Annotation of XML/HTMLparser.c, revision 1.68
1.1 daniel 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:
9: #ifdef WIN32
1.29 daniel 10: #include "win32config.h"
1.1 daniel 11: #else
1.13 daniel 12: #include "config.h"
1.1 daniel 13: #endif
1.13 daniel 14:
1.39 daniel 15: #include "xmlversion.h"
16: #ifdef LIBXML_HTML_ENABLED
17:
1.1 daniel 18: #include <stdio.h>
1.50 veillard 19: #include <string.h>
1.13 daniel 20: #ifdef HAVE_CTYPE_H
1.1 daniel 21: #include <ctype.h>
1.13 daniel 22: #endif
23: #ifdef HAVE_STDLIB_H
1.1 daniel 24: #include <stdlib.h>
1.13 daniel 25: #endif
26: #ifdef HAVE_SYS_STAT_H
1.1 daniel 27: #include <sys/stat.h>
1.13 daniel 28: #endif
1.1 daniel 29: #ifdef HAVE_FCNTL_H
30: #include <fcntl.h>
31: #endif
32: #ifdef HAVE_UNISTD_H
33: #include <unistd.h>
34: #endif
35: #ifdef HAVE_ZLIB_H
36: #include <zlib.h>
37: #endif
38:
1.39 daniel 39: #include <libxml/xmlmemory.h>
40: #include <libxml/tree.h>
41: #include <libxml/HTMLparser.h>
42: #include <libxml/entities.h>
43: #include <libxml/encoding.h>
1.50 veillard 44: #include <libxml/parser.h>
1.39 daniel 45: #include <libxml/valid.h>
46: #include <libxml/parserInternals.h>
47: #include <libxml/xmlIO.h>
1.31 daniel 48: #include "xml-error.h"
1.5 daniel 49:
50: #define HTML_MAX_NAMELEN 1000
1.53 veillard 51: #define HTML_PARSER_BIG_BUFFER_SIZE 1000
1.31 daniel 52: #define HTML_PARSER_BUFFER_SIZE 100
1.1 daniel 53:
54: /* #define DEBUG */
1.31 daniel 55: /* #define DEBUG_PUSH */
1.1 daniel 56:
57: /************************************************************************
58: * *
59: * Parser stacks related functions and macros *
60: * *
61: ************************************************************************/
62:
63: /*
64: * Generic function for accessing stacks in the Parser Context
65: */
66:
1.30 daniel 67: #define PUSH_AND_POP(scope, type, name) \
68: scope int html##name##Push(htmlParserCtxtPtr ctxt, type value) { \
1.1 daniel 69: if (ctxt->name##Nr >= ctxt->name##Max) { \
70: ctxt->name##Max *= 2; \
1.50 veillard 71: ctxt->name##Tab = (type *) xmlRealloc(ctxt->name##Tab, \
1.1 daniel 72: ctxt->name##Max * sizeof(ctxt->name##Tab[0])); \
73: if (ctxt->name##Tab == NULL) { \
74: fprintf(stderr, "realloc failed !\n"); \
1.33 daniel 75: return(0); \
1.1 daniel 76: } \
77: } \
78: ctxt->name##Tab[ctxt->name##Nr] = value; \
79: ctxt->name = value; \
80: return(ctxt->name##Nr++); \
81: } \
1.30 daniel 82: scope type html##name##Pop(htmlParserCtxtPtr ctxt) { \
1.1 daniel 83: type ret; \
1.18 daniel 84: if (ctxt->name##Nr < 0) return(0); \
1.1 daniel 85: ctxt->name##Nr--; \
1.18 daniel 86: if (ctxt->name##Nr < 0) return(0); \
1.1 daniel 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:
1.30 daniel 96: PUSH_AND_POP(extern, xmlNodePtr, node)
97: PUSH_AND_POP(extern, xmlChar*, name)
1.1 daniel 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: *
1.14 daniel 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
1.1 daniel 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.
1.14 daniel 111: * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
1.1 daniel 112: * to compare on ASCII based substring.
1.14 daniel 113: * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR
1.1 daniel 114: * it should be used only to compare on ASCII based substring.
1.14 daniel 115: * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
1.1 daniel 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))
1.36 daniel 128:
1.26 daniel 129: #define SKIP(val) ctxt->nbChars += (val),ctxt->input->cur += (val)
1.36 daniel 130:
1.1 daniel 131: #define NXT(val) ctxt->input->cur[(val)]
1.36 daniel 132:
1.1 daniel 133: #define UPP(val) (toupper(ctxt->input->cur[(val)]))
1.36 daniel 134:
1.1 daniel 135: #define CUR_PTR ctxt->input->cur
1.36 daniel 136:
1.5 daniel 137: #define SHRINK xmlParserInputShrink(ctxt->input)
1.36 daniel 138:
1.5 daniel 139: #define GROW xmlParserInputGrow(ctxt->input, INPUT_CHUNK)
1.1 daniel 140:
1.36 daniel 141: #define CURRENT ((int) (*ctxt->input->cur))
1.1 daniel 142:
1.53 veillard 143: #define SKIP_BLANKS htmlSkipBlankChars(ctxt);
144:
145: #if 0
146: #define CUR ((int) (*ctxt->input->cur))
1.36 daniel 147: #define NEXT htmlNextChar(ctxt);
1.53 veillard 148: #else
149: /* Inported from XML */
150:
151: /* #define CUR (ctxt->token ? ctxt->token : (int) (*ctxt->input->cur)) */
152: #define CUR ((int) (*ctxt->input->cur))
153: #define NEXT xmlNextChar(ctxt);ctxt->nbChars++;
154:
155: #define RAW (ctxt->token ? -1 : (*ctxt->input->cur))
156: #define NXT(val) ctxt->input->cur[(val)]
157: #define CUR_PTR ctxt->input->cur
158:
159:
160: #define NEXTL(l) \
161: if (*(ctxt->input->cur) == '\n') { \
162: ctxt->input->line++; ctxt->input->col = 1; \
163: } else ctxt->input->col++; \
164: ctxt->token = 0; ctxt->input->cur += l; ctxt->nbChars++;
165:
166: /************
167: \
168: if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
169: if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt);
170: ************/
171:
172: #define CUR_CHAR(l) htmlCurrentChar(ctxt, &l);
173: #define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l);
174:
175: #define COPY_BUF(l,b,i,v) \
176: if (l == 1) b[i++] = (xmlChar) v; \
177: else i += xmlCopyChar(l,&b[i],v);
178: #endif
179:
180: /**
181: * htmlCurrentChar:
182: * @ctxt: the HTML parser context
183: * @len: pointer to the length of the char read
184: *
185: * The current char value, if using UTF-8 this may actaully span multiple
186: * bytes in the input buffer. Implement the end of line normalization:
187: * 2.11 End-of-Line Handling
188: * If the encoding is unspecified, in the case we find an ISO-Latin-1
189: * char, then the encoding converter is plugged in automatically.
190: *
191: * Returns the current char value and its lenght
192: */
193:
194: int
195: htmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
196: if (ctxt->instate == XML_PARSER_EOF)
197: return(0);
1.35 daniel 198:
1.53 veillard 199: if (ctxt->token != 0) {
200: *len = 0;
201: return(ctxt->token);
202: }
203: if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
204: /*
205: * We are supposed to handle UTF8, check it's valid
206: * From rfc2044: encoding of the Unicode values on UTF-8:
207: *
208: * UCS-4 range (hex.) UTF-8 octet sequence (binary)
209: * 0000 0000-0000 007F 0xxxxxxx
210: * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
211: * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
212: *
213: * Check for the 0x110000 limit too
214: */
215: const unsigned char *cur = ctxt->input->cur;
216: unsigned char c;
217: unsigned int val;
218:
219: c = *cur;
220: if (c & 0x80) {
221: if (cur[1] == 0)
222: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
223: if ((cur[1] & 0xc0) != 0x80)
224: goto encoding_error;
225: if ((c & 0xe0) == 0xe0) {
226:
227: if (cur[2] == 0)
228: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
229: if ((cur[2] & 0xc0) != 0x80)
230: goto encoding_error;
231: if ((c & 0xf0) == 0xf0) {
232: if (cur[3] == 0)
233: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
234: if (((c & 0xf8) != 0xf0) ||
235: ((cur[3] & 0xc0) != 0x80))
236: goto encoding_error;
237: /* 4-byte code */
238: *len = 4;
239: val = (cur[0] & 0x7) << 18;
240: val |= (cur[1] & 0x3f) << 12;
241: val |= (cur[2] & 0x3f) << 6;
242: val |= cur[3] & 0x3f;
243: } else {
244: /* 3-byte code */
245: *len = 3;
246: val = (cur[0] & 0xf) << 12;
247: val |= (cur[1] & 0x3f) << 6;
248: val |= cur[2] & 0x3f;
249: }
250: } else {
251: /* 2-byte code */
252: *len = 2;
253: val = (cur[0] & 0x1f) << 6;
254: val |= cur[1] & 0x3f;
255: }
256: if (!IS_CHAR(val)) {
1.67 veillard 257: ctxt->errNo = XML_ERR_INVALID_ENCODING;
1.53 veillard 258: if ((ctxt->sax != NULL) &&
259: (ctxt->sax->error != NULL))
260: ctxt->sax->error(ctxt->userData,
261: "Char 0x%X out of allowed range\n", val);
262: ctxt->wellFormed = 0;
263: ctxt->disableSAX = 1;
264: }
265: return(val);
266: } else {
267: /* 1-byte code */
268: *len = 1;
269: return((int) *ctxt->input->cur);
270: }
271: }
272: /*
273: * Assume it's a fixed lenght encoding (1) with
274: * a compatibke encoding for the ASCII set, since
275: * XML constructs only use < 128 chars
276: */
277: *len = 1;
278: if ((int) *ctxt->input->cur < 0x80)
279: return((int) *ctxt->input->cur);
280:
281: /*
282: * Humm this is bad, do an automatic flow conversion
283: */
284: xmlSwitchEncoding(ctxt, XML_CHAR_ENCODING_8859_1);
285: ctxt->charset = XML_CHAR_ENCODING_UTF8;
286: return(xmlCurrentChar(ctxt, len));
287:
288: encoding_error:
289: /*
290: * If we detect an UTF8 error that probably mean that the
291: * input encoding didn't get properly advertized in the
292: * declaration header. Report the error and switch the encoding
293: * to ISO-Latin-1 (if you don't like this policy, just declare the
294: * encoding !)
295: */
1.67 veillard 296: ctxt->errNo = XML_ERR_INVALID_ENCODING;
1.53 veillard 297: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) {
298: ctxt->sax->error(ctxt->userData,
299: "Input is not proper UTF-8, indicate encoding !\n");
300: ctxt->sax->error(ctxt->userData, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
301: ctxt->input->cur[0], ctxt->input->cur[1],
302: ctxt->input->cur[2], ctxt->input->cur[3]);
303: }
304:
305: ctxt->charset = XML_CHAR_ENCODING_8859_1;
306: *len = 1;
307: return((int) *ctxt->input->cur);
308: }
1.35 daniel 309:
310: /**
311: * htmlNextChar:
312: * @ctxt: the HTML parser context
313: *
314: * Skip to the next char input char.
315: */
316:
317: void
318: htmlNextChar(htmlParserCtxtPtr ctxt) {
1.44 daniel 319: if (ctxt->instate == XML_PARSER_EOF)
320: return;
1.35 daniel 321: if ((*ctxt->input->cur == 0) &&
322: (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
323: xmlPopInput(ctxt);
324: } else {
325: if (*(ctxt->input->cur) == '\n') {
326: ctxt->input->line++; ctxt->input->col = 1;
327: } else ctxt->input->col++;
328: ctxt->input->cur++;
329: ctxt->nbChars++;
330: if (*ctxt->input->cur == 0)
331: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
332: }
333: }
1.5 daniel 334:
1.36 daniel 335: /**
336: * htmlSkipBlankChars:
337: * @ctxt: the HTML parser context
338: *
339: * skip all blanks character found at that point in the input streams.
340: *
341: * Returns the number of space chars skipped
342: */
343:
344: int
345: htmlSkipBlankChars(xmlParserCtxtPtr ctxt) {
346: int res = 0;
347:
348: while (IS_BLANK(*(ctxt->input->cur))) {
349: if ((*ctxt->input->cur == 0) &&
350: (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
351: xmlPopInput(ctxt);
352: } else {
353: if (*(ctxt->input->cur) == '\n') {
354: ctxt->input->line++; ctxt->input->col = 1;
355: } else ctxt->input->col++;
356: ctxt->input->cur++;
357: ctxt->nbChars++;
358: if (*ctxt->input->cur == 0)
359: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
360: }
361: res++;
362: }
363: return(res);
364: }
1.1 daniel 365:
366:
1.5 daniel 367:
1.1 daniel 368: /************************************************************************
369: * *
370: * The list of HTML elements and their properties *
371: * *
372: ************************************************************************/
373:
374: /*
375: * Start Tag: 1 means the start tag can be ommited
376: * End Tag: 1 means the end tag can be ommited
377: * 2 means it's forbidden (empty elements)
378: * Depr: this element is deprecated
379: * DTD: 1 means that this element is valid only in the Loose DTD
380: * 2 means that this element is valid only in the Frameset DTD
381: *
382: * Name,Start Tag,End Tag, Empty, Depr., DTD, Description
383: */
384: htmlElemDesc html40ElementTable[] = {
1.26 daniel 385: { "a", 0, 0, 0, 0, 0, "anchor " },
386: { "abbr", 0, 0, 0, 0, 0, "abbreviated form" },
387: { "acronym", 0, 0, 0, 0, 0, "" },
388: { "address", 0, 0, 0, 0, 0, "information on author " },
389: { "applet", 0, 0, 0, 1, 1, "java applet " },
390: { "area", 0, 2, 1, 0, 0, "client-side image map area " },
391: { "b", 0, 0, 0, 0, 0, "bold text style" },
392: { "base", 0, 2, 1, 0, 0, "document base uri " },
393: { "basefont", 0, 2, 1, 1, 1, "base font size " },
394: { "bdo", 0, 0, 0, 0, 0, "i18n bidi over-ride " },
395: { "big", 0, 0, 0, 0, 0, "large text style" },
396: { "blockquote", 0, 0, 0, 0, 0, "long quotation " },
397: { "body", 1, 1, 0, 0, 0, "document body " },
398: { "br", 0, 2, 1, 0, 0, "forced line break " },
399: { "button", 0, 0, 0, 0, 0, "push button " },
400: { "caption", 0, 0, 0, 0, 0, "table caption " },
401: { "center", 0, 0, 0, 1, 1, "shorthand for div align=center " },
402: { "cite", 0, 0, 0, 0, 0, "citation" },
403: { "code", 0, 0, 0, 0, 0, "computer code fragment" },
404: { "col", 0, 2, 1, 0, 0, "table column " },
405: { "colgroup", 0, 1, 0, 0, 0, "table column group " },
406: { "dd", 0, 1, 0, 0, 0, "definition description " },
407: { "del", 0, 0, 0, 0, 0, "deleted text " },
408: { "dfn", 0, 0, 0, 0, 0, "instance definition" },
409: { "dir", 0, 0, 0, 1, 1, "directory list" },
410: { "div", 0, 0, 0, 0, 0, "generic language/style container"},
411: { "dl", 0, 0, 0, 0, 0, "definition list " },
412: { "dt", 0, 1, 0, 0, 0, "definition term " },
413: { "em", 0, 0, 0, 0, 0, "emphasis" },
414: { "fieldset", 0, 0, 0, 0, 0, "form control group " },
415: { "font", 0, 0, 0, 1, 1, "local change to font " },
416: { "form", 0, 0, 0, 0, 0, "interactive form " },
417: { "frame", 0, 2, 1, 0, 2, "subwindow " },
418: { "frameset", 0, 0, 0, 0, 2, "window subdivision" },
419: { "h1", 0, 0, 0, 0, 0, "heading " },
420: { "h2", 0, 0, 0, 0, 0, "heading " },
421: { "h3", 0, 0, 0, 0, 0, "heading " },
422: { "h4", 0, 0, 0, 0, 0, "heading " },
423: { "h5", 0, 0, 0, 0, 0, "heading " },
424: { "h6", 0, 0, 0, 0, 0, "heading " },
425: { "head", 1, 1, 0, 0, 0, "document head " },
426: { "hr", 0, 2, 1, 0, 0, "horizontal rule " },
427: { "html", 1, 1, 0, 0, 0, "document root element " },
428: { "i", 0, 0, 0, 0, 0, "italic text style" },
429: { "iframe", 0, 0, 0, 0, 1, "inline subwindow " },
430: { "img", 0, 2, 1, 0, 0, "embedded image " },
431: { "input", 0, 2, 1, 0, 0, "form control " },
432: { "ins", 0, 0, 0, 0, 0, "inserted text" },
433: { "isindex", 0, 2, 1, 1, 1, "single line prompt " },
434: { "kbd", 0, 0, 0, 0, 0, "text to be entered by the user" },
435: { "label", 0, 0, 0, 0, 0, "form field label text " },
436: { "legend", 0, 0, 0, 0, 0, "fieldset legend " },
437: { "li", 0, 1, 0, 0, 0, "list item " },
438: { "link", 0, 2, 1, 0, 0, "a media-independent link " },
439: { "map", 0, 0, 0, 0, 0, "client-side image map " },
440: { "menu", 0, 0, 0, 1, 1, "menu list " },
441: { "meta", 0, 2, 1, 0, 0, "generic metainformation " },
442: { "noframes", 0, 0, 0, 0, 2, "alternate content container for non frame-based rendering " },
443: { "noscript", 0, 0, 0, 0, 0, "alternate content container for non script-based rendering " },
444: { "object", 0, 0, 0, 0, 0, "generic embedded object " },
445: { "ol", 0, 0, 0, 0, 0, "ordered list " },
446: { "optgroup", 0, 0, 0, 0, 0, "option group " },
447: { "option", 0, 1, 0, 0, 0, "selectable choice " },
448: { "p", 0, 1, 0, 0, 0, "paragraph " },
449: { "param", 0, 2, 1, 0, 0, "named property value " },
450: { "pre", 0, 0, 0, 0, 0, "preformatted text " },
451: { "q", 0, 0, 0, 0, 0, "short inline quotation " },
452: { "s", 0, 0, 0, 1, 1, "strike-through text style" },
453: { "samp", 0, 0, 0, 0, 0, "sample program output, scripts, etc." },
454: { "script", 0, 0, 0, 0, 0, "script statements " },
455: { "select", 0, 0, 0, 0, 0, "option selector " },
456: { "small", 0, 0, 0, 0, 0, "small text style" },
457: { "span", 0, 0, 0, 0, 0, "generic language/style container " },
458: { "strike", 0, 0, 0, 1, 1, "strike-through text" },
459: { "strong", 0, 0, 0, 0, 0, "strong emphasis" },
460: { "style", 0, 0, 0, 0, 0, "style info " },
461: { "sub", 0, 0, 0, 0, 0, "subscript" },
462: { "sup", 0, 0, 0, 0, 0, "superscript " },
463: { "table", 0, 0, 0, 0, 0, " " },
464: { "tbody", 1, 1, 0, 0, 0, "table body " },
465: { "td", 0, 1, 0, 0, 0, "table data cell" },
466: { "textarea", 0, 0, 0, 0, 0, "multi-line text field " },
467: { "tfoot", 0, 1, 0, 0, 0, "table footer " },
468: { "th", 0, 1, 0, 0, 0, "table header cell" },
469: { "thead", 0, 1, 0, 0, 0, "table header " },
470: { "title", 0, 0, 0, 0, 0, "document title " },
471: { "tr", 0, 1, 0, 0, 0, "table row " },
472: { "tt", 0, 0, 0, 0, 0, "teletype or monospaced text style" },
473: { "u", 0, 0, 0, 1, 1, "underlined text style" },
474: { "ul", 0, 0, 0, 0, 0, "unordered list " },
475: { "var", 0, 0, 0, 0, 0, "instance of a variable or program argument" },
1.1 daniel 476: };
477:
478: /*
479: * start tags that imply the end of a current element
480: * any tag of each line implies the end of the current element if the type of
481: * that element is in the same line
482: */
1.8 daniel 483: char *htmlEquEnd[] = {
1.26 daniel 484: "dt", "dd", "li", "option", NULL,
485: "h1", "h2", "h3", "h4", "h5", "h6", NULL,
486: "ol", "menu", "dir", "address", "pre", "listing", "xmp", NULL,
1.1 daniel 487: NULL
488: };
489: /*
490: * acording the HTML DTD, HR should be added to the 2nd line above, as it
491: * is not allowed within a H1, H2, H3, etc. But we should tolerate that case
492: * because many documents contain rules in headings...
493: */
494:
495: /*
496: * start tags that imply the end of current element
497: */
1.8 daniel 498: char *htmlStartClose[] = {
1.26 daniel 499: "form", "form", "p", "hr", "h1", "h2", "h3", "h4", "h5", "h6",
500: "dl", "ul", "ol", "menu", "dir", "address", "pre",
501: "listing", "xmp", "head", NULL,
502: "head", "p", NULL,
503: "title", "p", NULL,
504: "body", "head", "style", "link", "title", "p", NULL,
505: "li", "p", "h1", "h2", "h3", "h4", "h5", "h6", "dl", "address",
506: "pre", "listing", "xmp", "head", "li", NULL,
507: "hr", "p", "head", NULL,
508: "h1", "p", "head", NULL,
509: "h2", "p", "head", NULL,
510: "h3", "p", "head", NULL,
511: "h4", "p", "head", NULL,
512: "h5", "p", "head", NULL,
513: "h6", "p", "head", NULL,
514: "dir", "p", "head", NULL,
515: "address", "p", "head", "ul", NULL,
516: "pre", "p", "head", "ul", NULL,
517: "listing", "p", "head", NULL,
518: "xmp", "p", "head", NULL,
519: "blockquote", "p", "head", NULL,
520: "dl", "p", "dt", "menu", "dir", "address", "pre", "listing",
521: "xmp", "head", NULL,
522: "dt", "p", "menu", "dir", "address", "pre", "listing", "xmp",
523: "head", "dd", NULL,
524: "dd", "p", "menu", "dir", "address", "pre", "listing", "xmp",
525: "head", "dt", NULL,
526: "ul", "p", "head", "ol", "menu", "dir", "address", "pre",
527: "listing", "xmp", NULL,
528: "ol", "p", "head", "ul", NULL,
529: "menu", "p", "head", "ul", NULL,
530: "p", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", NULL,
531: "div", "p", "head", NULL,
532: "noscript", "p", "head", NULL,
533: "center", "font", "b", "i", "p", "head", NULL,
534: "a", "a", NULL,
535: "caption", "p", NULL,
536: "colgroup", "caption", "colgroup", "col", "p", NULL,
537: "col", "caption", "col", "p", NULL,
538: "table", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", "pre",
539: "listing", "xmp", "a", NULL,
540: "th", "th", "td", NULL,
541: "td", "th", "td", "p", NULL,
542: "tr", "th", "td", "tr", "caption", "col", "colgroup", "p", NULL,
543: "thead", "caption", "col", "colgroup", NULL,
544: "tfoot", "th", "td", "tr", "caption", "col", "colgroup", "thead",
545: "tbody", "p", NULL,
546: "tbody", "th", "td", "tr", "caption", "col", "colgroup", "thead",
547: "tfoot", "tbody", "p", NULL,
548: "optgroup", "option", NULL,
549: "fieldset", "legend", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6",
550: "pre", "listing", "xmp", "a", NULL,
1.1 daniel 551: NULL
552: };
553:
1.59 veillard 554: /*
555: * The list of HTML elements which are supposed not to have
556: * CDATA content and where a p element will be implied
557: *
558: * TODO: extend that list by reading the HTML SGML DtD on
559: * implied paragraph
560: */
561: static char *htmlNoContentElements[] = {
562: "html",
563: "head",
564: "body",
565: NULL
566: };
567:
1.43 daniel 568:
1.8 daniel 569: static char** htmlStartCloseIndex[100];
1.1 daniel 570: static int htmlStartCloseIndexinitialized = 0;
571:
572: /************************************************************************
573: * *
574: * functions to handle HTML specific data *
575: * *
576: ************************************************************************/
577:
578: /**
579: * htmlInitAutoClose:
580: *
581: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
582: *
583: */
584: void
585: htmlInitAutoClose(void) {
586: int index, i = 0;
587:
588: if (htmlStartCloseIndexinitialized) return;
589:
590: for (index = 0;index < 100;index ++) htmlStartCloseIndex[index] = NULL;
591: index = 0;
592: while ((htmlStartClose[i] != NULL) && (index < 100 - 1)) {
593: htmlStartCloseIndex[index++] = &htmlStartClose[i];
594: while (htmlStartClose[i] != NULL) i++;
595: i++;
596: }
597: }
598:
599: /**
600: * htmlTagLookup:
601: * @tag: The tag name
602: *
603: * Lookup the HTML tag in the ElementTable
604: *
605: * Returns the related htmlElemDescPtr or NULL if not found.
606: */
607: htmlElemDescPtr
1.14 daniel 608: htmlTagLookup(const xmlChar *tag) {
1.61 veillard 609: int i;
1.1 daniel 610:
611: for (i = 0; i < (sizeof(html40ElementTable) /
612: sizeof(html40ElementTable[0]));i++) {
1.8 daniel 613: if (!xmlStrcmp(tag, BAD_CAST html40ElementTable[i].name))
1.1 daniel 614: return(&html40ElementTable[i]);
615: }
616: return(NULL);
617: }
618:
619: /**
620: * htmlCheckAutoClose:
1.50 veillard 621: * @newtag: The new tag name
622: * @oldtag: The old tag name
1.1 daniel 623: *
624: * Checks wether the new tag is one of the registered valid tags for closing old.
625: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
626: *
627: * Returns 0 if no, 1 if yes.
628: */
629: int
1.50 veillard 630: htmlCheckAutoClose(const xmlChar *newtag, const xmlChar *oldtag) {
1.1 daniel 631: int i, index;
1.64 veillard 632: char **close = NULL;
1.1 daniel 633:
634: if (htmlStartCloseIndexinitialized == 0) htmlInitAutoClose();
635:
636: /* inefficient, but not a big deal */
637: for (index = 0; index < 100;index++) {
638: close = htmlStartCloseIndex[index];
639: if (close == NULL) return(0);
1.50 veillard 640: if (!xmlStrcmp(BAD_CAST *close, newtag)) break;
1.1 daniel 641: }
642:
643: i = close - htmlStartClose;
644: i++;
645: while (htmlStartClose[i] != NULL) {
1.50 veillard 646: if (!xmlStrcmp(BAD_CAST htmlStartClose[i], oldtag)) {
1.1 daniel 647: return(1);
648: }
649: i++;
650: }
651: return(0);
652: }
653:
654: /**
1.50 veillard 655: * htmlAutoCloseOnClose:
656: * @ctxt: an HTML parser context
657: * @newtag: The new tag name
658: *
659: * The HTmL DtD allows an ending tag to implicitely close other tags.
660: */
661: void
662: htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
663: htmlElemDescPtr info;
664: xmlChar *oldname;
665: int i;
666:
667: #ifdef DEBUG
668: fprintf(stderr,"Close of %s stack: %d elements\n", newtag, ctxt->nameNr);
669: for (i = 0;i < ctxt->nameNr;i++)
670: fprintf(stderr,"%d : %s\n", i, ctxt->nameTab[i]);
671: #endif
672:
673: for (i = (ctxt->nameNr - 1);i >= 0;i--) {
674: if (!xmlStrcmp(newtag, ctxt->nameTab[i])) break;
675: }
676: if (i < 0) return;
677:
678: while (xmlStrcmp(newtag, ctxt->name)) {
679: info = htmlTagLookup(ctxt->name);
680: if ((info == NULL) || (info->endTag == 1)) {
681: #ifdef DEBUG
682: fprintf(stderr,"htmlAutoCloseOnClose: %s closes %s\n", newtag, ctxt->name);
683: #endif
684: } else {
685: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
686: ctxt->sax->error(ctxt->userData,
687: "Opening and ending tag mismatch: %s and %s\n",
688: newtag, ctxt->name);
689: ctxt->wellFormed = 0;
690: }
691: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
692: ctxt->sax->endElement(ctxt->userData, ctxt->name);
693: oldname = htmlnamePop(ctxt);
694: if (oldname != NULL) {
695: #ifdef DEBUG
696: fprintf(stderr,"htmlAutoCloseOnClose: popped %s\n", oldname);
697: #endif
698: xmlFree(oldname);
699: }
700: }
701: }
702:
703: /**
1.1 daniel 704: * htmlAutoClose:
705: * @ctxt: an HTML parser context
1.50 veillard 706: * @newtag: The new tag name or NULL
1.1 daniel 707: *
708: * The HTmL DtD allows a tag to implicitely close other tags.
709: * The list is kept in htmlStartClose array. This function is
710: * called when a new tag has been detected and generates the
711: * appropriates closes if possible/needed.
1.50 veillard 712: * If newtag is NULL this mean we are at the end of the resource
1.47 daniel 713: * and we should check
1.1 daniel 714: */
715: void
1.50 veillard 716: htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
1.15 daniel 717: xmlChar *oldname;
1.50 veillard 718: while ((newtag != NULL) && (ctxt->name != NULL) &&
719: (htmlCheckAutoClose(newtag, ctxt->name))) {
1.1 daniel 720: #ifdef DEBUG
1.50 veillard 721: fprintf(stderr,"htmlAutoClose: %s closes %s\n", newtag, ctxt->name);
1.1 daniel 722: #endif
723: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1.15 daniel 724: ctxt->sax->endElement(ctxt->userData, ctxt->name);
1.24 daniel 725: oldname = htmlnamePop(ctxt);
1.18 daniel 726: if (oldname != NULL) {
727: #ifdef DEBUG
728: fprintf(stderr,"htmlAutoClose: popped %s\n", oldname);
729: #endif
1.17 daniel 730: xmlFree(oldname);
1.18 daniel 731: }
1.1 daniel 732: }
1.50 veillard 733: if (newtag == NULL) {
1.49 daniel 734: htmlAutoCloseOnClose(ctxt, BAD_CAST"head");
735: htmlAutoCloseOnClose(ctxt, BAD_CAST"body");
736: htmlAutoCloseOnClose(ctxt, BAD_CAST"html");
737: }
1.50 veillard 738: while ((newtag == NULL) && (ctxt->name != NULL) &&
1.47 daniel 739: ((!xmlStrcmp(ctxt->name, BAD_CAST"head")) ||
740: (!xmlStrcmp(ctxt->name, BAD_CAST"body")) ||
741: (!xmlStrcmp(ctxt->name, BAD_CAST"html")))) {
742: #ifdef DEBUG
743: fprintf(stderr,"htmlAutoClose: EOF closes %s\n", ctxt->name);
744: #endif
745: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
746: ctxt->sax->endElement(ctxt->userData, ctxt->name);
747: oldname = htmlnamePop(ctxt);
748: if (oldname != NULL) {
749: #ifdef DEBUG
750: fprintf(stderr,"htmlAutoClose: popped %s\n", oldname);
751: #endif
752: xmlFree(oldname);
753: }
754: }
755:
1.1 daniel 756: }
757:
758: /**
1.28 daniel 759: * htmlAutoCloseTag:
760: * @doc: the HTML document
761: * @name: The tag name
762: * @elem: the HTML element
763: *
764: * The HTmL DtD allows a tag to implicitely close other tags.
765: * The list is kept in htmlStartClose array. This function checks
766: * if the element or one of it's children would autoclose the
767: * given tag.
768: *
769: * Returns 1 if autoclose, 0 otherwise
770: */
771: int
772: htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) {
773: htmlNodePtr child;
774:
775: if (elem == NULL) return(1);
776: if (!xmlStrcmp(name, elem->name)) return(0);
777: if (htmlCheckAutoClose(elem->name, name)) return(1);
1.37 daniel 778: child = elem->children;
1.28 daniel 779: while (child != NULL) {
780: if (htmlAutoCloseTag(doc, name, child)) return(1);
781: child = child->next;
782: }
783: return(0);
784: }
785:
786: /**
787: * htmlIsAutoClosed:
788: * @doc: the HTML document
789: * @elem: the HTML element
790: *
791: * The HTmL DtD allows a tag to implicitely close other tags.
792: * The list is kept in htmlStartClose array. This function checks
793: * if a tag is autoclosed by one of it's child
794: *
795: * Returns 1 if autoclosed, 0 otherwise
796: */
797: int
798: htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) {
799: htmlNodePtr child;
800:
801: if (elem == NULL) return(1);
1.37 daniel 802: child = elem->children;
1.28 daniel 803: while (child != NULL) {
804: if (htmlAutoCloseTag(doc, elem->name, child)) return(1);
805: child = child->next;
806: }
807: return(0);
808: }
809:
810: /**
1.43 daniel 811: * htmlCheckImplied:
812: * @ctxt: an HTML parser context
1.50 veillard 813: * @newtag: The new tag name
1.43 daniel 814: *
815: * The HTmL DtD allows a tag to exists only implicitely
816: * called when a new tag has been detected and generates the
817: * appropriates implicit tags if missing
818: */
819: void
1.50 veillard 820: htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
821: if (!xmlStrcmp(newtag, BAD_CAST"html"))
1.43 daniel 822: return;
823: if (ctxt->nameNr <= 0) {
824: #ifdef DEBUG
825: fprintf(stderr,"Implied element html: pushed html\n");
826: #endif
827: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"html"));
828: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
829: ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL);
830: }
1.50 veillard 831: if ((!xmlStrcmp(newtag, BAD_CAST"body")) || (!xmlStrcmp(newtag, BAD_CAST"head")))
1.43 daniel 832: return;
833: if (ctxt->nameNr <= 1) {
1.50 veillard 834: if ((!xmlStrcmp(newtag, BAD_CAST"script")) ||
835: (!xmlStrcmp(newtag, BAD_CAST"style")) ||
836: (!xmlStrcmp(newtag, BAD_CAST"meta")) ||
837: (!xmlStrcmp(newtag, BAD_CAST"link")) ||
838: (!xmlStrcmp(newtag, BAD_CAST"title")) ||
839: (!xmlStrcmp(newtag, BAD_CAST"base"))) {
1.43 daniel 840: /*
841: * dropped OBJECT ... i you put it first BODY will be
842: * assumed !
843: */
844: #ifdef DEBUG
845: fprintf(stderr,"Implied element head: pushed head\n");
846: #endif
847: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"head"));
848: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
849: ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL);
850: } else {
851: #ifdef DEBUG
852: fprintf(stderr,"Implied element body: pushed body\n");
853: #endif
854: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"body"));
855: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
856: ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL);
857: }
858: }
859: }
860:
1.59 veillard 861: /**
862: * htmlCheckParagraph
863: * @ctxt: an HTML parser context
864: *
865: * Check whether a p element need to be implied before inserting
866: * characters in the current element.
867: *
868: * Returns 1 if a paragraph has been inserted, 0 if not and -1
869: * in case of error.
870: */
871:
872: int
873: htmlCheckParagraph(htmlParserCtxtPtr ctxt) {
874: const xmlChar *tag;
875: int i;
876:
877: if (ctxt == NULL)
878: return(-1);
879: tag = ctxt->name;
880: if (tag == NULL) {
881: htmlAutoClose(ctxt, BAD_CAST"p");
882: htmlCheckImplied(ctxt, BAD_CAST"p");
883: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p"));
884: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
885: ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
886: return(1);
887: }
888: for (i = 0; htmlNoContentElements[i] != NULL; i++) {
889: if (!xmlStrcmp(tag, BAD_CAST htmlNoContentElements[i])) {
890: #ifdef DEBUG
891: fprintf(stderr,"Implied element paragraph\n");
892: #endif
893: htmlAutoClose(ctxt, BAD_CAST"p");
894: htmlCheckImplied(ctxt, BAD_CAST"p");
895: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p"));
896: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
897: ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
898: return(1);
899: }
900: }
901: return(0);
902: }
903:
1.1 daniel 904: /************************************************************************
905: * *
906: * The list of HTML predefined entities *
907: * *
908: ************************************************************************/
909:
910:
911: htmlEntityDesc html40EntitiesTable[] = {
912: /*
1.61 veillard 913: * the 4 absolute ones, plus apostrophe.
1.1 daniel 914: */
915: { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
916: { 38, "amp", "ampersand, U+0026 ISOnum" },
1.61 veillard 917: { 39, "apos", "single quote" },
1.1 daniel 918: { 60, "lt", "less-than sign, U+003C ISOnum" },
919: { 62, "gt", "greater-than sign, U+003E ISOnum" },
920:
921: /*
922: * A bunch still in the 128-255 range
923: * Replacing them depend really on the charset used.
924: */
925: { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
926: { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
927: { 162, "cent", "cent sign, U+00A2 ISOnum" },
928: { 163, "pound","pound sign, U+00A3 ISOnum" },
929: { 164, "curren","currency sign, U+00A4 ISOnum" },
930: { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
931: { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
932: { 167, "sect", "section sign, U+00A7 ISOnum" },
933: { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
934: { 169, "copy", "copyright sign, U+00A9 ISOnum" },
935: { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
936: { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
937: { 172, "not", "not sign, U+00AC ISOnum" },
938: { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
939: { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
940: { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
941: { 176, "deg", "degree sign, U+00B0 ISOnum" },
942: { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
943: { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
944: { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
945: { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
946: { 181, "micro","micro sign, U+00B5 ISOnum" },
947: { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1.7 daniel 948: { 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1.1 daniel 949: { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
950: { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
951: { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1.7 daniel 952: { 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1.1 daniel 953: { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
954: { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
955: { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
956: { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
957: { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
958: { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
959: { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
960: { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
961: { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
962: { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
963: { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
964: { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
965: { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
966: { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
967: { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
968: { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
969: { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
970: { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
971: { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
972: { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
973: { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
974: { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
975: { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
976: { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
977: { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
978: { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
979: { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
980: { 215, "times","multiplication sign, U+00D7 ISOnum" },
1.7 daniel 981: { 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1.1 daniel 982: { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
983: { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
984: { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
985: { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
986: { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
987: { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
988: { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
989: { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
990: { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
991: { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
992: { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
993: { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
994: { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
995: { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
996: { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
997: { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
998: { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
999: { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
1000: { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
1001: { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
1002: { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
1003: { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
1004: { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
1005: { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
1006: { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
1007: { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
1008: { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
1009: { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
1010: { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
1011: { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
1012: { 247, "divide","division sign, U+00F7 ISOnum" },
1013: { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
1014: { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
1015: { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
1016: { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
1017: { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
1018: { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
1019: { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
1020: { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
1021:
1.61 veillard 1022: { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
1023: { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
1024: { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
1025: { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
1026: { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
1027:
1.1 daniel 1028: /*
1029: * Anything below should really be kept as entities references
1030: */
1031: { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
1032:
1.61 veillard 1033: { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
1034: { 732, "tilde","small tilde, U+02DC ISOdia" },
1035:
1.1 daniel 1036: { 913, "Alpha","greek capital letter alpha, U+0391" },
1037: { 914, "Beta", "greek capital letter beta, U+0392" },
1038: { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
1039: { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
1040: { 917, "Epsilon","greek capital letter epsilon, U+0395" },
1041: { 918, "Zeta", "greek capital letter zeta, U+0396" },
1042: { 919, "Eta", "greek capital letter eta, U+0397" },
1043: { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
1044: { 921, "Iota", "greek capital letter iota, U+0399" },
1045: { 922, "Kappa","greek capital letter kappa, U+039A" },
1046: { 923, "Lambda""greek capital letter lambda, U+039B ISOgrk3" },
1047: { 924, "Mu", "greek capital letter mu, U+039C" },
1048: { 925, "Nu", "greek capital letter nu, U+039D" },
1049: { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
1050: { 927, "Omicron","greek capital letter omicron, U+039F" },
1051: { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
1052: { 929, "Rho", "greek capital letter rho, U+03A1" },
1053: { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
1054: { 932, "Tau", "greek capital letter tau, U+03A4" },
1055: { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
1056: { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
1057: { 935, "Chi", "greek capital letter chi, U+03A7" },
1058: { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
1059: { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
1060:
1061: { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
1062: { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
1063: { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
1064: { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
1065: { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
1066: { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
1067: { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
1068: { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
1069: { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
1070: { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
1071: { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
1072: { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
1073: { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
1074: { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
1075: { 959, "omicron","greek small letter omicron, U+03BF NEW" },
1076: { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
1077: { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
1078: { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
1079: { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
1080: { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
1081: { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
1082: { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
1083: { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
1084: { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
1085: { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
1086: { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
1087: { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
1088: { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
1089:
1.61 veillard 1090: { 8194, "ensp", "en space, U+2002 ISOpub" },
1091: { 8195, "emsp", "em space, U+2003 ISOpub" },
1092: { 8201, "thinsp","thin space, U+2009 ISOpub" },
1093: { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
1094: { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
1095: { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
1096: { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
1097: { 8211, "ndash","en dash, U+2013 ISOpub" },
1098: { 8212, "mdash","em dash, U+2014 ISOpub" },
1099: { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
1100: { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
1101: { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
1102: { 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
1103: { 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
1104: { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
1105: { 8224, "dagger","dagger, U+2020 ISOpub" },
1106: { 8225, "Dagger","double dagger, U+2021 ISOpub" },
1107:
1.1 daniel 1108: { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
1109: { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
1.61 veillard 1110:
1111: { 8240, "permil","per mille sign, U+2030 ISOtech" },
1112:
1.1 daniel 1113: { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
1114: { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
1.61 veillard 1115:
1116: { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
1117: { 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
1118:
1.1 daniel 1119: { 8254, "oline","overline = spacing overscore, U+203E NEW" },
1120: { 8260, "frasl","fraction slash, U+2044 NEW" },
1121:
1.61 veillard 1122: { 8364, "euro", "euro sign, U+20AC NEW" },
1123:
1124: { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
1.7 daniel 1125: { 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
1.1 daniel 1126: { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
1127: { 8482, "trade","trade mark sign, U+2122 ISOnum" },
1128: { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
1129: { 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
1130: { 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
1131: { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
1132: { 8595, "darr", "downwards arrow, U+2193 ISOnum" },
1133: { 8596, "harr", "left right arrow, U+2194 ISOamsa" },
1134: { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
1135: { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
1136: { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
1137: { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
1138: { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
1139: { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
1140:
1141: { 8704, "forall","for all, U+2200 ISOtech" },
1142: { 8706, "part", "partial differential, U+2202 ISOtech" },
1143: { 8707, "exist","there exists, U+2203 ISOtech" },
1144: { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
1145: { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
1146: { 8712, "isin", "element of, U+2208 ISOtech" },
1147: { 8713, "notin","not an element of, U+2209 ISOtech" },
1148: { 8715, "ni", "contains as member, U+220B ISOtech" },
1149: { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
1150: { 8721, "sum", "n-ary sumation, U+2211 ISOamsb" },
1151: { 8722, "minus","minus sign, U+2212 ISOtech" },
1152: { 8727, "lowast","asterisk operator, U+2217 ISOtech" },
1153: { 8730, "radic","square root = radical sign, U+221A ISOtech" },
1154: { 8733, "prop", "proportional to, U+221D ISOtech" },
1155: { 8734, "infin","infinity, U+221E ISOtech" },
1156: { 8736, "ang", "angle, U+2220 ISOamso" },
1157: { 8743, "and", "logical and = wedge, U+2227 ISOtech" },
1158: { 8744, "or", "logical or = vee, U+2228 ISOtech" },
1159: { 8745, "cap", "intersection = cap, U+2229 ISOtech" },
1160: { 8746, "cup", "union = cup, U+222A ISOtech" },
1161: { 8747, "int", "integral, U+222B ISOtech" },
1162: { 8756, "there4","therefore, U+2234 ISOtech" },
1163: { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
1164: { 8773, "cong", "approximately equal to, U+2245 ISOtech" },
1165: { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
1166: { 8800, "ne", "not equal to, U+2260 ISOtech" },
1167: { 8801, "equiv","identical to, U+2261 ISOtech" },
1168: { 8804, "le", "less-than or equal to, U+2264 ISOtech" },
1169: { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
1170: { 8834, "sub", "subset of, U+2282 ISOtech" },
1171: { 8835, "sup", "superset of, U+2283 ISOtech" },
1172: { 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
1173: { 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
1174: { 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
1175: { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
1176: { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
1177: { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
1178: { 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
1179: { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
1180: { 8969, "rceil","right ceiling, U+2309 ISOamsc" },
1181: { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
1182: { 8971, "rfloor","right floor, U+230B ISOamsc" },
1183: { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
1184: { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
1185: { 9674, "loz", "lozenge, U+25CA ISOpub" },
1186:
1187: { 9824, "spades","black spade suit, U+2660 ISOpub" },
1188: { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
1189: { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
1190: { 9830, "diams","black diamond suit, U+2666 ISOpub" },
1191:
1192: };
1193:
1194: /************************************************************************
1195: * *
1196: * Commodity functions to handle entities *
1197: * *
1198: ************************************************************************/
1199:
1200: /*
1201: * Macro used to grow the current buffer.
1202: */
1203: #define growBuffer(buffer) { \
1204: buffer##_size *= 2; \
1.14 daniel 1205: buffer = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
1.1 daniel 1206: if (buffer == NULL) { \
1207: perror("realloc failed"); \
1.33 daniel 1208: return(NULL); \
1.1 daniel 1209: } \
1210: }
1211:
1212: /**
1213: * htmlEntityLookup:
1214: * @name: the entity name
1215: *
1216: * Lookup the given entity in EntitiesTable
1217: *
1218: * TODO: the linear scan is really ugly, an hash table is really needed.
1219: *
1220: * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1221: */
1222: htmlEntityDescPtr
1.14 daniel 1223: htmlEntityLookup(const xmlChar *name) {
1.1 daniel 1224: int i;
1225:
1226: for (i = 0;i < (sizeof(html40EntitiesTable)/
1227: sizeof(html40EntitiesTable[0]));i++) {
1.8 daniel 1228: if (!xmlStrcmp(name, BAD_CAST html40EntitiesTable[i].name)) {
1.1 daniel 1229: #ifdef DEBUG
1.18 daniel 1230: fprintf(stderr,"Found entity %s\n", name);
1.1 daniel 1231: #endif
1232: return(&html40EntitiesTable[i]);
1233: }
1234: }
1235: return(NULL);
1236: }
1237:
1.53 veillard 1238: /**
1.61 veillard 1239: * htmlEntityValueLookup:
1240: * @value: the entity's unicode value
1241: *
1242: * Lookup the given entity in EntitiesTable
1243: *
1244: * TODO: the linear scan is really ugly, an hash table is really needed.
1245: *
1246: * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1247: */
1248: htmlEntityDescPtr
1249: htmlEntityValueLookup(int value) {
1250: int i;
1251: #ifdef DEBUG
1252: int lv = 0;
1253: #endif
1254:
1255: for (i = 0;i < (sizeof(html40EntitiesTable)/
1256: sizeof(html40EntitiesTable[0]));i++) {
1257: if (html40EntitiesTable[i].value >= value) {
1258: if (html40EntitiesTable[i].value > value)
1259: break;
1260: #ifdef DEBUG
1261: fprintf(stderr,"Found entity %s\n", html40EntitiesTable[i].name);
1262: #endif
1263: return(&html40EntitiesTable[i]);
1264: }
1265: #ifdef DEBUG
1266: if (lv > html40EntitiesTable[i].value) {
1267: fprintf(stderr, "html40EntitiesTable[] is not sorted (%d > %d)!\n",
1268: lv, html40EntitiesTable[i].value);
1269: }
1270: lv = html40EntitiesTable[i].value;
1271: #endif
1272: }
1273: return(NULL);
1274: }
1275:
1276: /**
1.53 veillard 1277: * UTF8ToHtml:
1278: * @out: a pointer to an array of bytes to store the result
1279: * @outlen: the length of @out
1280: * @in: a pointer to an array of UTF-8 chars
1281: * @inlen: the length of @in
1282: *
1283: * Take a block of UTF-8 chars in and try to convert it to an ASCII
1284: * plus HTML entities block of chars out.
1285: *
1286: * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1287: * The value of @inlen after return is the number of octets consumed
1288: * as the return value is positive, else unpredictiable.
1289: * The value of @outlen after return is the number of octets consumed.
1290: */
1291: int
1292: UTF8ToHtml(unsigned char* out, int *outlen,
1293: const unsigned char* in, int *inlen) {
1294: const unsigned char* processed = in;
1295: const unsigned char* outend;
1296: const unsigned char* outstart = out;
1297: const unsigned char* instart = in;
1298: const unsigned char* inend;
1299: unsigned int c, d;
1300: int trailing;
1301:
1302: if (in == NULL) {
1303: /*
1304: * initialization nothing to do
1305: */
1306: *outlen = 0;
1307: *inlen = 0;
1308: return(0);
1309: }
1310: inend = in + (*inlen);
1311: outend = out + (*outlen);
1312: while (in < inend) {
1313: d = *in++;
1314: if (d < 0x80) { c= d; trailing= 0; }
1315: else if (d < 0xC0) {
1316: /* trailing byte in leading position */
1317: *outlen = out - outstart;
1318: *inlen = processed - instart;
1319: return(-2);
1320: } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
1321: else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
1322: else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
1323: else {
1324: /* no chance for this in Ascii */
1325: *outlen = out - outstart;
1326: *inlen = processed - instart;
1327: return(-2);
1328: }
1329:
1330: if (inend - in < trailing) {
1331: break;
1332: }
1333:
1334: for ( ; trailing; trailing--) {
1335: if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
1336: break;
1337: c <<= 6;
1338: c |= d & 0x3F;
1339: }
1340:
1341: /* assertion: c is a single UTF-4 value */
1342: if (c < 0x80) {
1.62 veillard 1343: if (out + 1 >= outend)
1.53 veillard 1344: break;
1345: *out++ = c;
1346: } else {
1.61 veillard 1347: int len;
1348: htmlEntityDescPtr ent;
1349:
1.53 veillard 1350: /*
1351: * Try to lookup a predefined HTML entity for it
1352: */
1353:
1.61 veillard 1354: ent = htmlEntityValueLookup(c);
1355: if (ent == NULL) {
1356: /* no chance for this in Ascii */
1357: *outlen = out - outstart;
1358: *inlen = processed - instart;
1359: return(-2);
1.53 veillard 1360: }
1.61 veillard 1361: len = strlen(ent->name);
1.62 veillard 1362: if (out + 2 + len >= outend)
1.53 veillard 1363: break;
1364: *out++ = '&';
1.61 veillard 1365: memcpy(out, ent->name, len);
1366: out += len;
1.53 veillard 1367: *out++ = ';';
1368: }
1369: processed = in;
1370: }
1371: *outlen = out - outstart;
1372: *inlen = processed - instart;
1373: return(0);
1374: }
1375:
1.62 veillard 1376: /**
1377: * htmlEncodeEntities:
1378: * @out: a pointer to an array of bytes to store the result
1379: * @outlen: the length of @out
1380: * @in: a pointer to an array of UTF-8 chars
1381: * @inlen: the length of @in
1382: * @quoteChar: the quote character to escape (' or ") or zero.
1383: *
1384: * Take a block of UTF-8 chars in and try to convert it to an ASCII
1385: * plus HTML entities block of chars out.
1386: *
1387: * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1388: * The value of @inlen after return is the number of octets consumed
1389: * as the return value is positive, else unpredictiable.
1390: * The value of @outlen after return is the number of octets consumed.
1391: */
1392: int
1393: htmlEncodeEntities(unsigned char* out, int *outlen,
1394: const unsigned char* in, int *inlen, int quoteChar) {
1395: const unsigned char* processed = in;
1396: const unsigned char* outend = out + (*outlen);
1397: const unsigned char* outstart = out;
1398: const unsigned char* instart = in;
1399: const unsigned char* inend = in + (*inlen);
1400: unsigned int c, d;
1401: int trailing;
1402:
1403: while (in < inend) {
1404: d = *in++;
1405: if (d < 0x80) { c= d; trailing= 0; }
1406: else if (d < 0xC0) {
1407: /* trailing byte in leading position */
1408: *outlen = out - outstart;
1409: *inlen = processed - instart;
1410: return(-2);
1411: } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
1412: else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
1413: else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
1414: else {
1415: /* no chance for this in Ascii */
1416: *outlen = out - outstart;
1417: *inlen = processed - instart;
1418: return(-2);
1419: }
1420:
1421: if (inend - in < trailing)
1422: break;
1423:
1424: while (trailing--) {
1425: if (((d= *in++) & 0xC0) != 0x80) {
1426: *outlen = out - outstart;
1427: *inlen = processed - instart;
1428: return(-2);
1429: }
1430: c <<= 6;
1431: c |= d & 0x3F;
1432: }
1433:
1434: /* assertion: c is a single UTF-4 value */
1435: if (c < 0x80 && c != quoteChar && c != '&' && c != '<' && c != '>') {
1436: if (out >= outend)
1437: break;
1438: *out++ = c;
1439: } else {
1440: htmlEntityDescPtr ent;
1441: const char *cp;
1442: char nbuf[16];
1443: int len;
1444:
1445: /*
1446: * Try to lookup a predefined HTML entity for it
1447: */
1448: ent = htmlEntityValueLookup(c);
1449: if (ent == NULL) {
1450: sprintf(nbuf, "#%u", c);
1451: cp = nbuf;
1452: }
1453: else
1454: cp = ent->name;
1455: len = strlen(cp);
1456: if (out + 2 + len > outend)
1457: break;
1458: *out++ = '&';
1459: memcpy(out, cp, len);
1460: out += len;
1461: *out++ = ';';
1462: }
1463: processed = in;
1464: }
1465: *outlen = out - outstart;
1466: *inlen = processed - instart;
1467: return(0);
1468: }
1.1 daniel 1469:
1470: /**
1471: * htmlDecodeEntities:
1472: * @ctxt: the parser context
1473: * @len: the len to decode (in bytes !), -1 for no size limit
1.14 daniel 1474: * @end: an end marker xmlChar, 0 if none
1475: * @end2: an end marker xmlChar, 0 if none
1476: * @end3: an end marker xmlChar, 0 if none
1.1 daniel 1477: *
1478: * Subtitute the HTML entities by their value
1479: *
1.19 daniel 1480: * DEPRECATED !!!!
1.1 daniel 1481: *
1482: * Returns A newly allocated string with the substitution done. The caller
1483: * must deallocate it !
1484: */
1.14 daniel 1485: xmlChar *
1.1 daniel 1486: htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
1.14 daniel 1487: xmlChar end, xmlChar end2, xmlChar end3) {
1.53 veillard 1488: xmlChar *name = NULL;
1.14 daniel 1489: xmlChar *buffer = NULL;
1.53 veillard 1490: unsigned int buffer_size = 0;
1491: unsigned int nbchars = 0;
1.1 daniel 1492: htmlEntityDescPtr ent;
1493: unsigned int max = (unsigned int) len;
1.53 veillard 1494: int c,l;
1495:
1496: if (ctxt->depth > 40) {
1.67 veillard 1497: ctxt->errNo = XML_ERR_ENTITY_LOOP;
1.53 veillard 1498: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1499: ctxt->sax->error(ctxt->userData,
1500: "Detected entity reference loop\n");
1501: ctxt->wellFormed = 0;
1502: ctxt->disableSAX = 1;
1503: return(NULL);
1504: }
1.1 daniel 1505:
1506: /*
1507: * allocate a translation buffer.
1508: */
1.31 daniel 1509: buffer_size = HTML_PARSER_BIG_BUFFER_SIZE;
1.14 daniel 1510: buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
1.1 daniel 1511: if (buffer == NULL) {
1.53 veillard 1512: perror("xmlDecodeEntities: malloc failed");
1.1 daniel 1513: return(NULL);
1514: }
1515:
1516: /*
1517: * Ok loop until we reach one of the ending char or a size limit.
1518: */
1.53 veillard 1519: c = CUR_CHAR(l);
1520: while ((nbchars < max) && (c != end) &&
1521: (c != end2) && (c != end3)) {
1522:
1523: if (c == 0) break;
1524: if (((c == '&') && (ctxt->token != '&')) && (NXT(1) == '#')) {
1525: int val = htmlParseCharRef(ctxt);
1526: COPY_BUF(0,buffer,nbchars,val);
1527: NEXTL(l);
1528: } else if ((c == '&') && (ctxt->token != '&')) {
1529: ent = htmlParseEntityRef(ctxt, &name);
1530: if (name != NULL) {
1531: if (ent != NULL) {
1532: int val = ent->value;
1533: COPY_BUF(0,buffer,nbchars,val);
1534: NEXTL(l);
1535: } else {
1536: const xmlChar *cur = name;
1.1 daniel 1537:
1.53 veillard 1538: buffer[nbchars++] = '&';
1539: if (nbchars > buffer_size - HTML_PARSER_BUFFER_SIZE) {
1540: growBuffer(buffer);
1541: }
1542: while (*cur != 0) {
1543: buffer[nbchars++] = *cur++;
1.1 daniel 1544: }
1.53 veillard 1545: buffer[nbchars++] = ';';
1.1 daniel 1546: }
1547: }
1548: } else {
1.53 veillard 1549: COPY_BUF(l,buffer,nbchars,c);
1550: NEXTL(l);
1551: if (nbchars > buffer_size - HTML_PARSER_BUFFER_SIZE) {
1552: growBuffer(buffer);
1.1 daniel 1553: }
1554: }
1.53 veillard 1555: c = CUR_CHAR(l);
1.1 daniel 1556: }
1.53 veillard 1557: buffer[nbchars++] = 0;
1.1 daniel 1558: return(buffer);
1559: }
1560:
1.31 daniel 1561: /************************************************************************
1562: * *
1563: * Commodity functions to handle streams *
1564: * *
1565: ************************************************************************/
1566:
1567: /**
1568: * htmlFreeInputStream:
1569: * @input: an htmlParserInputPtr
1570: *
1571: * Free up an input stream.
1572: */
1573: void
1574: htmlFreeInputStream(htmlParserInputPtr input) {
1575: if (input == NULL) return;
1576:
1577: if (input->filename != NULL) xmlFree((char *) input->filename);
1578: if (input->directory != NULL) xmlFree((char *) input->directory);
1579: if ((input->free != NULL) && (input->base != NULL))
1580: input->free((xmlChar *) input->base);
1581: if (input->buf != NULL)
1582: xmlFreeParserInputBuffer(input->buf);
1583: memset(input, -1, sizeof(htmlParserInput));
1584: xmlFree(input);
1585: }
1586:
1587: /**
1588: * htmlNewInputStream:
1589: * @ctxt: an HTML parser context
1590: *
1591: * Create a new input stream structure
1592: * Returns the new input stream or NULL
1593: */
1594: htmlParserInputPtr
1595: htmlNewInputStream(htmlParserCtxtPtr ctxt) {
1596: htmlParserInputPtr input;
1597:
1598: input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1599: if (input == NULL) {
1600: ctxt->errNo = XML_ERR_NO_MEMORY;
1601: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1602: ctxt->sax->error(ctxt->userData,
1603: "malloc: couldn't allocate a new input stream\n");
1604: return(NULL);
1605: }
1.51 veillard 1606: memset(input, 0, sizeof(htmlParserInput));
1.31 daniel 1607: input->filename = NULL;
1608: input->directory = NULL;
1609: input->base = NULL;
1610: input->cur = NULL;
1611: input->buf = NULL;
1612: input->line = 1;
1613: input->col = 1;
1614: input->buf = NULL;
1615: input->free = NULL;
1.51 veillard 1616: input->version = NULL;
1.31 daniel 1617: input->consumed = 0;
1618: input->length = 0;
1619: return(input);
1620: }
1621:
1.1 daniel 1622:
1623: /************************************************************************
1624: * *
1625: * Commodity functions, cleanup needed ? *
1626: * *
1627: ************************************************************************/
1628:
1629: /**
1630: * areBlanks:
1631: * @ctxt: an HTML parser context
1.14 daniel 1632: * @str: a xmlChar *
1.1 daniel 1633: * @len: the size of @str
1634: *
1635: * Is this a sequence of blank chars that one can ignore ?
1636: *
1637: * Returns 1 if ignorable 0 otherwise.
1638: */
1639:
1.14 daniel 1640: static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
1.1 daniel 1641: int i;
1642: xmlNodePtr lastChild;
1643:
1644: for (i = 0;i < len;i++)
1645: if (!(IS_BLANK(str[i]))) return(0);
1646:
1.48 daniel 1647: if (CUR == 0) return(1);
1.1 daniel 1648: if (CUR != '<') return(0);
1.62 veillard 1649: if (ctxt->name == NULL)
1650: return(1);
1.63 veillard 1651: if (!xmlStrcmp(ctxt->name, BAD_CAST"html"))
1652: return(1);
1.62 veillard 1653: if (!xmlStrcmp(ctxt->name, BAD_CAST"head"))
1654: return(1);
1655: if (!xmlStrcmp(ctxt->name, BAD_CAST"body"))
1656: return(1);
1.1 daniel 1657: if (ctxt->node == NULL) return(0);
1658: lastChild = xmlGetLastChild(ctxt->node);
1659: if (lastChild == NULL) {
1660: if (ctxt->node->content != NULL) return(0);
1661: } else if (xmlNodeIsText(lastChild))
1662: return(0);
1663: return(1);
1664: }
1665:
1666: /**
1667: * htmlHandleEntity:
1668: * @ctxt: an HTML parser context
1669: * @entity: an XML entity pointer.
1670: *
1671: * Default handling of an HTML entity, call the parser with the
1672: * substitution string
1673: */
1674:
1675: void
1676: htmlHandleEntity(htmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1677: int len;
1678:
1679: if (entity->content == NULL) {
1680: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1681: ctxt->sax->error(ctxt->userData, "htmlHandleEntity %s: content == NULL\n",
1682: entity->name);
1683: ctxt->wellFormed = 0;
1684: return;
1685: }
1686: len = xmlStrlen(entity->content);
1687:
1688: /*
1689: * Just handle the content as a set of chars.
1690: */
1.59 veillard 1691: htmlCheckParagraph(ctxt);
1.1 daniel 1692: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1693: ctxt->sax->characters(ctxt->userData, entity->content, len);
1694:
1695: }
1696:
1697: /**
1.59 veillard 1698: * htmlNewDocNoDtD:
1.1 daniel 1699: * @URI: URI for the dtd, or NULL
1700: * @ExternalID: the external ID of the DTD, or NULL
1701: *
1.59 veillard 1702: * Returns a new document, do not intialize the DTD if not provided
1.1 daniel 1703: */
1704: htmlDocPtr
1.59 veillard 1705: htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) {
1.1 daniel 1706: xmlDocPtr cur;
1707:
1708: /*
1709: * Allocate a new document and fill the fields.
1710: */
1.11 daniel 1711: cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
1.1 daniel 1712: if (cur == NULL) {
1713: fprintf(stderr, "xmlNewDoc : malloc failed\n");
1714: return(NULL);
1715: }
1.10 daniel 1716: memset(cur, 0, sizeof(xmlDoc));
1.1 daniel 1717:
1.20 daniel 1718: cur->type = XML_HTML_DOCUMENT_NODE;
1.1 daniel 1719: cur->version = NULL;
1720: cur->intSubset = NULL;
1.59 veillard 1721: if ((ExternalID != NULL) ||
1722: (URI != NULL))
1.28 daniel 1723: xmlCreateIntSubset(cur, BAD_CAST "HTML", ExternalID, URI);
1.41 daniel 1724: cur->doc = cur;
1.1 daniel 1725: cur->name = NULL;
1.37 daniel 1726: cur->children = NULL;
1.1 daniel 1727: cur->extSubset = NULL;
1728: cur->oldNs = NULL;
1729: cur->encoding = NULL;
1730: cur->standalone = 1;
1731: cur->compression = 0;
1.12 daniel 1732: cur->ids = NULL;
1733: cur->refs = NULL;
1.1 daniel 1734: #ifndef XML_WITHOUT_CORBA
1735: cur->_private = NULL;
1736: #endif
1737: return(cur);
1738: }
1739:
1.59 veillard 1740: /**
1741: * htmlNewDoc:
1742: * @URI: URI for the dtd, or NULL
1743: * @ExternalID: the external ID of the DTD, or NULL
1744: *
1745: * Returns a new document
1746: */
1747: htmlDocPtr
1748: htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
1749: if ((URI == NULL) && (ExternalID == NULL))
1750: return(htmlNewDocNoDtD(
1751: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
1752: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd"));
1753:
1754: return(htmlNewDocNoDtD(URI, ExternalID));
1755: }
1756:
1.1 daniel 1757:
1758: /************************************************************************
1759: * *
1760: * The parser itself *
1761: * Relates to http://www.w3.org/TR/html40 *
1762: * *
1763: ************************************************************************/
1764:
1765: /************************************************************************
1766: * *
1767: * The parser itself *
1768: * *
1769: ************************************************************************/
1770:
1771: /**
1772: * htmlParseHTMLName:
1773: * @ctxt: an HTML parser context
1774: *
1.26 daniel 1775: * parse an HTML tag or attribute name, note that we convert it to lowercase
1.1 daniel 1776: * since HTML names are not case-sensitive.
1777: *
1778: * Returns the Tag Name parsed or NULL
1779: */
1780:
1.14 daniel 1781: xmlChar *
1.1 daniel 1782: htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1783: xmlChar *ret = NULL;
1.1 daniel 1784: int i = 0;
1.31 daniel 1785: xmlChar loc[HTML_PARSER_BUFFER_SIZE];
1.1 daniel 1786:
1787: if (!IS_LETTER(CUR) && (CUR != '_') &&
1788: (CUR != ':')) return(NULL);
1789:
1.31 daniel 1790: while ((i < HTML_PARSER_BUFFER_SIZE) &&
1.45 daniel 1791: ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1792: (CUR == ':') || (CUR == '_'))) {
1.26 daniel 1793: if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20;
1.1 daniel 1794: else loc[i] = CUR;
1795: i++;
1796:
1797: NEXT;
1798: }
1799:
1800: ret = xmlStrndup(loc, i);
1801:
1802: return(ret);
1803: }
1804:
1805: /**
1806: * htmlParseName:
1807: * @ctxt: an HTML parser context
1808: *
1809: * parse an HTML name, this routine is case sensistive.
1810: *
1811: * Returns the Name parsed or NULL
1812: */
1813:
1.14 daniel 1814: xmlChar *
1.1 daniel 1815: htmlParseName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1816: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1817: int len = 0;
1.1 daniel 1818:
1.5 daniel 1819: GROW;
1820: if (!IS_LETTER(CUR) && (CUR != '_')) {
1821: return(NULL);
1822: }
1.1 daniel 1823:
1824: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1825: (CUR == '.') || (CUR == '-') ||
1826: (CUR == '_') || (CUR == ':') ||
1827: (IS_COMBINING(CUR)) ||
1.5 daniel 1828: (IS_EXTENDER(CUR))) {
1829: buf[len++] = CUR;
1.1 daniel 1830: NEXT;
1.5 daniel 1831: if (len >= HTML_MAX_NAMELEN) {
1832: fprintf(stderr,
1833: "htmlParseName: reached HTML_MAX_NAMELEN limit\n");
1834: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1835: (CUR == '.') || (CUR == '-') ||
1836: (CUR == '_') || (CUR == ':') ||
1837: (IS_COMBINING(CUR)) ||
1838: (IS_EXTENDER(CUR)))
1839: NEXT;
1840: break;
1841: }
1842: }
1843: return(xmlStrndup(buf, len));
1.1 daniel 1844: }
1845:
1846: /**
1847: * htmlParseHTMLAttribute:
1848: * @ctxt: an HTML parser context
1.19 daniel 1849: * @stop: a char stop value
1.1 daniel 1850: *
1.19 daniel 1851: * parse an HTML attribute value till the stop (quote), if
1852: * stop is 0 then it stops at the first space
1.1 daniel 1853: *
1.19 daniel 1854: * Returns the attribute parsed or NULL
1.1 daniel 1855: */
1856:
1.14 daniel 1857: xmlChar *
1.19 daniel 1858: htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) {
1.32 daniel 1859: #if 0
1.14 daniel 1860: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1861: int len = 0;
1.1 daniel 1862:
1.5 daniel 1863: GROW;
1.19 daniel 1864: while ((CUR != 0) && (CUR != stop) && (CUR != '>')) {
1865: if ((stop == 0) && (IS_BLANK(CUR))) break;
1.5 daniel 1866: buf[len++] = CUR;
1.1 daniel 1867: NEXT;
1.5 daniel 1868: if (len >= HTML_MAX_NAMELEN) {
1869: fprintf(stderr,
1870: "htmlParseHTMLAttribute: reached HTML_MAX_NAMELEN limit\n");
1871: while ((!IS_BLANK(CUR)) && (CUR != '<') &&
1.19 daniel 1872: (CUR != '>') &&
1.5 daniel 1873: (CUR != '\'') && (CUR != '"'))
1874: NEXT;
1875: break;
1876: }
1877: }
1878: return(xmlStrndup(buf, len));
1.32 daniel 1879: #else
1880: xmlChar *buffer = NULL;
1881: int buffer_size = 0;
1882: xmlChar *out = NULL;
1883: xmlChar *name = NULL;
1884:
1885: xmlChar *cur = NULL;
1886: htmlEntityDescPtr ent;
1887:
1888: /*
1889: * allocate a translation buffer.
1890: */
1891: buffer_size = HTML_PARSER_BIG_BUFFER_SIZE;
1892: buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
1893: if (buffer == NULL) {
1894: perror("htmlParseHTMLAttribute: malloc failed");
1895: return(NULL);
1896: }
1897: out = buffer;
1898:
1899: /*
1900: * Ok loop until we reach one of the ending chars
1901: */
1902: while ((CUR != 0) && (CUR != stop) && (CUR != '>')) {
1903: if ((stop == 0) && (IS_BLANK(CUR))) break;
1904: if (CUR == '&') {
1905: if (NXT(1) == '#') {
1.52 veillard 1906: unsigned int c;
1907: int bits;
1908:
1909: c = htmlParseCharRef(ctxt);
1910: if (c < 0x80)
1911: { *out++ = c; bits= -6; }
1912: else if (c < 0x800)
1913: { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
1914: else if (c < 0x10000)
1915: { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
1916: else
1917: { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
1918:
1919: for ( ; bits >= 0; bits-= 6) {
1920: *out++ = ((c >> bits) & 0x3F) | 0x80;
1921: }
1.32 daniel 1922: } else {
1923: ent = htmlParseEntityRef(ctxt, &name);
1924: if (name == NULL) {
1925: *out++ = '&';
1926: if (out - buffer > buffer_size - 100) {
1927: int index = out - buffer;
1928:
1929: growBuffer(buffer);
1930: out = &buffer[index];
1931: }
1.52 veillard 1932: } else if (ent == NULL) {
1.32 daniel 1933: *out++ = '&';
1934: cur = name;
1935: while (*cur != 0) {
1936: if (out - buffer > buffer_size - 100) {
1937: int index = out - buffer;
1938:
1939: growBuffer(buffer);
1940: out = &buffer[index];
1941: }
1942: *out++ = *cur++;
1943: }
1944: xmlFree(name);
1945: } else {
1.52 veillard 1946: unsigned int c;
1947: int bits;
1948:
1.32 daniel 1949: if (out - buffer > buffer_size - 100) {
1950: int index = out - buffer;
1951:
1952: growBuffer(buffer);
1953: out = &buffer[index];
1954: }
1.52 veillard 1955: c = (xmlChar)ent->value;
1956: if (c < 0x80)
1957: { *out++ = c; bits= -6; }
1958: else if (c < 0x800)
1959: { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
1960: else if (c < 0x10000)
1961: { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
1962: else
1963: { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
1964:
1965: for ( ; bits >= 0; bits-= 6) {
1966: *out++ = ((c >> bits) & 0x3F) | 0x80;
1967: }
1.32 daniel 1968: xmlFree(name);
1969: }
1970: }
1971: } else {
1.52 veillard 1972: unsigned int c;
1.68 ! veillard 1973: int bits, l;
1.52 veillard 1974:
1.32 daniel 1975: if (out - buffer > buffer_size - 100) {
1.52 veillard 1976: int index = out - buffer;
1977:
1978: growBuffer(buffer);
1979: out = &buffer[index];
1980: }
1.68 ! veillard 1981: c = CUR_CHAR(l);
1.52 veillard 1982: if (c < 0x80)
1983: { *out++ = c; bits= -6; }
1984: else if (c < 0x800)
1985: { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
1986: else if (c < 0x10000)
1987: { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
1988: else
1989: { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
1990:
1991: for ( ; bits >= 0; bits-= 6) {
1992: *out++ = ((c >> bits) & 0x3F) | 0x80;
1.32 daniel 1993: }
1994: NEXT;
1995: }
1996: }
1997: *out++ = 0;
1998: return(buffer);
1999: #endif
1.1 daniel 2000: }
2001:
2002: /**
2003: * htmlParseNmtoken:
2004: * @ctxt: an HTML parser context
2005: *
2006: * parse an HTML Nmtoken.
2007: *
2008: * Returns the Nmtoken parsed or NULL
2009: */
2010:
1.14 daniel 2011: xmlChar *
1.1 daniel 2012: htmlParseNmtoken(htmlParserCtxtPtr ctxt) {
1.14 daniel 2013: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 2014: int len = 0;
1.1 daniel 2015:
1.5 daniel 2016: GROW;
1.1 daniel 2017: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
2018: (CUR == '.') || (CUR == '-') ||
2019: (CUR == '_') || (CUR == ':') ||
2020: (IS_COMBINING(CUR)) ||
1.5 daniel 2021: (IS_EXTENDER(CUR))) {
2022: buf[len++] = CUR;
1.1 daniel 2023: NEXT;
1.5 daniel 2024: if (len >= HTML_MAX_NAMELEN) {
2025: fprintf(stderr,
2026: "htmlParseNmtoken: reached HTML_MAX_NAMELEN limit\n");
2027: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
2028: (CUR == '.') || (CUR == '-') ||
2029: (CUR == '_') || (CUR == ':') ||
2030: (IS_COMBINING(CUR)) ||
2031: (IS_EXTENDER(CUR)))
2032: NEXT;
2033: break;
2034: }
2035: }
2036: return(xmlStrndup(buf, len));
1.1 daniel 2037: }
2038:
2039: /**
2040: * htmlParseEntityRef:
2041: * @ctxt: an HTML parser context
2042: * @str: location to store the entity name
2043: *
2044: * parse an HTML ENTITY references
2045: *
2046: * [68] EntityRef ::= '&' Name ';'
2047: *
2048: * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
2049: * if non-NULL *str will have to be freed by the caller.
2050: */
2051: htmlEntityDescPtr
1.14 daniel 2052: htmlParseEntityRef(htmlParserCtxtPtr ctxt, xmlChar **str) {
2053: xmlChar *name;
1.1 daniel 2054: htmlEntityDescPtr ent = NULL;
2055: *str = NULL;
2056:
2057: if (CUR == '&') {
2058: NEXT;
2059: name = htmlParseName(ctxt);
2060: if (name == NULL) {
2061: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2062: ctxt->sax->error(ctxt->userData, "htmlParseEntityRef: no name\n");
2063: ctxt->wellFormed = 0;
2064: } else {
1.5 daniel 2065: GROW;
1.1 daniel 2066: if (CUR == ';') {
2067: *str = name;
2068:
2069: /*
2070: * Lookup the entity in the table.
2071: */
2072: ent = htmlEntityLookup(name);
1.32 daniel 2073: if (ent != NULL) /* OK that's ugly !!! */
2074: NEXT;
1.1 daniel 2075: } else {
2076: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2077: ctxt->sax->error(ctxt->userData,
2078: "htmlParseEntityRef: expecting ';'\n");
1.32 daniel 2079: *str = name;
1.1 daniel 2080: }
2081: }
2082: }
2083: return(ent);
2084: }
2085:
2086: /**
2087: * htmlParseAttValue:
2088: * @ctxt: an HTML parser context
2089: *
2090: * parse a value for an attribute
2091: * Note: the parser won't do substitution of entities here, this
2092: * will be handled later in xmlStringGetNodeList, unless it was
2093: * asked for ctxt->replaceEntities != 0
2094: *
2095: * Returns the AttValue parsed or NULL.
2096: */
2097:
1.14 daniel 2098: xmlChar *
1.1 daniel 2099: htmlParseAttValue(htmlParserCtxtPtr ctxt) {
1.14 daniel 2100: xmlChar *ret = NULL;
1.1 daniel 2101:
2102: if (CUR == '"') {
2103: NEXT;
1.19 daniel 2104: ret = htmlParseHTMLAttribute(ctxt, '"');
1.1 daniel 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 if (CUR == '\'') {
2112: NEXT;
1.19 daniel 2113: ret = htmlParseHTMLAttribute(ctxt, '\'');
1.1 daniel 2114: if (CUR != '\'') {
2115: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2116: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
2117: ctxt->wellFormed = 0;
2118: } else
2119: NEXT;
2120: } else {
2121: /*
2122: * That's an HTMLism, the attribute value may not be quoted
2123: */
1.19 daniel 2124: ret = htmlParseHTMLAttribute(ctxt, 0);
1.1 daniel 2125: if (ret == NULL) {
2126: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2127: ctxt->sax->error(ctxt->userData, "AttValue: no value found\n");
2128: ctxt->wellFormed = 0;
2129: }
2130: }
2131: return(ret);
2132: }
2133:
2134: /**
2135: * htmlParseSystemLiteral:
2136: * @ctxt: an HTML parser context
2137: *
2138: * parse an HTML Literal
2139: *
2140: * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
2141: *
2142: * Returns the SystemLiteral parsed or NULL
2143: */
2144:
1.14 daniel 2145: xmlChar *
1.1 daniel 2146: htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 2147: const xmlChar *q;
2148: xmlChar *ret = NULL;
1.1 daniel 2149:
2150: if (CUR == '"') {
2151: NEXT;
2152: q = CUR_PTR;
2153: while ((IS_CHAR(CUR)) && (CUR != '"'))
2154: NEXT;
2155: if (!IS_CHAR(CUR)) {
2156: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2157: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
2158: ctxt->wellFormed = 0;
2159: } else {
2160: ret = xmlStrndup(q, CUR_PTR - q);
2161: NEXT;
2162: }
2163: } else if (CUR == '\'') {
2164: NEXT;
2165: q = CUR_PTR;
2166: while ((IS_CHAR(CUR)) && (CUR != '\''))
2167: NEXT;
2168: if (!IS_CHAR(CUR)) {
2169: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2170: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
2171: ctxt->wellFormed = 0;
2172: } else {
2173: ret = xmlStrndup(q, CUR_PTR - q);
2174: NEXT;
2175: }
2176: } else {
2177: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.38 daniel 2178: ctxt->sax->error(ctxt->userData,
2179: "SystemLiteral \" or ' expected\n");
1.1 daniel 2180: ctxt->wellFormed = 0;
2181: }
2182:
2183: return(ret);
2184: }
2185:
2186: /**
2187: * htmlParsePubidLiteral:
2188: * @ctxt: an HTML parser context
2189: *
2190: * parse an HTML public literal
2191: *
2192: * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
2193: *
2194: * Returns the PubidLiteral parsed or NULL.
2195: */
2196:
1.14 daniel 2197: xmlChar *
1.1 daniel 2198: htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 2199: const xmlChar *q;
2200: xmlChar *ret = NULL;
1.1 daniel 2201: /*
2202: * Name ::= (Letter | '_') (NameChar)*
2203: */
2204: if (CUR == '"') {
2205: NEXT;
2206: q = CUR_PTR;
2207: while (IS_PUBIDCHAR(CUR)) NEXT;
2208: if (CUR != '"') {
2209: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2210: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
2211: ctxt->wellFormed = 0;
2212: } else {
2213: ret = xmlStrndup(q, CUR_PTR - q);
2214: NEXT;
2215: }
2216: } else if (CUR == '\'') {
2217: NEXT;
2218: q = CUR_PTR;
2219: while ((IS_LETTER(CUR)) && (CUR != '\''))
2220: NEXT;
2221: if (!IS_LETTER(CUR)) {
2222: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2223: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
2224: ctxt->wellFormed = 0;
2225: } else {
2226: ret = xmlStrndup(q, CUR_PTR - q);
2227: NEXT;
2228: }
2229: } else {
2230: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2231: ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n");
2232: ctxt->wellFormed = 0;
2233: }
2234:
2235: return(ret);
2236: }
2237:
2238: /**
2239: * htmlParseCharData:
2240: * @ctxt: an HTML parser context
2241: * @cdata: int indicating whether we are within a CDATA section
2242: *
2243: * parse a CharData section.
2244: * if we are within a CDATA section ']]>' marks an end of section.
2245: *
2246: * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
2247: */
2248:
2249: void
2250: htmlParseCharData(htmlParserCtxtPtr ctxt, int cdata) {
1.53 veillard 2251: xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
2252: int nbchar = 0;
2253: int cur, l;
2254:
2255: SHRINK;
2256: cur = CUR_CHAR(l);
2257: while (((cur != '<') || (ctxt->token == '<')) &&
2258: ((cur != '&') || (ctxt->token == '&')) &&
2259: (IS_CHAR(cur))) {
2260: COPY_BUF(l,buf,nbchar,cur);
2261: if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2262: /*
2263: * Ok the segment is to be consumed as chars.
2264: */
2265: if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2266: if (areBlanks(ctxt, buf, nbchar)) {
2267: if (ctxt->sax->ignorableWhitespace != NULL)
2268: ctxt->sax->ignorableWhitespace(ctxt->userData,
2269: buf, nbchar);
2270: } else {
1.59 veillard 2271: htmlCheckParagraph(ctxt);
1.53 veillard 2272: if (ctxt->sax->characters != NULL)
2273: ctxt->sax->characters(ctxt->userData, buf, nbchar);
2274: }
1.1 daniel 2275: }
1.53 veillard 2276: nbchar = 0;
1.1 daniel 2277: }
1.53 veillard 2278: NEXTL(l);
2279: cur = CUR_CHAR(l);
2280: }
2281: if (nbchar != 0) {
2282: /*
2283: * Ok the segment is to be consumed as chars.
2284: */
2285: if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2286: if (areBlanks(ctxt, buf, nbchar)) {
2287: if (ctxt->sax->ignorableWhitespace != NULL)
2288: ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar);
2289: } else {
1.59 veillard 2290: htmlCheckParagraph(ctxt);
1.53 veillard 2291: if (ctxt->sax->characters != NULL)
2292: ctxt->sax->characters(ctxt->userData, buf, nbchar);
1.25 daniel 2293: }
2294: }
1.1 daniel 2295: }
2296: }
2297:
2298: /**
2299: * htmlParseExternalID:
2300: * @ctxt: an HTML parser context
1.14 daniel 2301: * @publicID: a xmlChar** receiving PubidLiteral
1.1 daniel 2302: * @strict: indicate whether we should restrict parsing to only
2303: * production [75], see NOTE below
2304: *
2305: * Parse an External ID or a Public ID
2306: *
2307: * NOTE: Productions [75] and [83] interract badly since [75] can generate
2308: * 'PUBLIC' S PubidLiteral S SystemLiteral
2309: *
2310: * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
2311: * | 'PUBLIC' S PubidLiteral S SystemLiteral
2312: *
2313: * [83] PublicID ::= 'PUBLIC' S PubidLiteral
2314: *
2315: * Returns the function returns SystemLiteral and in the second
2316: * case publicID receives PubidLiteral, is strict is off
2317: * it is possible to return NULL and have publicID set.
2318: */
2319:
1.14 daniel 2320: xmlChar *
2321: htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID, int strict) {
2322: xmlChar *URI = NULL;
1.1 daniel 2323:
2324: if ((UPPER == 'S') && (UPP(1) == 'Y') &&
2325: (UPP(2) == 'S') && (UPP(3) == 'T') &&
2326: (UPP(4) == 'E') && (UPP(5) == 'M')) {
2327: SKIP(6);
2328: if (!IS_BLANK(CUR)) {
2329: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2330: ctxt->sax->error(ctxt->userData,
2331: "Space required after 'SYSTEM'\n");
2332: ctxt->wellFormed = 0;
2333: }
2334: SKIP_BLANKS;
2335: URI = htmlParseSystemLiteral(ctxt);
2336: if (URI == NULL) {
2337: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2338: ctxt->sax->error(ctxt->userData,
2339: "htmlParseExternalID: SYSTEM, no URI\n");
2340: ctxt->wellFormed = 0;
2341: }
2342: } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
2343: (UPP(2) == 'B') && (UPP(3) == 'L') &&
2344: (UPP(4) == 'I') && (UPP(5) == 'C')) {
2345: SKIP(6);
2346: if (!IS_BLANK(CUR)) {
2347: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2348: ctxt->sax->error(ctxt->userData,
2349: "Space required after 'PUBLIC'\n");
2350: ctxt->wellFormed = 0;
2351: }
2352: SKIP_BLANKS;
2353: *publicID = htmlParsePubidLiteral(ctxt);
2354: if (*publicID == NULL) {
2355: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2356: ctxt->sax->error(ctxt->userData,
2357: "htmlParseExternalID: PUBLIC, no Public Identifier\n");
2358: ctxt->wellFormed = 0;
2359: }
1.5 daniel 2360: SKIP_BLANKS;
2361: if ((CUR == '"') || (CUR == '\'')) {
2362: URI = htmlParseSystemLiteral(ctxt);
1.1 daniel 2363: }
2364: }
2365: return(URI);
2366: }
2367:
2368: /**
2369: * htmlParseComment:
2370: * @ctxt: an HTML parser context
2371: *
2372: * Parse an XML (SGML) comment <!-- .... -->
2373: *
2374: * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
2375: */
2376: void
1.31 daniel 2377: htmlParseComment(htmlParserCtxtPtr ctxt) {
1.25 daniel 2378: xmlChar *buf = NULL;
1.56 veillard 2379: int len;
1.31 daniel 2380: int size = HTML_PARSER_BUFFER_SIZE;
1.56 veillard 2381: int q, ql;
2382: int r, rl;
2383: int cur, l;
2384: xmlParserInputState state;
1.1 daniel 2385:
2386: /*
2387: * Check that there is a comment right here.
2388: */
1.56 veillard 2389: if ((RAW != '<') || (NXT(1) != '!') ||
1.1 daniel 2390: (NXT(2) != '-') || (NXT(3) != '-')) return;
2391:
1.56 veillard 2392: state = ctxt->instate;
2393: ctxt->instate = XML_PARSER_COMMENT;
2394: SHRINK;
2395: SKIP(4);
1.25 daniel 2396: buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar));
2397: if (buf == NULL) {
2398: fprintf(stderr, "malloc of %d byte failed\n", size);
1.56 veillard 2399: ctxt->instate = state;
1.25 daniel 2400: return;
2401: }
1.56 veillard 2402: q = CUR_CHAR(ql);
2403: NEXTL(ql);
2404: r = CUR_CHAR(rl);
2405: NEXTL(rl);
2406: cur = CUR_CHAR(l);
2407: len = 0;
2408: while (IS_CHAR(cur) &&
2409: ((cur != '>') ||
2410: (r != '-') || (q != '-'))) {
2411: if (len + 5 >= size) {
1.25 daniel 2412: size *= 2;
1.50 veillard 2413: buf = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
1.25 daniel 2414: if (buf == NULL) {
2415: fprintf(stderr, "realloc of %d byte failed\n", size);
1.56 veillard 2416: ctxt->instate = state;
1.25 daniel 2417: return;
2418: }
2419: }
1.56 veillard 2420: COPY_BUF(ql,buf,len,q);
1.25 daniel 2421: q = r;
1.56 veillard 2422: ql = rl;
2423: r = cur;
2424: rl = l;
2425: NEXTL(l);
2426: cur = CUR_CHAR(l);
2427: if (cur == 0) {
2428: SHRINK;
2429: GROW;
2430: cur = CUR_CHAR(l);
2431: }
1.1 daniel 2432: }
1.56 veillard 2433: buf[len] = 0;
2434: if (!IS_CHAR(cur)) {
1.67 veillard 2435: ctxt->errNo = XML_ERR_COMMENT_NOT_FINISHED;
1.1 daniel 2436: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.56 veillard 2437: ctxt->sax->error(ctxt->userData,
2438: "Comment not terminated \n<!--%.50s\n", buf);
1.1 daniel 2439: ctxt->wellFormed = 0;
1.56 veillard 2440: xmlFree(buf);
1.1 daniel 2441: } else {
2442: NEXT;
1.56 veillard 2443: if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) &&
2444: (!ctxt->disableSAX))
1.31 daniel 2445: ctxt->sax->comment(ctxt->userData, buf);
1.56 veillard 2446: xmlFree(buf);
1.1 daniel 2447: }
1.56 veillard 2448: ctxt->instate = state;
1.1 daniel 2449: }
2450:
2451: /**
2452: * htmlParseCharRef:
2453: * @ctxt: an HTML parser context
2454: *
2455: * parse Reference declarations
2456: *
2457: * [66] CharRef ::= '&#' [0-9]+ ';' |
2458: * '&#x' [0-9a-fA-F]+ ';'
2459: *
2460: * Returns the value parsed (as an int)
2461: */
2462: int
2463: htmlParseCharRef(htmlParserCtxtPtr ctxt) {
2464: int val = 0;
2465:
2466: if ((CUR == '&') && (NXT(1) == '#') &&
2467: (NXT(2) == 'x')) {
2468: SKIP(3);
2469: while (CUR != ';') {
2470: if ((CUR >= '0') && (CUR <= '9'))
2471: val = val * 16 + (CUR - '0');
2472: else if ((CUR >= 'a') && (CUR <= 'f'))
2473: val = val * 16 + (CUR - 'a') + 10;
2474: else if ((CUR >= 'A') && (CUR <= 'F'))
2475: val = val * 16 + (CUR - 'A') + 10;
2476: else {
2477: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2478: ctxt->sax->error(ctxt->userData,
2479: "htmlParseCharRef: invalid hexadecimal value\n");
2480: ctxt->wellFormed = 0;
2481: val = 0;
2482: break;
2483: }
2484: NEXT;
2485: }
2486: if (CUR == ';')
2487: NEXT;
2488: } else if ((CUR == '&') && (NXT(1) == '#')) {
2489: SKIP(2);
2490: while (CUR != ';') {
2491: if ((CUR >= '0') && (CUR <= '9'))
2492: val = val * 10 + (CUR - '0');
2493: else {
2494: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2495: ctxt->sax->error(ctxt->userData,
2496: "htmlParseCharRef: invalid decimal value\n");
2497: ctxt->wellFormed = 0;
2498: val = 0;
2499: break;
2500: }
2501: NEXT;
2502: }
2503: if (CUR == ';')
2504: NEXT;
2505: } else {
2506: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2507: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid value\n");
2508: ctxt->wellFormed = 0;
2509: }
2510: /*
2511: * Check the value IS_CHAR ...
2512: */
2513: if (IS_CHAR(val)) {
2514: return(val);
2515: } else {
2516: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.14 daniel 2517: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid xmlChar value %d\n",
1.1 daniel 2518: val);
2519: ctxt->wellFormed = 0;
2520: }
2521: return(0);
2522: }
2523:
2524:
2525: /**
2526: * htmlParseDocTypeDecl :
2527: * @ctxt: an HTML parser context
2528: *
2529: * parse a DOCTYPE declaration
2530: *
2531: * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
2532: * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
2533: */
2534:
2535: void
2536: htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
1.14 daniel 2537: xmlChar *name;
2538: xmlChar *ExternalID = NULL;
2539: xmlChar *URI = NULL;
1.1 daniel 2540:
2541: /*
2542: * We know that '<!DOCTYPE' has been detected.
2543: */
2544: SKIP(9);
2545:
2546: SKIP_BLANKS;
2547:
2548: /*
2549: * Parse the DOCTYPE name.
2550: */
2551: name = htmlParseName(ctxt);
2552: if (name == NULL) {
2553: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2554: ctxt->sax->error(ctxt->userData, "htmlParseDocTypeDecl : no DOCTYPE name !\n");
2555: ctxt->wellFormed = 0;
2556: }
2557: /*
2558: * Check that upper(name) == "HTML" !!!!!!!!!!!!!
2559: */
2560:
2561: SKIP_BLANKS;
2562:
2563: /*
2564: * Check for SystemID and ExternalID
2565: */
1.5 daniel 2566: URI = htmlParseExternalID(ctxt, &ExternalID, 0);
1.1 daniel 2567: SKIP_BLANKS;
2568:
2569: /*
2570: * We should be at the end of the DOCTYPE declaration.
2571: */
2572: if (CUR != '>') {
2573: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2574: ctxt->sax->error(ctxt->userData, "DOCTYPE unproperly terminated\n");
2575: ctxt->wellFormed = 0;
2576: /* We shouldn't try to resynchronize ... */
2577: }
2578: NEXT;
2579:
2580: /*
1.46 daniel 2581: * Create or update the document accordingly to the DOCTYPE
1.1 daniel 2582: */
1.46 daniel 2583: if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) &&
2584: (!ctxt->disableSAX))
2585: ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI);
1.1 daniel 2586:
2587: /*
2588: * Cleanup, since we don't use all those identifiers
2589: */
1.11 daniel 2590: if (URI != NULL) xmlFree(URI);
2591: if (ExternalID != NULL) xmlFree(ExternalID);
2592: if (name != NULL) xmlFree(name);
1.1 daniel 2593: }
2594:
2595: /**
2596: * htmlParseAttribute:
2597: * @ctxt: an HTML parser context
1.14 daniel 2598: * @value: a xmlChar ** used to store the value of the attribute
1.1 daniel 2599: *
2600: * parse an attribute
2601: *
2602: * [41] Attribute ::= Name Eq AttValue
2603: *
2604: * [25] Eq ::= S? '=' S?
2605: *
2606: * With namespace:
2607: *
2608: * [NS 11] Attribute ::= QName Eq AttValue
2609: *
2610: * Also the case QName == xmlns:??? is handled independently as a namespace
2611: * definition.
2612: *
2613: * Returns the attribute name, and the value in *value.
2614: */
2615:
1.14 daniel 2616: xmlChar *
2617: htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
1.31 daniel 2618: xmlChar *name, *val = NULL;
1.1 daniel 2619:
2620: *value = NULL;
2621: name = htmlParseName(ctxt);
2622: if (name == NULL) {
2623: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2624: ctxt->sax->error(ctxt->userData, "error parsing attribute name\n");
2625: ctxt->wellFormed = 0;
2626: return(NULL);
2627: }
2628:
2629: /*
2630: * read the value
2631: */
2632: SKIP_BLANKS;
2633: if (CUR == '=') {
2634: NEXT;
2635: SKIP_BLANKS;
2636: val = htmlParseAttValue(ctxt);
1.42 daniel 2637: /******
1.1 daniel 2638: } else {
1.42 daniel 2639: * TODO : some attribute must have values, some may not
1.1 daniel 2640: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.31 daniel 2641: ctxt->sax->warning(ctxt->userData,
1.42 daniel 2642: "No value for attribute %s\n", name); */
1.1 daniel 2643: }
2644:
2645: *value = val;
2646: return(name);
2647: }
2648:
2649: /**
1.47 daniel 2650: * htmlCheckEncoding:
2651: * @ctxt: an HTML parser context
2652: * @attvalue: the attribute value
2653: *
2654: * Checks an http-equiv attribute from a Meta tag to detect
2655: * the encoding
2656: * If a new encoding is detected the parser is switched to decode
2657: * it and pass UTF8
2658: */
2659: void
2660: htmlCheckEncoding(htmlParserCtxtPtr ctxt, const xmlChar *attvalue) {
2661: const xmlChar *encoding;
2662:
2663: if ((ctxt == NULL) || (attvalue == NULL))
2664: return;
2665:
2666: encoding = xmlStrstr(attvalue, BAD_CAST"charset=");
2667: if (encoding == NULL)
2668: encoding = xmlStrstr(attvalue, BAD_CAST"Charset=");
2669: if (encoding == NULL)
2670: encoding = xmlStrstr(attvalue, BAD_CAST"CHARSET=");
2671: if (encoding != NULL) {
2672: encoding += 8;
2673: } else {
2674: encoding = xmlStrstr(attvalue, BAD_CAST"charset =");
2675: if (encoding == NULL)
2676: encoding = xmlStrstr(attvalue, BAD_CAST"Charset =");
2677: if (encoding == NULL)
2678: encoding = xmlStrstr(attvalue, BAD_CAST"CHARSET =");
2679: if (encoding != NULL)
2680: encoding += 9;
2681: }
2682: if (encoding != NULL) {
2683: xmlCharEncoding enc;
2684: xmlCharEncodingHandlerPtr handler;
2685:
2686: while ((*encoding == ' ') || (*encoding == '\t')) encoding++;
2687:
2688: if (ctxt->input->encoding != NULL)
2689: xmlFree((xmlChar *) ctxt->input->encoding);
2690: ctxt->input->encoding = xmlStrdup(encoding);
2691:
2692: enc = xmlParseCharEncoding((const char *) encoding);
2693: /*
2694: * registered set of known encodings
2695: */
2696: if (enc != XML_CHAR_ENCODING_ERROR) {
2697: xmlSwitchEncoding(ctxt, enc);
1.53 veillard 2698: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1.47 daniel 2699: } else {
2700: /*
2701: * fallback for unknown encodings
2702: */
2703: handler = xmlFindCharEncodingHandler((const char *) encoding);
2704: if (handler != NULL) {
2705: xmlSwitchToEncoding(ctxt, handler);
1.54 veillard 2706: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1.47 daniel 2707: } else {
2708: ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
2709: }
2710: }
1.54 veillard 2711:
2712: if ((ctxt->input->buf != NULL) &&
2713: (ctxt->input->buf->encoder != NULL) &&
2714: (ctxt->input->buf->raw != NULL) &&
2715: (ctxt->input->buf->buffer != NULL)) {
2716: int nbchars;
1.56 veillard 2717: int processed;
1.54 veillard 2718:
2719: /*
2720: * convert as much as possible to the parser reading buffer.
2721: */
1.56 veillard 2722: processed = ctxt->input->cur - ctxt->input->base;
2723: xmlBufferShrink(ctxt->input->buf->buffer, processed);
1.54 veillard 2724: nbchars = xmlCharEncInFunc(ctxt->input->buf->encoder,
2725: ctxt->input->buf->buffer,
2726: ctxt->input->buf->raw);
2727: if (nbchars < 0) {
1.67 veillard 2728: ctxt->errNo = XML_ERR_INVALID_ENCODING;
1.54 veillard 2729: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2730: ctxt->sax->error(ctxt->userData,
2731: "htmlCheckEncoding: encoder error\n");
2732: }
1.56 veillard 2733: ctxt->input->base =
2734: ctxt->input->cur = ctxt->input->buf->buffer->content;
1.54 veillard 2735: }
1.47 daniel 2736: }
2737: }
2738:
2739: /**
2740: * htmlCheckMeta:
2741: * @ctxt: an HTML parser context
2742: * @atts: the attributes values
2743: *
2744: * Checks an attributes from a Meta tag
2745: */
2746: void
2747: htmlCheckMeta(htmlParserCtxtPtr ctxt, const xmlChar **atts) {
2748: int i;
2749: const xmlChar *att, *value;
2750: int http = 0;
2751: const xmlChar *content = NULL;
2752:
2753: if ((ctxt == NULL) || (atts == NULL))
2754: return;
2755:
2756: i = 0;
2757: att = atts[i++];
2758: while (att != NULL) {
2759: value = atts[i++];
2760: if ((value != NULL) &&
2761: ((!xmlStrcmp(att, BAD_CAST"http-equiv")) ||
2762: (!xmlStrcmp(att, BAD_CAST"Http-Equiv")) ||
2763: (!xmlStrcmp(att, BAD_CAST"HTTP-EQUIV"))) &&
2764: ((!xmlStrcmp(value, BAD_CAST"Content-Type")) ||
2765: (!xmlStrcmp(value, BAD_CAST"content-type")) ||
2766: (!xmlStrcmp(value, BAD_CAST"CONTENT-TYPE"))))
2767: http = 1;
2768: else if ((value != NULL) &&
2769: ((!xmlStrcmp(att, BAD_CAST"content")) ||
2770: (!xmlStrcmp(att, BAD_CAST"Content")) ||
2771: (!xmlStrcmp(att, BAD_CAST"CONTENT"))))
2772: content = value;
2773: att = atts[i++];
2774: }
2775: if ((http) && (content != NULL))
2776: htmlCheckEncoding(ctxt, content);
2777:
2778: }
2779:
2780: /**
1.1 daniel 2781: * htmlParseStartTag:
2782: * @ctxt: an HTML parser context
2783: *
2784: * parse a start of tag either for rule element or
2785: * EmptyElement. In both case we don't parse the tag closing chars.
2786: *
2787: * [40] STag ::= '<' Name (S Attribute)* S? '>'
2788: *
2789: * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
2790: *
2791: * With namespace:
2792: *
2793: * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
2794: *
2795: * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
2796: *
2797: */
2798:
1.18 daniel 2799: void
1.1 daniel 2800: htmlParseStartTag(htmlParserCtxtPtr ctxt) {
1.14 daniel 2801: xmlChar *name;
2802: xmlChar *attname;
2803: xmlChar *attvalue;
2804: const xmlChar **atts = NULL;
1.1 daniel 2805: int nbatts = 0;
2806: int maxatts = 0;
1.47 daniel 2807: int meta = 0;
1.1 daniel 2808: int i;
2809:
1.18 daniel 2810: if (CUR != '<') return;
1.1 daniel 2811: NEXT;
2812:
1.19 daniel 2813: GROW;
1.1 daniel 2814: name = htmlParseHTMLName(ctxt);
2815: if (name == NULL) {
2816: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2817: ctxt->sax->error(ctxt->userData,
2818: "htmlParseStartTag: invalid element name\n");
2819: ctxt->wellFormed = 0;
1.18 daniel 2820: return;
1.1 daniel 2821: }
1.47 daniel 2822: if (!xmlStrcmp(name, BAD_CAST"meta"))
2823: meta = 1;
1.1 daniel 2824:
2825: /*
2826: * Check for auto-closure of HTML elements.
2827: */
2828: htmlAutoClose(ctxt, name);
1.43 daniel 2829:
2830: /*
2831: * Check for implied HTML elements.
2832: */
2833: htmlCheckImplied(ctxt, name);
1.1 daniel 2834:
2835: /*
2836: * Now parse the attributes, it ends up with the ending
2837: *
2838: * (S Attribute)* S?
2839: */
2840: SKIP_BLANKS;
2841: while ((IS_CHAR(CUR)) &&
2842: (CUR != '>') &&
2843: ((CUR != '/') || (NXT(1) != '>'))) {
1.26 daniel 2844: long cons = ctxt->nbChars;
1.1 daniel 2845:
1.19 daniel 2846: GROW;
1.1 daniel 2847: attname = htmlParseAttribute(ctxt, &attvalue);
1.31 daniel 2848: if (attname != NULL) {
1.47 daniel 2849:
1.1 daniel 2850: /*
2851: * Well formedness requires at most one declaration of an attribute
2852: */
2853: for (i = 0; i < nbatts;i += 2) {
2854: if (!xmlStrcmp(atts[i], attname)) {
2855: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.19 daniel 2856: ctxt->sax->error(ctxt->userData,
2857: "Attribute %s redefined\n",
2858: attname);
1.1 daniel 2859: ctxt->wellFormed = 0;
1.11 daniel 2860: xmlFree(attname);
1.31 daniel 2861: if (attvalue != NULL)
2862: xmlFree(attvalue);
1.19 daniel 2863: goto failed;
1.1 daniel 2864: }
2865: }
2866:
2867: /*
2868: * Add the pair to atts
2869: */
2870: if (atts == NULL) {
2871: maxatts = 10;
1.14 daniel 2872: atts = (const xmlChar **) xmlMalloc(maxatts * sizeof(xmlChar *));
1.1 daniel 2873: if (atts == NULL) {
2874: fprintf(stderr, "malloc of %ld byte failed\n",
1.14 daniel 2875: maxatts * (long)sizeof(xmlChar *));
1.18 daniel 2876: if (name != NULL) xmlFree(name);
2877: return;
1.1 daniel 2878: }
1.23 daniel 2879: } else if (nbatts + 4 > maxatts) {
1.1 daniel 2880: maxatts *= 2;
1.14 daniel 2881: atts = (const xmlChar **) xmlRealloc(atts, maxatts * sizeof(xmlChar *));
1.1 daniel 2882: if (atts == NULL) {
2883: fprintf(stderr, "realloc of %ld byte failed\n",
1.14 daniel 2884: maxatts * (long)sizeof(xmlChar *));
1.18 daniel 2885: if (name != NULL) xmlFree(name);
2886: return;
1.1 daniel 2887: }
2888: }
2889: atts[nbatts++] = attname;
2890: atts[nbatts++] = attvalue;
2891: atts[nbatts] = NULL;
2892: atts[nbatts + 1] = NULL;
2893: }
2894:
1.19 daniel 2895: failed:
1.1 daniel 2896: SKIP_BLANKS;
1.26 daniel 2897: if (cons == ctxt->nbChars) {
1.1 daniel 2898: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2899: ctxt->sax->error(ctxt->userData,
2900: "htmlParseStartTag: problem parsing attributes\n");
2901: ctxt->wellFormed = 0;
2902: break;
2903: }
2904: }
2905:
2906: /*
1.47 daniel 2907: * Handle specific association to the META tag
2908: */
2909: if (meta)
2910: htmlCheckMeta(ctxt, atts);
2911:
2912: /*
1.1 daniel 2913: * SAX: Start of Element !
2914: */
1.15 daniel 2915: htmlnamePush(ctxt, xmlStrdup(name));
1.18 daniel 2916: #ifdef DEBUG
2917: fprintf(stderr,"Start of element %s: pushed %s\n", name, ctxt->name);
2918: #endif
1.1 daniel 2919: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
2920: ctxt->sax->startElement(ctxt->userData, name, atts);
2921:
2922: if (atts != NULL) {
1.31 daniel 2923: for (i = 0;i < nbatts;i++) {
2924: if (atts[i] != NULL)
2925: xmlFree((xmlChar *) atts[i]);
2926: }
1.45 daniel 2927: xmlFree((void *) atts);
1.1 daniel 2928: }
1.18 daniel 2929: if (name != NULL) xmlFree(name);
1.1 daniel 2930: }
2931:
2932: /**
2933: * htmlParseEndTag:
2934: * @ctxt: an HTML parser context
2935: *
2936: * parse an end of tag
2937: *
2938: * [42] ETag ::= '</' Name S? '>'
2939: *
2940: * With namespace
2941: *
2942: * [NS 9] ETag ::= '</' QName S? '>'
2943: */
2944:
2945: void
1.18 daniel 2946: htmlParseEndTag(htmlParserCtxtPtr ctxt) {
1.14 daniel 2947: xmlChar *name;
1.15 daniel 2948: xmlChar *oldname;
1.1 daniel 2949: int i;
2950:
2951: if ((CUR != '<') || (NXT(1) != '/')) {
2952: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2953: ctxt->sax->error(ctxt->userData, "htmlParseEndTag: '</' not found\n");
2954: ctxt->wellFormed = 0;
2955: return;
2956: }
2957: SKIP(2);
2958:
2959: name = htmlParseHTMLName(ctxt);
1.24 daniel 2960: if (name == NULL) return;
1.1 daniel 2961:
2962: /*
2963: * We should definitely be at the ending "S? '>'" part
2964: */
2965: SKIP_BLANKS;
2966: if ((!IS_CHAR(CUR)) || (CUR != '>')) {
2967: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2968: ctxt->sax->error(ctxt->userData, "End tag : expected '>'\n");
2969: ctxt->wellFormed = 0;
2970: } else
2971: NEXT;
2972:
2973: /*
1.18 daniel 2974: * If the name read is not one of the element in the parsing stack
2975: * then return, it's just an error.
1.1 daniel 2976: */
1.18 daniel 2977: for (i = (ctxt->nameNr - 1);i >= 0;i--) {
2978: if (!xmlStrcmp(name, ctxt->nameTab[i])) break;
1.1 daniel 2979: }
2980: if (i < 0) {
2981: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.18 daniel 2982: ctxt->sax->error(ctxt->userData,
2983: "Unexpected end tag : %s\n", name);
1.11 daniel 2984: xmlFree(name);
1.1 daniel 2985: ctxt->wellFormed = 0;
2986: return;
2987: }
2988:
1.18 daniel 2989:
1.1 daniel 2990: /*
2991: * Check for auto-closure of HTML elements.
2992: */
1.18 daniel 2993:
1.1 daniel 2994: htmlAutoCloseOnClose(ctxt, name);
2995:
2996: /*
2997: * Well formedness constraints, opening and closing must match.
2998: * With the exception that the autoclose may have popped stuff out
2999: * of the stack.
3000: */
1.18 daniel 3001: if (xmlStrcmp(name, ctxt->name)) {
3002: #ifdef DEBUG
3003: fprintf(stderr,"End of tag %s: expecting %s\n", name, ctxt->name);
3004: #endif
1.15 daniel 3005: if ((ctxt->name != NULL) &&
3006: (xmlStrcmp(ctxt->name, name))) {
1.1 daniel 3007: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3008: ctxt->sax->error(ctxt->userData,
3009: "Opening and ending tag mismatch: %s and %s\n",
1.15 daniel 3010: name, ctxt->name);
1.1 daniel 3011: ctxt->wellFormed = 0;
3012: }
3013: }
3014:
3015: /*
3016: * SAX: End of Tag
3017: */
1.15 daniel 3018: oldname = ctxt->name;
1.24 daniel 3019: if ((oldname != NULL) && (!xmlStrcmp(oldname, name))) {
1.18 daniel 3020: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3021: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 3022: oldname = htmlnamePop(ctxt);
1.18 daniel 3023: if (oldname != NULL) {
3024: #ifdef DEBUG
3025: fprintf(stderr,"End of tag %s: popping out %s\n", name, oldname);
3026: #endif
3027: xmlFree(oldname);
3028: #ifdef DEBUG
3029: } else {
3030: fprintf(stderr,"End of tag %s: stack empty !!!\n", name);
3031: #endif
3032: }
3033: }
1.1 daniel 3034:
3035: if (name != NULL)
1.11 daniel 3036: xmlFree(name);
1.1 daniel 3037:
3038: return;
3039: }
3040:
3041:
3042: /**
3043: * htmlParseReference:
3044: * @ctxt: an HTML parser context
3045: *
3046: * parse and handle entity references in content,
3047: * this will end-up in a call to character() since this is either a
3048: * CharRef, or a predefined entity.
3049: */
3050: void
3051: htmlParseReference(htmlParserCtxtPtr ctxt) {
3052: htmlEntityDescPtr ent;
1.52 veillard 3053: xmlChar out[6];
1.14 daniel 3054: xmlChar *name;
1.1 daniel 3055: if (CUR != '&') return;
3056:
3057: if (NXT(1) == '#') {
1.52 veillard 3058: unsigned int c;
3059: int bits, i = 0;
3060:
3061: c = htmlParseCharRef(ctxt);
3062: if (c < 0x80) { out[i++]= c; bits= -6; }
3063: else if (c < 0x800) { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3064: else if (c < 0x10000) { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3065: else { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3066:
3067: for ( ; bits >= 0; bits-= 6) {
3068: out[i++]= ((c >> bits) & 0x3F) | 0x80;
3069: }
3070: out[i] = 0;
3071:
1.59 veillard 3072: htmlCheckParagraph(ctxt);
1.1 daniel 3073: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1.52 veillard 3074: ctxt->sax->characters(ctxt->userData, out, i);
1.1 daniel 3075: } else {
3076: ent = htmlParseEntityRef(ctxt, &name);
1.32 daniel 3077: if (name == NULL) {
1.59 veillard 3078: htmlCheckParagraph(ctxt);
1.58 veillard 3079: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3080: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
1.32 daniel 3081: return;
3082: }
1.52 veillard 3083: if ((ent == NULL) || (ent->value <= 0)) {
1.59 veillard 3084: htmlCheckParagraph(ctxt);
1.1 daniel 3085: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
1.8 daniel 3086: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
1.1 daniel 3087: ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
1.32 daniel 3088: /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
1.1 daniel 3089: }
3090: } else {
1.52 veillard 3091: unsigned int c;
3092: int bits, i = 0;
3093:
3094: c = ent->value;
3095: if (c < 0x80)
3096: { out[i++]= c; bits= -6; }
3097: else if (c < 0x800)
3098: { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3099: else if (c < 0x10000)
3100: { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3101: else
3102: { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3103:
3104: for ( ; bits >= 0; bits-= 6) {
3105: out[i++]= ((c >> bits) & 0x3F) | 0x80;
3106: }
3107: out[i] = 0;
3108:
1.59 veillard 3109: htmlCheckParagraph(ctxt);
1.1 daniel 3110: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1.52 veillard 3111: ctxt->sax->characters(ctxt->userData, out, i);
1.1 daniel 3112: }
1.11 daniel 3113: xmlFree(name);
1.1 daniel 3114: }
3115: }
3116:
3117: /**
3118: * htmlParseContent:
3119: * @ctxt: an HTML parser context
3120: * @name: the node name
3121: *
3122: * Parse a content: comment, sub-element, reference or text.
3123: *
3124: */
3125:
3126: void
1.18 daniel 3127: htmlParseContent(htmlParserCtxtPtr ctxt) {
1.15 daniel 3128: xmlChar *currentNode;
1.18 daniel 3129: int depth;
1.1 daniel 3130:
1.26 daniel 3131: currentNode = xmlStrdup(ctxt->name);
1.18 daniel 3132: depth = ctxt->nameNr;
3133: while (1) {
1.26 daniel 3134: long cons = ctxt->nbChars;
1.1 daniel 3135:
1.18 daniel 3136: GROW;
3137: /*
3138: * Our tag or one of it's parent or children is ending.
3139: */
3140: if ((CUR == '<') && (NXT(1) == '/')) {
3141: htmlParseEndTag(ctxt);
1.26 daniel 3142: if (currentNode != NULL) xmlFree(currentNode);
1.18 daniel 3143: return;
3144: }
3145:
3146: /*
3147: * Has this node been popped out during parsing of
3148: * the next element
3149: */
1.26 daniel 3150: if ((xmlStrcmp(currentNode, ctxt->name)) &&
3151: (depth >= ctxt->nameNr)) {
3152: if (currentNode != NULL) xmlFree(currentNode);
3153: return;
3154: }
1.18 daniel 3155:
1.1 daniel 3156: /*
1.59 veillard 3157: * Sometimes DOCTYPE arrives in the middle of the document
3158: */
3159: if ((CUR == '<') && (NXT(1) == '!') &&
3160: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3161: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3162: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3163: (UPP(8) == 'E')) {
3164: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3165: ctxt->sax->error(ctxt->userData,
3166: "Misplaced DOCTYPE declaration\n");
3167: ctxt->wellFormed = 0;
3168: htmlParseDocTypeDecl(ctxt);
3169: }
3170:
3171: /*
1.1 daniel 3172: * First case : a comment
3173: */
3174: if ((CUR == '<') && (NXT(1) == '!') &&
3175: (NXT(2) == '-') && (NXT(3) == '-')) {
1.31 daniel 3176: htmlParseComment(ctxt);
1.1 daniel 3177: }
3178:
3179: /*
3180: * Second case : a sub-element.
3181: */
3182: else if (CUR == '<') {
3183: htmlParseElement(ctxt);
3184: }
3185:
3186: /*
3187: * Third case : a reference. If if has not been resolved,
3188: * parsing returns it's Name, create the node
3189: */
3190: else if (CUR == '&') {
3191: htmlParseReference(ctxt);
3192: }
3193:
3194: /*
1.47 daniel 3195: * Fourth : end of the resource
3196: */
3197: else if (CUR == 0) {
3198: htmlAutoClose(ctxt, NULL);
3199: }
3200:
3201: /*
1.1 daniel 3202: * Last case, text. Note that References are handled directly.
3203: */
3204: else {
3205: htmlParseCharData(ctxt, 0);
3206: }
3207:
1.26 daniel 3208: if (cons == ctxt->nbChars) {
1.22 daniel 3209: if (ctxt->node != NULL) {
3210: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3211: ctxt->sax->error(ctxt->userData,
3212: "detected an error in element content\n");
3213: ctxt->wellFormed = 0;
3214: }
1.1 daniel 3215: break;
3216: }
1.17 daniel 3217:
1.5 daniel 3218: GROW;
1.1 daniel 3219: }
1.26 daniel 3220: if (currentNode != NULL) xmlFree(currentNode);
1.1 daniel 3221: }
3222:
3223: /**
3224: * htmlParseElement:
3225: * @ctxt: an HTML parser context
3226: *
3227: * parse an HTML element, this is highly recursive
3228: *
3229: * [39] element ::= EmptyElemTag | STag content ETag
3230: *
3231: * [41] Attribute ::= Name Eq AttValue
3232: */
3233:
3234: void
3235: htmlParseElement(htmlParserCtxtPtr ctxt) {
1.14 daniel 3236: xmlChar *name;
1.16 daniel 3237: xmlChar *currentNode = NULL;
1.1 daniel 3238: htmlElemDescPtr info;
1.10 daniel 3239: htmlParserNodeInfo node_info;
1.31 daniel 3240: xmlChar *oldname;
1.18 daniel 3241: int depth = ctxt->nameNr;
1.1 daniel 3242:
3243: /* Capture start position */
1.10 daniel 3244: if (ctxt->record_info) {
3245: node_info.begin_pos = ctxt->input->consumed +
3246: (CUR_PTR - ctxt->input->base);
3247: node_info.begin_line = ctxt->input->line;
3248: }
1.1 daniel 3249:
1.26 daniel 3250: oldname = xmlStrdup(ctxt->name);
1.18 daniel 3251: htmlParseStartTag(ctxt);
3252: name = ctxt->name;
1.19 daniel 3253: #ifdef DEBUG
3254: if (oldname == NULL)
3255: fprintf(stderr, "Start of element %s\n", name);
3256: else if (name == NULL)
3257: fprintf(stderr, "Start of element failed, was %s\n", oldname);
3258: else
3259: fprintf(stderr, "Start of element %s, was %s\n", name, oldname);
3260: #endif
1.26 daniel 3261: if (((depth == ctxt->nameNr) && (!xmlStrcmp(oldname, ctxt->name))) ||
1.18 daniel 3262: (name == NULL)) {
1.19 daniel 3263: if (CUR == '>')
3264: NEXT;
1.26 daniel 3265: if (oldname != NULL)
3266: xmlFree(oldname);
1.1 daniel 3267: return;
3268: }
1.26 daniel 3269: if (oldname != NULL)
3270: xmlFree(oldname);
1.1 daniel 3271:
3272: /*
3273: * Lookup the info for that element.
3274: */
3275: info = htmlTagLookup(name);
3276: if (info == NULL) {
3277: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3278: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
3279: name);
3280: ctxt->wellFormed = 0;
3281: } else if (info->depr) {
3282: /***************************
3283: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
3284: ctxt->sax->warning(ctxt->userData, "Tag %s is deprecated\n",
3285: name);
3286: ***************************/
3287: }
3288:
3289: /*
3290: * Check for an Empty Element labelled the XML/SGML way
3291: */
3292: if ((CUR == '/') && (NXT(1) == '>')) {
3293: SKIP(2);
3294: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3295: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 3296: oldname = htmlnamePop(ctxt);
1.18 daniel 3297: #ifdef DEBUG
3298: fprintf(stderr,"End of tag the XML way: popping out %s\n", oldname);
3299: #endif
1.17 daniel 3300: if (oldname != NULL)
3301: xmlFree(oldname);
1.1 daniel 3302: return;
3303: }
3304:
1.5 daniel 3305: if (CUR == '>') {
3306: NEXT;
3307: } else {
1.1 daniel 3308: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.56 veillard 3309: ctxt->sax->error(ctxt->userData,
3310: "Couldn't find end of Start Tag %s\n",
3311: name);
1.1 daniel 3312: ctxt->wellFormed = 0;
3313:
3314: /*
3315: * end of parsing of this node.
3316: */
1.18 daniel 3317: if (!xmlStrcmp(name, ctxt->name)) {
3318: nodePop(ctxt);
1.24 daniel 3319: oldname = htmlnamePop(ctxt);
1.18 daniel 3320: #ifdef DEBUG
3321: fprintf(stderr,"End of start tag problem: popping out %s\n", oldname);
3322: #endif
3323: if (oldname != NULL)
3324: xmlFree(oldname);
3325: }
1.10 daniel 3326:
3327: /*
3328: * Capture end position and add node
3329: */
3330: if ( currentNode != NULL && ctxt->record_info ) {
3331: node_info.end_pos = ctxt->input->consumed +
3332: (CUR_PTR - ctxt->input->base);
3333: node_info.end_line = ctxt->input->line;
1.15 daniel 3334: node_info.node = ctxt->node;
1.10 daniel 3335: xmlParserAddNodeInfo(ctxt, &node_info);
3336: }
1.1 daniel 3337: return;
3338: }
3339:
3340: /*
3341: * Check for an Empty Element from DTD definition
3342: */
3343: if ((info != NULL) && (info->empty)) {
3344: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3345: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 3346: oldname = htmlnamePop(ctxt);
1.18 daniel 3347: #ifdef DEBUG
3348: fprintf(stderr,"End of empty tag %s : popping out %s\n", name, oldname);
3349: #endif
1.17 daniel 3350: if (oldname != NULL)
3351: xmlFree(oldname);
1.1 daniel 3352: return;
3353: }
3354:
3355: /*
3356: * Parse the content of the element:
3357: */
1.26 daniel 3358: currentNode = xmlStrdup(ctxt->name);
1.18 daniel 3359: depth = ctxt->nameNr;
3360: while (IS_CHAR(CUR)) {
3361: htmlParseContent(ctxt);
3362: if (ctxt->nameNr < depth) break;
3363: }
1.1 daniel 3364:
3365: if (!IS_CHAR(CUR)) {
1.49 daniel 3366: /************
1.1 daniel 3367: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3368: ctxt->sax->error(ctxt->userData,
1.18 daniel 3369: "Premature end of data in tag %s\n", currentNode);
1.1 daniel 3370: ctxt->wellFormed = 0;
1.49 daniel 3371: *************/
1.1 daniel 3372:
3373: /*
3374: * end of parsing of this node.
3375: */
3376: nodePop(ctxt);
1.24 daniel 3377: oldname = htmlnamePop(ctxt);
1.18 daniel 3378: #ifdef DEBUG
3379: fprintf(stderr,"Premature end of tag %s : popping out %s\n", name, oldname);
3380: #endif
1.17 daniel 3381: if (oldname != NULL)
3382: xmlFree(oldname);
1.26 daniel 3383: if (currentNode != NULL)
3384: xmlFree(currentNode);
1.1 daniel 3385: return;
3386: }
1.10 daniel 3387:
3388: /*
3389: * Capture end position and add node
3390: */
3391: if ( currentNode != NULL && ctxt->record_info ) {
3392: node_info.end_pos = ctxt->input->consumed +
3393: (CUR_PTR - ctxt->input->base);
3394: node_info.end_line = ctxt->input->line;
1.15 daniel 3395: node_info.node = ctxt->node;
1.10 daniel 3396: xmlParserAddNodeInfo(ctxt, &node_info);
3397: }
1.26 daniel 3398: if (currentNode != NULL)
3399: xmlFree(currentNode);
1.1 daniel 3400: }
3401:
3402: /**
3403: * htmlParseDocument :
3404: * @ctxt: an HTML parser context
3405: *
3406: * parse an HTML document (and build a tree if using the standard SAX
3407: * interface).
3408: *
3409: * Returns 0, -1 in case of error. the parser context is augmented
3410: * as a result of the parsing.
3411: */
3412:
3413: int
3414: htmlParseDocument(htmlParserCtxtPtr ctxt) {
1.59 veillard 3415: xmlDtdPtr dtd;
3416:
1.1 daniel 3417: htmlDefaultSAXHandlerInit();
3418: ctxt->html = 1;
3419:
1.5 daniel 3420: GROW;
1.1 daniel 3421: /*
1.9 daniel 3422: * SAX: beginning of the document processing.
1.1 daniel 3423: */
3424: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
3425: ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
3426:
3427: /*
3428: * Wipe out everything which is before the first '<'
3429: */
1.22 daniel 3430: SKIP_BLANKS;
1.1 daniel 3431: if (CUR == 0) {
3432: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3433: ctxt->sax->error(ctxt->userData, "Document is empty\n");
3434: ctxt->wellFormed = 0;
3435: }
3436:
1.40 daniel 3437: if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX))
3438: ctxt->sax->startDocument(ctxt->userData);
3439:
3440:
1.22 daniel 3441: /*
3442: * Parse possible comments before any content
3443: */
3444: while ((CUR == '<') && (NXT(1) == '!') &&
3445: (NXT(2) == '-') && (NXT(3) == '-')) {
1.31 daniel 3446: htmlParseComment(ctxt);
1.22 daniel 3447: SKIP_BLANKS;
3448: }
3449:
1.1 daniel 3450:
3451: /*
3452: * Then possibly doc type declaration(s) and more Misc
3453: * (doctypedecl Misc*)?
3454: */
3455: if ((CUR == '<') && (NXT(1) == '!') &&
3456: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3457: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3458: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3459: (UPP(8) == 'E')) {
3460: htmlParseDocTypeDecl(ctxt);
3461: }
3462: SKIP_BLANKS;
3463:
3464: /*
1.55 veillard 3465: * Parse possible comments before any content
3466: */
3467: while ((CUR == '<') && (NXT(1) == '!') &&
3468: (NXT(2) == '-') && (NXT(3) == '-')) {
3469: htmlParseComment(ctxt);
3470: SKIP_BLANKS;
3471: }
3472:
3473: /*
1.1 daniel 3474: * Time to start parsing the tree itself
3475: */
1.22 daniel 3476: htmlParseContent(ctxt);
1.1 daniel 3477:
3478: /*
1.47 daniel 3479: * autoclose
3480: */
3481: if (CUR == 0)
3482: htmlAutoClose(ctxt, NULL);
3483:
3484:
3485: /*
1.1 daniel 3486: * SAX: end of the document processing.
3487: */
3488: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3489: ctxt->sax->endDocument(ctxt->userData);
1.59 veillard 3490:
3491: if (ctxt->myDoc != NULL) {
3492: dtd = xmlGetIntSubset(ctxt->myDoc);
3493: if (dtd == NULL)
3494: ctxt->myDoc->intSubset =
3495: xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML",
3496: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
3497: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
3498: }
1.1 daniel 3499: if (! ctxt->wellFormed) return(-1);
3500: return(0);
3501: }
3502:
3503:
1.30 daniel 3504: /************************************************************************
3505: * *
3506: * Parser contexts handling *
3507: * *
3508: ************************************************************************/
1.1 daniel 3509:
3510: /**
3511: * xmlInitParserCtxt:
3512: * @ctxt: an HTML parser context
3513: *
3514: * Initialize a parser context
3515: */
3516:
3517: void
3518: htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
3519: {
3520: htmlSAXHandler *sax;
3521:
1.21 daniel 3522: if (ctxt == NULL) return;
3523: memset(ctxt, 0, sizeof(htmlParserCtxt));
3524:
1.11 daniel 3525: sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
1.1 daniel 3526: if (sax == NULL) {
3527: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
3528: }
1.68 ! veillard 3529: else
! 3530: memset(sax, 0, sizeof(htmlSAXHandler));
1.1 daniel 3531:
3532: /* Allocate the Input stack */
1.19 daniel 3533: ctxt->inputTab = (htmlParserInputPtr *)
3534: xmlMalloc(5 * sizeof(htmlParserInputPtr));
3535: if (ctxt->inputTab == NULL) {
3536: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
1.65 veillard 3537: ctxt->inputNr = 0;
3538: ctxt->inputMax = 0;
3539: ctxt->input = NULL;
3540: return;
1.19 daniel 3541: }
1.1 daniel 3542: ctxt->inputNr = 0;
3543: ctxt->inputMax = 5;
3544: ctxt->input = NULL;
3545: ctxt->version = NULL;
3546: ctxt->encoding = NULL;
3547: ctxt->standalone = -1;
1.30 daniel 3548: ctxt->instate = XML_PARSER_START;
1.1 daniel 3549:
3550: /* Allocate the Node stack */
1.11 daniel 3551: ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
1.65 veillard 3552: if (ctxt->nodeTab == NULL) {
3553: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
3554: ctxt->nodeNr = 0;
3555: ctxt->nodeMax = 0;
3556: ctxt->node = NULL;
3557: ctxt->inputNr = 0;
3558: ctxt->inputMax = 0;
3559: ctxt->input = NULL;
3560: return;
3561: }
1.1 daniel 3562: ctxt->nodeNr = 0;
3563: ctxt->nodeMax = 10;
3564: ctxt->node = NULL;
3565:
1.15 daniel 3566: /* Allocate the Name stack */
3567: ctxt->nameTab = (xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
1.65 veillard 3568: if (ctxt->nameTab == NULL) {
3569: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
3570: ctxt->nameNr = 0;
3571: ctxt->nameMax = 10;
3572: ctxt->name = NULL;
3573: ctxt->nodeNr = 0;
3574: ctxt->nodeMax = 0;
3575: ctxt->node = NULL;
3576: ctxt->inputNr = 0;
3577: ctxt->inputMax = 0;
3578: ctxt->input = NULL;
3579: return;
3580: }
1.15 daniel 3581: ctxt->nameNr = 0;
3582: ctxt->nameMax = 10;
3583: ctxt->name = NULL;
3584:
1.1 daniel 3585: if (sax == NULL) ctxt->sax = &htmlDefaultSAXHandler;
3586: else {
3587: ctxt->sax = sax;
3588: memcpy(sax, &htmlDefaultSAXHandler, sizeof(htmlSAXHandler));
3589: }
3590: ctxt->userData = ctxt;
3591: ctxt->myDoc = NULL;
3592: ctxt->wellFormed = 1;
3593: ctxt->replaceEntities = 0;
3594: ctxt->html = 1;
3595: ctxt->record_info = 0;
1.21 daniel 3596: ctxt->validate = 0;
1.26 daniel 3597: ctxt->nbChars = 0;
1.30 daniel 3598: ctxt->checkIndex = 0;
1.1 daniel 3599: xmlInitNodeInfoSeq(&ctxt->node_seq);
3600: }
3601:
3602: /**
3603: * htmlFreeParserCtxt:
3604: * @ctxt: an HTML parser context
3605: *
3606: * Free all the memory used by a parser context. However the parsed
3607: * document in ctxt->myDoc is not freed.
3608: */
3609:
3610: void
3611: htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
3612: {
1.47 daniel 3613: xmlFreeParserCtxt(ctxt);
1.1 daniel 3614: }
3615:
3616: /**
3617: * htmlCreateDocParserCtxt :
1.14 daniel 3618: * @cur: a pointer to an array of xmlChar
1.1 daniel 3619: * @encoding: a free form C string describing the HTML document encoding, or NULL
3620: *
3621: * Create a parser context for an HTML document.
3622: *
3623: * Returns the new parser context or NULL
3624: */
3625: htmlParserCtxtPtr
1.14 daniel 3626: htmlCreateDocParserCtxt(xmlChar *cur, const char *encoding) {
1.1 daniel 3627: htmlParserCtxtPtr ctxt;
3628: htmlParserInputPtr input;
3629: /* htmlCharEncoding enc; */
3630:
1.11 daniel 3631: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 3632: if (ctxt == NULL) {
3633: perror("malloc");
3634: return(NULL);
3635: }
3636: htmlInitParserCtxt(ctxt);
1.11 daniel 3637: input = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 3638: if (input == NULL) {
3639: perror("malloc");
1.11 daniel 3640: xmlFree(ctxt);
1.1 daniel 3641: return(NULL);
3642: }
1.19 daniel 3643: memset(input, 0, sizeof(htmlParserInput));
1.1 daniel 3644:
3645: input->line = 1;
3646: input->col = 1;
3647: input->base = cur;
3648: input->cur = cur;
3649:
3650: inputPush(ctxt, input);
3651: return(ctxt);
3652: }
3653:
1.31 daniel 3654: /************************************************************************
3655: * *
3656: * Progressive parsing interfaces *
3657: * *
3658: ************************************************************************/
3659:
3660: /**
3661: * htmlParseLookupSequence:
3662: * @ctxt: an HTML parser context
3663: * @first: the first char to lookup
3664: * @next: the next char to lookup or zero
3665: * @third: the next char to lookup or zero
3666: *
3667: * Try to find if a sequence (first, next, third) or just (first next) or
3668: * (first) is available in the input stream.
3669: * This function has a side effect of (possibly) incrementing ctxt->checkIndex
3670: * to avoid rescanning sequences of bytes, it DOES change the state of the
3671: * parser, do not use liberally.
3672: * This is basically similar to xmlParseLookupSequence()
3673: *
3674: * Returns the index to the current parsing point if the full sequence
3675: * is available, -1 otherwise.
3676: */
3677: int
3678: htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first,
3679: xmlChar next, xmlChar third) {
3680: int base, len;
3681: htmlParserInputPtr in;
3682: const xmlChar *buf;
3683:
3684: in = ctxt->input;
3685: if (in == NULL) return(-1);
3686: base = in->cur - in->base;
3687: if (base < 0) return(-1);
3688: if (ctxt->checkIndex > base)
3689: base = ctxt->checkIndex;
3690: if (in->buf == NULL) {
3691: buf = in->base;
3692: len = in->length;
3693: } else {
3694: buf = in->buf->buffer->content;
3695: len = in->buf->buffer->use;
3696: }
3697: /* take into account the sequence length */
3698: if (third) len -= 2;
3699: else if (next) len --;
3700: for (;base < len;base++) {
3701: if (buf[base] == first) {
3702: if (third != 0) {
3703: if ((buf[base + 1] != next) ||
3704: (buf[base + 2] != third)) continue;
3705: } else if (next != 0) {
3706: if (buf[base + 1] != next) continue;
3707: }
3708: ctxt->checkIndex = 0;
3709: #ifdef DEBUG_PUSH
3710: if (next == 0)
3711: fprintf(stderr, "HPP: lookup '%c' found at %d\n",
3712: first, base);
3713: else if (third == 0)
3714: fprintf(stderr, "HPP: lookup '%c%c' found at %d\n",
3715: first, next, base);
3716: else
3717: fprintf(stderr, "HPP: lookup '%c%c%c' found at %d\n",
3718: first, next, third, base);
3719: #endif
3720: return(base - (in->cur - in->base));
3721: }
3722: }
3723: ctxt->checkIndex = base;
3724: #ifdef DEBUG_PUSH
3725: if (next == 0)
3726: fprintf(stderr, "HPP: lookup '%c' failed\n", first);
3727: else if (third == 0)
3728: fprintf(stderr, "HPP: lookup '%c%c' failed\n", first, next);
3729: else
3730: fprintf(stderr, "HPP: lookup '%c%c%c' failed\n", first, next, third);
3731: #endif
3732: return(-1);
3733: }
3734:
3735: /**
1.32 daniel 3736: * htmlParseTryOrFinish:
1.31 daniel 3737: * @ctxt: an HTML parser context
1.32 daniel 3738: * @terminate: last chunk indicator
1.31 daniel 3739: *
3740: * Try to progress on parsing
3741: *
3742: * Returns zero if no parsing was possible
3743: */
3744: int
1.32 daniel 3745: htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
1.31 daniel 3746: int ret = 0;
3747: htmlParserInputPtr in;
1.47 daniel 3748: int avail = 0;
1.31 daniel 3749: xmlChar cur, next;
3750:
3751: #ifdef DEBUG_PUSH
3752: switch (ctxt->instate) {
3753: case XML_PARSER_EOF:
3754: fprintf(stderr, "HPP: try EOF\n"); break;
3755: case XML_PARSER_START:
3756: fprintf(stderr, "HPP: try START\n"); break;
3757: case XML_PARSER_MISC:
3758: fprintf(stderr, "HPP: try MISC\n");break;
3759: case XML_PARSER_COMMENT:
3760: fprintf(stderr, "HPP: try COMMENT\n");break;
3761: case XML_PARSER_PROLOG:
3762: fprintf(stderr, "HPP: try PROLOG\n");break;
3763: case XML_PARSER_START_TAG:
3764: fprintf(stderr, "HPP: try START_TAG\n");break;
3765: case XML_PARSER_CONTENT:
3766: fprintf(stderr, "HPP: try CONTENT\n");break;
3767: case XML_PARSER_CDATA_SECTION:
3768: fprintf(stderr, "HPP: try CDATA_SECTION\n");break;
3769: case XML_PARSER_END_TAG:
3770: fprintf(stderr, "HPP: try END_TAG\n");break;
3771: case XML_PARSER_ENTITY_DECL:
3772: fprintf(stderr, "HPP: try ENTITY_DECL\n");break;
3773: case XML_PARSER_ENTITY_VALUE:
3774: fprintf(stderr, "HPP: try ENTITY_VALUE\n");break;
3775: case XML_PARSER_ATTRIBUTE_VALUE:
3776: fprintf(stderr, "HPP: try ATTRIBUTE_VALUE\n");break;
3777: case XML_PARSER_DTD:
3778: fprintf(stderr, "HPP: try DTD\n");break;
3779: case XML_PARSER_EPILOG:
3780: fprintf(stderr, "HPP: try EPILOG\n");break;
3781: case XML_PARSER_PI:
3782: fprintf(stderr, "HPP: try PI\n");break;
3783: }
3784: #endif
3785:
3786: while (1) {
3787:
3788: in = ctxt->input;
3789: if (in == NULL) break;
3790: if (in->buf == NULL)
3791: avail = in->length - (in->cur - in->base);
3792: else
3793: avail = in->buf->buffer->use - (in->cur - in->base);
1.47 daniel 3794: if ((avail == 0) && (terminate)) {
3795: htmlAutoClose(ctxt, NULL);
1.54 veillard 3796: if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
3797: /*
3798: * SAX: end of the document processing.
3799: */
1.47 daniel 3800: ctxt->instate = XML_PARSER_EOF;
1.54 veillard 3801: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3802: ctxt->sax->endDocument(ctxt->userData);
3803: }
1.47 daniel 3804: }
1.31 daniel 3805: if (avail < 1)
3806: goto done;
3807: switch (ctxt->instate) {
3808: case XML_PARSER_EOF:
3809: /*
3810: * Document parsing is done !
3811: */
3812: goto done;
3813: case XML_PARSER_START:
3814: /*
3815: * Very first chars read from the document flow.
3816: */
3817: cur = in->cur[0];
3818: if (IS_BLANK(cur)) {
3819: SKIP_BLANKS;
3820: if (in->buf == NULL)
3821: avail = in->length - (in->cur - in->base);
3822: else
3823: avail = in->buf->buffer->use - (in->cur - in->base);
3824: }
3825: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
3826: ctxt->sax->setDocumentLocator(ctxt->userData,
3827: &xmlDefaultSAXLocator);
1.46 daniel 3828: if ((ctxt->sax) && (ctxt->sax->startDocument) &&
3829: (!ctxt->disableSAX))
3830: ctxt->sax->startDocument(ctxt->userData);
3831:
1.31 daniel 3832: cur = in->cur[0];
3833: next = in->cur[1];
3834: if ((cur == '<') && (next == '!') &&
3835: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3836: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3837: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3838: (UPP(8) == 'E')) {
1.32 daniel 3839: if ((!terminate) &&
3840: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3841: goto done;
3842: #ifdef DEBUG_PUSH
3843: fprintf(stderr, "HPP: Parsing internal subset\n");
3844: #endif
3845: htmlParseDocTypeDecl(ctxt);
3846: ctxt->instate = XML_PARSER_PROLOG;
3847: #ifdef DEBUG_PUSH
3848: fprintf(stderr, "HPP: entering PROLOG\n");
3849: #endif
3850: } else {
3851: ctxt->instate = XML_PARSER_MISC;
3852: }
3853: #ifdef DEBUG_PUSH
3854: fprintf(stderr, "HPP: entering MISC\n");
3855: #endif
3856: break;
3857: case XML_PARSER_MISC:
3858: SKIP_BLANKS;
3859: if (in->buf == NULL)
3860: avail = in->length - (in->cur - in->base);
3861: else
3862: avail = in->buf->buffer->use - (in->cur - in->base);
3863: if (avail < 2)
3864: goto done;
3865: cur = in->cur[0];
3866: next = in->cur[1];
3867: if ((cur == '<') && (next == '!') &&
3868: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3869: if ((!terminate) &&
3870: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3871: goto done;
3872: #ifdef DEBUG_PUSH
3873: fprintf(stderr, "HPP: Parsing Comment\n");
3874: #endif
3875: htmlParseComment(ctxt);
3876: ctxt->instate = XML_PARSER_MISC;
3877: } else if ((cur == '<') && (next == '!') &&
3878: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3879: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3880: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3881: (UPP(8) == 'E')) {
1.32 daniel 3882: if ((!terminate) &&
3883: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3884: goto done;
3885: #ifdef DEBUG_PUSH
3886: fprintf(stderr, "HPP: Parsing internal subset\n");
3887: #endif
3888: htmlParseDocTypeDecl(ctxt);
3889: ctxt->instate = XML_PARSER_PROLOG;
3890: #ifdef DEBUG_PUSH
3891: fprintf(stderr, "HPP: entering PROLOG\n");
3892: #endif
3893: } else if ((cur == '<') && (next == '!') &&
3894: (avail < 9)) {
3895: goto done;
3896: } else {
3897: ctxt->instate = XML_PARSER_START_TAG;
3898: #ifdef DEBUG_PUSH
3899: fprintf(stderr, "HPP: entering START_TAG\n");
3900: #endif
3901: }
3902: break;
3903: case XML_PARSER_PROLOG:
3904: SKIP_BLANKS;
3905: if (in->buf == NULL)
3906: avail = in->length - (in->cur - in->base);
3907: else
3908: avail = in->buf->buffer->use - (in->cur - in->base);
3909: if (avail < 2)
3910: goto done;
3911: cur = in->cur[0];
3912: next = in->cur[1];
3913: if ((cur == '<') && (next == '!') &&
3914: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3915: if ((!terminate) &&
3916: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3917: goto done;
3918: #ifdef DEBUG_PUSH
3919: fprintf(stderr, "HPP: Parsing Comment\n");
3920: #endif
3921: htmlParseComment(ctxt);
3922: ctxt->instate = XML_PARSER_PROLOG;
3923: } else if ((cur == '<') && (next == '!') &&
3924: (avail < 4)) {
3925: goto done;
3926: } else {
3927: ctxt->instate = XML_PARSER_START_TAG;
3928: #ifdef DEBUG_PUSH
3929: fprintf(stderr, "HPP: entering START_TAG\n");
3930: #endif
3931: }
3932: break;
3933: case XML_PARSER_EPILOG:
3934: if (in->buf == NULL)
3935: avail = in->length - (in->cur - in->base);
3936: else
3937: avail = in->buf->buffer->use - (in->cur - in->base);
1.55 veillard 3938: if (avail < 1)
3939: goto done;
3940: cur = in->cur[0];
3941: if (IS_BLANK(cur)) {
3942: htmlParseCharData(ctxt, 0);
3943: goto done;
3944: }
1.31 daniel 3945: if (avail < 2)
3946: goto done;
3947: next = in->cur[1];
3948: if ((cur == '<') && (next == '!') &&
3949: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3950: if ((!terminate) &&
3951: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3952: goto done;
3953: #ifdef DEBUG_PUSH
3954: fprintf(stderr, "HPP: Parsing Comment\n");
3955: #endif
3956: htmlParseComment(ctxt);
3957: ctxt->instate = XML_PARSER_EPILOG;
3958: } else if ((cur == '<') && (next == '!') &&
3959: (avail < 4)) {
3960: goto done;
3961: } else {
1.67 veillard 3962: ctxt->errNo = XML_ERR_DOCUMENT_END;
1.31 daniel 3963: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3964: ctxt->sax->error(ctxt->userData,
3965: "Extra content at the end of the document\n");
3966: ctxt->wellFormed = 0;
3967: ctxt->instate = XML_PARSER_EOF;
3968: #ifdef DEBUG_PUSH
3969: fprintf(stderr, "HPP: entering EOF\n");
3970: #endif
3971: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3972: ctxt->sax->endDocument(ctxt->userData);
3973: goto done;
3974: }
3975: break;
3976: case XML_PARSER_START_TAG: {
3977: xmlChar *name, *oldname;
3978: int depth = ctxt->nameNr;
3979: htmlElemDescPtr info;
3980:
3981: if (avail < 2)
3982: goto done;
3983: cur = in->cur[0];
3984: if (cur != '<') {
3985: ctxt->instate = XML_PARSER_CONTENT;
3986: #ifdef DEBUG_PUSH
3987: fprintf(stderr, "HPP: entering CONTENT\n");
3988: #endif
3989: break;
3990: }
1.32 daniel 3991: if ((!terminate) &&
3992: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3993: goto done;
3994:
3995: oldname = xmlStrdup(ctxt->name);
3996: htmlParseStartTag(ctxt);
3997: name = ctxt->name;
3998: #ifdef DEBUG
3999: if (oldname == NULL)
4000: fprintf(stderr, "Start of element %s\n", name);
4001: else if (name == NULL)
4002: fprintf(stderr, "Start of element failed, was %s\n",
4003: oldname);
4004: else
4005: fprintf(stderr, "Start of element %s, was %s\n",
4006: name, oldname);
4007: #endif
4008: if (((depth == ctxt->nameNr) &&
4009: (!xmlStrcmp(oldname, ctxt->name))) ||
4010: (name == NULL)) {
4011: if (CUR == '>')
4012: NEXT;
4013: if (oldname != NULL)
4014: xmlFree(oldname);
4015: break;
4016: }
4017: if (oldname != NULL)
4018: xmlFree(oldname);
4019:
4020: /*
4021: * Lookup the info for that element.
4022: */
4023: info = htmlTagLookup(name);
4024: if (info == NULL) {
4025: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4026: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
4027: name);
4028: ctxt->wellFormed = 0;
4029: } else if (info->depr) {
4030: /***************************
4031: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
4032: ctxt->sax->warning(ctxt->userData,
4033: "Tag %s is deprecated\n",
4034: name);
4035: ***************************/
4036: }
4037:
4038: /*
4039: * Check for an Empty Element labelled the XML/SGML way
4040: */
4041: if ((CUR == '/') && (NXT(1) == '>')) {
4042: SKIP(2);
4043: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4044: ctxt->sax->endElement(ctxt->userData, name);
4045: oldname = htmlnamePop(ctxt);
4046: #ifdef DEBUG
4047: fprintf(stderr,"End of tag the XML way: popping out %s\n",
4048: oldname);
4049: #endif
4050: if (oldname != NULL)
4051: xmlFree(oldname);
4052: ctxt->instate = XML_PARSER_CONTENT;
4053: #ifdef DEBUG_PUSH
4054: fprintf(stderr, "HPP: entering CONTENT\n");
4055: #endif
4056: break;
4057: }
4058:
4059: if (CUR == '>') {
4060: NEXT;
4061: } else {
4062: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4063: ctxt->sax->error(ctxt->userData,
4064: "Couldn't find end of Start Tag %s\n",
4065: name);
4066: ctxt->wellFormed = 0;
4067:
4068: /*
4069: * end of parsing of this node.
4070: */
4071: if (!xmlStrcmp(name, ctxt->name)) {
4072: nodePop(ctxt);
4073: oldname = htmlnamePop(ctxt);
4074: #ifdef DEBUG
4075: fprintf(stderr,
4076: "End of start tag problem: popping out %s\n", oldname);
4077: #endif
4078: if (oldname != NULL)
4079: xmlFree(oldname);
4080: }
4081:
4082: ctxt->instate = XML_PARSER_CONTENT;
4083: #ifdef DEBUG_PUSH
4084: fprintf(stderr, "HPP: entering CONTENT\n");
4085: #endif
4086: break;
4087: }
4088:
4089: /*
4090: * Check for an Empty Element from DTD definition
4091: */
4092: if ((info != NULL) && (info->empty)) {
4093: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4094: ctxt->sax->endElement(ctxt->userData, name);
4095: oldname = htmlnamePop(ctxt);
4096: #ifdef DEBUG
4097: fprintf(stderr,"End of empty tag %s : popping out %s\n", name, oldname);
4098: #endif
4099: if (oldname != NULL)
4100: xmlFree(oldname);
4101: }
4102: ctxt->instate = XML_PARSER_CONTENT;
4103: #ifdef DEBUG_PUSH
4104: fprintf(stderr, "HPP: entering CONTENT\n");
4105: #endif
4106: break;
4107: }
1.56 veillard 4108: case XML_PARSER_CONTENT: {
4109: long cons;
1.31 daniel 4110: /*
4111: * Handle preparsed entities and charRef
4112: */
4113: if (ctxt->token != 0) {
1.47 daniel 4114: xmlChar chr[2] = { 0 , 0 } ;
1.31 daniel 4115:
1.47 daniel 4116: chr[0] = (xmlChar) ctxt->token;
1.59 veillard 4117: htmlCheckParagraph(ctxt);
1.31 daniel 4118: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1.47 daniel 4119: ctxt->sax->characters(ctxt->userData, chr, 1);
1.31 daniel 4120: ctxt->token = 0;
4121: ctxt->checkIndex = 0;
4122: }
1.47 daniel 4123: if ((avail == 1) && (terminate)) {
4124: cur = in->cur[0];
4125: if ((cur != '<') && (cur != '&')) {
1.48 daniel 4126: if (ctxt->sax != NULL) {
4127: if (IS_BLANK(cur)) {
4128: if (ctxt->sax->ignorableWhitespace != NULL)
4129: ctxt->sax->ignorableWhitespace(
4130: ctxt->userData, &cur, 1);
4131: } else {
1.59 veillard 4132: htmlCheckParagraph(ctxt);
1.48 daniel 4133: if (ctxt->sax->characters != NULL)
4134: ctxt->sax->characters(
4135: ctxt->userData, &cur, 1);
4136: }
4137: }
1.47 daniel 4138: ctxt->token = 0;
4139: ctxt->checkIndex = 0;
4140: NEXT;
4141: }
4142: break;
4143: }
1.31 daniel 4144: if (avail < 2)
4145: goto done;
4146: cur = in->cur[0];
4147: next = in->cur[1];
1.56 veillard 4148: cons = ctxt->nbChars;
1.59 veillard 4149: /*
4150: * Sometimes DOCTYPE arrives in the middle of the document
4151: */
4152: if ((cur == '<') && (next == '!') &&
4153: (UPP(2) == 'D') && (UPP(3) == 'O') &&
4154: (UPP(4) == 'C') && (UPP(5) == 'T') &&
4155: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4156: (UPP(8) == 'E')) {
4157: if ((!terminate) &&
4158: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
4159: goto done;
4160: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4161: ctxt->sax->error(ctxt->userData,
4162: "Misplaced DOCTYPE declaration\n");
4163: ctxt->wellFormed = 0;
4164: htmlParseDocTypeDecl(ctxt);
4165: } else if ((cur == '<') && (next == '!') &&
1.31 daniel 4166: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 4167: if ((!terminate) &&
4168: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 4169: goto done;
4170: #ifdef DEBUG_PUSH
4171: fprintf(stderr, "HPP: Parsing Comment\n");
4172: #endif
4173: htmlParseComment(ctxt);
4174: ctxt->instate = XML_PARSER_CONTENT;
4175: } else if ((cur == '<') && (next == '!') && (avail < 4)) {
4176: goto done;
4177: } else if ((cur == '<') && (next == '/')) {
4178: ctxt->instate = XML_PARSER_END_TAG;
4179: ctxt->checkIndex = 0;
4180: #ifdef DEBUG_PUSH
4181: fprintf(stderr, "HPP: entering END_TAG\n");
4182: #endif
4183: break;
4184: } else if (cur == '<') {
4185: ctxt->instate = XML_PARSER_START_TAG;
4186: ctxt->checkIndex = 0;
4187: #ifdef DEBUG_PUSH
4188: fprintf(stderr, "HPP: entering START_TAG\n");
4189: #endif
4190: break;
4191: } else if (cur == '&') {
1.32 daniel 4192: if ((!terminate) &&
4193: (htmlParseLookupSequence(ctxt, ';', 0, 0) < 0))
1.31 daniel 4194: goto done;
4195: #ifdef DEBUG_PUSH
4196: fprintf(stderr, "HPP: Parsing Reference\n");
4197: #endif
4198: /* TODO: check generation of subtrees if noent !!! */
4199: htmlParseReference(ctxt);
4200: } else {
4201: /* TODO Avoid the extra copy, handle directly !!!!!! */
4202: /*
4203: * Goal of the following test is :
4204: * - minimize calls to the SAX 'character' callback
4205: * when they are mergeable
4206: */
4207: if ((ctxt->inputNr == 1) &&
4208: (avail < HTML_PARSER_BIG_BUFFER_SIZE)) {
1.32 daniel 4209: if ((!terminate) &&
4210: (htmlParseLookupSequence(ctxt, '<', 0, 0) < 0))
1.31 daniel 4211: goto done;
4212: }
4213: ctxt->checkIndex = 0;
4214: #ifdef DEBUG_PUSH
4215: fprintf(stderr, "HPP: Parsing char data\n");
4216: #endif
4217: htmlParseCharData(ctxt, 0);
4218: }
1.56 veillard 4219: if (cons == ctxt->nbChars) {
4220: if (ctxt->node != NULL) {
4221: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4222: ctxt->sax->error(ctxt->userData,
4223: "detected an error in element content\n");
4224: ctxt->wellFormed = 0;
4225: NEXT;
4226: }
4227: break;
4228: }
4229:
1.31 daniel 4230: break;
1.56 veillard 4231: }
1.31 daniel 4232: case XML_PARSER_END_TAG:
4233: if (avail < 2)
4234: goto done;
1.32 daniel 4235: if ((!terminate) &&
4236: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 4237: goto done;
4238: htmlParseEndTag(ctxt);
4239: if (ctxt->nameNr == 0) {
4240: ctxt->instate = XML_PARSER_EPILOG;
4241: } else {
4242: ctxt->instate = XML_PARSER_CONTENT;
4243: }
4244: ctxt->checkIndex = 0;
4245: #ifdef DEBUG_PUSH
4246: fprintf(stderr, "HPP: entering CONTENT\n");
4247: #endif
4248: break;
4249: case XML_PARSER_CDATA_SECTION:
4250: fprintf(stderr, "HPP: internal error, state == CDATA\n");
4251: ctxt->instate = XML_PARSER_CONTENT;
4252: ctxt->checkIndex = 0;
4253: #ifdef DEBUG_PUSH
4254: fprintf(stderr, "HPP: entering CONTENT\n");
4255: #endif
4256: break;
4257: case XML_PARSER_DTD:
4258: fprintf(stderr, "HPP: internal error, state == DTD\n");
4259: ctxt->instate = XML_PARSER_CONTENT;
4260: ctxt->checkIndex = 0;
4261: #ifdef DEBUG_PUSH
4262: fprintf(stderr, "HPP: entering CONTENT\n");
4263: #endif
4264: break;
4265: case XML_PARSER_COMMENT:
4266: fprintf(stderr, "HPP: internal error, state == COMMENT\n");
4267: ctxt->instate = XML_PARSER_CONTENT;
4268: ctxt->checkIndex = 0;
4269: #ifdef DEBUG_PUSH
4270: fprintf(stderr, "HPP: entering CONTENT\n");
4271: #endif
4272: break;
4273: case XML_PARSER_PI:
4274: fprintf(stderr, "HPP: internal error, state == PI\n");
4275: ctxt->instate = XML_PARSER_CONTENT;
4276: ctxt->checkIndex = 0;
4277: #ifdef DEBUG_PUSH
4278: fprintf(stderr, "HPP: entering CONTENT\n");
4279: #endif
4280: break;
4281: case XML_PARSER_ENTITY_DECL:
4282: fprintf(stderr, "HPP: internal error, state == ENTITY_DECL\n");
4283: ctxt->instate = XML_PARSER_CONTENT;
4284: ctxt->checkIndex = 0;
4285: #ifdef DEBUG_PUSH
4286: fprintf(stderr, "HPP: entering CONTENT\n");
4287: #endif
4288: break;
4289: case XML_PARSER_ENTITY_VALUE:
4290: fprintf(stderr, "HPP: internal error, state == ENTITY_VALUE\n");
4291: ctxt->instate = XML_PARSER_CONTENT;
4292: ctxt->checkIndex = 0;
4293: #ifdef DEBUG_PUSH
4294: fprintf(stderr, "HPP: entering DTD\n");
4295: #endif
4296: break;
4297: case XML_PARSER_ATTRIBUTE_VALUE:
4298: fprintf(stderr, "HPP: internal error, state == ATTRIBUTE_VALUE\n");
4299: ctxt->instate = XML_PARSER_START_TAG;
4300: ctxt->checkIndex = 0;
4301: #ifdef DEBUG_PUSH
4302: fprintf(stderr, "HPP: entering START_TAG\n");
1.53 veillard 4303: #endif
4304: break;
4305: case XML_PARSER_SYSTEM_LITERAL:
4306: fprintf(stderr, "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n");
4307: ctxt->instate = XML_PARSER_CONTENT;
4308: ctxt->checkIndex = 0;
4309: #ifdef DEBUG_PUSH
4310: fprintf(stderr, "HPP: entering CONTENT\n");
1.31 daniel 4311: #endif
4312: break;
4313: }
4314: }
4315: done:
1.47 daniel 4316: if ((avail == 0) && (terminate)) {
4317: htmlAutoClose(ctxt, NULL);
1.54 veillard 4318: if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
4319: /*
4320: * SAX: end of the document processing.
4321: */
1.47 daniel 4322: ctxt->instate = XML_PARSER_EOF;
1.54 veillard 4323: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4324: ctxt->sax->endDocument(ctxt->userData);
4325: }
1.59 veillard 4326: }
4327: if ((ctxt->myDoc != NULL) &&
4328: ((terminate) || (ctxt->instate == XML_PARSER_EOF) ||
4329: (ctxt->instate == XML_PARSER_EPILOG))) {
4330: xmlDtdPtr dtd;
4331: dtd = xmlGetIntSubset(ctxt->myDoc);
4332: if (dtd == NULL)
4333: ctxt->myDoc->intSubset =
4334: xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML",
4335: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
4336: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
1.47 daniel 4337: }
1.31 daniel 4338: #ifdef DEBUG_PUSH
4339: fprintf(stderr, "HPP: done %d\n", ret);
4340: #endif
4341: return(ret);
4342: }
4343:
4344: /**
1.32 daniel 4345: * htmlParseTry:
4346: * @ctxt: an HTML parser context
4347: *
4348: * Try to progress on parsing
4349: *
4350: * Returns zero if no parsing was possible
4351: */
4352: int
4353: htmlParseTry(htmlParserCtxtPtr ctxt) {
4354: return(htmlParseTryOrFinish(ctxt, 0));
4355: }
4356:
4357: /**
1.31 daniel 4358: * htmlParseChunk:
4359: * @ctxt: an XML parser context
4360: * @chunk: an char array
4361: * @size: the size in byte of the chunk
4362: * @terminate: last chunk indicator
4363: *
4364: * Parse a Chunk of memory
4365: *
4366: * Returns zero if no error, the xmlParserErrors otherwise.
4367: */
4368: int
4369: htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
4370: int terminate) {
4371: if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
4372: (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) {
4373: int base = ctxt->input->base - ctxt->input->buf->buffer->content;
4374: int cur = ctxt->input->cur - ctxt->input->base;
4375:
4376: xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
4377: ctxt->input->base = ctxt->input->buf->buffer->content + base;
4378: ctxt->input->cur = ctxt->input->base + cur;
4379: #ifdef DEBUG_PUSH
4380: fprintf(stderr, "HPP: pushed %d\n", size);
4381: #endif
4382:
1.34 daniel 4383: if ((terminate) || (ctxt->input->buf->buffer->use > 80))
4384: htmlParseTryOrFinish(ctxt, terminate);
1.60 veillard 4385: } else if (ctxt->instate != XML_PARSER_EOF) {
4386: xmlParserInputBufferPush(ctxt->input->buf, 0, "");
1.32 daniel 4387: htmlParseTryOrFinish(ctxt, terminate);
1.60 veillard 4388: }
1.31 daniel 4389: if (terminate) {
4390: if ((ctxt->instate != XML_PARSER_EOF) &&
4391: (ctxt->instate != XML_PARSER_EPILOG) &&
4392: (ctxt->instate != XML_PARSER_MISC)) {
1.67 veillard 4393: ctxt->errNo = XML_ERR_DOCUMENT_END;
1.31 daniel 4394: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4395: ctxt->sax->error(ctxt->userData,
4396: "Extra content at the end of the document\n");
4397: ctxt->wellFormed = 0;
4398: }
4399: if (ctxt->instate != XML_PARSER_EOF) {
4400: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4401: ctxt->sax->endDocument(ctxt->userData);
4402: }
4403: ctxt->instate = XML_PARSER_EOF;
4404: }
4405: return((xmlParserErrors) ctxt->errNo);
4406: }
4407:
4408: /************************************************************************
4409: * *
4410: * User entry points *
4411: * *
4412: ************************************************************************/
4413:
4414: /**
4415: * htmlCreatePushParserCtxt :
4416: * @sax: a SAX handler
4417: * @user_data: The user data returned on SAX callbacks
4418: * @chunk: a pointer to an array of chars
4419: * @size: number of chars in the array
4420: * @filename: an optional file name or URI
4421: * @enc: an optional encoding
4422: *
4423: * Create a parser context for using the HTML parser in push mode
4424: * To allow content encoding detection, @size should be >= 4
4425: * The value of @filename is used for fetching external entities
4426: * and error/warning reports.
4427: *
4428: * Returns the new parser context or NULL
4429: */
4430: htmlParserCtxtPtr
4431: htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data,
4432: const char *chunk, int size, const char *filename,
4433: xmlCharEncoding enc) {
4434: htmlParserCtxtPtr ctxt;
4435: htmlParserInputPtr inputStream;
4436: xmlParserInputBufferPtr buf;
4437:
4438: buf = xmlAllocParserInputBuffer(enc);
4439: if (buf == NULL) return(NULL);
4440:
4441: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
4442: if (ctxt == NULL) {
4443: xmlFree(buf);
4444: return(NULL);
4445: }
4446: memset(ctxt, 0, sizeof(htmlParserCtxt));
4447: htmlInitParserCtxt(ctxt);
4448: if (sax != NULL) {
4449: if (ctxt->sax != &htmlDefaultSAXHandler)
4450: xmlFree(ctxt->sax);
4451: ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler));
4452: if (ctxt->sax == NULL) {
4453: xmlFree(buf);
4454: xmlFree(ctxt);
4455: return(NULL);
4456: }
4457: memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler));
4458: if (user_data != NULL)
4459: ctxt->userData = user_data;
4460: }
4461: if (filename == NULL) {
4462: ctxt->directory = NULL;
4463: } else {
4464: ctxt->directory = xmlParserGetDirectory(filename);
4465: }
4466:
4467: inputStream = htmlNewInputStream(ctxt);
4468: if (inputStream == NULL) {
4469: xmlFreeParserCtxt(ctxt);
4470: return(NULL);
4471: }
4472:
4473: if (filename == NULL)
4474: inputStream->filename = NULL;
4475: else
4476: inputStream->filename = xmlMemStrdup(filename);
4477: inputStream->buf = buf;
4478: inputStream->base = inputStream->buf->buffer->content;
4479: inputStream->cur = inputStream->buf->buffer->content;
4480:
4481: inputPush(ctxt, inputStream);
4482:
4483: if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
4484: (ctxt->input->buf != NULL)) {
4485: xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
4486: #ifdef DEBUG_PUSH
4487: fprintf(stderr, "HPP: pushed %d\n", size);
4488: #endif
4489: }
4490:
4491: return(ctxt);
4492: }
1.1 daniel 4493:
4494: /**
4495: * htmlSAXParseDoc :
1.14 daniel 4496: * @cur: a pointer to an array of xmlChar
1.1 daniel 4497: * @encoding: a free form C string describing the HTML document encoding, or NULL
4498: * @sax: the SAX handler block
4499: * @userData: if using SAX, this pointer will be provided on callbacks.
4500: *
4501: * parse an HTML in-memory document and build a tree.
4502: * It use the given SAX function block to handle the parsing callback.
4503: * If sax is NULL, fallback to the default DOM tree building routines.
4504: *
4505: * Returns the resulting document tree
4506: */
4507:
4508: htmlDocPtr
1.14 daniel 4509: htmlSAXParseDoc(xmlChar *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) {
1.1 daniel 4510: htmlDocPtr ret;
4511: htmlParserCtxtPtr ctxt;
4512:
4513: if (cur == NULL) return(NULL);
4514:
4515:
4516: ctxt = htmlCreateDocParserCtxt(cur, encoding);
4517: if (ctxt == NULL) return(NULL);
4518: if (sax != NULL) {
4519: ctxt->sax = sax;
4520: ctxt->userData = userData;
4521: }
4522:
4523: htmlParseDocument(ctxt);
4524: ret = ctxt->myDoc;
4525: if (sax != NULL) {
4526: ctxt->sax = NULL;
4527: ctxt->userData = NULL;
4528: }
4529: htmlFreeParserCtxt(ctxt);
4530:
4531: return(ret);
4532: }
4533:
4534: /**
4535: * htmlParseDoc :
1.14 daniel 4536: * @cur: a pointer to an array of xmlChar
1.1 daniel 4537: * @encoding: a free form C string describing the HTML document encoding, or NULL
4538: *
4539: * parse an HTML in-memory document and build a tree.
4540: *
4541: * Returns the resulting document tree
4542: */
4543:
4544: htmlDocPtr
1.14 daniel 4545: htmlParseDoc(xmlChar *cur, const char *encoding) {
1.1 daniel 4546: return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
4547: }
4548:
4549:
4550: /**
4551: * htmlCreateFileParserCtxt :
4552: * @filename: the filename
4553: * @encoding: a free form C string describing the HTML document encoding, or NULL
4554: *
4555: * Create a parser context for a file content.
4556: * Automatic support for ZLIB/Compress compressed document is provided
4557: * by default if found at compile-time.
4558: *
4559: * Returns the new parser context or NULL
4560: */
4561: htmlParserCtxtPtr
4562: htmlCreateFileParserCtxt(const char *filename, const char *encoding)
4563: {
4564: htmlParserCtxtPtr ctxt;
4565: htmlParserInputPtr inputStream;
1.5 daniel 4566: xmlParserInputBufferPtr buf;
1.1 daniel 4567: /* htmlCharEncoding enc; */
4568:
1.5 daniel 4569: buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
4570: if (buf == NULL) return(NULL);
1.1 daniel 4571:
1.11 daniel 4572: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 4573: if (ctxt == NULL) {
4574: perror("malloc");
4575: return(NULL);
4576: }
1.19 daniel 4577: memset(ctxt, 0, sizeof(htmlParserCtxt));
1.1 daniel 4578: htmlInitParserCtxt(ctxt);
1.11 daniel 4579: inputStream = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 4580: if (inputStream == NULL) {
4581: perror("malloc");
1.11 daniel 4582: xmlFree(ctxt);
1.1 daniel 4583: return(NULL);
4584: }
1.19 daniel 4585: memset(inputStream, 0, sizeof(htmlParserInput));
1.1 daniel 4586:
1.11 daniel 4587: inputStream->filename = xmlMemStrdup(filename);
1.1 daniel 4588: inputStream->line = 1;
4589: inputStream->col = 1;
1.5 daniel 4590: inputStream->buf = buf;
1.21 daniel 4591: inputStream->directory = NULL;
1.1 daniel 4592:
1.5 daniel 4593: inputStream->base = inputStream->buf->buffer->content;
4594: inputStream->cur = inputStream->buf->buffer->content;
4595: inputStream->free = NULL;
1.1 daniel 4596:
4597: inputPush(ctxt, inputStream);
4598: return(ctxt);
4599: }
4600:
4601: /**
4602: * htmlSAXParseFile :
4603: * @filename: the filename
4604: * @encoding: a free form C string describing the HTML document encoding, or NULL
4605: * @sax: the SAX handler block
4606: * @userData: if using SAX, this pointer will be provided on callbacks.
4607: *
4608: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
4609: * compressed document is provided by default if found at compile-time.
4610: * It use the given SAX function block to handle the parsing callback.
4611: * If sax is NULL, fallback to the default DOM tree building routines.
4612: *
4613: * Returns the resulting document tree
4614: */
4615:
4616: htmlDocPtr
4617: htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
4618: void *userData) {
4619: htmlDocPtr ret;
4620: htmlParserCtxtPtr ctxt;
1.57 veillard 4621: htmlSAXHandlerPtr oldsax = NULL;
1.1 daniel 4622:
4623: ctxt = htmlCreateFileParserCtxt(filename, encoding);
4624: if (ctxt == NULL) return(NULL);
4625: if (sax != NULL) {
1.55 veillard 4626: oldsax = ctxt->sax;
1.1 daniel 4627: ctxt->sax = sax;
4628: ctxt->userData = userData;
4629: }
4630:
4631: htmlParseDocument(ctxt);
4632:
4633: ret = ctxt->myDoc;
4634: if (sax != NULL) {
1.55 veillard 4635: ctxt->sax = oldsax;
1.1 daniel 4636: ctxt->userData = NULL;
4637: }
4638: htmlFreeParserCtxt(ctxt);
4639:
4640: return(ret);
4641: }
4642:
4643: /**
4644: * htmlParseFile :
4645: * @filename: the filename
4646: * @encoding: a free form C string describing the HTML document encoding, or NULL
4647: *
4648: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
4649: * compressed document is provided by default if found at compile-time.
4650: *
4651: * Returns the resulting document tree
4652: */
4653:
4654: htmlDocPtr
4655: htmlParseFile(const char *filename, const char *encoding) {
4656: return(htmlSAXParseFile(filename, encoding, NULL, NULL));
4657: }
1.39 daniel 4658:
4659: #endif /* LIBXML_HTML_ENABLED */
Webmaster