 | { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" }, |
| 1271 | { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" }, |
| 1272 | { 9830, "diams","black diamond suit, U+2666 ISOpub" }, |
| 1273 | |
| 1274 | }; |
| 1275 | |
| 1276 | /************************************************************************ |
| 1277 | * * |
| 1278 | * Commodity functions to handle entities * |
| 1279 | * * |
| 1280 | ************************************************************************/ |
| 1281 | |
| 1282 | /* |
| 1283 | * Macro used to grow the current buffer. |
| 1284 | */ |
| 1285 | #define growBuffer(buffer) { \ |
| 1286 | buffer##_size *= 2; \ |
| 1287 | buffer = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \ |
| 1288 | if (buffer == NULL) { \ |
| 1289 | perror("realloc failed"); \ |
| 1290 | return(NULL); \ |
| 1291 | } \ |
| 1292 | } |
| 1293 | |
| 1294 | /** |
| 1295 | * htmlEntityLookup: |
| 1296 | * @name: the entity name |
| 1297 | * |
| 1298 | * Lookup the given entity in EntitiesTable |
| 1299 | * |
| 1300 | * TODO: the linear scan is really ugly, an hash table is really needed. |
| 1301 | * |
| 1302 | * Returns the associated htmlEntityDescPtr if found, NULL otherwise. |
| 1303 | */ |
| 1304 | htmlEntityDescPtr |
| 1305 | htmlEntityLookup(const xmlChar *name) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1306 | unsigned int i; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1307 | |
| 1308 | for (i = 0;i < (sizeof(html40EntitiesTable)/ |
| 1309 | sizeof(html40EntitiesTable[0]));i++) { |
| 1310 | if (xmlStrEqual(name, BAD_CAST html40EntitiesTable[i].name)) { |
| 1311 | #ifdef DEBUG |
| 1312 | xmlGenericError(xmlGenericErrorContext,"Found entity %s\n", name); |
| 1313 | #endif |
| 1314 | return(&html40EntitiesTable[i]); |
| 1315 | } |
| 1316 | } |
| 1317 | return(NULL); |
| 1318 | } |
| 1319 | |
| 1320 | /** |
| 1321 | * htmlEntityValueLookup: |
| 1322 | * @value: the entity's unicode value |
| 1323 | * |
| 1324 | * Lookup the given entity in EntitiesTable |
| 1325 | * |
| 1326 | * TODO: the linear scan is really ugly, an hash table is really needed. |
| 1327 | * |
| 1328 | * Returns the associated htmlEntityDescPtr if found, NULL otherwise. |
| 1329 | */ |
| 1330 | htmlEntityDescPtr |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1331 | htmlEntityValueLookup(unsigned int value) { |
| 1332 | unsigned int i; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1333 | #ifdef DEBUG |
| 1334 | int lv = 0; |
| 1335 | #endif |
| 1336 | |
| 1337 | for (i = 0;i < (sizeof(html40EntitiesTable)/ |
| 1338 | sizeof(html40EntitiesTable[0]));i++) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1339 | if (html40EntitiesTable[i].value >= value) { |
| 1340 | if (html40EntitiesTable[i].value > value) |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1341 | break; |
| 1342 | #ifdef DEBUG |
| 1343 | xmlGenericError(xmlGenericErrorContext,"Found entity %s\n", html40EntitiesTable[i].name); |
| 1344 | #endif |
| 1345 | return(&html40EntitiesTable[i]); |
| 1346 | } |
| 1347 | #ifdef DEBUG |
| 1348 | if (lv > html40EntitiesTable[i].value) { |
| 1349 | xmlGenericError(xmlGenericErrorContext, |
| 1350 | "html40EntitiesTable[] is not sorted (%d > %d)!\n", |
| 1351 | lv, html40EntitiesTable[i].value); |
| 1352 | } |
| 1353 | lv = html40EntitiesTable[i].value; |
| 1354 | #endif |
| 1355 | } |
| 1356 | return(NULL); |
| 1357 | } |
| 1358 | |
| 1359 | /** |
| 1360 | * UTF8ToHtml: |
| 1361 | * @out: a pointer to an array of bytes to store the result |
| 1362 | * @outlen: the length of @out |
| 1363 | * @in: a pointer to an array of UTF-8 chars |
| 1364 | * @inlen: the length of @in |
| 1365 | * |
| 1366 | * Take a block of UTF-8 chars in and try to convert it to an ASCII |
| 1367 | * plus HTML entities block of chars out. |
| 1368 | * |
| 1369 | * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise |
| 1370 | * The value of @inlen after return is the number of octets consumed |
| 1371 | * as the return value is positive, else unpredictiable. |
| 1372 | * The value of @outlen after return is the number of octets consumed. |
| 1373 | */ |
| 1374 | int |
| 1375 | UTF8ToHtml(unsigned char* out, int *outlen, |
| 1376 | const unsigned char* in, int *inlen) { |
| 1377 | const unsigned char* processed = in; |
| 1378 | const unsigned char* outend; |
| 1379 | const unsigned char* outstart = out; |
| 1380 | const unsigned char* instart = in; |
| 1381 | const unsigned char* inend; |
| 1382 | unsigned int c, d; |
| 1383 | int trailing; |
| 1384 | |
| 1385 | if (in == NULL) { |
| 1386 | /* |
| 1387 | * initialization nothing to do |
| 1388 | */ |
| 1389 | *outlen = 0; |
| 1390 | *inlen = 0; |
| 1391 | return(0); |
| 1392 | } |
| 1393 | inend = in + (*inlen); |
| 1394 | outend = out + (*outlen); |
| 1395 | while (in < inend) { |
| 1396 | d = *in++; |
| 1397 | if (d < 0x80) { c= d; trailing= 0; } |
| 1398 | else if (d < 0xC0) { |
| 1399 | /* trailing byte in leading position */ |
| 1400 | *outlen = out - outstart; |
| 1401 | *inlen = processed - instart; |
| 1402 | return(-2); |
| 1403 | } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; } |
| 1404 | else if (d < 0xF0) { c= d & 0x0F; trailing= 2; } |
| 1405 | else if (d < 0xF8) { c= d & 0x07; trailing= 3; } |
| 1406 | else { |
| 1407 | /* no chance for this in Ascii */ |
| 1408 | *outlen = out - outstart; |
| 1409 | *inlen = processed - instart; |
| 1410 | return(-2); |
| 1411 | } |
| 1412 | |
| 1413 | if (inend - in < trailing) { |
| 1414 | break; |
| 1415 | } |
| 1416 | |
| 1417 | for ( ; trailing; trailing--) { |
| 1418 | if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80)) |
| 1419 | break; |
| 1420 | c <<= 6; |
| 1421 | c |= d & 0x3F; |
| 1422 | } |
| 1423 | |
| 1424 | /* assertion: c is a single UTF-4 value */ |
| 1425 | if (c < 0x80) { |
| 1426 | if (out + 1 >= outend) |
| 1427 | break; |
| 1428 | *out++ = c; |
| 1429 | } else { |
| 1430 | int len; |
| 1431 | htmlEntityDescPtr ent; |
| 1432 | |
| 1433 | /* |
| 1434 | * Try to lookup a predefined HTML entity for it |
| 1435 | */ |
| 1436 | |
| 1437 | ent = htmlEntityValueLookup(c); |
| 1438 | if (ent == NULL) { |
| 1439 | /* no chance for this in Ascii */ |
| 1440 | *outlen = out - outstart; |
| 1441 | *inlen = processed - instart; |
| 1442 | return(-2); |
| 1443 | } |
| 1444 | len = strlen(ent->name); |
| 1445 | if (out + 2 + len >= outend) |
| 1446 | break; |
| 1447 | *out++ = '&'; |
| 1448 | memcpy(out, ent->name, len); |
| 1449 | out += len; |
| 1450 | *out++ = ';'; |
| 1451 | } |
| 1452 | processed = in; |
| 1453 | } |
| 1454 | *outlen = out - outstart; |
| 1455 | *inlen = processed - instart; |
| 1456 | return(0); |
| 1457 | } |
| 1458 | |
| 1459 | /** |
| 1460 | * htmlEncodeEntities: |
| 1461 | * @out: a pointer to an array of bytes to store the result |
| 1462 | * @outlen: the length of @out |
| 1463 | * @in: a pointer to an array of UTF-8 chars |
| 1464 | * @inlen: the length of @in |
| 1465 | * @quoteChar: the quote character to escape (' or ") or zero. |
| 1466 | * |
| 1467 | * Take a block of UTF-8 chars in and try to convert it to an ASCII |
| 1468 | * plus HTML entities block of chars out. |
| 1469 | * |
| 1470 | * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise |
| 1471 | * The value of @inlen after return is the number of octets consumed |
| 1472 | * as the return value is positive, else unpredictiable. |
| 1473 | * The value of @outlen after return is the number of octets consumed. |
| 1474 | */ |
| 1475 | int |
| 1476 | htmlEncodeEntities(unsigned char* out, int *outlen, |
| 1477 | const unsigned char* in, int *inlen, int quoteChar) { |
| 1478 | const unsigned char* processed = in; |
| 1479 | const unsigned char* outend = out + (*outlen); |
| 1480 | const unsigned char* outstart = out; |
| 1481 | const unsigned char* instart = in; |
| 1482 | const unsigned char* inend = in + (*inlen); |
| 1483 | unsigned int c, d; |
| 1484 | int trailing; |
| 1485 | |
| 1486 | while (in < inend) { |
| 1487 | d = *in++; |
| 1488 | if (d < 0x80) { c= d; trailing= 0; } |
| 1489 | else if (d < 0xC0) { |
| 1490 | /* trailing byte in leading position */ |
| 1491 | *outlen = out - outstart; |
| 1492 | *inlen = processed - instart; |
| 1493 | return(-2); |
| 1494 | } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; } |
| 1495 | else if (d < 0xF0) { c= d & 0x0F; trailing= 2; } |
| 1496 | else if (d < 0xF8) { c= d & 0x07; trailing= 3; } |
| 1497 | else { |
| 1498 | /* no chance for this in Ascii */ |
| 1499 | *outlen = out - outstart; |
| 1500 | *inlen = processed - instart; |
| 1501 | return(-2); |
| 1502 | } |
| 1503 | |
| 1504 | if (inend - in < trailing) |
| 1505 | break; |
| 1506 | |
| 1507 | while (trailing--) { |
| 1508 | if (((d= *in++) & 0xC0) != 0x80) { |
| 1509 | *outlen = out - outstart; |
| 1510 | *inlen = processed - instart; |
| 1511 | return(-2); |
| 1512 | } |
| 1513 | c <<= 6; |
| 1514 | c |= d & 0x3F; |
| 1515 | } |
| 1516 | |
| 1517 | /* assertion: c is a single UTF-4 value */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1518 | if ((c < 0x80) && (c != (unsigned int) quoteChar) && |
| 1519 | (c != '&') && (c != '<') && (c != '>')) { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1520 | if (out >= outend) |
| 1521 | break; |
| 1522 | *out++ = c; |
| 1523 | } else { |
| 1524 | htmlEntityDescPtr ent; |
| 1525 | const char *cp; |
| 1526 | char nbuf[16]; |
| 1527 | int len; |
| 1528 | |
| 1529 | /* |
| 1530 | * Try to lookup a predefined HTML entity for it |
| 1531 | */ |
| 1532 | ent = htmlEntityValueLookup(c); |
| 1533 | if (ent == NULL) { |
| 1534 | sprintf(nbuf, "#%u", c); |
| 1535 | cp = nbuf; |
| 1536 | } |
| 1537 | else |
| 1538 | cp = ent->name; |
| 1539 | len = strlen(cp); |
| 1540 | if (out + 2 + len > outend) |
| 1541 | break; |
| 1542 | *out++ = '&'; |
| 1543 | memcpy(out, cp, len); |
| 1544 | out += len; |
| 1545 | *out++ = ';'; |
| 1546 | } |
| 1547 | processed = in; |
| 1548 | } |
| 1549 | *outlen = out - outstart; |
| 1550 | *inlen = processed - instart; |
| 1551 | return(0); |
| 1552 | } |
| 1553 | |
| 1554 | /** |
| 1555 | * htmlDecodeEntities: |
| 1556 | * @ctxt: the parser context |
| 1557 | * @len: the len to decode (in bytes !), -1 for no size limit |
| 1558 | * @end: an end marker xmlChar, 0 if none |
| 1559 | * @end2: an end marker xmlChar, 0 if none |
| 1560 | * @end3: an end marker xmlChar, 0 if none |
| 1561 | * |
| 1562 | * Subtitute the HTML entities by their value |
| 1563 | * |
| 1564 | * DEPRECATED !!!! |
| 1565 | * |
| 1566 | * Returns A newly allocated string with the substitution done. The caller |
| 1567 | * must deallocate it ! |
| 1568 | */ |
| 1569 | xmlChar * |
Daniel Veillard | c86a4fa | 2001-03-26 16:28:29 +0000 | [diff] [blame] | 1570 | htmlDecodeEntities(htmlParserCtxtPtr ctxt ATTRIBUTE_UNUSED, int len ATTRIBUTE_UNUSED, |
| 1571 | xmlChar end ATTRIBUTE_UNUSED, xmlChar end2 ATTRIBUTE_UNUSED, xmlChar end3 ATTRIBUTE_UNUSED) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1572 | static int deprecated = 0; |
| 1573 | if (!deprecated) { |
| 1574 | xmlGenericError(xmlGenericErrorContext, |
| 1575 | "htmlDecodeEntities() deprecated function reached\n"); |
| 1576 | deprecated = 1; |
| 1577 | } |
| 1578 | return(NULL); |
| 1579 | #if 0 |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1580 | xmlChar *name = NULL; |
| 1581 | xmlChar *buffer = NULL; |
| 1582 | unsigned int buffer_size = 0; |
| 1583 | unsigned int nbchars = 0; |
| 1584 | htmlEntityDescPtr ent; |
| 1585 | unsigned int max = (unsigned int) len; |
| 1586 | int c,l; |
| 1587 | |
| 1588 | if (ctxt->depth > 40) { |
| 1589 | ctxt->errNo = XML_ERR_ENTITY_LOOP; |
| 1590 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 1591 | ctxt->sax->error(ctxt->userData, |
| 1592 | "Detected entity reference loop\n"); |
| 1593 | ctxt->wellFormed = 0; |
| 1594 | ctxt->disableSAX = 1; |
| 1595 | return(NULL); |
| 1596 | } |
| 1597 | |
| 1598 | /* |
| 1599 | * allocate a translation buffer. |
| 1600 | */ |
| 1601 | buffer_size = HTML_PARSER_BIG_BUFFER_SIZE; |
| 1602 | buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar)); |
| 1603 | if (buffer == NULL) { |
| 1604 | perror("xmlDecodeEntities: malloc failed"); |
| 1605 | return(NULL); |
| 1606 | } |
| 1607 | |
| 1608 | /* |
| 1609 | * Ok loop until we reach one of the ending char or a size limit. |
| 1610 | */ |
| 1611 | c = CUR_CHAR(l); |
| 1612 | while ((nbchars < max) && (c != end) && |
| 1613 | (c != end2) && (c != end3)) { |
| 1614 | |
| 1615 | if (c == 0) break; |
| 1616 | if (((c == '&') && (ctxt->token != '&')) && (NXT(1) == '#')) { |
| 1617 | int val = htmlParseCharRef(ctxt); |
| 1618 | COPY_BUF(0,buffer,nbchars,val); |
| 1619 | NEXTL(l); |
| 1620 | } else if ((c == '&') && (ctxt->token != '&')) { |
| 1621 | ent = htmlParseEntityRef(ctxt, &name); |
| 1622 | if (name != NULL) { |
| 1623 | if (ent != NULL) { |
| 1624 | int val = ent->value; |
| 1625 | COPY_BUF(0,buffer,nbchars,val); |
| 1626 | NEXTL(l); |
| 1627 | } else { |
| 1628 | const xmlChar *cur = name; |
| 1629 | |
| 1630 | buffer[nbchars++] = '&'; |
| 1631 | if (nbchars > buffer_size - HTML_PARSER_BUFFER_SIZE) { |
| 1632 | growBuffer(buffer); |
| 1633 | } |
| 1634 | while (*cur != 0) { |
| 1635 | buffer[nbchars++] = *cur++; |
| 1636 | } |
| 1637 | buffer[nbchars++] = ';'; |
| 1638 | } |
| 1639 | } |
| 1640 | } else { |
| 1641 | COPY_BUF(l,buffer,nbchars,c); |
| 1642 | NEXTL(l); |
| 1643 | if (nbchars > buffer_size - HTML_PARSER_BUFFER_SIZE) { |
| 1644 | growBuffer(buffer); |
| 1645 | } |
| 1646 | } |
| 1647 | c = CUR_CHAR(l); |
| 1648 | } |
| 1649 | buffer[nbchars++] = 0; |
| 1650 | return(buffer); |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1651 | #endif |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1652 | } |
| 1653 | |
| 1654 | /************************************************************************ |
| 1655 | * * |
| 1656 | * Commodity functions to handle streams * |
| 1657 | * * |
| 1658 | ************************************************************************/ |
| 1659 | |
| 1660 | /** |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1661 | * htmlNewInputStream: |
| 1662 | * @ctxt: an HTML parser context |
| 1663 | * |
| 1664 | * Create a new input stream structure |
| 1665 | * Returns the new input stream or NULL |
| 1666 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1667 | static htmlParserInputPtr |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1668 | htmlNewInputStream(htmlParserCtxtPtr ctxt) { |
| 1669 | htmlParserInputPtr input; |
| 1670 | |
| 1671 | input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput)); |
| 1672 | if (input == NULL) { |
| 1673 | ctxt->errNo = XML_ERR_NO_MEMORY; |
| 1674 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 1675 | ctxt->sax->error(ctxt->userData, |
| 1676 | "malloc: couldn't allocate a new input stream\n"); |
| 1677 | return(NULL); |
| 1678 | } |
| 1679 | memset(input, 0, sizeof(htmlParserInput)); |
| 1680 | input->filename = NULL; |
| 1681 | input->directory = NULL; |
| 1682 | input->base = NULL; |
| 1683 | input->cur = NULL; |
| 1684 | input->buf = NULL; |
| 1685 | input->line = 1; |
| 1686 | input->col = 1; |
| 1687 | input->buf = NULL; |
| 1688 | input->free = NULL; |
| 1689 | input->version = NULL; |
| 1690 | input->consumed = 0; |
| 1691 | input->length = 0; |
| 1692 | return(input); |
| 1693 | } |
| 1694 | |
| 1695 | |
| 1696 | /************************************************************************ |
| 1697 | * * |
| 1698 | * Commodity functions, cleanup needed ? * |
| 1699 | * * |
| 1700 | ************************************************************************/ |
| 1701 | |
| 1702 | /** |
| 1703 | * areBlanks: |
| 1704 | * @ctxt: an HTML parser context |
| 1705 | * @str: a xmlChar * |
| 1706 | * @len: the size of @str |
| 1707 | * |
| 1708 | * Is this a sequence of blank chars that one can ignore ? |
| 1709 | * |
| 1710 | * Returns 1 if ignorable 0 otherwise. |
| 1711 | */ |
| 1712 | |
| 1713 | static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) { |
| 1714 | int i; |
| 1715 | xmlNodePtr lastChild; |
| 1716 | |
| 1717 | for (i = 0;i < len;i++) |
| 1718 | if (!(IS_BLANK(str[i]))) return(0); |
| 1719 | |
| 1720 | if (CUR == 0) return(1); |
| 1721 | if (CUR != '<') return(0); |
| 1722 | if (ctxt->name == NULL) |
| 1723 | return(1); |
| 1724 | if (xmlStrEqual(ctxt->name, BAD_CAST"html")) |
| 1725 | return(1); |
| 1726 | if (xmlStrEqual(ctxt->name, BAD_CAST"head")) |
| 1727 | return(1); |
| 1728 | if (xmlStrEqual(ctxt->name, BAD_CAST"body")) |
| 1729 | return(1); |
| 1730 | if (ctxt->node == NULL) return(0); |
| 1731 | lastChild = xmlGetLastChild(ctxt->node); |
| 1732 | if (lastChild == NULL) { |
| 1733 | if (ctxt->node->content != NULL) return(0); |
| 1734 | } else if (xmlNodeIsText(lastChild)) { |
| 1735 | return(0); |
| 1736 | } else if (xmlStrEqual(lastChild->name, BAD_CAST"b")) { |
| 1737 | return(0); |
| 1738 | } else if (xmlStrEqual(lastChild->name, BAD_CAST"bold")) { |
| 1739 | return(0); |
| 1740 | } else if (xmlStrEqual(lastChild->name, BAD_CAST"em")) { |
| 1741 | return(0); |
| 1742 | } |
| 1743 | return(1); |
| 1744 | } |
| 1745 | |
| 1746 | /** |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1747 | * htmlNewDocNoDtD: |
| 1748 | * @URI: URI for the dtd, or NULL |
| 1749 | * @ExternalID: the external ID of the DTD, or NULL |
| 1750 | * |
| 1751 | * Returns a new document, do not intialize the DTD if not provided |
| 1752 | */ |
| 1753 | htmlDocPtr |
| 1754 | htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) { |
| 1755 | xmlDocPtr cur; |
| 1756 | |
| 1757 | /* |
| 1758 | * Allocate a new document and fill the fields. |
| 1759 | */ |
| 1760 | cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc)); |
| 1761 | if (cur == NULL) { |
| 1762 | xmlGenericError(xmlGenericErrorContext, |
| 1763 | "xmlNewDoc : malloc failed\n"); |
| 1764 | return(NULL); |
| 1765 | } |
| 1766 | memset(cur, 0, sizeof(xmlDoc)); |
| 1767 | |
| 1768 | cur->type = XML_HTML_DOCUMENT_NODE; |
| 1769 | cur->version = NULL; |
| 1770 | cur->intSubset = NULL; |
| 1771 | if ((ExternalID != NULL) || |
| 1772 | (URI != NULL)) |
| 1773 | xmlCreateIntSubset(cur, BAD_CAST "HTML", ExternalID, URI); |
| 1774 | cur->doc = cur; |
| 1775 | cur->name = NULL; |
| 1776 | cur->children = NULL; |
| 1777 | cur->extSubset = NULL; |
| 1778 | cur->oldNs = NULL; |
| 1779 | cur->encoding = NULL; |
| 1780 | cur->standalone = 1; |
| 1781 | cur->compression = 0; |
| 1782 | cur->ids = NULL; |
| 1783 | cur->refs = NULL; |
| 1784 | #ifndef XML_WITHOUT_CORBA |
| 1785 | cur->_private = NULL; |
| 1786 | #endif |
| 1787 | return(cur); |
| 1788 | } |
| 1789 | |
| 1790 | /** |
| 1791 | * htmlNewDoc: |
| 1792 | * @URI: URI for the dtd, or NULL |
| 1793 | * @ExternalID: the external ID of the DTD, or NULL |
| 1794 | * |
| 1795 | * Returns a new document |
| 1796 | */ |
| 1797 | htmlDocPtr |
| 1798 | htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) { |
| 1799 | if ((URI == NULL) && (ExternalID == NULL)) |
| 1800 | return(htmlNewDocNoDtD( |
| 1801 | BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN", |
| 1802 | BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd")); |
| 1803 | |
| 1804 | return(htmlNewDocNoDtD(URI, ExternalID)); |
| 1805 | } |
| 1806 | |
| 1807 | |
| 1808 | /************************************************************************ |
| 1809 | * * |
| 1810 | * The parser itself * |
| 1811 | * Relates to http://www.w3.org/TR/html40 * |
| 1812 | * * |
| 1813 | ************************************************************************/ |
| 1814 | |
| 1815 | /************************************************************************ |
| 1816 | * * |
| 1817 | * The parser itself * |
| 1818 | * * |
| 1819 | ************************************************************************/ |
| 1820 | |
| 1821 | /** |
| 1822 | * htmlParseHTMLName: |
| 1823 | * @ctxt: an HTML parser context |
| 1824 | * |
| 1825 | * parse an HTML tag or attribute name, note that we convert it to lowercase |
| 1826 | * since HTML names are not case-sensitive. |
| 1827 | * |
| 1828 | * Returns the Tag Name parsed or NULL |
| 1829 | */ |
| 1830 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1831 | static xmlChar * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1832 | htmlParseHTMLName(htmlParserCtxtPtr ctxt) { |
| 1833 | xmlChar *ret = NULL; |
| 1834 | int i = 0; |
| 1835 | xmlChar loc[HTML_PARSER_BUFFER_SIZE]; |
| 1836 | |
| 1837 | if (!IS_LETTER(CUR) && (CUR != '_') && |
| 1838 | (CUR != ':')) return(NULL); |
| 1839 | |
| 1840 | while ((i < HTML_PARSER_BUFFER_SIZE) && |
| 1841 | ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || |
| 1842 | (CUR == ':') || (CUR == '-') || (CUR == '_'))) { |
| 1843 | if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20; |
| 1844 | else loc[i] = CUR; |
| 1845 | i++; |
| 1846 | |
| 1847 | NEXT; |
| 1848 | } |
| 1849 | |
| 1850 | ret = xmlStrndup(loc, i); |
| 1851 | |
| 1852 | return(ret); |
| 1853 | } |
| 1854 | |
| 1855 | /** |
| 1856 | * htmlParseName: |
| 1857 | * @ctxt: an HTML parser context |
| 1858 | * |
| 1859 | * parse an HTML name, this routine is case sensistive. |
| 1860 | * |
| 1861 | * Returns the Name parsed or NULL |
| 1862 | */ |
| 1863 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1864 | static xmlChar * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1865 | htmlParseName(htmlParserCtxtPtr ctxt) { |
| 1866 | xmlChar buf[HTML_MAX_NAMELEN]; |
| 1867 | int len = 0; |
| 1868 | |
| 1869 | GROW; |
| 1870 | if (!IS_LETTER(CUR) && (CUR != '_')) { |
| 1871 | return(NULL); |
| 1872 | } |
| 1873 | |
| 1874 | while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || |
| 1875 | (CUR == '.') || (CUR == '-') || |
| 1876 | (CUR == '_') || (CUR == ':') || |
| 1877 | (IS_COMBINING(CUR)) || |
| 1878 | (IS_EXTENDER(CUR))) { |
| 1879 | buf[len++] = CUR; |
| 1880 | NEXT; |
| 1881 | if (len >= HTML_MAX_NAMELEN) { |
| 1882 | xmlGenericError(xmlGenericErrorContext, |
| 1883 | "htmlParseName: reached HTML_MAX_NAMELEN limit\n"); |
| 1884 | while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || |
| 1885 | (CUR == '.') || (CUR == '-') || |
| 1886 | (CUR == '_') || (CUR == ':') || |
| 1887 | (IS_COMBINING(CUR)) || |
| 1888 | (IS_EXTENDER(CUR))) |
| 1889 | NEXT; |
| 1890 | break; |
| 1891 | } |
| 1892 | } |
| 1893 | return(xmlStrndup(buf, len)); |
| 1894 | } |
| 1895 | |
| 1896 | /** |
| 1897 | * htmlParseHTMLAttribute: |
| 1898 | * @ctxt: an HTML parser context |
| 1899 | * @stop: a char stop value |
| 1900 | * |
| 1901 | * parse an HTML attribute value till the stop (quote), if |
| 1902 | * stop is 0 then it stops at the first space |
| 1903 | * |
| 1904 | * Returns the attribute parsed or NULL |
| 1905 | */ |
| 1906 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1907 | static xmlChar * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1908 | htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) { |
| 1909 | xmlChar *buffer = NULL; |
| 1910 | int buffer_size = 0; |
| 1911 | xmlChar *out = NULL; |
| 1912 | xmlChar *name = NULL; |
| 1913 | |
| 1914 | xmlChar *cur = NULL; |
| 1915 | htmlEntityDescPtr ent; |
| 1916 | |
| 1917 | /* |
| 1918 | * allocate a translation buffer. |
| 1919 | */ |
| 1920 | buffer_size = HTML_PARSER_BUFFER_SIZE; |
| 1921 | buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar)); |
| 1922 | if (buffer == NULL) { |
| 1923 | perror("htmlParseHTMLAttribute: malloc failed"); |
| 1924 | return(NULL); |
| 1925 | } |
| 1926 | out = buffer; |
| 1927 | |
| 1928 | /* |
| 1929 | * Ok loop until we reach one of the ending chars |
| 1930 | */ |
| 1931 | while ((CUR != 0) && (CUR != stop) && (CUR != '>')) { |
| 1932 | if ((stop == 0) && (IS_BLANK(CUR))) break; |
| 1933 | if (CUR == '&') { |
| 1934 | if (NXT(1) == '#') { |
| 1935 | unsigned int c; |
| 1936 | int bits; |
| 1937 | |
| 1938 | c = htmlParseCharRef(ctxt); |
| 1939 | if (c < 0x80) |
| 1940 | { *out++ = c; bits= -6; } |
| 1941 | else if (c < 0x800) |
| 1942 | { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; } |
| 1943 | else if (c < 0x10000) |
| 1944 | { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; } |
| 1945 | else |
| 1946 | { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; } |
| 1947 | |
| 1948 | for ( ; bits >= 0; bits-= 6) { |
| 1949 | *out++ = ((c >> bits) & 0x3F) | 0x80; |
| 1950 | } |
| 1951 | } else { |
| 1952 | ent = htmlParseEntityRef(ctxt, &name); |
| 1953 | if (name == NULL) { |
| 1954 | *out++ = '&'; |
| 1955 | if (out - buffer > buffer_size - 100) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1956 | int indx = out - buffer; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1957 | |
| 1958 | growBuffer(buffer); |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1959 | out = &buffer[indx]; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1960 | } |
| 1961 | } else if (ent == NULL) { |
| 1962 | *out++ = '&'; |
| 1963 | cur = name; |
| 1964 | while (*cur != 0) { |
| 1965 | if (out - buffer > buffer_size - 100) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1966 | int indx = out - buffer; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1967 | |
| 1968 | growBuffer(buffer); |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1969 | out = &buffer[indx]; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1970 | } |
| 1971 | *out++ = *cur++; |
| 1972 | } |
| 1973 | xmlFree(name); |
| 1974 | } else { |
| 1975 | unsigned int c; |
| 1976 | int bits; |
| 1977 | |
| 1978 | if (out - buffer > buffer_size - 100) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1979 | int indx = out - buffer; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1980 | |
| 1981 | growBuffer(buffer); |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 1982 | out = &buffer[indx]; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 1983 | } |
| 1984 | c = (xmlChar)ent->value; |
| 1985 | if (c < 0x80) |
| 1986 | { *out++ = c; bits= -6; } |
| 1987 | else if (c < 0x800) |
| 1988 | { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; } |
| 1989 | else if (c < 0x10000) |
| 1990 | { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; } |
| 1991 | else |
| 1992 | { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; } |
| 1993 | |
| 1994 | for ( ; bits >= 0; bits-= 6) { |
| 1995 | *out++ = ((c >> bits) & 0x3F) | 0x80; |
| 1996 | } |
| 1997 | xmlFree(name); |
| 1998 | } |
| 1999 | } |
| 2000 | } else { |
| 2001 | unsigned int c; |
| 2002 | int bits, l; |
| 2003 | |
| 2004 | if (out - buffer > buffer_size - 100) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2005 | int indx = out - buffer; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2006 | |
| 2007 | growBuffer(buffer); |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2008 | out = &buffer[indx]; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2009 | } |
| 2010 | c = CUR_CHAR(l); |
| 2011 | if (c < 0x80) |
| 2012 | { *out++ = c; bits= -6; } |
| 2013 | else if (c < 0x800) |
| 2014 | { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; } |
| 2015 | else if (c < 0x10000) |
| 2016 | { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; } |
| 2017 | else |
| 2018 | { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; } |
| 2019 | |
| 2020 | for ( ; bits >= 0; bits-= 6) { |
| 2021 | *out++ = ((c >> bits) & 0x3F) | 0x80; |
| 2022 | } |
| 2023 | NEXT; |
| 2024 | } |
| 2025 | } |
| 2026 | *out++ = 0; |
| 2027 | return(buffer); |
| 2028 | } |
| 2029 | |
| 2030 | /** |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2031 | * htmlParseEntityRef: |
| 2032 | * @ctxt: an HTML parser context |
| 2033 | * @str: location to store the entity name |
| 2034 | * |
| 2035 | * parse an HTML ENTITY references |
| 2036 | * |
| 2037 | * [68] EntityRef ::= '&' Name ';' |
| 2038 | * |
| 2039 | * Returns the associated htmlEntityDescPtr if found, or NULL otherwise, |
| 2040 | * if non-NULL *str will have to be freed by the caller. |
| 2041 | */ |
| 2042 | htmlEntityDescPtr |
| 2043 | htmlParseEntityRef(htmlParserCtxtPtr ctxt, xmlChar **str) { |
| 2044 | xmlChar *name; |
| 2045 | htmlEntityDescPtr ent = NULL; |
| 2046 | *str = NULL; |
| 2047 | |
| 2048 | if (CUR == '&') { |
| 2049 | NEXT; |
| 2050 | name = htmlParseName(ctxt); |
| 2051 | if (name == NULL) { |
| 2052 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2053 | ctxt->sax->error(ctxt->userData, "htmlParseEntityRef: no name\n"); |
| 2054 | ctxt->wellFormed = 0; |
| 2055 | } else { |
| 2056 | GROW; |
| 2057 | if (CUR == ';') { |
| 2058 | *str = name; |
| 2059 | |
| 2060 | /* |
| 2061 | * Lookup the entity in the table. |
| 2062 | */ |
| 2063 | ent = htmlEntityLookup(name); |
| 2064 | if (ent != NULL) /* OK that's ugly !!! */ |
| 2065 | NEXT; |
| 2066 | } else { |
| 2067 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2068 | ctxt->sax->error(ctxt->userData, |
| 2069 | "htmlParseEntityRef: expecting ';'\n"); |
| 2070 | *str = name; |
| 2071 | } |
| 2072 | } |
| 2073 | } |
| 2074 | return(ent); |
| 2075 | } |
| 2076 | |
| 2077 | /** |
| 2078 | * htmlParseAttValue: |
| 2079 | * @ctxt: an HTML parser context |
| 2080 | * |
| 2081 | * parse a value for an attribute |
| 2082 | * Note: the parser won't do substitution of entities here, this |
| 2083 | * will be handled later in xmlStringGetNodeList, unless it was |
| 2084 | * asked for ctxt->replaceEntities != 0 |
| 2085 | * |
| 2086 | * Returns the AttValue parsed or NULL. |
| 2087 | */ |
| 2088 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2089 | static xmlChar * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2090 | htmlParseAttValue(htmlParserCtxtPtr ctxt) { |
| 2091 | xmlChar *ret = NULL; |
| 2092 | |
| 2093 | if (CUR == '"') { |
| 2094 | NEXT; |
| 2095 | ret = htmlParseHTMLAttribute(ctxt, '"'); |
| 2096 | if (CUR != '"') { |
| 2097 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2098 | ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n"); |
| 2099 | ctxt->wellFormed = 0; |
| 2100 | } else |
| 2101 | NEXT; |
| 2102 | } else if (CUR == '\'') { |
| 2103 | NEXT; |
| 2104 | ret = htmlParseHTMLAttribute(ctxt, '\''); |
| 2105 | if (CUR != '\'') { |
| 2106 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2107 | ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n"); |
| 2108 | ctxt->wellFormed = 0; |
| 2109 | } else |
| 2110 | NEXT; |
| 2111 | } else { |
| 2112 | /* |
| 2113 | * That's an HTMLism, the attribute value may not be quoted |
| 2114 | */ |
| 2115 | ret = htmlParseHTMLAttribute(ctxt, 0); |
| 2116 | if (ret == NULL) { |
| 2117 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2118 | ctxt->sax->error(ctxt->userData, "AttValue: no value found\n"); |
| 2119 | ctxt->wellFormed = 0; |
| 2120 | } |
| 2121 | } |
| 2122 | return(ret); |
| 2123 | } |
| 2124 | |
| 2125 | /** |
| 2126 | * htmlParseSystemLiteral: |
| 2127 | * @ctxt: an HTML parser context |
| 2128 | * |
| 2129 | * parse an HTML Literal |
| 2130 | * |
| 2131 | * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'") |
| 2132 | * |
| 2133 | * Returns the SystemLiteral parsed or NULL |
| 2134 | */ |
| 2135 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2136 | static xmlChar * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2137 | htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) { |
| 2138 | const xmlChar *q; |
| 2139 | xmlChar *ret = NULL; |
| 2140 | |
| 2141 | if (CUR == '"') { |
| 2142 | NEXT; |
| 2143 | q = CUR_PTR; |
| 2144 | while ((IS_CHAR(CUR)) && (CUR != '"')) |
| 2145 | NEXT; |
| 2146 | if (!IS_CHAR(CUR)) { |
| 2147 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2148 | ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n"); |
| 2149 | ctxt->wellFormed = 0; |
| 2150 | } else { |
| 2151 | ret = xmlStrndup(q, CUR_PTR - q); |
| 2152 | NEXT; |
| 2153 | } |
| 2154 | } else if (CUR == '\'') { |
| 2155 | NEXT; |
| 2156 | q = CUR_PTR; |
| 2157 | while ((IS_CHAR(CUR)) && (CUR != '\'')) |
| 2158 | NEXT; |
| 2159 | if (!IS_CHAR(CUR)) { |
| 2160 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2161 | ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n"); |
| 2162 | ctxt->wellFormed = 0; |
| 2163 | } else { |
| 2164 | ret = xmlStrndup(q, CUR_PTR - q); |
| 2165 | NEXT; |
| 2166 | } |
| 2167 | } else { |
| 2168 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2169 | ctxt->sax->error(ctxt->userData, |
| 2170 | "SystemLiteral \" or ' expected\n"); |
| 2171 | ctxt->wellFormed = 0; |
| 2172 | } |
| 2173 | |
| 2174 | return(ret); |
| 2175 | } |
| 2176 | |
| 2177 | /** |
| 2178 | * htmlParsePubidLiteral: |
| 2179 | * @ctxt: an HTML parser context |
| 2180 | * |
| 2181 | * parse an HTML public literal |
| 2182 | * |
| 2183 | * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'" |
| 2184 | * |
| 2185 | * Returns the PubidLiteral parsed or NULL. |
| 2186 | */ |
| 2187 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2188 | static xmlChar * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2189 | htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) { |
| 2190 | const xmlChar *q; |
| 2191 | xmlChar *ret = NULL; |
| 2192 | /* |
| 2193 | * Name ::= (Letter | '_') (NameChar)* |
| 2194 | */ |
| 2195 | if (CUR == '"') { |
| 2196 | NEXT; |
| 2197 | q = CUR_PTR; |
| 2198 | while (IS_PUBIDCHAR(CUR)) NEXT; |
| 2199 | if (CUR != '"') { |
| 2200 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2201 | ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n"); |
| 2202 | ctxt->wellFormed = 0; |
| 2203 | } else { |
| 2204 | ret = xmlStrndup(q, CUR_PTR - q); |
| 2205 | NEXT; |
| 2206 | } |
| 2207 | } else if (CUR == '\'') { |
| 2208 | NEXT; |
| 2209 | q = CUR_PTR; |
| 2210 | while ((IS_LETTER(CUR)) && (CUR != '\'')) |
| 2211 | NEXT; |
| 2212 | if (!IS_LETTER(CUR)) { |
| 2213 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2214 | ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n"); |
| 2215 | ctxt->wellFormed = 0; |
| 2216 | } else { |
| 2217 | ret = xmlStrndup(q, CUR_PTR - q); |
| 2218 | NEXT; |
| 2219 | } |
| 2220 | } else { |
| 2221 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2222 | ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n"); |
| 2223 | ctxt->wellFormed = 0; |
| 2224 | } |
| 2225 | |
| 2226 | return(ret); |
| 2227 | } |
| 2228 | |
| 2229 | /** |
| 2230 | * htmlParseScript: |
| 2231 | * @ctxt: an HTML parser context |
| 2232 | * |
| 2233 | * parse the content of an HTML SCRIPT or STYLE element |
| 2234 | * http://www.w3.org/TR/html4/sgml/dtd.html#Script |
| 2235 | * http://www.w3.org/TR/html4/sgml/dtd.html#StyleSheet |
| 2236 | * http://www.w3.org/TR/html4/types.html#type-script |
| 2237 | * http://www.w3.org/TR/html4/types.html#h-6.15 |
| 2238 | * http://www.w3.org/TR/html4/appendix/notes.html#h-B.3.2.1 |
| 2239 | * |
| 2240 | * Script data ( %Script; in the DTD) can be the content of the SCRIPT |
| 2241 | * element and the value of intrinsic event attributes. User agents must |
| 2242 | * not evaluate script data as HTML markup but instead must pass it on as |
| 2243 | * data to a script engine. |
| 2244 | * NOTES: |
| 2245 | * - The content is passed like CDATA |
| 2246 | * - the attributes for style and scripting "onXXX" are also described |
| 2247 | * as CDATA but SGML allows entities references in attributes so their |
| 2248 | * processing is identical as other attributes |
| 2249 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2250 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2251 | htmlParseScript(htmlParserCtxtPtr ctxt) { |
| 2252 | xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 1]; |
| 2253 | int nbchar = 0; |
| 2254 | xmlChar cur; |
| 2255 | |
| 2256 | SHRINK; |
| 2257 | cur = CUR; |
| 2258 | while (IS_CHAR(cur)) { |
| 2259 | if ((cur == '<') && (NXT(1) == '/')) { |
| 2260 | /* |
| 2261 | * One should break here, the specification is clear: |
| 2262 | * Authors should therefore escape "</" within the content. |
| 2263 | * Escape mechanisms are specific to each scripting or |
| 2264 | * style sheet language. |
| 2265 | */ |
| 2266 | if (((NXT(2) >= 'A') && (NXT(2) <= 'Z')) || |
| 2267 | ((NXT(2) >= 'a') && (NXT(2) <= 'z'))) |
| 2268 | break; /* while */ |
| 2269 | } |
| 2270 | buf[nbchar++] = cur; |
| 2271 | if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) { |
| 2272 | if (ctxt->sax->cdataBlock!= NULL) { |
| 2273 | /* |
| 2274 | * Insert as CDATA, which is the same as HTML_PRESERVE_NODE |
| 2275 | */ |
| 2276 | ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar); |
| 2277 | } |
| 2278 | nbchar = 0; |
| 2279 | } |
| 2280 | NEXT; |
| 2281 | cur = CUR; |
| 2282 | } |
| 2283 | if (!(IS_CHAR(cur))) { |
| 2284 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2285 | ctxt->sax->error(ctxt->userData, |
| 2286 | "Invalid char in CDATA 0x%X\n", cur); |
| 2287 | ctxt->wellFormed = 0; |
| 2288 | NEXT; |
| 2289 | } |
| 2290 | |
| 2291 | if ((nbchar != 0) && (ctxt->sax != NULL) && (!ctxt->disableSAX)) { |
| 2292 | if (ctxt->sax->cdataBlock!= NULL) { |
| 2293 | /* |
| 2294 | * Insert as CDATA, which is the same as HTML_PRESERVE_NODE |
| 2295 | */ |
| 2296 | ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar); |
| 2297 | } |
| 2298 | } |
| 2299 | } |
| 2300 | |
| 2301 | |
| 2302 | /** |
| 2303 | * htmlParseCharData: |
| 2304 | * @ctxt: an HTML parser context |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2305 | * |
| 2306 | * parse a CharData section. |
| 2307 | * if we are within a CDATA section ']]>' marks an end of section. |
| 2308 | * |
| 2309 | * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*) |
| 2310 | */ |
| 2311 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2312 | static void |
| 2313 | htmlParseCharData(htmlParserCtxtPtr ctxt) { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2314 | xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5]; |
| 2315 | int nbchar = 0; |
| 2316 | int cur, l; |
| 2317 | |
| 2318 | SHRINK; |
| 2319 | cur = CUR_CHAR(l); |
| 2320 | while (((cur != '<') || (ctxt->token == '<')) && |
| 2321 | ((cur != '&') || (ctxt->token == '&')) && |
| 2322 | (IS_CHAR(cur))) { |
| 2323 | COPY_BUF(l,buf,nbchar,cur); |
| 2324 | if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) { |
| 2325 | /* |
| 2326 | * Ok the segment is to be consumed as chars. |
| 2327 | */ |
| 2328 | if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) { |
| 2329 | if (areBlanks(ctxt, buf, nbchar)) { |
| 2330 | if (ctxt->sax->ignorableWhitespace != NULL) |
| 2331 | ctxt->sax->ignorableWhitespace(ctxt->userData, |
| 2332 | buf, nbchar); |
| 2333 | } else { |
| 2334 | htmlCheckParagraph(ctxt); |
| 2335 | if (ctxt->sax->characters != NULL) |
| 2336 | ctxt->sax->characters(ctxt->userData, buf, nbchar); |
| 2337 | } |
| 2338 | } |
| 2339 | nbchar = 0; |
| 2340 | } |
| 2341 | NEXTL(l); |
| 2342 | cur = CUR_CHAR(l); |
| 2343 | } |
| 2344 | if (nbchar != 0) { |
| 2345 | /* |
| 2346 | * Ok the segment is to be consumed as chars. |
| 2347 | */ |
| 2348 | if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) { |
| 2349 | if (areBlanks(ctxt, buf, nbchar)) { |
| 2350 | if (ctxt->sax->ignorableWhitespace != NULL) |
| 2351 | ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar); |
| 2352 | } else { |
| 2353 | htmlCheckParagraph(ctxt); |
| 2354 | if (ctxt->sax->characters != NULL) |
| 2355 | ctxt->sax->characters(ctxt->userData, buf, nbchar); |
| 2356 | } |
| 2357 | } |
| 2358 | } |
| 2359 | } |
| 2360 | |
| 2361 | /** |
| 2362 | * htmlParseExternalID: |
| 2363 | * @ctxt: an HTML parser context |
| 2364 | * @publicID: a xmlChar** receiving PubidLiteral |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2365 | * |
| 2366 | * Parse an External ID or a Public ID |
| 2367 | * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2368 | * [75] ExternalID ::= 'SYSTEM' S SystemLiteral |
| 2369 | * | 'PUBLIC' S PubidLiteral S SystemLiteral |
| 2370 | * |
| 2371 | * [83] PublicID ::= 'PUBLIC' S PubidLiteral |
| 2372 | * |
| 2373 | * Returns the function returns SystemLiteral and in the second |
| 2374 | * case publicID receives PubidLiteral, is strict is off |
| 2375 | * it is possible to return NULL and have publicID set. |
| 2376 | */ |
| 2377 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2378 | static xmlChar * |
| 2379 | htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID) { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2380 | xmlChar *URI = NULL; |
| 2381 | |
| 2382 | if ((UPPER == 'S') && (UPP(1) == 'Y') && |
| 2383 | (UPP(2) == 'S') && (UPP(3) == 'T') && |
| 2384 | (UPP(4) == 'E') && (UPP(5) == 'M')) { |
| 2385 | SKIP(6); |
| 2386 | if (!IS_BLANK(CUR)) { |
| 2387 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2388 | ctxt->sax->error(ctxt->userData, |
| 2389 | "Space required after 'SYSTEM'\n"); |
| 2390 | ctxt->wellFormed = 0; |
| 2391 | } |
| 2392 | SKIP_BLANKS; |
| 2393 | URI = htmlParseSystemLiteral(ctxt); |
| 2394 | if (URI == NULL) { |
| 2395 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2396 | ctxt->sax->error(ctxt->userData, |
| 2397 | "htmlParseExternalID: SYSTEM, no URI\n"); |
| 2398 | ctxt->wellFormed = 0; |
| 2399 | } |
| 2400 | } else if ((UPPER == 'P') && (UPP(1) == 'U') && |
| 2401 | (UPP(2) == 'B') && (UPP(3) == 'L') && |
| 2402 | (UPP(4) == 'I') && (UPP(5) == 'C')) { |
| 2403 | SKIP(6); |
| 2404 | if (!IS_BLANK(CUR)) { |
| 2405 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2406 | ctxt->sax->error(ctxt->userData, |
| 2407 | "Space required after 'PUBLIC'\n"); |
| 2408 | ctxt->wellFormed = 0; |
| 2409 | } |
| 2410 | SKIP_BLANKS; |
| 2411 | *publicID = htmlParsePubidLiteral(ctxt); |
| 2412 | if (*publicID == NULL) { |
| 2413 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2414 | ctxt->sax->error(ctxt->userData, |
| 2415 | "htmlParseExternalID: PUBLIC, no Public Identifier\n"); |
| 2416 | ctxt->wellFormed = 0; |
| 2417 | } |
| 2418 | SKIP_BLANKS; |
| 2419 | if ((CUR == '"') || (CUR == '\'')) { |
| 2420 | URI = htmlParseSystemLiteral(ctxt); |
| 2421 | } |
| 2422 | } |
| 2423 | return(URI); |
| 2424 | } |
| 2425 | |
| 2426 | /** |
| 2427 | * htmlParseComment: |
| 2428 | * @ctxt: an HTML parser context |
| 2429 | * |
| 2430 | * Parse an XML (SGML) comment <!-- .... --> |
| 2431 | * |
| 2432 | * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->' |
| 2433 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2434 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2435 | htmlParseComment(htmlParserCtxtPtr ctxt) { |
| 2436 | xmlChar *buf = NULL; |
| 2437 | int len; |
| 2438 | int size = HTML_PARSER_BUFFER_SIZE; |
| 2439 | int q, ql; |
| 2440 | int r, rl; |
| 2441 | int cur, l; |
| 2442 | xmlParserInputState state; |
| 2443 | |
| 2444 | /* |
| 2445 | * Check that there is a comment right here. |
| 2446 | */ |
| 2447 | if ((RAW != '<') || (NXT(1) != '!') || |
| 2448 | (NXT(2) != '-') || (NXT(3) != '-')) return; |
| 2449 | |
| 2450 | state = ctxt->instate; |
| 2451 | ctxt->instate = XML_PARSER_COMMENT; |
| 2452 | SHRINK; |
| 2453 | SKIP(4); |
| 2454 | buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar)); |
| 2455 | if (buf == NULL) { |
| 2456 | xmlGenericError(xmlGenericErrorContext, |
| 2457 | "malloc of %d byte failed\n", size); |
| 2458 | ctxt->instate = state; |
| 2459 | return; |
| 2460 | } |
| 2461 | q = CUR_CHAR(ql); |
| 2462 | NEXTL(ql); |
| 2463 | r = CUR_CHAR(rl); |
| 2464 | NEXTL(rl); |
| 2465 | cur = CUR_CHAR(l); |
| 2466 | len = 0; |
| 2467 | while (IS_CHAR(cur) && |
| 2468 | ((cur != '>') || |
| 2469 | (r != '-') || (q != '-'))) { |
| 2470 | if (len + 5 >= size) { |
| 2471 | size *= 2; |
| 2472 | buf = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar)); |
| 2473 | if (buf == NULL) { |
| 2474 | xmlGenericError(xmlGenericErrorContext, |
| 2475 | "realloc of %d byte failed\n", size); |
| 2476 | ctxt->instate = state; |
| 2477 | return; |
| 2478 | } |
| 2479 | } |
| 2480 | COPY_BUF(ql,buf,len,q); |
| 2481 | q = r; |
| 2482 | ql = rl; |
| 2483 | r = cur; |
| 2484 | rl = l; |
| 2485 | NEXTL(l); |
| 2486 | cur = CUR_CHAR(l); |
| 2487 | if (cur == 0) { |
| 2488 | SHRINK; |
| 2489 | GROW; |
| 2490 | cur = CUR_CHAR(l); |
| 2491 | } |
| 2492 | } |
| 2493 | buf[len] = 0; |
| 2494 | if (!IS_CHAR(cur)) { |
| 2495 | ctxt->errNo = XML_ERR_COMMENT_NOT_FINISHED; |
| 2496 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2497 | ctxt->sax->error(ctxt->userData, |
| 2498 | "Comment not terminated \n<!--%.50s\n", buf); |
| 2499 | ctxt->wellFormed = 0; |
| 2500 | xmlFree(buf); |
| 2501 | } else { |
| 2502 | NEXT; |
| 2503 | if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) && |
| 2504 | (!ctxt->disableSAX)) |
| 2505 | ctxt->sax->comment(ctxt->userData, buf); |
| 2506 | xmlFree(buf); |
| 2507 | } |
| 2508 | ctxt->instate = state; |
| 2509 | } |
| 2510 | |
| 2511 | /** |
| 2512 | * htmlParseCharRef: |
| 2513 | * @ctxt: an HTML parser context |
| 2514 | * |
| 2515 | * parse Reference declarations |
| 2516 | * |
| 2517 | * [66] CharRef ::= '&#' [0-9]+ ';' | |
| 2518 | * '&#x' [0-9a-fA-F]+ ';' |
| 2519 | * |
| 2520 | * Returns the value parsed (as an int) |
| 2521 | */ |
| 2522 | int |
| 2523 | htmlParseCharRef(htmlParserCtxtPtr ctxt) { |
| 2524 | int val = 0; |
| 2525 | |
| 2526 | if ((CUR == '&') && (NXT(1) == '#') && |
| 2527 | (NXT(2) == 'x')) { |
| 2528 | SKIP(3); |
| 2529 | while (CUR != ';') { |
| 2530 | if ((CUR >= '0') && (CUR <= '9')) |
| 2531 | val = val * 16 + (CUR - '0'); |
| 2532 | else if ((CUR >= 'a') && (CUR <= 'f')) |
| 2533 | val = val * 16 + (CUR - 'a') + 10; |
| 2534 | else if ((CUR >= 'A') && (CUR <= 'F')) |
| 2535 | val = val * 16 + (CUR - 'A') + 10; |
| 2536 | else { |
| 2537 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2538 | ctxt->sax->error(ctxt->userData, |
| 2539 | "htmlParseCharRef: invalid hexadecimal value\n"); |
| 2540 | ctxt->wellFormed = 0; |
| 2541 | return(0); |
| 2542 | } |
| 2543 | NEXT; |
| 2544 | } |
| 2545 | if (CUR == ';') |
| 2546 | NEXT; |
| 2547 | } else if ((CUR == '&') && (NXT(1) == '#')) { |
| 2548 | SKIP(2); |
| 2549 | while (CUR != ';') { |
| 2550 | if ((CUR >= '0') && (CUR <= '9')) |
| 2551 | val = val * 10 + (CUR - '0'); |
| 2552 | else { |
| 2553 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2554 | ctxt->sax->error(ctxt->userData, |
| 2555 | "htmlParseCharRef: invalid decimal value\n"); |
| 2556 | ctxt->wellFormed = 0; |
| 2557 | return(0); |
| 2558 | } |
| 2559 | NEXT; |
| 2560 | } |
| 2561 | if (CUR == ';') |
| 2562 | NEXT; |
| 2563 | } else { |
| 2564 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2565 | ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid value\n"); |
| 2566 | ctxt->wellFormed = 0; |
| 2567 | } |
| 2568 | /* |
| 2569 | * Check the value IS_CHAR ... |
| 2570 | */ |
| 2571 | if (IS_CHAR(val)) { |
| 2572 | return(val); |
| 2573 | } else { |
| 2574 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2575 | ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid xmlChar value %d\n", |
| 2576 | val); |
| 2577 | ctxt->wellFormed = 0; |
| 2578 | } |
| 2579 | return(0); |
| 2580 | } |
| 2581 | |
| 2582 | |
| 2583 | /** |
| 2584 | * htmlParseDocTypeDecl : |
| 2585 | * @ctxt: an HTML parser context |
| 2586 | * |
| 2587 | * parse a DOCTYPE declaration |
| 2588 | * |
| 2589 | * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S? |
| 2590 | * ('[' (markupdecl | PEReference | S)* ']' S?)? '>' |
| 2591 | */ |
| 2592 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2593 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2594 | htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) { |
| 2595 | xmlChar *name; |
| 2596 | xmlChar *ExternalID = NULL; |
| 2597 | xmlChar *URI = NULL; |
| 2598 | |
| 2599 | /* |
| 2600 | * We know that '<!DOCTYPE' has been detected. |
| 2601 | */ |
| 2602 | SKIP(9); |
| 2603 | |
| 2604 | SKIP_BLANKS; |
| 2605 | |
| 2606 | /* |
| 2607 | * Parse the DOCTYPE name. |
| 2608 | */ |
| 2609 | name = htmlParseName(ctxt); |
| 2610 | if (name == NULL) { |
| 2611 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2612 | ctxt->sax->error(ctxt->userData, "htmlParseDocTypeDecl : no DOCTYPE name !\n"); |
| 2613 | ctxt->wellFormed = 0; |
| 2614 | } |
| 2615 | /* |
| 2616 | * Check that upper(name) == "HTML" !!!!!!!!!!!!! |
| 2617 | */ |
| 2618 | |
| 2619 | SKIP_BLANKS; |
| 2620 | |
| 2621 | /* |
| 2622 | * Check for SystemID and ExternalID |
| 2623 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2624 | URI = htmlParseExternalID(ctxt, &ExternalID); |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2625 | SKIP_BLANKS; |
| 2626 | |
| 2627 | /* |
| 2628 | * We should be at the end of the DOCTYPE declaration. |
| 2629 | */ |
| 2630 | if (CUR != '>') { |
| 2631 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2632 | ctxt->sax->error(ctxt->userData, "DOCTYPE unproperly terminated\n"); |
| 2633 | ctxt->wellFormed = 0; |
| 2634 | /* We shouldn't try to resynchronize ... */ |
| 2635 | } |
| 2636 | NEXT; |
| 2637 | |
| 2638 | /* |
| 2639 | * Create or update the document accordingly to the DOCTYPE |
| 2640 | */ |
| 2641 | if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) && |
| 2642 | (!ctxt->disableSAX)) |
| 2643 | ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI); |
| 2644 | |
| 2645 | /* |
| 2646 | * Cleanup, since we don't use all those identifiers |
| 2647 | */ |
| 2648 | if (URI != NULL) xmlFree(URI); |
| 2649 | if (ExternalID != NULL) xmlFree(ExternalID); |
| 2650 | if (name != NULL) xmlFree(name); |
| 2651 | } |
| 2652 | |
| 2653 | /** |
| 2654 | * htmlParseAttribute: |
| 2655 | * @ctxt: an HTML parser context |
| 2656 | * @value: a xmlChar ** used to store the value of the attribute |
| 2657 | * |
| 2658 | * parse an attribute |
| 2659 | * |
| 2660 | * [41] Attribute ::= Name Eq AttValue |
| 2661 | * |
| 2662 | * [25] Eq ::= S? '=' S? |
| 2663 | * |
| 2664 | * With namespace: |
| 2665 | * |
| 2666 | * [NS 11] Attribute ::= QName Eq AttValue |
| 2667 | * |
| 2668 | * Also the case QName == xmlns:??? is handled independently as a namespace |
| 2669 | * definition. |
| 2670 | * |
| 2671 | * Returns the attribute name, and the value in *value. |
| 2672 | */ |
| 2673 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2674 | static xmlChar * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2675 | htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) { |
| 2676 | xmlChar *name, *val = NULL; |
| 2677 | |
| 2678 | *value = NULL; |
| 2679 | name = htmlParseHTMLName(ctxt); |
| 2680 | if (name == NULL) { |
| 2681 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2682 | ctxt->sax->error(ctxt->userData, "error parsing attribute name\n"); |
| 2683 | ctxt->wellFormed = 0; |
| 2684 | return(NULL); |
| 2685 | } |
| 2686 | |
| 2687 | /* |
| 2688 | * read the value |
| 2689 | */ |
| 2690 | SKIP_BLANKS; |
| 2691 | if (CUR == '=') { |
| 2692 | NEXT; |
| 2693 | SKIP_BLANKS; |
| 2694 | val = htmlParseAttValue(ctxt); |
| 2695 | /****** |
| 2696 | } else { |
| 2697 | * TODO : some attribute must have values, some may not |
| 2698 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2699 | ctxt->sax->warning(ctxt->userData, |
| 2700 | "No value for attribute %s\n", name); */ |
| 2701 | } |
| 2702 | |
| 2703 | *value = val; |
| 2704 | return(name); |
| 2705 | } |
| 2706 | |
| 2707 | /** |
| 2708 | * htmlCheckEncoding: |
| 2709 | * @ctxt: an HTML parser context |
| 2710 | * @attvalue: the attribute value |
| 2711 | * |
| 2712 | * Checks an http-equiv attribute from a Meta tag to detect |
| 2713 | * the encoding |
| 2714 | * If a new encoding is detected the parser is switched to decode |
| 2715 | * it and pass UTF8 |
| 2716 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2717 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2718 | htmlCheckEncoding(htmlParserCtxtPtr ctxt, const xmlChar *attvalue) { |
| 2719 | const xmlChar *encoding; |
| 2720 | |
| 2721 | if ((ctxt == NULL) || (attvalue == NULL)) |
| 2722 | return; |
| 2723 | |
| 2724 | /* do not change encoding */ |
| 2725 | if (ctxt->input->encoding != NULL) |
| 2726 | return; |
| 2727 | |
| 2728 | encoding = xmlStrcasestr(attvalue, BAD_CAST"charset="); |
| 2729 | if (encoding != NULL) { |
| 2730 | encoding += 8; |
| 2731 | } else { |
| 2732 | encoding = xmlStrcasestr(attvalue, BAD_CAST"charset ="); |
| 2733 | if (encoding != NULL) |
| 2734 | encoding += 9; |
| 2735 | } |
| 2736 | if (encoding != NULL) { |
| 2737 | xmlCharEncoding enc; |
| 2738 | xmlCharEncodingHandlerPtr handler; |
| 2739 | |
| 2740 | while ((*encoding == ' ') || (*encoding == '\t')) encoding++; |
| 2741 | |
| 2742 | if (ctxt->input->encoding != NULL) |
| 2743 | xmlFree((xmlChar *) ctxt->input->encoding); |
| 2744 | ctxt->input->encoding = xmlStrdup(encoding); |
| 2745 | |
| 2746 | enc = xmlParseCharEncoding((const char *) encoding); |
| 2747 | /* |
| 2748 | * registered set of known encodings |
| 2749 | */ |
| 2750 | if (enc != XML_CHAR_ENCODING_ERROR) { |
| 2751 | xmlSwitchEncoding(ctxt, enc); |
| 2752 | ctxt->charset = XML_CHAR_ENCODING_UTF8; |
| 2753 | } else { |
| 2754 | /* |
| 2755 | * fallback for unknown encodings |
| 2756 | */ |
| 2757 | handler = xmlFindCharEncodingHandler((const char *) encoding); |
| 2758 | if (handler != NULL) { |
| 2759 | xmlSwitchToEncoding(ctxt, handler); |
| 2760 | ctxt->charset = XML_CHAR_ENCODING_UTF8; |
| 2761 | } else { |
| 2762 | ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING; |
| 2763 | } |
| 2764 | } |
| 2765 | |
| 2766 | if ((ctxt->input->buf != NULL) && |
| 2767 | (ctxt->input->buf->encoder != NULL) && |
| 2768 | (ctxt->input->buf->raw != NULL) && |
| 2769 | (ctxt->input->buf->buffer != NULL)) { |
| 2770 | int nbchars; |
| 2771 | int processed; |
| 2772 | |
| 2773 | /* |
| 2774 | * convert as much as possible to the parser reading buffer. |
| 2775 | */ |
| 2776 | processed = ctxt->input->cur - ctxt->input->base; |
| 2777 | xmlBufferShrink(ctxt->input->buf->buffer, processed); |
| 2778 | nbchars = xmlCharEncInFunc(ctxt->input->buf->encoder, |
| 2779 | ctxt->input->buf->buffer, |
| 2780 | ctxt->input->buf->raw); |
| 2781 | if (nbchars < 0) { |
| 2782 | ctxt->errNo = XML_ERR_INVALID_ENCODING; |
| 2783 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2784 | ctxt->sax->error(ctxt->userData, |
| 2785 | "htmlCheckEncoding: encoder error\n"); |
| 2786 | } |
| 2787 | ctxt->input->base = |
| 2788 | ctxt->input->cur = ctxt->input->buf->buffer->content; |
| 2789 | } |
| 2790 | } |
| 2791 | } |
| 2792 | |
| 2793 | /** |
| 2794 | * htmlCheckMeta: |
| 2795 | * @ctxt: an HTML parser context |
| 2796 | * @atts: the attributes values |
| 2797 | * |
| 2798 | * Checks an attributes from a Meta tag |
| 2799 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2800 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2801 | htmlCheckMeta(htmlParserCtxtPtr ctxt, const xmlChar **atts) { |
| 2802 | int i; |
| 2803 | const xmlChar *att, *value; |
| 2804 | int http = 0; |
| 2805 | const xmlChar *content = NULL; |
| 2806 | |
| 2807 | if ((ctxt == NULL) || (atts == NULL)) |
| 2808 | return; |
| 2809 | |
| 2810 | i = 0; |
| 2811 | att = atts[i++]; |
| 2812 | while (att != NULL) { |
| 2813 | value = atts[i++]; |
| 2814 | if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"http-equiv")) |
| 2815 | && (!xmlStrcasecmp(value, BAD_CAST"Content-Type"))) |
| 2816 | http = 1; |
| 2817 | else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"content"))) |
| 2818 | content = value; |
| 2819 | att = atts[i++]; |
| 2820 | } |
| 2821 | if ((http) && (content != NULL)) |
| 2822 | htmlCheckEncoding(ctxt, content); |
| 2823 | |
| 2824 | } |
| 2825 | |
| 2826 | /** |
| 2827 | * htmlParseStartTag: |
| 2828 | * @ctxt: an HTML parser context |
| 2829 | * |
| 2830 | * parse a start of tag either for rule element or |
| 2831 | * EmptyElement. In both case we don't parse the tag closing chars. |
| 2832 | * |
| 2833 | * [40] STag ::= '<' Name (S Attribute)* S? '>' |
| 2834 | * |
| 2835 | * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>' |
| 2836 | * |
| 2837 | * With namespace: |
| 2838 | * |
| 2839 | * [NS 8] STag ::= '<' QName (S Attribute)* S? '>' |
| 2840 | * |
| 2841 | * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>' |
| 2842 | * |
| 2843 | */ |
| 2844 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2845 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2846 | htmlParseStartTag(htmlParserCtxtPtr ctxt) { |
| 2847 | xmlChar *name; |
| 2848 | xmlChar *attname; |
| 2849 | xmlChar *attvalue; |
| 2850 | const xmlChar **atts = NULL; |
| 2851 | int nbatts = 0; |
| 2852 | int maxatts = 0; |
| 2853 | int meta = 0; |
| 2854 | int i; |
| 2855 | |
| 2856 | if (CUR != '<') return; |
| 2857 | NEXT; |
| 2858 | |
| 2859 | GROW; |
| 2860 | name = htmlParseHTMLName(ctxt); |
| 2861 | if (name == NULL) { |
| 2862 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2863 | ctxt->sax->error(ctxt->userData, |
| 2864 | "htmlParseStartTag: invalid element name\n"); |
| 2865 | ctxt->wellFormed = 0; |
| 2866 | /* Dump the bogus tag like browsers do */ |
| 2867 | while ((IS_CHAR(CUR)) && (CUR != '>')) |
| 2868 | NEXT; |
| 2869 | return; |
| 2870 | } |
| 2871 | if (xmlStrEqual(name, BAD_CAST"meta")) |
| 2872 | meta = 1; |
| 2873 | |
| 2874 | /* |
| 2875 | * Check for auto-closure of HTML elements. |
| 2876 | */ |
| 2877 | htmlAutoClose(ctxt, name); |
| 2878 | |
| 2879 | /* |
| 2880 | * Check for implied HTML elements. |
| 2881 | */ |
| 2882 | htmlCheckImplied(ctxt, name); |
| 2883 | |
| 2884 | /* |
| 2885 | * Avoid html at any level > 0, head at any level != 1 |
| 2886 | * or any attempt to recurse body |
| 2887 | */ |
| 2888 | if ((ctxt->nameNr > 0) && (xmlStrEqual(name, BAD_CAST"html"))) { |
| 2889 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2890 | ctxt->sax->error(ctxt->userData, |
| 2891 | "htmlParseStartTag: misplaced <html> tag\n"); |
| 2892 | ctxt->wellFormed = 0; |
| 2893 | xmlFree(name); |
| 2894 | return; |
| 2895 | } |
| 2896 | if ((ctxt->nameNr != 1) && |
| 2897 | (xmlStrEqual(name, BAD_CAST"head"))) { |
| 2898 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2899 | ctxt->sax->error(ctxt->userData, |
| 2900 | "htmlParseStartTag: misplaced <head> tag\n"); |
| 2901 | ctxt->wellFormed = 0; |
| 2902 | xmlFree(name); |
| 2903 | return; |
| 2904 | } |
| 2905 | if (xmlStrEqual(name, BAD_CAST"body")) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 2906 | int indx; |
| 2907 | for (indx = 0;indx < ctxt->nameNr;indx++) { |
| 2908 | if (xmlStrEqual(ctxt->nameTab[indx], BAD_CAST"body")) { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 2909 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2910 | ctxt->sax->error(ctxt->userData, |
| 2911 | "htmlParseStartTag: misplaced <body> tag\n"); |
| 2912 | ctxt->wellFormed = 0; |
| 2913 | xmlFree(name); |
| 2914 | return; |
| 2915 | } |
| 2916 | } |
| 2917 | } |
| 2918 | |
| 2919 | /* |
| 2920 | * Now parse the attributes, it ends up with the ending |
| 2921 | * |
| 2922 | * (S Attribute)* S? |
| 2923 | */ |
| 2924 | SKIP_BLANKS; |
| 2925 | while ((IS_CHAR(CUR)) && |
| 2926 | (CUR != '>') && |
| 2927 | ((CUR != '/') || (NXT(1) != '>'))) { |
| 2928 | long cons = ctxt->nbChars; |
| 2929 | |
| 2930 | GROW; |
| 2931 | attname = htmlParseAttribute(ctxt, &attvalue); |
| 2932 | if (attname != NULL) { |
| 2933 | |
| 2934 | /* |
| 2935 | * Well formedness requires at most one declaration of an attribute |
| 2936 | */ |
| 2937 | for (i = 0; i < nbatts;i += 2) { |
| 2938 | if (xmlStrEqual(atts[i], attname)) { |
| 2939 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2940 | ctxt->sax->error(ctxt->userData, |
| 2941 | "Attribute %s redefined\n", |
| 2942 | attname); |
| 2943 | ctxt->wellFormed = 0; |
| 2944 | xmlFree(attname); |
| 2945 | if (attvalue != NULL) |
| 2946 | xmlFree(attvalue); |
| 2947 | goto failed; |
| 2948 | } |
| 2949 | } |
| 2950 | |
| 2951 | /* |
| 2952 | * Add the pair to atts |
| 2953 | */ |
| 2954 | if (atts == NULL) { |
| 2955 | maxatts = 10; |
| 2956 | atts = (const xmlChar **) xmlMalloc(maxatts * sizeof(xmlChar *)); |
| 2957 | if (atts == NULL) { |
| 2958 | xmlGenericError(xmlGenericErrorContext, |
| 2959 | "malloc of %ld byte failed\n", |
| 2960 | maxatts * (long)sizeof(xmlChar *)); |
| 2961 | if (name != NULL) xmlFree(name); |
| 2962 | return; |
| 2963 | } |
| 2964 | } else if (nbatts + 4 > maxatts) { |
| 2965 | maxatts *= 2; |
| 2966 | atts = (const xmlChar **) xmlRealloc((void *) atts, |
| 2967 | maxatts * sizeof(xmlChar *)); |
| 2968 | if (atts == NULL) { |
| 2969 | xmlGenericError(xmlGenericErrorContext, |
| 2970 | "realloc of %ld byte failed\n", |
| 2971 | maxatts * (long)sizeof(xmlChar *)); |
| 2972 | if (name != NULL) xmlFree(name); |
| 2973 | return; |
| 2974 | } |
| 2975 | } |
| 2976 | atts[nbatts++] = attname; |
| 2977 | atts[nbatts++] = attvalue; |
| 2978 | atts[nbatts] = NULL; |
| 2979 | atts[nbatts + 1] = NULL; |
| 2980 | } |
| 2981 | else { |
| 2982 | /* Dump the bogus attribute string up to the next blank or |
| 2983 | * the end of the tag. */ |
| 2984 | while ((IS_CHAR(CUR)) && !(IS_BLANK(CUR)) && (CUR != '>') |
| 2985 | && ((CUR != '/') || (NXT(1) != '>'))) |
| 2986 | NEXT; |
| 2987 | } |
| 2988 | |
| 2989 | failed: |
| 2990 | SKIP_BLANKS; |
| 2991 | if (cons == ctxt->nbChars) { |
| 2992 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 2993 | ctxt->sax->error(ctxt->userData, |
| 2994 | "htmlParseStartTag: problem parsing attributes\n"); |
| 2995 | ctxt->wellFormed = 0; |
| 2996 | break; |
| 2997 | } |
| 2998 | } |
| 2999 | |
| 3000 | /* |
| 3001 | * Handle specific association to the META tag |
| 3002 | */ |
| 3003 | if (meta) |
| 3004 | htmlCheckMeta(ctxt, atts); |
| 3005 | |
| 3006 | /* |
| 3007 | * SAX: Start of Element ! |
| 3008 | */ |
| 3009 | htmlnamePush(ctxt, xmlStrdup(name)); |
| 3010 | #ifdef DEBUG |
| 3011 | xmlGenericError(xmlGenericErrorContext,"Start of element %s: pushed %s\n", name, ctxt->name); |
| 3012 | #endif |
| 3013 | if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) |
| 3014 | ctxt->sax->startElement(ctxt->userData, name, atts); |
| 3015 | |
| 3016 | if (atts != NULL) { |
| 3017 | for (i = 0;i < nbatts;i++) { |
| 3018 | if (atts[i] != NULL) |
| 3019 | xmlFree((xmlChar *) atts[i]); |
| 3020 | } |
| 3021 | xmlFree((void *) atts); |
| 3022 | } |
| 3023 | if (name != NULL) xmlFree(name); |
| 3024 | } |
| 3025 | |
| 3026 | /** |
| 3027 | * htmlParseEndTag: |
| 3028 | * @ctxt: an HTML parser context |
| 3029 | * |
| 3030 | * parse an end of tag |
| 3031 | * |
| 3032 | * [42] ETag ::= '</' Name S? '>' |
| 3033 | * |
| 3034 | * With namespace |
| 3035 | * |
| 3036 | * [NS 9] ETag ::= '</' QName S? '>' |
| 3037 | */ |
| 3038 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3039 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3040 | htmlParseEndTag(htmlParserCtxtPtr ctxt) { |
| 3041 | xmlChar *name; |
| 3042 | xmlChar *oldname; |
| 3043 | int i; |
| 3044 | |
| 3045 | if ((CUR != '<') || (NXT(1) != '/')) { |
| 3046 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 3047 | ctxt->sax->error(ctxt->userData, "htmlParseEndTag: '</' not found\n"); |
| 3048 | ctxt->wellFormed = 0; |
| 3049 | return; |
| 3050 | } |
| 3051 | SKIP(2); |
| 3052 | |
| 3053 | name = htmlParseHTMLName(ctxt); |
| 3054 | if (name == NULL) return; |
| 3055 | |
| 3056 | /* |
| 3057 | * We should definitely be at the ending "S? '>'" part |
| 3058 | */ |
| 3059 | SKIP_BLANKS; |
| 3060 | if ((!IS_CHAR(CUR)) || (CUR != '>')) { |
| 3061 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 3062 | ctxt->sax->error(ctxt->userData, "End tag : expected '>'\n"); |
| 3063 | ctxt->wellFormed = 0; |
| 3064 | } else |
| 3065 | NEXT; |
| 3066 | |
| 3067 | /* |
| 3068 | * If the name read is not one of the element in the parsing stack |
| 3069 | * then return, it's just an error. |
| 3070 | */ |
| 3071 | for (i = (ctxt->nameNr - 1);i >= 0;i--) { |
| 3072 | if (xmlStrEqual(name, ctxt->nameTab[i])) break; |
| 3073 | } |
| 3074 | if (i < 0) { |
| 3075 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 3076 | ctxt->sax->error(ctxt->userData, |
| 3077 | "Unexpected end tag : %s\n", name); |
| 3078 | xmlFree(name); |
| 3079 | ctxt->wellFormed = 0; |
| 3080 | return; |
| 3081 | } |
| 3082 | |
| 3083 | |
| 3084 | /* |
| 3085 | * Check for auto-closure of HTML elements. |
| 3086 | */ |
| 3087 | |
| 3088 | htmlAutoCloseOnClose(ctxt, name); |
| 3089 | |
| 3090 | /* |
| 3091 | * Well formedness constraints, opening and closing must match. |
| 3092 | * With the exception that the autoclose may have popped stuff out |
| 3093 | * of the stack. |
| 3094 | */ |
| 3095 | if (!xmlStrEqual(name, ctxt->name)) { |
| 3096 | #ifdef DEBUG |
| 3097 | xmlGenericError(xmlGenericErrorContext,"End of tag %s: expecting %s\n", name, ctxt->name); |
| 3098 | #endif |
| 3099 | if ((ctxt->name != NULL) && |
| 3100 | (!xmlStrEqual(ctxt->name, name))) { |
| 3101 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 3102 | ctxt->sax->error(ctxt->userData, |
| 3103 | "Opening and ending tag mismatch: %s and %s\n", |
| 3104 | name, ctxt->name); |
| 3105 | ctxt->wellFormed = 0; |
| 3106 | } |
| 3107 | } |
| 3108 | |
| 3109 | /* |
| 3110 | * SAX: End of Tag |
| 3111 | */ |
| 3112 | oldname = ctxt->name; |
| 3113 | if ((oldname != NULL) && (xmlStrEqual(oldname, name))) { |
| 3114 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL)) |
| 3115 | ctxt->sax->endElement(ctxt->userData, name); |
| 3116 | oldname = htmlnamePop(ctxt); |
| 3117 | if (oldname != NULL) { |
| 3118 | #ifdef DEBUG |
| 3119 | xmlGenericError(xmlGenericErrorContext,"End of tag %s: popping out %s\n", name, oldname); |
| 3120 | #endif |
| 3121 | xmlFree(oldname); |
| 3122 | #ifdef DEBUG |
| 3123 | } else { |
| 3124 | xmlGenericError(xmlGenericErrorContext,"End of tag %s: stack empty !!!\n", name); |
| 3125 | #endif |
| 3126 | } |
| 3127 | } |
| 3128 | |
| 3129 | if (name != NULL) |
| 3130 | xmlFree(name); |
| 3131 | |
| 3132 | return; |
| 3133 | } |
| 3134 | |
| 3135 | |
| 3136 | /** |
| 3137 | * htmlParseReference: |
| 3138 | * @ctxt: an HTML parser context |
| 3139 | * |
| 3140 | * parse and handle entity references in content, |
| 3141 | * this will end-up in a call to character() since this is either a |
| 3142 | * CharRef, or a predefined entity. |
| 3143 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3144 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3145 | htmlParseReference(htmlParserCtxtPtr ctxt) { |
| 3146 | htmlEntityDescPtr ent; |
| 3147 | xmlChar out[6]; |
| 3148 | xmlChar *name; |
| 3149 | if (CUR != '&') return; |
| 3150 | |
| 3151 | if (NXT(1) == '#') { |
| 3152 | unsigned int c; |
| 3153 | int bits, i = 0; |
| 3154 | |
| 3155 | c = htmlParseCharRef(ctxt); |
| 3156 | if (c == 0) |
| 3157 | return; |
| 3158 | |
| 3159 | if (c < 0x80) { out[i++]= c; bits= -6; } |
| 3160 | else if (c < 0x800) { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; } |
| 3161 | else if (c < 0x10000) { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; } |
| 3162 | else { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; } |
| 3163 | |
| 3164 | for ( ; bits >= 0; bits-= 6) { |
| 3165 | out[i++]= ((c >> bits) & 0x3F) | 0x80; |
| 3166 | } |
| 3167 | out[i] = 0; |
| 3168 | |
| 3169 | htmlCheckParagraph(ctxt); |
| 3170 | if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) |
| 3171 | ctxt->sax->characters(ctxt->userData, out, i); |
| 3172 | } else { |
| 3173 | ent = htmlParseEntityRef(ctxt, &name); |
| 3174 | if (name == NULL) { |
| 3175 | htmlCheckParagraph(ctxt); |
| 3176 | if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) |
| 3177 | ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1); |
| 3178 | return; |
| 3179 | } |
| 3180 | if ((ent == NULL) || (ent->value <= 0)) { |
| 3181 | htmlCheckParagraph(ctxt); |
| 3182 | if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) { |
| 3183 | ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1); |
| 3184 | ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name)); |
| 3185 | /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */ |
| 3186 | } |
| 3187 | } else { |
| 3188 | unsigned int c; |
| 3189 | int bits, i = 0; |
| 3190 | |
| 3191 | c = ent->value; |
| 3192 | if (c < 0x80) |
| 3193 | { out[i++]= c; bits= -6; } |
| 3194 | else if (c < 0x800) |
| 3195 | { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; } |
| 3196 | else if (c < 0x10000) |
| 3197 | { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; } |
| 3198 | else |
| 3199 | { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; } |
| 3200 | |
| 3201 | for ( ; bits >= 0; bits-= 6) { |
| 3202 | out[i++]= ((c >> bits) & 0x3F) | 0x80; |
| 3203 | } |
| 3204 | out[i] = 0; |
| 3205 | |
| 3206 | htmlCheckParagraph(ctxt); |
| 3207 | if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) |
| 3208 | ctxt->sax->characters(ctxt->userData, out, i); |
| 3209 | } |
| 3210 | xmlFree(name); |
| 3211 | } |
| 3212 | } |
| 3213 | |
| 3214 | /** |
| 3215 | * htmlParseContent: |
| 3216 | * @ctxt: an HTML parser context |
| 3217 | * @name: the node name |
| 3218 | * |
| 3219 | * Parse a content: comment, sub-element, reference or text. |
| 3220 | * |
| 3221 | */ |
| 3222 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3223 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3224 | htmlParseContent(htmlParserCtxtPtr ctxt) { |
| 3225 | xmlChar *currentNode; |
| 3226 | int depth; |
| 3227 | |
| 3228 | currentNode = xmlStrdup(ctxt->name); |
| 3229 | depth = ctxt->nameNr; |
| 3230 | while (1) { |
| 3231 | long cons = ctxt->nbChars; |
| 3232 | |
| 3233 | GROW; |
| 3234 | /* |
| 3235 | * Our tag or one of it's parent or children is ending. |
| 3236 | */ |
| 3237 | if ((CUR == '<') && (NXT(1) == '/')) { |
| 3238 | htmlParseEndTag(ctxt); |
| 3239 | if (currentNode != NULL) xmlFree(currentNode); |
| 3240 | return; |
| 3241 | } |
| 3242 | |
| 3243 | /* |
| 3244 | * Has this node been popped out during parsing of |
| 3245 | * the next element |
| 3246 | */ |
| 3247 | if ((!xmlStrEqual(currentNode, ctxt->name)) && |
| 3248 | (depth >= ctxt->nameNr)) { |
| 3249 | if (currentNode != NULL) xmlFree(currentNode); |
| 3250 | return; |
| 3251 | } |
| 3252 | |
Daniel Veillard | f9533d1 | 2001-03-03 10:04:57 +0000 | [diff] [blame] | 3253 | if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) || |
| 3254 | (xmlStrEqual(currentNode, BAD_CAST"style")))) { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3255 | /* |
| 3256 | * Handle SCRIPT/STYLE separately |
| 3257 | */ |
| 3258 | htmlParseScript(ctxt); |
| 3259 | } else { |
| 3260 | /* |
| 3261 | * Sometimes DOCTYPE arrives in the middle of the document |
| 3262 | */ |
| 3263 | if ((CUR == '<') && (NXT(1) == '!') && |
| 3264 | (UPP(2) == 'D') && (UPP(3) == 'O') && |
| 3265 | (UPP(4) == 'C') && (UPP(5) == 'T') && |
| 3266 | (UPP(6) == 'Y') && (UPP(7) == 'P') && |
| 3267 | (UPP(8) == 'E')) { |
| 3268 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 3269 | ctxt->sax->error(ctxt->userData, |
| 3270 | "Misplaced DOCTYPE declaration\n"); |
| 3271 | ctxt->wellFormed = 0; |
| 3272 | htmlParseDocTypeDecl(ctxt); |
| 3273 | } |
| 3274 | |
| 3275 | /* |
| 3276 | * First case : a comment |
| 3277 | */ |
| 3278 | if ((CUR == '<') && (NXT(1) == '!') && |
| 3279 | (NXT(2) == '-') && (NXT(3) == '-')) { |
| 3280 | htmlParseComment(ctxt); |
| 3281 | } |
| 3282 | |
| 3283 | /* |
| 3284 | * Second case : a sub-element. |
| 3285 | */ |
| 3286 | else if (CUR == '<') { |
| 3287 | htmlParseElement(ctxt); |
| 3288 | } |
| 3289 | |
| 3290 | /* |
| 3291 | * Third case : a reference. If if has not been resolved, |
| 3292 | * parsing returns it's Name, create the node |
| 3293 | */ |
| 3294 | else if (CUR == '&') { |
| 3295 | htmlParseReference(ctxt); |
| 3296 | } |
| 3297 | |
| 3298 | /* |
| 3299 | * Fourth : end of the resource |
| 3300 | */ |
| 3301 | else if (CUR == 0) { |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 3302 | htmlAutoCloseOnEnd(ctxt); |
| 3303 | break; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3304 | } |
| 3305 | |
| 3306 | /* |
| 3307 | * Last case, text. Note that References are handled directly. |
| 3308 | */ |
| 3309 | else { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3310 | htmlParseCharData(ctxt); |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3311 | } |
| 3312 | |
| 3313 | if (cons == ctxt->nbChars) { |
| 3314 | if (ctxt->node != NULL) { |
| 3315 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 3316 | ctxt->sax->error(ctxt->userData, |
| 3317 | "detected an error in element content\n"); |
| 3318 | ctxt->wellFormed = 0; |
| 3319 | } |
| 3320 | break; |
| 3321 | } |
| 3322 | } |
| 3323 | GROW; |
| 3324 | } |
| 3325 | if (currentNode != NULL) xmlFree(currentNode); |
| 3326 | } |
| 3327 | |
| 3328 | /** |
| 3329 | * htmlParseElement: |
| 3330 | * @ctxt: an HTML parser context |
| 3331 | * |
| 3332 | * parse an HTML element, this is highly recursive |
| 3333 | * |
| 3334 | * [39] element ::= EmptyElemTag | STag content ETag |
| 3335 | * |
| 3336 | * [41] Attribute ::= Name Eq AttValue |
| 3337 | */ |
| 3338 | |
| 3339 | void |
| 3340 | htmlParseElement(htmlParserCtxtPtr ctxt) { |
| 3341 | xmlChar *name; |
| 3342 | xmlChar *currentNode = NULL; |
| 3343 | htmlElemDescPtr info; |
| 3344 | htmlParserNodeInfo node_info; |
| 3345 | xmlChar *oldname; |
| 3346 | int depth = ctxt->nameNr; |
| 3347 | |
| 3348 | /* Capture start position */ |
| 3349 | if (ctxt->record_info) { |
| 3350 | node_info.begin_pos = ctxt->input->consumed + |
| 3351 | (CUR_PTR - ctxt->input->base); |
| 3352 | node_info.begin_line = ctxt->input->line; |
| 3353 | } |
| 3354 | |
| 3355 | oldname = xmlStrdup(ctxt->name); |
| 3356 | htmlParseStartTag(ctxt); |
| 3357 | name = ctxt->name; |
| 3358 | #ifdef DEBUG |
| 3359 | if (oldname == NULL) |
| 3360 | xmlGenericError(xmlGenericErrorContext, |
| 3361 | "Start of element %s\n", name); |
| 3362 | else if (name == NULL) |
| 3363 | xmlGenericError(xmlGenericErrorContext, |
| 3364 | "Start of element failed, was %s\n", oldname); |
| 3365 | else |
| 3366 | xmlGenericError(xmlGenericErrorContext, |
| 3367 | "Start of element %s, was %s\n", name, oldname); |
| 3368 | #endif |
| 3369 | if (((depth == ctxt->nameNr) && (xmlStrEqual(oldname, ctxt->name))) || |
| 3370 | (name == NULL)) { |
| 3371 | if (CUR == '>') |
| 3372 | NEXT; |
| 3373 | if (oldname != NULL) |
| 3374 | xmlFree(oldname); |
| 3375 | return; |
| 3376 | } |
| 3377 | if (oldname != NULL) |
| 3378 | xmlFree(oldname); |
| 3379 | |
| 3380 | /* |
| 3381 | * Lookup the info for that element. |
| 3382 | */ |
| 3383 | info = htmlTagLookup(name); |
| 3384 | if (info == NULL) { |
| 3385 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 3386 | ctxt->sax->error(ctxt->userData, "Tag %s invalid\n", |
| 3387 | name); |
| 3388 | ctxt->wellFormed = 0; |
| 3389 | } else if (info->depr) { |
| 3390 | /*************************** |
| 3391 | if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL)) |
| 3392 | ctxt->sax->warning(ctxt->userData, "Tag %s is deprecated\n", |
| 3393 | name); |
| 3394 | ***************************/ |
| 3395 | } |
| 3396 | |
| 3397 | /* |
| 3398 | * Check for an Empty Element labelled the XML/SGML way |
| 3399 | */ |
| 3400 | if ((CUR == '/') && (NXT(1) == '>')) { |
| 3401 | SKIP(2); |
| 3402 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL)) |
| 3403 | ctxt->sax->endElement(ctxt->userData, name); |
| 3404 | oldname = htmlnamePop(ctxt); |
| 3405 | #ifdef DEBUG |
| 3406 | xmlGenericError(xmlGenericErrorContext,"End of tag the XML way: popping out %s\n", oldname); |
| 3407 | #endif |
| 3408 | if (oldname != NULL) |
| 3409 | xmlFree(oldname); |
| 3410 | return; |
| 3411 | } |
| 3412 | |
| 3413 | if (CUR == '>') { |
| 3414 | NEXT; |
| 3415 | } else { |
| 3416 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 3417 | ctxt->sax->error(ctxt->userData, |
| 3418 | "Couldn't find end of Start Tag %s\n", |
| 3419 | name); |
| 3420 | ctxt->wellFormed = 0; |
| 3421 | |
| 3422 | /* |
| 3423 | * end of parsing of this node. |
| 3424 | */ |
| 3425 | if (xmlStrEqual(name, ctxt->name)) { |
| 3426 | nodePop(ctxt); |
| 3427 | oldname = htmlnamePop(ctxt); |
| 3428 | #ifdef DEBUG |
| 3429 | xmlGenericError(xmlGenericErrorContext,"End of start tag problem: popping out %s\n", oldname); |
| 3430 | #endif |
| 3431 | if (oldname != NULL) |
| 3432 | xmlFree(oldname); |
| 3433 | } |
| 3434 | |
| 3435 | /* |
| 3436 | * Capture end position and add node |
| 3437 | */ |
| 3438 | if ( currentNode != NULL && ctxt->record_info ) { |
| 3439 | node_info.end_pos = ctxt->input->consumed + |
| 3440 | (CUR_PTR - ctxt->input->base); |
| 3441 | node_info.end_line = ctxt->input->line; |
| 3442 | node_info.node = ctxt->node; |
| 3443 | xmlParserAddNodeInfo(ctxt, &node_info); |
| 3444 | } |
| 3445 | return; |
| 3446 | } |
| 3447 | |
| 3448 | /* |
| 3449 | * Check for an Empty Element from DTD definition |
| 3450 | */ |
| 3451 | if ((info != NULL) && (info->empty)) { |
| 3452 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL)) |
| 3453 | ctxt->sax->endElement(ctxt->userData, name); |
| 3454 | oldname = htmlnamePop(ctxt); |
| 3455 | #ifdef DEBUG |
| 3456 | xmlGenericError(xmlGenericErrorContext,"End of empty tag %s : popping out %s\n", name, oldname); |
| 3457 | #endif |
| 3458 | if (oldname != NULL) |
| 3459 | xmlFree(oldname); |
| 3460 | return; |
| 3461 | } |
| 3462 | |
| 3463 | /* |
| 3464 | * Parse the content of the element: |
| 3465 | */ |
| 3466 | currentNode = xmlStrdup(ctxt->name); |
| 3467 | depth = ctxt->nameNr; |
| 3468 | while (IS_CHAR(CUR)) { |
| 3469 | htmlParseContent(ctxt); |
| 3470 | if (ctxt->nameNr < depth) break; |
| 3471 | } |
| 3472 | |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3473 | /* |
| 3474 | * Capture end position and add node |
| 3475 | */ |
| 3476 | if ( currentNode != NULL && ctxt->record_info ) { |
| 3477 | node_info.end_pos = ctxt->input->consumed + |
| 3478 | (CUR_PTR - ctxt->input->base); |
| 3479 | node_info.end_line = ctxt->input->line; |
| 3480 | node_info.node = ctxt->node; |
| 3481 | xmlParserAddNodeInfo(ctxt, &node_info); |
| 3482 | } |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 3483 | if (!IS_CHAR(CUR)) { |
| 3484 | htmlAutoCloseOnEnd(ctxt); |
| 3485 | } |
| 3486 | |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3487 | if (currentNode != NULL) |
| 3488 | xmlFree(currentNode); |
| 3489 | } |
| 3490 | |
| 3491 | /** |
| 3492 | * htmlParseDocument : |
| 3493 | * @ctxt: an HTML parser context |
| 3494 | * |
| 3495 | * parse an HTML document (and build a tree if using the standard SAX |
| 3496 | * interface). |
| 3497 | * |
| 3498 | * Returns 0, -1 in case of error. the parser context is augmented |
| 3499 | * as a result of the parsing. |
| 3500 | */ |
| 3501 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3502 | static int |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3503 | htmlParseDocument(htmlParserCtxtPtr ctxt) { |
| 3504 | xmlDtdPtr dtd; |
| 3505 | |
| 3506 | htmlDefaultSAXHandlerInit(); |
| 3507 | ctxt->html = 1; |
| 3508 | |
| 3509 | GROW; |
| 3510 | /* |
| 3511 | * SAX: beginning of the document processing. |
| 3512 | */ |
| 3513 | if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) |
| 3514 | ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator); |
| 3515 | |
| 3516 | /* |
| 3517 | * Wipe out everything which is before the first '<' |
| 3518 | */ |
| 3519 | SKIP_BLANKS; |
| 3520 | if (CUR == 0) { |
| 3521 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 3522 | ctxt->sax->error(ctxt->userData, "Document is empty\n"); |
| 3523 | ctxt->wellFormed = 0; |
| 3524 | } |
| 3525 | |
| 3526 | if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX)) |
| 3527 | ctxt->sax->startDocument(ctxt->userData); |
| 3528 | |
| 3529 | |
| 3530 | /* |
| 3531 | * Parse possible comments before any content |
| 3532 | */ |
| 3533 | while ((CUR == '<') && (NXT(1) == '!') && |
| 3534 | (NXT(2) == '-') && (NXT(3) == '-')) { |
| 3535 | htmlParseComment(ctxt); |
| 3536 | SKIP_BLANKS; |
| 3537 | } |
| 3538 | |
| 3539 | |
| 3540 | /* |
| 3541 | * Then possibly doc type declaration(s) and more Misc |
| 3542 | * (doctypedecl Misc*)? |
| 3543 | */ |
| 3544 | if ((CUR == '<') && (NXT(1) == '!') && |
| 3545 | (UPP(2) == 'D') && (UPP(3) == 'O') && |
| 3546 | (UPP(4) == 'C') && (UPP(5) == 'T') && |
| 3547 | (UPP(6) == 'Y') && (UPP(7) == 'P') && |
| 3548 | (UPP(8) == 'E')) { |
| 3549 | htmlParseDocTypeDecl(ctxt); |
| 3550 | } |
| 3551 | SKIP_BLANKS; |
| 3552 | |
| 3553 | /* |
| 3554 | * Parse possible comments before any content |
| 3555 | */ |
| 3556 | while ((CUR == '<') && (NXT(1) == '!') && |
| 3557 | (NXT(2) == '-') && (NXT(3) == '-')) { |
| 3558 | htmlParseComment(ctxt); |
| 3559 | SKIP_BLANKS; |
| 3560 | } |
| 3561 | |
| 3562 | /* |
| 3563 | * Time to start parsing the tree itself |
| 3564 | */ |
| 3565 | htmlParseContent(ctxt); |
| 3566 | |
| 3567 | /* |
| 3568 | * autoclose |
| 3569 | */ |
| 3570 | if (CUR == 0) |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 3571 | htmlAutoCloseOnEnd(ctxt); |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3572 | |
| 3573 | |
| 3574 | /* |
| 3575 | * SAX: end of the document processing. |
| 3576 | */ |
| 3577 | if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) |
| 3578 | ctxt->sax->endDocument(ctxt->userData); |
| 3579 | |
| 3580 | if (ctxt->myDoc != NULL) { |
| 3581 | dtd = xmlGetIntSubset(ctxt->myDoc); |
| 3582 | if (dtd == NULL) |
| 3583 | ctxt->myDoc->intSubset = |
| 3584 | xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML", |
| 3585 | BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN", |
| 3586 | BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd"); |
| 3587 | } |
| 3588 | if (! ctxt->wellFormed) return(-1); |
| 3589 | return(0); |
| 3590 | } |
| 3591 | |
| 3592 | |
| 3593 | /************************************************************************ |
| 3594 | * * |
| 3595 | * Parser contexts handling * |
| 3596 | * * |
| 3597 | ************************************************************************/ |
| 3598 | |
| 3599 | /** |
| 3600 | * xmlInitParserCtxt: |
| 3601 | * @ctxt: an HTML parser context |
| 3602 | * |
| 3603 | * Initialize a parser context |
| 3604 | */ |
| 3605 | |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3606 | static void |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3607 | htmlInitParserCtxt(htmlParserCtxtPtr ctxt) |
| 3608 | { |
| 3609 | htmlSAXHandler *sax; |
| 3610 | |
| 3611 | if (ctxt == NULL) return; |
| 3612 | memset(ctxt, 0, sizeof(htmlParserCtxt)); |
| 3613 | |
| 3614 | sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler)); |
| 3615 | if (sax == NULL) { |
| 3616 | xmlGenericError(xmlGenericErrorContext, |
| 3617 | "htmlInitParserCtxt: out of memory\n"); |
| 3618 | } |
| 3619 | else |
| 3620 | memset(sax, 0, sizeof(htmlSAXHandler)); |
| 3621 | |
| 3622 | /* Allocate the Input stack */ |
| 3623 | ctxt->inputTab = (htmlParserInputPtr *) |
| 3624 | xmlMalloc(5 * sizeof(htmlParserInputPtr)); |
| 3625 | if (ctxt->inputTab == NULL) { |
| 3626 | xmlGenericError(xmlGenericErrorContext, |
| 3627 | "htmlInitParserCtxt: out of memory\n"); |
| 3628 | ctxt->inputNr = 0; |
| 3629 | ctxt->inputMax = 0; |
| 3630 | ctxt->input = NULL; |
| 3631 | return; |
| 3632 | } |
| 3633 | ctxt->inputNr = 0; |
| 3634 | ctxt->inputMax = 5; |
| 3635 | ctxt->input = NULL; |
| 3636 | ctxt->version = NULL; |
| 3637 | ctxt->encoding = NULL; |
| 3638 | ctxt->standalone = -1; |
| 3639 | ctxt->instate = XML_PARSER_START; |
| 3640 | |
| 3641 | /* Allocate the Node stack */ |
| 3642 | ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr)); |
| 3643 | if (ctxt->nodeTab == NULL) { |
| 3644 | xmlGenericError(xmlGenericErrorContext, |
| 3645 | "htmlInitParserCtxt: out of memory\n"); |
| 3646 | ctxt->nodeNr = 0; |
| 3647 | ctxt->nodeMax = 0; |
| 3648 | ctxt->node = NULL; |
| 3649 | ctxt->inputNr = 0; |
| 3650 | ctxt->inputMax = 0; |
| 3651 | ctxt->input = NULL; |
| 3652 | return; |
| 3653 | } |
| 3654 | ctxt->nodeNr = 0; |
| 3655 | ctxt->nodeMax = 10; |
| 3656 | ctxt->node = NULL; |
| 3657 | |
| 3658 | /* Allocate the Name stack */ |
| 3659 | ctxt->nameTab = (xmlChar **) xmlMalloc(10 * sizeof(xmlChar *)); |
| 3660 | if (ctxt->nameTab == NULL) { |
| 3661 | xmlGenericError(xmlGenericErrorContext, |
| 3662 | "htmlInitParserCtxt: out of memory\n"); |
| 3663 | ctxt->nameNr = 0; |
| 3664 | ctxt->nameMax = 10; |
| 3665 | ctxt->name = NULL; |
| 3666 | ctxt->nodeNr = 0; |
| 3667 | ctxt->nodeMax = 0; |
| 3668 | ctxt->node = NULL; |
| 3669 | ctxt->inputNr = 0; |
| 3670 | ctxt->inputMax = 0; |
| 3671 | ctxt->input = NULL; |
| 3672 | return; |
| 3673 | } |
| 3674 | ctxt->nameNr = 0; |
| 3675 | ctxt->nameMax = 10; |
| 3676 | ctxt->name = NULL; |
| 3677 | |
| 3678 | if (sax == NULL) ctxt->sax = &htmlDefaultSAXHandler; |
| 3679 | else { |
| 3680 | ctxt->sax = sax; |
| 3681 | memcpy(sax, &htmlDefaultSAXHandler, sizeof(htmlSAXHandler)); |
| 3682 | } |
| 3683 | ctxt->userData = ctxt; |
| 3684 | ctxt->myDoc = NULL; |
| 3685 | ctxt->wellFormed = 1; |
| 3686 | ctxt->replaceEntities = 0; |
| 3687 | ctxt->html = 1; |
| 3688 | ctxt->record_info = 0; |
| 3689 | ctxt->validate = 0; |
| 3690 | ctxt->nbChars = 0; |
| 3691 | ctxt->checkIndex = 0; |
| 3692 | xmlInitNodeInfoSeq(&ctxt->node_seq); |
| 3693 | } |
| 3694 | |
| 3695 | /** |
| 3696 | * htmlFreeParserCtxt: |
| 3697 | * @ctxt: an HTML parser context |
| 3698 | * |
| 3699 | * Free all the memory used by a parser context. However the parsed |
| 3700 | * document in ctxt->myDoc is not freed. |
| 3701 | */ |
| 3702 | |
| 3703 | void |
| 3704 | htmlFreeParserCtxt(htmlParserCtxtPtr ctxt) |
| 3705 | { |
| 3706 | xmlFreeParserCtxt(ctxt); |
| 3707 | } |
| 3708 | |
| 3709 | /** |
| 3710 | * htmlCreateDocParserCtxt : |
| 3711 | * @cur: a pointer to an array of xmlChar |
| 3712 | * @encoding: a free form C string describing the HTML document encoding, or NULL |
| 3713 | * |
| 3714 | * Create a parser context for an HTML document. |
| 3715 | * |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3716 | * TODO: check the need to add encoding handling there |
| 3717 | * |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3718 | * Returns the new parser context or NULL |
| 3719 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3720 | static htmlParserCtxtPtr |
Daniel Veillard | c86a4fa | 2001-03-26 16:28:29 +0000 | [diff] [blame] | 3721 | htmlCreateDocParserCtxt(xmlChar *cur, const char *encoding ATTRIBUTE_UNUSED) { |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3722 | htmlParserCtxtPtr ctxt; |
| 3723 | htmlParserInputPtr input; |
| 3724 | /* htmlCharEncoding enc; */ |
| 3725 | |
| 3726 | ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt)); |
| 3727 | if (ctxt == NULL) { |
| 3728 | perror("malloc"); |
| 3729 | return(NULL); |
| 3730 | } |
| 3731 | htmlInitParserCtxt(ctxt); |
| 3732 | input = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput)); |
| 3733 | if (input == NULL) { |
| 3734 | perror("malloc"); |
| 3735 | xmlFree(ctxt); |
| 3736 | return(NULL); |
| 3737 | } |
| 3738 | memset(input, 0, sizeof(htmlParserInput)); |
| 3739 | |
| 3740 | input->line = 1; |
| 3741 | input->col = 1; |
| 3742 | input->base = cur; |
| 3743 | input->cur = cur; |
| 3744 | |
| 3745 | inputPush(ctxt, input); |
| 3746 | return(ctxt); |
| 3747 | } |
| 3748 | |
| 3749 | /************************************************************************ |
| 3750 | * * |
| 3751 | * Progressive parsing interfaces * |
| 3752 | * * |
| 3753 | ************************************************************************/ |
| 3754 | |
| 3755 | /** |
| 3756 | * htmlParseLookupSequence: |
| 3757 | * @ctxt: an HTML parser context |
| 3758 | * @first: the first char to lookup |
| 3759 | * @next: the next char to lookup or zero |
| 3760 | * @third: the next char to lookup or zero |
| 3761 | * |
| 3762 | * Try to find if a sequence (first, next, third) or just (first next) or |
| 3763 | * (first) is available in the input stream. |
| 3764 | * This function has a side effect of (possibly) incrementing ctxt->checkIndex |
| 3765 | * to avoid rescanning sequences of bytes, it DOES change the state of the |
| 3766 | * parser, do not use liberally. |
| 3767 | * This is basically similar to xmlParseLookupSequence() |
| 3768 | * |
| 3769 | * Returns the index to the current parsing point if the full sequence |
| 3770 | * is available, -1 otherwise. |
| 3771 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3772 | static int |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3773 | htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first, |
| 3774 | xmlChar next, xmlChar third) { |
| 3775 | int base, len; |
| 3776 | htmlParserInputPtr in; |
| 3777 | const xmlChar *buf; |
| 3778 | |
| 3779 | in = ctxt->input; |
| 3780 | if (in == NULL) return(-1); |
| 3781 | base = in->cur - in->base; |
| 3782 | if (base < 0) return(-1); |
| 3783 | if (ctxt->checkIndex > base) |
| 3784 | base = ctxt->checkIndex; |
| 3785 | if (in->buf == NULL) { |
| 3786 | buf = in->base; |
| 3787 | len = in->length; |
| 3788 | } else { |
| 3789 | buf = in->buf->buffer->content; |
| 3790 | len = in->buf->buffer->use; |
| 3791 | } |
| 3792 | /* take into account the sequence length */ |
| 3793 | if (third) len -= 2; |
| 3794 | else if (next) len --; |
| 3795 | for (;base < len;base++) { |
| 3796 | if (buf[base] == first) { |
| 3797 | if (third != 0) { |
| 3798 | if ((buf[base + 1] != next) || |
| 3799 | (buf[base + 2] != third)) continue; |
| 3800 | } else if (next != 0) { |
| 3801 | if (buf[base + 1] != next) continue; |
| 3802 | } |
| 3803 | ctxt->checkIndex = 0; |
| 3804 | #ifdef DEBUG_PUSH |
| 3805 | if (next == 0) |
| 3806 | xmlGenericError(xmlGenericErrorContext, |
| 3807 | "HPP: lookup '%c' found at %d\n", |
| 3808 | first, base); |
| 3809 | else if (third == 0) |
| 3810 | xmlGenericError(xmlGenericErrorContext, |
| 3811 | "HPP: lookup '%c%c' found at %d\n", |
| 3812 | first, next, base); |
| 3813 | else |
| 3814 | xmlGenericError(xmlGenericErrorContext, |
| 3815 | "HPP: lookup '%c%c%c' found at %d\n", |
| 3816 | first, next, third, base); |
| 3817 | #endif |
| 3818 | return(base - (in->cur - in->base)); |
| 3819 | } |
| 3820 | } |
| 3821 | ctxt->checkIndex = base; |
| 3822 | #ifdef DEBUG_PUSH |
| 3823 | if (next == 0) |
| 3824 | xmlGenericError(xmlGenericErrorContext, |
| 3825 | "HPP: lookup '%c' failed\n", first); |
| 3826 | else if (third == 0) |
| 3827 | xmlGenericError(xmlGenericErrorContext, |
| 3828 | "HPP: lookup '%c%c' failed\n", first, next); |
| 3829 | else |
| 3830 | xmlGenericError(xmlGenericErrorContext, |
| 3831 | "HPP: lookup '%c%c%c' failed\n", first, next, third); |
| 3832 | #endif |
| 3833 | return(-1); |
| 3834 | } |
| 3835 | |
| 3836 | /** |
| 3837 | * htmlParseTryOrFinish: |
| 3838 | * @ctxt: an HTML parser context |
| 3839 | * @terminate: last chunk indicator |
| 3840 | * |
| 3841 | * Try to progress on parsing |
| 3842 | * |
| 3843 | * Returns zero if no parsing was possible |
| 3844 | */ |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 3845 | static int |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3846 | htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) { |
| 3847 | int ret = 0; |
| 3848 | htmlParserInputPtr in; |
| 3849 | int avail = 0; |
| 3850 | xmlChar cur, next; |
| 3851 | |
| 3852 | #ifdef DEBUG_PUSH |
| 3853 | switch (ctxt->instate) { |
| 3854 | case XML_PARSER_EOF: |
| 3855 | xmlGenericError(xmlGenericErrorContext, |
| 3856 | "HPP: try EOF\n"); break; |
| 3857 | case XML_PARSER_START: |
| 3858 | xmlGenericError(xmlGenericErrorContext, |
| 3859 | "HPP: try START\n"); break; |
| 3860 | case XML_PARSER_MISC: |
| 3861 | xmlGenericError(xmlGenericErrorContext, |
| 3862 | "HPP: try MISC\n");break; |
| 3863 | case XML_PARSER_COMMENT: |
| 3864 | xmlGenericError(xmlGenericErrorContext, |
| 3865 | "HPP: try COMMENT\n");break; |
| 3866 | case XML_PARSER_PROLOG: |
| 3867 | xmlGenericError(xmlGenericErrorContext, |
| 3868 | "HPP: try PROLOG\n");break; |
| 3869 | case XML_PARSER_START_TAG: |
| 3870 | xmlGenericError(xmlGenericErrorContext, |
| 3871 | "HPP: try START_TAG\n");break; |
| 3872 | case XML_PARSER_CONTENT: |
| 3873 | xmlGenericError(xmlGenericErrorContext, |
| 3874 | "HPP: try CONTENT\n");break; |
| 3875 | case XML_PARSER_CDATA_SECTION: |
| 3876 | xmlGenericError(xmlGenericErrorContext, |
| 3877 | "HPP: try CDATA_SECTION\n");break; |
| 3878 | case XML_PARSER_END_TAG: |
| 3879 | xmlGenericError(xmlGenericErrorContext, |
| 3880 | "HPP: try END_TAG\n");break; |
| 3881 | case XML_PARSER_ENTITY_DECL: |
| 3882 | xmlGenericError(xmlGenericErrorContext, |
| 3883 | "HPP: try ENTITY_DECL\n");break; |
| 3884 | case XML_PARSER_ENTITY_VALUE: |
| 3885 | xmlGenericError(xmlGenericErrorContext, |
| 3886 | "HPP: try ENTITY_VALUE\n");break; |
| 3887 | case XML_PARSER_ATTRIBUTE_VALUE: |
| 3888 | xmlGenericError(xmlGenericErrorContext, |
| 3889 | "HPP: try ATTRIBUTE_VALUE\n");break; |
| 3890 | case XML_PARSER_DTD: |
| 3891 | xmlGenericError(xmlGenericErrorContext, |
| 3892 | "HPP: try DTD\n");break; |
| 3893 | case XML_PARSER_EPILOG: |
| 3894 | xmlGenericError(xmlGenericErrorContext, |
| 3895 | "HPP: try EPILOG\n");break; |
| 3896 | case XML_PARSER_PI: |
| 3897 | xmlGenericError(xmlGenericErrorContext, |
| 3898 | "HPP: try PI\n");break; |
| 3899 | case XML_PARSER_SYSTEM_LITERAL: |
| 3900 | xmlGenericError(xmlGenericErrorContext, |
| 3901 | "HPP: try SYSTEM_LITERAL\n");break; |
| 3902 | } |
| 3903 | #endif |
| 3904 | |
| 3905 | while (1) { |
| 3906 | |
| 3907 | in = ctxt->input; |
| 3908 | if (in == NULL) break; |
| 3909 | if (in->buf == NULL) |
| 3910 | avail = in->length - (in->cur - in->base); |
| 3911 | else |
| 3912 | avail = in->buf->buffer->use - (in->cur - in->base); |
| 3913 | if ((avail == 0) && (terminate)) { |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 3914 | htmlAutoCloseOnEnd(ctxt); |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 3915 | if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) { |
| 3916 | /* |
| 3917 | * SAX: end of the document processing. |
| 3918 | */ |
| 3919 | ctxt->instate = XML_PARSER_EOF; |
| 3920 | if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) |
| 3921 | ctxt->sax->endDocument(ctxt->userData); |
| 3922 | } |
| 3923 | } |
| 3924 | if (avail < 1) |
| 3925 | goto done; |
| 3926 | switch (ctxt->instate) { |
| 3927 | case XML_PARSER_EOF: |
| 3928 | /* |
| 3929 | * Document parsing is done ! |
| 3930 | */ |
| 3931 | goto done; |
| 3932 | case XML_PARSER_START: |
| 3933 | /* |
| 3934 | * Very first chars read from the document flow. |
| 3935 | */ |
| 3936 | cur = in->cur[0]; |
| 3937 | if (IS_BLANK(cur)) { |
| 3938 | SKIP_BLANKS; |
| 3939 | if (in->buf == NULL) |
| 3940 | avail = in->length - (in->cur - in->base); |
| 3941 | else |
| 3942 | avail = in->buf->buffer->use - (in->cur - in->base); |
| 3943 | } |
| 3944 | if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) |
| 3945 | ctxt->sax->setDocumentLocator(ctxt->userData, |
| 3946 | &xmlDefaultSAXLocator); |
| 3947 | if ((ctxt->sax) && (ctxt->sax->startDocument) && |
| 3948 | (!ctxt->disableSAX)) |
| 3949 | ctxt->sax->startDocument(ctxt->userData); |
| 3950 | |
| 3951 | cur = in->cur[0]; |
| 3952 | next = in->cur[1]; |
| 3953 | if ((cur == '<') && (next == '!') && |
| 3954 | (UPP(2) == 'D') && (UPP(3) == 'O') && |
| 3955 | (UPP(4) == 'C') && (UPP(5) == 'T') && |
| 3956 | (UPP(6) == 'Y') && (UPP(7) == 'P') && |
| 3957 | (UPP(8) == 'E')) { |
| 3958 | if ((!terminate) && |
| 3959 | (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0)) |
| 3960 | goto done; |
| 3961 | #ifdef DEBUG_PUSH |
| 3962 | xmlGenericError(xmlGenericErrorContext, |
| 3963 | "HPP: Parsing internal subset\n"); |
| 3964 | #endif |
| 3965 | htmlParseDocTypeDecl(ctxt); |
| 3966 | ctxt->instate = XML_PARSER_PROLOG; |
| 3967 | #ifdef DEBUG_PUSH |
| 3968 | xmlGenericError(xmlGenericErrorContext, |
| 3969 | "HPP: entering PROLOG\n"); |
| 3970 | #endif |
| 3971 | } else { |
| 3972 | ctxt->instate = XML_PARSER_MISC; |
| 3973 | } |
| 3974 | #ifdef DEBUG_PUSH |
| 3975 | xmlGenericError(xmlGenericErrorContext, |
| 3976 | "HPP: entering MISC\n"); |
| 3977 | #endif |
| 3978 | break; |
| 3979 | case XML_PARSER_MISC: |
| 3980 | SKIP_BLANKS; |
| 3981 | if (in->buf == NULL) |
| 3982 | avail = in->length - (in->cur - in->base); |
| 3983 | else |
| 3984 | avail = in->buf->buffer->use - (in->cur - in->base); |
| 3985 | if (avail < 2) |
| 3986 | goto done; |
| 3987 | cur = in->cur[0]; |
| 3988 | next = in->cur[1]; |
| 3989 | if ((cur == '<') && (next == '!') && |
| 3990 | (in->cur[2] == '-') && (in->cur[3] == '-')) { |
| 3991 | if ((!terminate) && |
| 3992 | (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0)) |
| 3993 | goto done; |
| 3994 | #ifdef DEBUG_PUSH |
| 3995 | xmlGenericError(xmlGenericErrorContext, |
| 3996 | "HPP: Parsing Comment\n"); |
| 3997 | #endif |
| 3998 | htmlParseComment(ctxt); |
| 3999 | ctxt->instate = XML_PARSER_MISC; |
| 4000 | } else if ((cur == '<') && (next == '!') && |
| 4001 | (UPP(2) == 'D') && (UPP(3) == 'O') && |
| 4002 | (UPP(4) == 'C') && (UPP(5) == 'T') && |
| 4003 | (UPP(6) == 'Y') && (UPP(7) == 'P') && |
| 4004 | (UPP(8) == 'E')) { |
| 4005 | if ((!terminate) && |
| 4006 | (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0)) |
| 4007 | goto done; |
| 4008 | #ifdef DEBUG_PUSH |
| 4009 | xmlGenericError(xmlGenericErrorContext, |
| 4010 | "HPP: Parsing internal subset\n"); |
| 4011 | #endif |
| 4012 | htmlParseDocTypeDecl(ctxt); |
| 4013 | ctxt->instate = XML_PARSER_PROLOG; |
| 4014 | #ifdef DEBUG_PUSH |
| 4015 | xmlGenericError(xmlGenericErrorContext, |
| 4016 | "HPP: entering PROLOG\n"); |
| 4017 | #endif |
| 4018 | } else if ((cur == '<') && (next == '!') && |
| 4019 | (avail < 9)) { |
| 4020 | goto done; |
| 4021 | } else { |
| 4022 | ctxt->instate = XML_PARSER_START_TAG; |
| 4023 | #ifdef DEBUG_PUSH |
| 4024 | xmlGenericError(xmlGenericErrorContext, |
| 4025 | "HPP: entering START_TAG\n"); |
| 4026 | #endif |
| 4027 | } |
| 4028 | break; |
| 4029 | case XML_PARSER_PROLOG: |
| 4030 | SKIP_BLANKS; |
| 4031 | if (in->buf == NULL) |
| 4032 | avail = in->length - (in->cur - in->base); |
| 4033 | else |
| 4034 | avail = in->buf->buffer->use - (in->cur - in->base); |
| 4035 | if (avail < 2) |
| 4036 | goto done; |
| 4037 | cur = in->cur[0]; |
| 4038 | next = in->cur[1]; |
| 4039 | if ((cur == '<') && (next == '!') && |
| 4040 | (in->cur[2] == '-') && (in->cur[3] == '-')) { |
| 4041 | if ((!terminate) && |
| 4042 | (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0)) |
| 4043 | goto done; |
| 4044 | #ifdef DEBUG_PUSH |
| 4045 | xmlGenericError(xmlGenericErrorContext, |
| 4046 | "HPP: Parsing Comment\n"); |
| 4047 | #endif |
| 4048 | htmlParseComment(ctxt); |
| 4049 | ctxt->instate = XML_PARSER_PROLOG; |
| 4050 | } else if ((cur == '<') && (next == '!') && |
| 4051 | (avail < 4)) { |
| 4052 | goto done; |
| 4053 | } else { |
| 4054 | ctxt->instate = XML_PARSER_START_TAG; |
| 4055 | #ifdef DEBUG_PUSH |
| 4056 | xmlGenericError(xmlGenericErrorContext, |
| 4057 | "HPP: entering START_TAG\n"); |
| 4058 | #endif |
| 4059 | } |
| 4060 | break; |
| 4061 | case XML_PARSER_EPILOG: |
| 4062 | if (in->buf == NULL) |
| 4063 | avail = in->length - (in->cur - in->base); |
| 4064 | else |
| 4065 | avail = in->buf->buffer->use - (in->cur - in->base); |
| 4066 | if (avail < 1) |
| 4067 | goto done; |
| 4068 | cur = in->cur[0]; |
| 4069 | if (IS_BLANK(cur)) { |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 4070 | htmlParseCharData(ctxt); |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 4071 | goto done; |
| 4072 | } |
| 4073 | if (avail < 2) |
| 4074 | goto done; |
| 4075 | next = in->cur[1]; |
| 4076 | if ((cur == '<') && (next == '!') && |
| 4077 | (in->cur[2] == '-') && (in->cur[3] == '-')) { |
| 4078 | if ((!terminate) && |
| 4079 | (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0)) |
| 4080 | goto done; |
| 4081 | #ifdef DEBUG_PUSH |
| 4082 | xmlGenericError(xmlGenericErrorContext, |
| 4083 | "HPP: Parsing Comment\n"); |
| 4084 | #endif |
| 4085 | htmlParseComment(ctxt); |
| 4086 | ctxt->instate = XML_PARSER_EPILOG; |
| 4087 | } else if ((cur == '<') && (next == '!') && |
| 4088 | (avail < 4)) { |
| 4089 | goto done; |
| 4090 | } else { |
| 4091 | ctxt->errNo = XML_ERR_DOCUMENT_END; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 4092 | ctxt->wellFormed = 0; |
| 4093 | ctxt->instate = XML_PARSER_EOF; |
| 4094 | #ifdef DEBUG_PUSH |
| 4095 | xmlGenericError(xmlGenericErrorContext, |
| 4096 | "HPP: entering EOF\n"); |
| 4097 | #endif |
| 4098 | if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) |
| 4099 | ctxt->sax->endDocument(ctxt->userData); |
| 4100 | goto done; |
| 4101 | } |
| 4102 | break; |
| 4103 | case XML_PARSER_START_TAG: { |
| 4104 | xmlChar *name, *oldname; |
| 4105 | int depth = ctxt->nameNr; |
| 4106 | htmlElemDescPtr info; |
| 4107 | |
| 4108 | if (avail < 2) |
| 4109 | goto done; |
| 4110 | cur = in->cur[0]; |
| 4111 | if (cur != '<') { |
| 4112 | ctxt->instate = XML_PARSER_CONTENT; |
| 4113 | #ifdef DEBUG_PUSH |
| 4114 | xmlGenericError(xmlGenericErrorContext, |
| 4115 | "HPP: entering CONTENT\n"); |
| 4116 | #endif |
| 4117 | break; |
| 4118 | } |
| 4119 | if ((!terminate) && |
| 4120 | (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0)) |
| 4121 | goto done; |
| 4122 | |
| 4123 | oldname = xmlStrdup(ctxt->name); |
| 4124 | htmlParseStartTag(ctxt); |
| 4125 | name = ctxt->name; |
| 4126 | #ifdef DEBUG |
| 4127 | if (oldname == NULL) |
| 4128 | xmlGenericError(xmlGenericErrorContext, |
| 4129 | "Start of element %s\n", name); |
| 4130 | else if (name == NULL) |
| 4131 | xmlGenericError(xmlGenericErrorContext, |
| 4132 | "Start of element failed, was %s\n", |
| 4133 | oldname); |
| 4134 | else |
| 4135 | xmlGenericError(xmlGenericErrorContext, |
| 4136 | "Start of element %s, was %s\n", |
| 4137 | name, oldname); |
| 4138 | #endif |
| 4139 | if (((depth == ctxt->nameNr) && |
| 4140 | (xmlStrEqual(oldname, ctxt->name))) || |
| 4141 | (name == NULL)) { |
| 4142 | if (CUR == '>') |
| 4143 | NEXT; |
| 4144 | if (oldname != NULL) |
| 4145 | xmlFree(oldname); |
| 4146 | break; |
| 4147 | } |
| 4148 | if (oldname != NULL) |
| 4149 | xmlFree(oldname); |
| 4150 | |
| 4151 | /* |
| 4152 | * Lookup the info for that element. |
| 4153 | */ |
| 4154 | info = htmlTagLookup(name); |
| 4155 | if (info == NULL) { |
| 4156 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 4157 | ctxt->sax->error(ctxt->userData, "Tag %s invalid\n", |
| 4158 | name); |
| 4159 | ctxt->wellFormed = 0; |
| 4160 | } else if (info->depr) { |
| 4161 | /*************************** |
| 4162 | if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL)) |
| 4163 | ctxt->sax->warning(ctxt->userData, |
| 4164 | "Tag %s is deprecated\n", |
| 4165 | name); |
| 4166 | ***************************/ |
| 4167 | } |
| 4168 | |
| 4169 | /* |
| 4170 | * Check for an Empty Element labelled the XML/SGML way |
| 4171 | */ |
| 4172 | if ((CUR == '/') && (NXT(1) == '>')) { |
| 4173 | SKIP(2); |
| 4174 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL)) |
| 4175 | ctxt->sax->endElement(ctxt->userData, name); |
| 4176 | oldname = htmlnamePop(ctxt); |
| 4177 | #ifdef DEBUG |
| 4178 | xmlGenericError(xmlGenericErrorContext,"End of tag the XML way: popping out %s\n", |
| 4179 | oldname); |
| 4180 | #endif |
| 4181 | if (oldname != NULL) |
| 4182 | xmlFree(oldname); |
| 4183 | ctxt->instate = XML_PARSER_CONTENT; |
| 4184 | #ifdef DEBUG_PUSH |
| 4185 | xmlGenericError(xmlGenericErrorContext, |
| 4186 | "HPP: entering CONTENT\n"); |
| 4187 | #endif |
| 4188 | break; |
| 4189 | } |
| 4190 | |
| 4191 | if (CUR == '>') { |
| 4192 | NEXT; |
| 4193 | } else { |
| 4194 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 4195 | ctxt->sax->error(ctxt->userData, |
| 4196 | "Couldn't find end of Start Tag %s\n", |
| 4197 | name); |
| 4198 | ctxt->wellFormed = 0; |
| 4199 | |
| 4200 | /* |
| 4201 | * end of parsing of this node. |
| 4202 | */ |
| 4203 | if (xmlStrEqual(name, ctxt->name)) { |
| 4204 | nodePop(ctxt); |
| 4205 | oldname = htmlnamePop(ctxt); |
| 4206 | #ifdef DEBUG |
| 4207 | xmlGenericError(xmlGenericErrorContext, |
| 4208 | "End of start tag problem: popping out %s\n", oldname); |
| 4209 | #endif |
| 4210 | if (oldname != NULL) |
| 4211 | xmlFree(oldname); |
| 4212 | } |
| 4213 | |
| 4214 | ctxt->instate = XML_PARSER_CONTENT; |
| 4215 | #ifdef DEBUG_PUSH |
| 4216 | xmlGenericError(xmlGenericErrorContext, |
| 4217 | "HPP: entering CONTENT\n"); |
| 4218 | #endif |
| 4219 | break; |
| 4220 | } |
| 4221 | |
| 4222 | /* |
| 4223 | * Check for an Empty Element from DTD definition |
| 4224 | */ |
| 4225 | if ((info != NULL) && (info->empty)) { |
| 4226 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL)) |
| 4227 | ctxt->sax->endElement(ctxt->userData, name); |
| 4228 | oldname = htmlnamePop(ctxt); |
| 4229 | #ifdef DEBUG |
| 4230 | xmlGenericError(xmlGenericErrorContext,"End of empty tag %s : popping out %s\n", name, oldname); |
| 4231 | #endif |
| 4232 | if (oldname != NULL) |
| 4233 | xmlFree(oldname); |
| 4234 | } |
| 4235 | ctxt->instate = XML_PARSER_CONTENT; |
| 4236 | #ifdef DEBUG_PUSH |
| 4237 | xmlGenericError(xmlGenericErrorContext, |
| 4238 | "HPP: entering CONTENT\n"); |
| 4239 | #endif |
| 4240 | break; |
| 4241 | } |
| 4242 | case XML_PARSER_CONTENT: { |
| 4243 | long cons; |
| 4244 | /* |
| 4245 | * Handle preparsed entities and charRef |
| 4246 | */ |
| 4247 | if (ctxt->token != 0) { |
| 4248 | xmlChar chr[2] = { 0 , 0 } ; |
| 4249 | |
| 4250 | chr[0] = (xmlChar) ctxt->token; |
| 4251 | htmlCheckParagraph(ctxt); |
| 4252 | if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) |
| 4253 | ctxt->sax->characters(ctxt->userData, chr, 1); |
| 4254 | ctxt->token = 0; |
| 4255 | ctxt->checkIndex = 0; |
| 4256 | } |
| 4257 | if ((avail == 1) && (terminate)) { |
| 4258 | cur = in->cur[0]; |
| 4259 | if ((cur != '<') && (cur != '&')) { |
| 4260 | if (ctxt->sax != NULL) { |
| 4261 | if (IS_BLANK(cur)) { |
| 4262 | if (ctxt->sax->ignorableWhitespace != NULL) |
| 4263 | ctxt->sax->ignorableWhitespace( |
| 4264 | ctxt->userData, &cur, 1); |
| 4265 | } else { |
| 4266 | htmlCheckParagraph(ctxt); |
| 4267 | if (ctxt->sax->characters != NULL) |
| 4268 | ctxt->sax->characters( |
| 4269 | ctxt->userData, &cur, 1); |
| 4270 | } |
| 4271 | } |
| 4272 | ctxt->token = 0; |
| 4273 | ctxt->checkIndex = 0; |
| 4274 | NEXT; |
| 4275 | } |
| 4276 | break; |
| 4277 | } |
| 4278 | if (avail < 2) |
| 4279 | goto done; |
| 4280 | cur = in->cur[0]; |
| 4281 | next = in->cur[1]; |
| 4282 | cons = ctxt->nbChars; |
| 4283 | if ((xmlStrEqual(ctxt->name, BAD_CAST"script")) || |
| 4284 | (xmlStrEqual(ctxt->name, BAD_CAST"style"))) { |
| 4285 | /* |
| 4286 | * Handle SCRIPT/STYLE separately |
| 4287 | */ |
| 4288 | if ((!terminate) && |
| 4289 | (htmlParseLookupSequence(ctxt, '<', '/', 0) < 0)) |
| 4290 | goto done; |
| 4291 | htmlParseScript(ctxt); |
| 4292 | if ((cur == '<') && (next == '/')) { |
| 4293 | ctxt->instate = XML_PARSER_END_TAG; |
| 4294 | ctxt->checkIndex = 0; |
| 4295 | #ifdef DEBUG_PUSH |
| 4296 | xmlGenericError(xmlGenericErrorContext, |
| 4297 | "HPP: entering END_TAG\n"); |
| 4298 | #endif |
| 4299 | break; |
| 4300 | } |
| 4301 | } else { |
| 4302 | /* |
| 4303 | * Sometimes DOCTYPE arrives in the middle of the document |
| 4304 | */ |
| 4305 | if ((cur == '<') && (next == '!') && |
| 4306 | (UPP(2) == 'D') && (UPP(3) == 'O') && |
| 4307 | (UPP(4) == 'C') && (UPP(5) == 'T') && |
| 4308 | (UPP(6) == 'Y') && (UPP(7) == 'P') && |
| 4309 | (UPP(8) == 'E')) { |
| 4310 | if ((!terminate) && |
| 4311 | (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0)) |
| 4312 | goto done; |
| 4313 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 4314 | ctxt->sax->error(ctxt->userData, |
| 4315 | "Misplaced DOCTYPE declaration\n"); |
| 4316 | ctxt->wellFormed = 0; |
| 4317 | htmlParseDocTypeDecl(ctxt); |
| 4318 | } else if ((cur == '<') && (next == '!') && |
| 4319 | (in->cur[2] == '-') && (in->cur[3] == '-')) { |
| 4320 | if ((!terminate) && |
| 4321 | (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0)) |
| 4322 | goto done; |
| 4323 | #ifdef DEBUG_PUSH |
| 4324 | xmlGenericError(xmlGenericErrorContext, |
| 4325 | "HPP: Parsing Comment\n"); |
| 4326 | #endif |
| 4327 | htmlParseComment(ctxt); |
| 4328 | ctxt->instate = XML_PARSER_CONTENT; |
| 4329 | } else if ((cur == '<') && (next == '!') && (avail < 4)) { |
| 4330 | goto done; |
| 4331 | } else if ((cur == '<') && (next == '/')) { |
| 4332 | ctxt->instate = XML_PARSER_END_TAG; |
| 4333 | ctxt->checkIndex = 0; |
| 4334 | #ifdef DEBUG_PUSH |
| 4335 | xmlGenericError(xmlGenericErrorContext, |
| 4336 | "HPP: entering END_TAG\n"); |
| 4337 | #endif |
| 4338 | break; |
| 4339 | } else if (cur == '<') { |
| 4340 | ctxt->instate = XML_PARSER_START_TAG; |
| 4341 | ctxt->checkIndex = 0; |
| 4342 | #ifdef DEBUG_PUSH |
| 4343 | xmlGenericError(xmlGenericErrorContext, |
| 4344 | "HPP: entering START_TAG\n"); |
| 4345 | #endif |
| 4346 | break; |
| 4347 | } else if (cur == '&') { |
| 4348 | if ((!terminate) && |
| 4349 | (htmlParseLookupSequence(ctxt, ';', 0, 0) < 0)) |
| 4350 | goto done; |
| 4351 | #ifdef DEBUG_PUSH |
| 4352 | xmlGenericError(xmlGenericErrorContext, |
| 4353 | "HPP: Parsing Reference\n"); |
| 4354 | #endif |
| 4355 | /* TODO: check generation of subtrees if noent !!! */ |
| 4356 | htmlParseReference(ctxt); |
| 4357 | } else { |
| 4358 | /* TODO Avoid the extra copy, handle directly !!!!!! */ |
| 4359 | /* |
| 4360 | * Goal of the following test is : |
| 4361 | * - minimize calls to the SAX 'character' callback |
| 4362 | * when they are mergeable |
| 4363 | */ |
| 4364 | if ((ctxt->inputNr == 1) && |
| 4365 | (avail < HTML_PARSER_BIG_BUFFER_SIZE)) { |
| 4366 | if ((!terminate) && |
| 4367 | (htmlParseLookupSequence(ctxt, '<', 0, 0) < 0)) |
| 4368 | goto done; |
| 4369 | } |
| 4370 | ctxt->checkIndex = 0; |
| 4371 | #ifdef DEBUG_PUSH |
| 4372 | xmlGenericError(xmlGenericErrorContext, |
| 4373 | "HPP: Parsing char data\n"); |
| 4374 | #endif |
Daniel Veillard | 56a4cb8 | 2001-03-24 17:00:36 +0000 | [diff] [blame] | 4375 | htmlParseCharData(ctxt); |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 4376 | } |
| 4377 | } |
| 4378 | if (cons == ctxt->nbChars) { |
| 4379 | if (ctxt->node != NULL) { |
| 4380 | if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) |
| 4381 | ctxt->sax->error(ctxt->userData, |
| 4382 | "detected an error in element content\n"); |
| 4383 | ctxt->wellFormed = 0; |
| 4384 | } |
| 4385 | NEXT; |
| 4386 | break; |
| 4387 | } |
| 4388 | |
| 4389 | break; |
| 4390 | } |
| 4391 | case XML_PARSER_END_TAG: |
| 4392 | if (avail < 2) |
| 4393 | goto done; |
| 4394 | if ((!terminate) && |
| 4395 | (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0)) |
| 4396 | goto done; |
| 4397 | htmlParseEndTag(ctxt); |
| 4398 | if (ctxt->nameNr == 0) { |
| 4399 | ctxt->instate = XML_PARSER_EPILOG; |
| 4400 | } else { |
| 4401 | ctxt->instate = XML_PARSER_CONTENT; |
| 4402 | } |
| 4403 | ctxt->checkIndex = 0; |
| 4404 | #ifdef DEBUG_PUSH |
| 4405 | xmlGenericError(xmlGenericErrorContext, |
| 4406 | "HPP: entering CONTENT\n"); |
| 4407 | #endif |
| 4408 | break; |
| 4409 | case XML_PARSER_CDATA_SECTION: |
| 4410 | xmlGenericError(xmlGenericErrorContext, |
| 4411 | "HPP: internal error, state == CDATA\n"); |
| 4412 | ctxt->instate = XML_PARSER_CONTENT; |
| 4413 | ctxt->checkIndex = 0; |
| 4414 | #ifdef DEBUG_PUSH |
| 4415 | xmlGenericError(xmlGenericErrorContext, |
| 4416 | "HPP: entering CONTENT\n"); |
| 4417 | #endif |
| 4418 | break; |
| 4419 | case XML_PARSER_DTD: |
| 4420 | xmlGenericError(xmlGenericErrorContext, |
| 4421 | "HPP: internal error, state == DTD\n"); |
| 4422 | ctxt->instate = XML_PARSER_CONTENT; |
| 4423 | ctxt->checkIndex = 0; |
| 4424 | #ifdef DEBUG_PUSH |
| 4425 | xmlGenericError(xmlGenericErrorContext, |
| 4426 | "HPP: entering CONTENT\n"); |
| 4427 | #endif |
| 4428 | break; |
| 4429 | case XML_PARSER_COMMENT: |
| 4430 | xmlGenericError(xmlGenericErrorContext, |
| 4431 | "HPP: internal error, state == COMMENT\n"); |
| 4432 | ctxt->instate = XML_PARSER_CONTENT; |
| 4433 | ctxt->checkIndex = 0; |
| 4434 | #ifdef DEBUG_PUSH |
| 4435 | xmlGenericError(xmlGenericErrorContext, |
| 4436 | "HPP: entering CONTENT\n"); |
| 4437 | #endif |
| 4438 | break; |
| 4439 | case XML_PARSER_PI: |
| 4440 | xmlGenericError(xmlGenericErrorContext, |
| 4441 | "HPP: internal error, state == PI\n"); |
| 4442 | ctxt->instate = XML_PARSER_CONTENT; |
| 4443 | ctxt->checkIndex = 0; |
| 4444 | #ifdef DEBUG_PUSH |
| 4445 | xmlGenericError(xmlGenericErrorContext, |
| 4446 | "HPP: entering CONTENT\n"); |
| 4447 | #endif |
| 4448 | break; |
| 4449 | case XML_PARSER_ENTITY_DECL: |
| 4450 | xmlGenericError(xmlGenericErrorContext, |
| 4451 | "HPP: internal error, state == ENTITY_DECL\n"); |
| 4452 | ctxt->instate = XML_PARSER_CONTENT; |
| 4453 | ctxt->checkIndex = 0; |
| 4454 | #ifdef DEBUG_PUSH |
| 4455 | xmlGenericError(xmlGenericErrorContext, |
| 4456 | "HPP: entering CONTENT\n"); |
| 4457 | #endif |
| 4458 | break; |
| 4459 | case XML_PARSER_ENTITY_VALUE: |
| 4460 | xmlGenericError(xmlGenericErrorContext, |
| 4461 | "HPP: internal error, state == ENTITY_VALUE\n"); |
| 4462 | ctxt->instate = XML_PARSER_CONTENT; |
| 4463 | ctxt->checkIndex = 0; |
| 4464 | #ifdef DEBUG_PUSH |
| 4465 | xmlGenericError(xmlGenericErrorContext, |
| 4466 | "HPP: entering DTD\n"); |
| 4467 | #endif |
| 4468 | break; |
| 4469 | case XML_PARSER_ATTRIBUTE_VALUE: |
| 4470 | xmlGenericError(xmlGenericErrorContext, |
| 4471 | "HPP: internal error, state == ATTRIBUTE_VALUE\n"); |
| 4472 | ctxt->instate = XML_PARSER_START_TAG; |
| 4473 | ctxt->checkIndex = 0; |
| 4474 | #ifdef DEBUG_PUSH |
| 4475 | xmlGenericError(xmlGenericErrorContext, |
| 4476 | "HPP: entering START_TAG\n"); |
| 4477 | #endif |
| 4478 | break; |
| 4479 | case XML_PARSER_SYSTEM_LITERAL: |
| 4480 | xmlGenericError(xmlGenericErrorContext, |
| 4481 | "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n"); |
| 4482 | ctxt->instate = XML_PARSER_CONTENT; |
| 4483 | ctxt->checkIndex = 0; |
| 4484 | #ifdef DEBUG_PUSH |
| 4485 | xmlGenericError(xmlGenericErrorContext, |
| 4486 | "HPP: entering CONTENT\n"); |
| 4487 | #endif |
| 4488 | break; |
| 4489 | case XML_PARSER_IGNORE: |
| 4490 | xmlGenericError(xmlGenericErrorContext, |
| 4491 | "HPP: internal error, state == XML_PARSER_IGNORE\n"); |
| 4492 | ctxt->instate = XML_PARSER_CONTENT; |
| 4493 | ctxt->checkIndex = 0; |
| 4494 | #ifdef DEBUG_PUSH |
| 4495 | xmlGenericError(xmlGenericErrorContext, |
| 4496 | "HPP: entering CONTENT\n"); |
| 4497 | #endif |
| 4498 | break; |
| 4499 | } |
| 4500 | } |
| 4501 | done: |
| 4502 | if ((avail == 0) && (terminate)) { |
Daniel Veillard | a3bfca5 | 2001-04-12 15:42:58 +0000 | [diff] [blame] | 4503 | htmlAutoCloseOnEnd(ctxt); |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 4504 | if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) { |
| 4505 | /* |
| 4506 | * SAX: end of the document processing. |
| 4507 | */ |
| 4508 | ctxt->instate = XML_PARSER_EOF; |
| 4509 | if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) |
| 4510 | ctxt->sax->endDocument(ctxt->userData); |
| 4511 | } |
| 4512 | } |
| 4513 | if ((ctxt->myDoc != NULL) && |
| 4514 | ((terminate) || (ctxt->instate == XML_PARSER_EOF) || |
| 4515 | (ctxt->instate == XML_PARSER_EPILOG))) { |
| 4516 | xmlDtdPtr dtd; |
| 4517 | dtd = xmlGetIntSubset(ctxt->myDoc); |
| 4518 | if (dtd == NULL) |
| 4519 | ctxt->myDoc->intSubset = |
| 4520 | xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML", |
| 4521 | BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN", |
| 4522 | BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd"); |
| 4523 | } |
| 4524 | #ifdef DEBUG_PUSH |
| 4525 | xmlGenericError(xmlGenericErrorContext, "HPP: done %d\n", ret); |
| 4526 | #endif |
| 4527 | return(ret); |
| 4528 | } |
| 4529 | |
| 4530 | /** |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 4531 | * htmlParseChunk: |
| 4532 | * @ctxt: an XML parser context |
| 4533 | * @chunk: an char array |
| 4534 | * @size: the size in byte of the chunk |
| 4535 | * @terminate: last chunk indicator |
| 4536 | * |
| 4537 | * Parse a Chunk of memory |
| 4538 | * |
| 4539 | * Returns zero if no error, the xmlParserErrors otherwise. |
| 4540 | */ |
| 4541 | int |
| 4542 | htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size, |
| 4543 | int terminate) { |
| 4544 | if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) && |
| 4545 | (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) { |
| 4546 | int base = ctxt->input->base - ctxt->input->buf->buffer->content; |
| 4547 | int cur = ctxt->input->cur - ctxt->input->base; |
| 4548 | |
| 4549 | xmlParserInputBufferPush(ctxt->input->buf, size, chunk); |
| 4550 | ctxt->input->base = ctxt->input->buf->buffer->content + base; |
| 4551 | ctxt->input->cur = ctxt->input->base + cur; |
| 4552 | #ifdef DEBUG_PUSH |
| 4553 | xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size); |
| 4554 | #endif |
| 4555 | |
| 4556 | if ((terminate) || (ctxt->input->buf->buffer->use > 80)) |
| 4557 | htmlParseTryOrFinish(ctxt, terminate); |
| 4558 | } else if (ctxt->instate != XML_PARSER_EOF) { |
| 4559 | xmlParserInputBufferPush(ctxt->input->buf, 0, ""); |
| 4560 | htmlParseTryOrFinish(ctxt, terminate); |
| 4561 | } |
| 4562 | if (terminate) { |
| 4563 | if ((ctxt->instate != XML_PARSER_EOF) && |
| 4564 | (ctxt->instate != XML_PARSER_EPILOG) && |
| 4565 | (ctxt->instate != XML_PARSER_MISC)) { |
| 4566 | ctxt->errNo = XML_ERR_DOCUMENT_END; |
Owen Taylor | 3473f88 | 2001-02-23 17:55:21 +0000 | [diff] [blame] | 4567 | ctxt->wellFormed = 0; |
| 4568 | } |
| 4569 | if (ctxt->instate != XML_PARSER_EOF) { |
| 4570 | if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) |
| 4571 | ctxt->sax->endDocument(ctxt->userData); |
| 4572 | } |
| 4573 | ctxt->instate = XML_PARSER_EOF; |
| 4574 | } |
| 4575 | return((xmlParserErrors) ctxt->errNo); |
| 4576 | } |
| 4577 | |
| 4578 | /************************************************************************ |
| 4579 | * * |
| 4580 | * User entry points * |
| 4581 | * * |
| 4582 | ************************************************************************/ |
| 4583 | |
| 4584 | /** |
| 4585 | * htmlCreatePushParserCtxt : |
| 4586 | * @sax: a SAX handler |
| 4587 | * @user_data: The user data returned on SAX callbacks |
| 4588 | * @chunk: a pointer to an array of chars |
| 4589 | * @size: number of chars in the array |
| 4590 | * @filename: an optional file name or URI |
| 4591 | * @enc: an optional encoding |
| 4592 | * |
| 4593 | * Create a parser context for using the HTML parser in push mode |
| 4594 | * To allow content encoding detection, @size should be >= 4 |
| 4595 | * The value of @filename is used for fetching external entities |
| 4596 | * and error/warning reports. |
| 4597 | * |
| 4598 | * Returns the new parser context or NULL |
| 4599 | */ |
| 4600 | htmlParserCtxtPtr |
| 4601 | htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data, |
| 4602 | const char *chunk, int size, const char *filename, |
| 4603 | xmlCharEncoding enc) { |
| 4604 | htmlParserCtxtPtr ctxt; |
| 4605 | htmlParserInputPtr inputStream; |
| 4606 | xmlParserInputBufferPtr buf; |
| 4607 | |
| 4608 | buf = xmlAllocParserInputBuffer(enc); |
| 4609 | if (buf == NULL) return(NULL); |
| 4610 | |
| 4611 | ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt)); |
| 4612 | if (ctxt == NULL) { |
| 4613 | xmlFree(buf); |
| 4614 | return(NULL); |
| 4615 | } |
| 4616 | memset(ctxt, 0, sizeof(htmlParserCtxt)); |
| 4617 | htmlInitParserCtxt(ctxt); |
| 4618 | if (sax != NULL) { |
| 4619 | if (ctxt->sax != &htmlDefaultSAXHandler) |
| 4620 | xmlFree(ctxt->sax); |
| 4621 | ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler)); |
| 4622 | if (ctxt->sax == NULL) { |
| 4623 | xmlFree(buf); |
| 4624 | xmlFree(ctxt); |
| 4625 | return(NULL); |
| 4626 | } |
| 4627 | memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler)); |
| 4628 | if (user_data != NULL) |
| 4629 | ctxt->userData = user_data; |
| 4630 | } |
| 4631 | if (filename == NULL) { |
| 4632 | ctxt->directory = NULL; |
| 4633 | } else { |
| 4634 | ctxt->directory = xmlParserGetDirectory(filename); |
| 4635 | } |
| 4636 | |
| 4637 | inputStream = htmlNewInputStream(ctxt); |
| 4638 | if (inputStream == NULL) { |
| 4639 | xmlFreeParserCtxt(ctxt); |
| 4640 | return(NULL); |
| 4641 | } |
| 4642 | |
| 4643 | if (filename == NULL) |
| 4644 | inputStream->filename = NULL; |
| 4645 | else |
| 4646 | inputStream->filename = xmlMemStrdup(filename); |
| 4647 | inputStream->buf = buf; |
| 4648 | inputStream->base = inputStream->buf->buffer->content; |
| 4649 | inputStream->cur = inputStream->buf->buffer->content; |
| 4650 | |
| 4651 | inputPush(ctxt, inputStream); |
| 4652 | |
| 4653 | if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) && |
| 4654 | (ctxt->input->buf != NULL)) { |
| 4655 | xmlParserInputBufferPush(ctxt->input->buf, size, chunk); |
| 4656 | #ifdef DEBUG_PUSH |
| 4657 | xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size); |
| 4658 | #endif |
| 4659 | } |
| 4660 | |
| 4661 | return(ctxt); |
| 4662 | } |
| 4663 | |
| 4664 | /** |
| 4665 | * htmlSAXParseDoc : |
| 4666 | * @cur: a pointer to an array of xmlChar |
| 4667 | * @encoding: a free form C string describing the HTML document encoding, or NULL |
| 4668 | * @sax: the SAX handler block |
| 4669 | * @userData: if using SAX, this pointer will be provided on callbacks. |
| 4670 | * |
| 4671 | * parse an HTML in-memory document and build a tree. |
| 4672 | * It use the given SAX function block to handle the parsing callback. |
| 4673 | * If sax is NULL, fallback to the default DOM tree building routines. |
| 4674 | * |
| 4675 | * Returns the resulting document tree |
| 4676 | */ |
| 4677 | |
| 4678 | htmlDocPtr |
| 4679 | htmlSAXParseDoc(xmlChar *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) { |
| 4680 | htmlDocPtr ret; |
| 4681 | htmlParserCtxtPtr ctxt; |
| 4682 | |
| 4683 | if (cur == NULL) return(NULL); |
| 4684 | |
| 4685 | |
| 4686 | ctxt = htmlCreateDocParserCtxt(cur, encoding); |
| 4687 | if (ctxt == NULL) return(NULL); |
| 4688 | if (sax != NULL) { |
| 4689 | ctxt->sax = sax; |
| 4690 | ctxt->userData = userData; |
| 4691 | } |
| 4692 | |
| 4693 | htmlParseDocument(ctxt); |
| 4694 | ret = ctxt->myDoc; |
| 4695 | if (sax != NULL) { |
| 4696 | ctxt->sax = NULL; |
| 4697 | ctxt->userData = NULL; |
| 4698 | } |
| 4699 | htmlFreeParserCtxt(ctxt); |
| 4700 | |
| 4701 | return(ret); |
| 4702 | } |
| 4703 | |
| 4704 | /** |
| 4705 | * htmlParseDoc : |
| 4706 | * @cur: a pointer to an array of xmlChar |
| 4707 | * @encoding: a free form C string describing the HTML document encoding, or NULL |
| 4708 | * |
| 4709 | * parse an HTML in-memory document and build a tree. |
| 4710 | * |
| 4711 | * Returns the resulting document tree |
| 4712 | */ |
| 4713 | |
| 4714 | htmlDocPtr |
| 4715 | htmlParseDoc(xmlChar *cur, const char *encoding) { |
| 4716 | return(htmlSAXParseDoc(cur, encoding, NULL, NULL)); |
| 4717 | } |
| 4718 | |
| 4719 | |
| 4720 | /** |
| 4721 | * htmlCreateFileParserCtxt : |
| 4722 | * @filename: the filename |
| 4723 | * @encoding: a free form C string describing the HTML document encoding, or NULL |
| 4724 | * |
| 4725 | * Create a parser context for a file content. |
| 4726 | * Automatic support for ZLIB/Compress compressed document is provided |
| 4727 | * by default if found at compile-time. |
| 4728 | * |
| 4729 | * Returns the new parser context or NULL |
| 4730 | */ |
| 4731 | htmlParserCtxtPtr |
| 4732 | htmlCreateFileParserCtxt(const char *filename, const char *encoding) |
| 4733 | { |
| 4734 | htmlParserCtxtPtr ctxt; |
| 4735 | htmlParserInputPtr inputStream; |
| 4736 | xmlParserInputBufferPtr buf; |
| 4737 | /* htmlCharEncoding enc; */ |
| 4738 | xmlChar *content, *content_line = (xmlChar *) "charset="; |
| 4739 | |
| 4740 | buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE); |
| 4741 | if (buf == NULL) return(NULL); |
| 4742 | |
| 4743 | ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt)); |
| 4744 | if (ctxt == NULL) { |
| 4745 | perror("malloc"); |
| 4746 | return(NULL); |
| 4747 | } |
| 4748 | memset(ctxt, 0, sizeof(htmlParserCtxt)); |
| 4749 | htmlInitParserCtxt(ctxt); |
| 4750 | inputStream = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput)); |
| 4751 | if (inputStream == NULL) { |
| 4752 | perror("malloc"); |
| 4753 | xmlFree(ctxt); |
| 4754 | return(NULL); |
| 4755 | } |
| 4756 | memset(inputStream, 0, sizeof(htmlParserInput)); |
| 4757 | |
| 4758 | inputStream->filename = xmlMemStrdup(filename); |
| 4759 | inputStream->line = 1; |
| 4760 | inputStream->col = 1; |
| 4761 | inputStream->buf = buf; |
| 4762 | inputStream->directory = NULL; |
| 4763 | |
| 4764 | inputStream->base = inputStream->buf->buffer->content; |
| 4765 | inputStream->cur = inputStream->buf->buffer->content; |
| 4766 | inputStream->free = NULL; |
| 4767 | |
| 4768 | inputPush(ctxt, inputStream); |
| 4769 | |
| 4770 | /* set encoding */ |
| 4771 | if (encoding) { |
| 4772 | content = xmlMalloc (xmlStrlen(content_line) + strlen(encoding) + 1); |
| 4773 | if (content) { |
| 4774 | strcpy ((char *)content, (char *)content_line); |
| 4775 | strcat ((char *)content, (char *)encoding); |
| 4776 | htmlCheckEncoding (ctxt, content); |
| 4777 | xmlFree (content); |
| 4778 | } |
| 4779 | } |
| 4780 | |
| 4781 | return(ctxt); |
| 4782 | } |
| 4783 | |
| 4784 | /** |
| 4785 | * htmlSAXParseFile : |
| 4786 | * @filename: the filename |
| 4787 | * @encoding: a free form C string describing the HTML document encoding, or NULL |
| 4788 | * @sax: the SAX handler block |
| 4789 | * @userData: if using SAX, this pointer will be provided on callbacks. |
| 4790 | * |
| 4791 | * parse an HTML file and build a tree. Automatic support for ZLIB/Compress |
| 4792 | * compressed document is provided by default if found at compile-time. |
| 4793 | * It use the given SAX function block to handle the parsing callback. |
| 4794 | * If sax is NULL, fallback to the default DOM tree building routines. |
| 4795 | * |
| 4796 | * Returns the resulting document tree |
| 4797 | */ |
| 4798 | |
| 4799 | htmlDocPtr |
| 4800 | htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax, |
| 4801 | void *userData) { |
| 4802 | htmlDocPtr ret; |
| 4803 | htmlParserCtxtPtr ctxt; |
| 4804 | htmlSAXHandlerPtr oldsax = NULL; |
| 4805 | |
| 4806 | ctxt = htmlCreateFileParserCtxt(filename, encoding); |
| 4807 | if (ctxt == NULL) return(NULL); |
| 4808 | if (sax != NULL) { |
| 4809 | oldsax = ctxt->sax; |
| 4810 | ctxt->sax = sax; |
| 4811 | ctxt->userData = userData; |
| 4812 | } |
| 4813 | |
| 4814 | htmlParseDocument(ctxt); |
| 4815 | |
| 4816 | ret = ctxt->myDoc; |
| 4817 | if (sax != NULL) { |
| 4818 | ctxt->sax = oldsax; |
| 4819 | ctxt->userData = NULL; |
| 4820 | } |
| 4821 | htmlFreeParserCtxt(ctxt); |
| 4822 | |
| 4823 | return(ret); |
| 4824 | } |
| 4825 | |
| 4826 | /** |
| 4827 | * htmlParseFile : |
| 4828 | * @filename: the filename |
| 4829 | * @encoding: a free form C string describing the HTML document encoding, or NULL |
| 4830 | * |
| 4831 | * parse an HTML file and build a tree. Automatic support for ZLIB/Compress |
| 4832 | * compressed document is provided by default if found at compile-time. |
| 4833 | * |
| 4834 | * Returns the resulting document tree |
| 4835 | */ |
| 4836 | |
| 4837 | htmlDocPtr |
| 4838 | htmlParseFile(const char *filename, const char *encoding) { |
| 4839 | return(htmlSAXParseFile(filename, encoding, NULL, NULL)); |
| 4840 | } |
| 4841 | |
| 4842 | /** |
| 4843 | * htmlHandleOmittedElem: |
| 4844 | * @val: int 0 or 1 |
| 4845 | * |
| 4846 | * Set and return the previous value for handling HTML omitted tags. |
| 4847 | * |
| 4848 | * Returns the last value for 0 for no handling, 1 for auto insertion. |
| 4849 | */ |
| 4850 | |
| 4851 | int |
| 4852 | htmlHandleOmittedElem(int val) { |
| 4853 | int old = htmlOmittedDefaultValue; |
| 4854 | |
| 4855 | htmlOmittedDefaultValue = val; |
| 4856 | return(old); |
| 4857 | } |
| 4858 | |
| 4859 | #endif /* LIBXML_HTML_ENABLED */ |