Annotation of XML/HTMLparser.c, revision 1.73
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.
1.72 veillard 582: * This is not reentrant. Call xmlInitParser() once before processing in
583: * case of use in multithreaded programs.
1.1 daniel 584: */
585: void
586: htmlInitAutoClose(void) {
587: int index, i = 0;
588:
589: if (htmlStartCloseIndexinitialized) return;
590:
591: for (index = 0;index < 100;index ++) htmlStartCloseIndex[index] = NULL;
592: index = 0;
593: while ((htmlStartClose[i] != NULL) && (index < 100 - 1)) {
594: htmlStartCloseIndex[index++] = &htmlStartClose[i];
595: while (htmlStartClose[i] != NULL) i++;
596: i++;
597: }
1.72 veillard 598: htmlStartCloseIndexinitialized = 1;
1.1 daniel 599: }
600:
601: /**
602: * htmlTagLookup:
1.69 veillard 603: * @tag: The tag name in lowercase
1.1 daniel 604: *
605: * Lookup the HTML tag in the ElementTable
606: *
607: * Returns the related htmlElemDescPtr or NULL if not found.
608: */
609: htmlElemDescPtr
1.14 daniel 610: htmlTagLookup(const xmlChar *tag) {
1.61 veillard 611: int i;
1.1 daniel 612:
613: for (i = 0; i < (sizeof(html40ElementTable) /
614: sizeof(html40ElementTable[0]));i++) {
1.73 ! veillard 615: if (xmlStrEqual(tag, BAD_CAST html40ElementTable[i].name))
1.1 daniel 616: return(&html40ElementTable[i]);
617: }
618: return(NULL);
619: }
620:
621: /**
622: * htmlCheckAutoClose:
1.50 veillard 623: * @newtag: The new tag name
624: * @oldtag: The old tag name
1.1 daniel 625: *
626: * Checks wether the new tag is one of the registered valid tags for closing old.
627: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
628: *
629: * Returns 0 if no, 1 if yes.
630: */
631: int
1.50 veillard 632: htmlCheckAutoClose(const xmlChar *newtag, const xmlChar *oldtag) {
1.1 daniel 633: int i, index;
1.64 veillard 634: char **close = NULL;
1.1 daniel 635:
636: if (htmlStartCloseIndexinitialized == 0) htmlInitAutoClose();
637:
638: /* inefficient, but not a big deal */
639: for (index = 0; index < 100;index++) {
640: close = htmlStartCloseIndex[index];
641: if (close == NULL) return(0);
1.73 ! veillard 642: if (xmlStrEqual(BAD_CAST *close, newtag)) break;
1.1 daniel 643: }
644:
645: i = close - htmlStartClose;
646: i++;
647: while (htmlStartClose[i] != NULL) {
1.73 ! veillard 648: if (xmlStrEqual(BAD_CAST htmlStartClose[i], oldtag)) {
1.1 daniel 649: return(1);
650: }
651: i++;
652: }
653: return(0);
654: }
655:
656: /**
1.50 veillard 657: * htmlAutoCloseOnClose:
658: * @ctxt: an HTML parser context
659: * @newtag: The new tag name
660: *
661: * The HTmL DtD allows an ending tag to implicitely close other tags.
662: */
663: void
664: htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
665: htmlElemDescPtr info;
666: xmlChar *oldname;
667: int i;
668:
669: #ifdef DEBUG
670: fprintf(stderr,"Close of %s stack: %d elements\n", newtag, ctxt->nameNr);
671: for (i = 0;i < ctxt->nameNr;i++)
672: fprintf(stderr,"%d : %s\n", i, ctxt->nameTab[i]);
673: #endif
674:
675: for (i = (ctxt->nameNr - 1);i >= 0;i--) {
1.73 ! veillard 676: if (xmlStrEqual(newtag, ctxt->nameTab[i])) break;
1.50 veillard 677: }
678: if (i < 0) return;
679:
1.73 ! veillard 680: while (!xmlStrEqual(newtag, ctxt->name)) {
1.50 veillard 681: info = htmlTagLookup(ctxt->name);
682: if ((info == NULL) || (info->endTag == 1)) {
683: #ifdef DEBUG
684: fprintf(stderr,"htmlAutoCloseOnClose: %s closes %s\n", newtag, ctxt->name);
685: #endif
686: } else {
687: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
688: ctxt->sax->error(ctxt->userData,
689: "Opening and ending tag mismatch: %s and %s\n",
690: newtag, ctxt->name);
691: ctxt->wellFormed = 0;
692: }
693: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
694: ctxt->sax->endElement(ctxt->userData, ctxt->name);
695: oldname = htmlnamePop(ctxt);
696: if (oldname != NULL) {
697: #ifdef DEBUG
698: fprintf(stderr,"htmlAutoCloseOnClose: popped %s\n", oldname);
699: #endif
700: xmlFree(oldname);
701: }
702: }
703: }
704:
705: /**
1.1 daniel 706: * htmlAutoClose:
707: * @ctxt: an HTML parser context
1.50 veillard 708: * @newtag: The new tag name or NULL
1.1 daniel 709: *
710: * The HTmL DtD allows a tag to implicitely close other tags.
711: * The list is kept in htmlStartClose array. This function is
712: * called when a new tag has been detected and generates the
713: * appropriates closes if possible/needed.
1.50 veillard 714: * If newtag is NULL this mean we are at the end of the resource
1.47 daniel 715: * and we should check
1.1 daniel 716: */
717: void
1.50 veillard 718: htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
1.15 daniel 719: xmlChar *oldname;
1.50 veillard 720: while ((newtag != NULL) && (ctxt->name != NULL) &&
721: (htmlCheckAutoClose(newtag, ctxt->name))) {
1.1 daniel 722: #ifdef DEBUG
1.50 veillard 723: fprintf(stderr,"htmlAutoClose: %s closes %s\n", newtag, ctxt->name);
1.1 daniel 724: #endif
725: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1.15 daniel 726: ctxt->sax->endElement(ctxt->userData, ctxt->name);
1.24 daniel 727: oldname = htmlnamePop(ctxt);
1.18 daniel 728: if (oldname != NULL) {
729: #ifdef DEBUG
730: fprintf(stderr,"htmlAutoClose: popped %s\n", oldname);
731: #endif
1.17 daniel 732: xmlFree(oldname);
1.18 daniel 733: }
1.1 daniel 734: }
1.50 veillard 735: if (newtag == NULL) {
1.49 daniel 736: htmlAutoCloseOnClose(ctxt, BAD_CAST"head");
737: htmlAutoCloseOnClose(ctxt, BAD_CAST"body");
738: htmlAutoCloseOnClose(ctxt, BAD_CAST"html");
739: }
1.50 veillard 740: while ((newtag == NULL) && (ctxt->name != NULL) &&
1.73 ! veillard 741: ((xmlStrEqual(ctxt->name, BAD_CAST"head")) ||
! 742: (xmlStrEqual(ctxt->name, BAD_CAST"body")) ||
! 743: (xmlStrEqual(ctxt->name, BAD_CAST"html")))) {
1.47 daniel 744: #ifdef DEBUG
745: fprintf(stderr,"htmlAutoClose: EOF closes %s\n", ctxt->name);
746: #endif
747: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
748: ctxt->sax->endElement(ctxt->userData, ctxt->name);
749: oldname = htmlnamePop(ctxt);
750: if (oldname != NULL) {
751: #ifdef DEBUG
752: fprintf(stderr,"htmlAutoClose: popped %s\n", oldname);
753: #endif
754: xmlFree(oldname);
755: }
756: }
757:
1.1 daniel 758: }
759:
760: /**
1.28 daniel 761: * htmlAutoCloseTag:
762: * @doc: the HTML document
763: * @name: The tag name
764: * @elem: the HTML element
765: *
766: * The HTmL DtD allows a tag to implicitely close other tags.
767: * The list is kept in htmlStartClose array. This function checks
768: * if the element or one of it's children would autoclose the
769: * given tag.
770: *
771: * Returns 1 if autoclose, 0 otherwise
772: */
773: int
774: htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) {
775: htmlNodePtr child;
776:
777: if (elem == NULL) return(1);
1.73 ! veillard 778: if (xmlStrEqual(name, elem->name)) return(0);
1.28 daniel 779: if (htmlCheckAutoClose(elem->name, name)) return(1);
1.37 daniel 780: child = elem->children;
1.28 daniel 781: while (child != NULL) {
782: if (htmlAutoCloseTag(doc, name, child)) return(1);
783: child = child->next;
784: }
785: return(0);
786: }
787:
788: /**
789: * htmlIsAutoClosed:
790: * @doc: the HTML document
791: * @elem: the HTML element
792: *
793: * The HTmL DtD allows a tag to implicitely close other tags.
794: * The list is kept in htmlStartClose array. This function checks
795: * if a tag is autoclosed by one of it's child
796: *
797: * Returns 1 if autoclosed, 0 otherwise
798: */
799: int
800: htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) {
801: htmlNodePtr child;
802:
803: if (elem == NULL) return(1);
1.37 daniel 804: child = elem->children;
1.28 daniel 805: while (child != NULL) {
806: if (htmlAutoCloseTag(doc, elem->name, child)) return(1);
807: child = child->next;
808: }
809: return(0);
810: }
811:
812: /**
1.43 daniel 813: * htmlCheckImplied:
814: * @ctxt: an HTML parser context
1.50 veillard 815: * @newtag: The new tag name
1.43 daniel 816: *
817: * The HTmL DtD allows a tag to exists only implicitely
818: * called when a new tag has been detected and generates the
819: * appropriates implicit tags if missing
820: */
821: void
1.50 veillard 822: htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
1.73 ! veillard 823: if (xmlStrEqual(newtag, BAD_CAST"html"))
1.43 daniel 824: return;
825: if (ctxt->nameNr <= 0) {
826: #ifdef DEBUG
827: fprintf(stderr,"Implied element html: pushed html\n");
828: #endif
829: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"html"));
830: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
831: ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL);
832: }
1.73 ! veillard 833: if ((xmlStrEqual(newtag, BAD_CAST"body")) || (xmlStrEqual(newtag, BAD_CAST"head")))
1.43 daniel 834: return;
835: if (ctxt->nameNr <= 1) {
1.73 ! veillard 836: if ((xmlStrEqual(newtag, BAD_CAST"script")) ||
! 837: (xmlStrEqual(newtag, BAD_CAST"style")) ||
! 838: (xmlStrEqual(newtag, BAD_CAST"meta")) ||
! 839: (xmlStrEqual(newtag, BAD_CAST"link")) ||
! 840: (xmlStrEqual(newtag, BAD_CAST"title")) ||
! 841: (xmlStrEqual(newtag, BAD_CAST"base"))) {
1.43 daniel 842: /*
843: * dropped OBJECT ... i you put it first BODY will be
844: * assumed !
845: */
846: #ifdef DEBUG
847: fprintf(stderr,"Implied element head: pushed head\n");
848: #endif
849: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"head"));
850: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
851: ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL);
852: } else {
853: #ifdef DEBUG
854: fprintf(stderr,"Implied element body: pushed body\n");
855: #endif
856: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"body"));
857: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
858: ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL);
859: }
860: }
861: }
862:
1.59 veillard 863: /**
864: * htmlCheckParagraph
865: * @ctxt: an HTML parser context
866: *
867: * Check whether a p element need to be implied before inserting
868: * characters in the current element.
869: *
870: * Returns 1 if a paragraph has been inserted, 0 if not and -1
871: * in case of error.
872: */
873:
874: int
875: htmlCheckParagraph(htmlParserCtxtPtr ctxt) {
876: const xmlChar *tag;
877: int i;
878:
879: if (ctxt == NULL)
880: return(-1);
881: tag = ctxt->name;
882: if (tag == NULL) {
883: htmlAutoClose(ctxt, BAD_CAST"p");
884: htmlCheckImplied(ctxt, BAD_CAST"p");
885: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p"));
886: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
887: ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
888: return(1);
889: }
890: for (i = 0; htmlNoContentElements[i] != NULL; i++) {
1.73 ! veillard 891: if (xmlStrEqual(tag, BAD_CAST htmlNoContentElements[i])) {
1.59 veillard 892: #ifdef DEBUG
893: fprintf(stderr,"Implied element paragraph\n");
894: #endif
895: htmlAutoClose(ctxt, BAD_CAST"p");
896: htmlCheckImplied(ctxt, BAD_CAST"p");
897: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p"));
898: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
899: ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
900: return(1);
901: }
902: }
903: return(0);
904: }
905:
1.1 daniel 906: /************************************************************************
907: * *
908: * The list of HTML predefined entities *
909: * *
910: ************************************************************************/
911:
912:
913: htmlEntityDesc html40EntitiesTable[] = {
914: /*
1.61 veillard 915: * the 4 absolute ones, plus apostrophe.
1.1 daniel 916: */
917: { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
918: { 38, "amp", "ampersand, U+0026 ISOnum" },
1.61 veillard 919: { 39, "apos", "single quote" },
1.1 daniel 920: { 60, "lt", "less-than sign, U+003C ISOnum" },
921: { 62, "gt", "greater-than sign, U+003E ISOnum" },
922:
923: /*
924: * A bunch still in the 128-255 range
925: * Replacing them depend really on the charset used.
926: */
927: { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
928: { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
929: { 162, "cent", "cent sign, U+00A2 ISOnum" },
930: { 163, "pound","pound sign, U+00A3 ISOnum" },
931: { 164, "curren","currency sign, U+00A4 ISOnum" },
932: { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
933: { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
934: { 167, "sect", "section sign, U+00A7 ISOnum" },
935: { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
936: { 169, "copy", "copyright sign, U+00A9 ISOnum" },
937: { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
938: { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
939: { 172, "not", "not sign, U+00AC ISOnum" },
940: { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
941: { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
942: { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
943: { 176, "deg", "degree sign, U+00B0 ISOnum" },
944: { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
945: { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
946: { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
947: { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
948: { 181, "micro","micro sign, U+00B5 ISOnum" },
949: { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1.7 daniel 950: { 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1.1 daniel 951: { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
952: { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
953: { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1.7 daniel 954: { 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1.1 daniel 955: { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
956: { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
957: { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
958: { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
959: { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
960: { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
961: { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
962: { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
963: { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
964: { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
965: { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
966: { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
967: { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
968: { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
969: { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
970: { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
971: { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
972: { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
973: { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
974: { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
975: { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
976: { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
977: { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
978: { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
979: { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
980: { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
981: { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
982: { 215, "times","multiplication sign, U+00D7 ISOnum" },
1.7 daniel 983: { 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1.1 daniel 984: { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
985: { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
986: { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
987: { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
988: { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
989: { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
990: { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
991: { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
992: { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
993: { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
994: { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
995: { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
996: { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
997: { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
998: { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
999: { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
1000: { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
1001: { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
1002: { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
1003: { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
1004: { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
1005: { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
1006: { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
1007: { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
1008: { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
1009: { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
1010: { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
1011: { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
1012: { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
1013: { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
1014: { 247, "divide","division sign, U+00F7 ISOnum" },
1015: { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
1016: { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
1017: { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
1018: { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
1019: { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
1020: { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
1021: { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
1022: { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
1023:
1.61 veillard 1024: { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
1025: { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
1026: { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
1027: { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
1028: { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
1029:
1.1 daniel 1030: /*
1031: * Anything below should really be kept as entities references
1032: */
1033: { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
1034:
1.61 veillard 1035: { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
1036: { 732, "tilde","small tilde, U+02DC ISOdia" },
1037:
1.1 daniel 1038: { 913, "Alpha","greek capital letter alpha, U+0391" },
1039: { 914, "Beta", "greek capital letter beta, U+0392" },
1040: { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
1041: { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
1042: { 917, "Epsilon","greek capital letter epsilon, U+0395" },
1043: { 918, "Zeta", "greek capital letter zeta, U+0396" },
1044: { 919, "Eta", "greek capital letter eta, U+0397" },
1045: { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
1046: { 921, "Iota", "greek capital letter iota, U+0399" },
1047: { 922, "Kappa","greek capital letter kappa, U+039A" },
1048: { 923, "Lambda""greek capital letter lambda, U+039B ISOgrk3" },
1049: { 924, "Mu", "greek capital letter mu, U+039C" },
1050: { 925, "Nu", "greek capital letter nu, U+039D" },
1051: { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
1052: { 927, "Omicron","greek capital letter omicron, U+039F" },
1053: { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
1054: { 929, "Rho", "greek capital letter rho, U+03A1" },
1055: { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
1056: { 932, "Tau", "greek capital letter tau, U+03A4" },
1057: { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
1058: { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
1059: { 935, "Chi", "greek capital letter chi, U+03A7" },
1060: { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
1061: { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
1062:
1063: { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
1064: { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
1065: { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
1066: { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
1067: { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
1068: { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
1069: { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
1070: { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
1071: { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
1072: { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
1073: { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
1074: { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
1075: { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
1076: { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
1077: { 959, "omicron","greek small letter omicron, U+03BF NEW" },
1078: { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
1079: { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
1080: { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
1081: { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
1082: { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
1083: { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
1084: { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
1085: { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
1086: { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
1087: { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
1088: { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
1089: { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
1090: { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
1091:
1.61 veillard 1092: { 8194, "ensp", "en space, U+2002 ISOpub" },
1093: { 8195, "emsp", "em space, U+2003 ISOpub" },
1094: { 8201, "thinsp","thin space, U+2009 ISOpub" },
1095: { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
1096: { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
1097: { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
1098: { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
1099: { 8211, "ndash","en dash, U+2013 ISOpub" },
1100: { 8212, "mdash","em dash, U+2014 ISOpub" },
1101: { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
1102: { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
1103: { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
1104: { 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
1105: { 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
1106: { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
1107: { 8224, "dagger","dagger, U+2020 ISOpub" },
1108: { 8225, "Dagger","double dagger, U+2021 ISOpub" },
1109:
1.1 daniel 1110: { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
1111: { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
1.61 veillard 1112:
1113: { 8240, "permil","per mille sign, U+2030 ISOtech" },
1114:
1.1 daniel 1115: { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
1116: { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
1.61 veillard 1117:
1118: { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
1119: { 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
1120:
1.1 daniel 1121: { 8254, "oline","overline = spacing overscore, U+203E NEW" },
1122: { 8260, "frasl","fraction slash, U+2044 NEW" },
1123:
1.61 veillard 1124: { 8364, "euro", "euro sign, U+20AC NEW" },
1125:
1126: { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
1.7 daniel 1127: { 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
1.1 daniel 1128: { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
1129: { 8482, "trade","trade mark sign, U+2122 ISOnum" },
1130: { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
1131: { 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
1132: { 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
1133: { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
1134: { 8595, "darr", "downwards arrow, U+2193 ISOnum" },
1135: { 8596, "harr", "left right arrow, U+2194 ISOamsa" },
1136: { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
1137: { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
1138: { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
1139: { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
1140: { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
1141: { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
1142:
1143: { 8704, "forall","for all, U+2200 ISOtech" },
1144: { 8706, "part", "partial differential, U+2202 ISOtech" },
1145: { 8707, "exist","there exists, U+2203 ISOtech" },
1146: { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
1147: { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
1148: { 8712, "isin", "element of, U+2208 ISOtech" },
1149: { 8713, "notin","not an element of, U+2209 ISOtech" },
1150: { 8715, "ni", "contains as member, U+220B ISOtech" },
1151: { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
1152: { 8721, "sum", "n-ary sumation, U+2211 ISOamsb" },
1153: { 8722, "minus","minus sign, U+2212 ISOtech" },
1154: { 8727, "lowast","asterisk operator, U+2217 ISOtech" },
1155: { 8730, "radic","square root = radical sign, U+221A ISOtech" },
1156: { 8733, "prop", "proportional to, U+221D ISOtech" },
1157: { 8734, "infin","infinity, U+221E ISOtech" },
1158: { 8736, "ang", "angle, U+2220 ISOamso" },
1159: { 8743, "and", "logical and = wedge, U+2227 ISOtech" },
1160: { 8744, "or", "logical or = vee, U+2228 ISOtech" },
1161: { 8745, "cap", "intersection = cap, U+2229 ISOtech" },
1162: { 8746, "cup", "union = cup, U+222A ISOtech" },
1163: { 8747, "int", "integral, U+222B ISOtech" },
1164: { 8756, "there4","therefore, U+2234 ISOtech" },
1165: { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
1166: { 8773, "cong", "approximately equal to, U+2245 ISOtech" },
1167: { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
1168: { 8800, "ne", "not equal to, U+2260 ISOtech" },
1169: { 8801, "equiv","identical to, U+2261 ISOtech" },
1170: { 8804, "le", "less-than or equal to, U+2264 ISOtech" },
1171: { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
1172: { 8834, "sub", "subset of, U+2282 ISOtech" },
1173: { 8835, "sup", "superset of, U+2283 ISOtech" },
1174: { 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
1175: { 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
1176: { 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
1177: { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
1178: { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
1179: { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
1180: { 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
1181: { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
1182: { 8969, "rceil","right ceiling, U+2309 ISOamsc" },
1183: { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
1184: { 8971, "rfloor","right floor, U+230B ISOamsc" },
1185: { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
1186: { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
1187: { 9674, "loz", "lozenge, U+25CA ISOpub" },
1188:
1189: { 9824, "spades","black spade suit, U+2660 ISOpub" },
1190: { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
1191: { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
1192: { 9830, "diams","black diamond suit, U+2666 ISOpub" },
1193:
1194: };
1195:
1196: /************************************************************************
1197: * *
1198: * Commodity functions to handle entities *
1199: * *
1200: ************************************************************************/
1201:
1202: /*
1203: * Macro used to grow the current buffer.
1204: */
1205: #define growBuffer(buffer) { \
1206: buffer##_size *= 2; \
1.14 daniel 1207: buffer = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
1.1 daniel 1208: if (buffer == NULL) { \
1209: perror("realloc failed"); \
1.33 daniel 1210: return(NULL); \
1.1 daniel 1211: } \
1212: }
1213:
1214: /**
1215: * htmlEntityLookup:
1216: * @name: the entity name
1217: *
1218: * Lookup the given entity in EntitiesTable
1219: *
1220: * TODO: the linear scan is really ugly, an hash table is really needed.
1221: *
1222: * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1223: */
1224: htmlEntityDescPtr
1.14 daniel 1225: htmlEntityLookup(const xmlChar *name) {
1.1 daniel 1226: int i;
1227:
1228: for (i = 0;i < (sizeof(html40EntitiesTable)/
1229: sizeof(html40EntitiesTable[0]));i++) {
1.73 ! veillard 1230: if (xmlStrEqual(name, BAD_CAST html40EntitiesTable[i].name)) {
1.1 daniel 1231: #ifdef DEBUG
1.18 daniel 1232: fprintf(stderr,"Found entity %s\n", name);
1.1 daniel 1233: #endif
1234: return(&html40EntitiesTable[i]);
1235: }
1236: }
1237: return(NULL);
1238: }
1239:
1.53 veillard 1240: /**
1.61 veillard 1241: * htmlEntityValueLookup:
1242: * @value: the entity's unicode value
1243: *
1244: * Lookup the given entity in EntitiesTable
1245: *
1246: * TODO: the linear scan is really ugly, an hash table is really needed.
1247: *
1248: * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1249: */
1250: htmlEntityDescPtr
1251: htmlEntityValueLookup(int value) {
1252: int i;
1253: #ifdef DEBUG
1254: int lv = 0;
1255: #endif
1256:
1257: for (i = 0;i < (sizeof(html40EntitiesTable)/
1258: sizeof(html40EntitiesTable[0]));i++) {
1.71 veillard 1259: if ((unsigned int) html40EntitiesTable[i].value >= value) {
1260: if ((unsigned int) html40EntitiesTable[i].value > value)
1.61 veillard 1261: break;
1262: #ifdef DEBUG
1263: fprintf(stderr,"Found entity %s\n", html40EntitiesTable[i].name);
1264: #endif
1265: return(&html40EntitiesTable[i]);
1266: }
1267: #ifdef DEBUG
1268: if (lv > html40EntitiesTable[i].value) {
1269: fprintf(stderr, "html40EntitiesTable[] is not sorted (%d > %d)!\n",
1270: lv, html40EntitiesTable[i].value);
1271: }
1272: lv = html40EntitiesTable[i].value;
1273: #endif
1274: }
1275: return(NULL);
1276: }
1277:
1278: /**
1.53 veillard 1279: * UTF8ToHtml:
1280: * @out: a pointer to an array of bytes to store the result
1281: * @outlen: the length of @out
1282: * @in: a pointer to an array of UTF-8 chars
1283: * @inlen: the length of @in
1284: *
1285: * Take a block of UTF-8 chars in and try to convert it to an ASCII
1286: * plus HTML entities block of chars out.
1287: *
1288: * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1289: * The value of @inlen after return is the number of octets consumed
1290: * as the return value is positive, else unpredictiable.
1291: * The value of @outlen after return is the number of octets consumed.
1292: */
1293: int
1294: UTF8ToHtml(unsigned char* out, int *outlen,
1295: const unsigned char* in, int *inlen) {
1296: const unsigned char* processed = in;
1297: const unsigned char* outend;
1298: const unsigned char* outstart = out;
1299: const unsigned char* instart = in;
1300: const unsigned char* inend;
1301: unsigned int c, d;
1302: int trailing;
1303:
1304: if (in == NULL) {
1305: /*
1306: * initialization nothing to do
1307: */
1308: *outlen = 0;
1309: *inlen = 0;
1310: return(0);
1311: }
1312: inend = in + (*inlen);
1313: outend = out + (*outlen);
1314: while (in < inend) {
1315: d = *in++;
1316: if (d < 0x80) { c= d; trailing= 0; }
1317: else if (d < 0xC0) {
1318: /* trailing byte in leading position */
1319: *outlen = out - outstart;
1320: *inlen = processed - instart;
1321: return(-2);
1322: } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
1323: else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
1324: else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
1325: else {
1326: /* no chance for this in Ascii */
1327: *outlen = out - outstart;
1328: *inlen = processed - instart;
1329: return(-2);
1330: }
1331:
1332: if (inend - in < trailing) {
1333: break;
1334: }
1335:
1336: for ( ; trailing; trailing--) {
1337: if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
1338: break;
1339: c <<= 6;
1340: c |= d & 0x3F;
1341: }
1342:
1343: /* assertion: c is a single UTF-4 value */
1344: if (c < 0x80) {
1.62 veillard 1345: if (out + 1 >= outend)
1.53 veillard 1346: break;
1347: *out++ = c;
1348: } else {
1.61 veillard 1349: int len;
1350: htmlEntityDescPtr ent;
1351:
1.53 veillard 1352: /*
1353: * Try to lookup a predefined HTML entity for it
1354: */
1355:
1.61 veillard 1356: ent = htmlEntityValueLookup(c);
1357: if (ent == NULL) {
1358: /* no chance for this in Ascii */
1359: *outlen = out - outstart;
1360: *inlen = processed - instart;
1361: return(-2);
1.53 veillard 1362: }
1.61 veillard 1363: len = strlen(ent->name);
1.62 veillard 1364: if (out + 2 + len >= outend)
1.53 veillard 1365: break;
1366: *out++ = '&';
1.61 veillard 1367: memcpy(out, ent->name, len);
1368: out += len;
1.53 veillard 1369: *out++ = ';';
1370: }
1371: processed = in;
1372: }
1373: *outlen = out - outstart;
1374: *inlen = processed - instart;
1375: return(0);
1376: }
1377:
1.62 veillard 1378: /**
1379: * htmlEncodeEntities:
1380: * @out: a pointer to an array of bytes to store the result
1381: * @outlen: the length of @out
1382: * @in: a pointer to an array of UTF-8 chars
1383: * @inlen: the length of @in
1384: * @quoteChar: the quote character to escape (' or ") or zero.
1385: *
1386: * Take a block of UTF-8 chars in and try to convert it to an ASCII
1387: * plus HTML entities block of chars out.
1388: *
1389: * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1390: * The value of @inlen after return is the number of octets consumed
1391: * as the return value is positive, else unpredictiable.
1392: * The value of @outlen after return is the number of octets consumed.
1393: */
1394: int
1395: htmlEncodeEntities(unsigned char* out, int *outlen,
1396: const unsigned char* in, int *inlen, int quoteChar) {
1397: const unsigned char* processed = in;
1398: const unsigned char* outend = out + (*outlen);
1399: const unsigned char* outstart = out;
1400: const unsigned char* instart = in;
1401: const unsigned char* inend = in + (*inlen);
1402: unsigned int c, d;
1403: int trailing;
1404:
1405: while (in < inend) {
1406: d = *in++;
1407: if (d < 0x80) { c= d; trailing= 0; }
1408: else if (d < 0xC0) {
1409: /* trailing byte in leading position */
1410: *outlen = out - outstart;
1411: *inlen = processed - instart;
1412: return(-2);
1413: } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
1414: else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
1415: else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
1416: else {
1417: /* no chance for this in Ascii */
1418: *outlen = out - outstart;
1419: *inlen = processed - instart;
1420: return(-2);
1421: }
1422:
1423: if (inend - in < trailing)
1424: break;
1425:
1426: while (trailing--) {
1427: if (((d= *in++) & 0xC0) != 0x80) {
1428: *outlen = out - outstart;
1429: *inlen = processed - instart;
1430: return(-2);
1431: }
1432: c <<= 6;
1433: c |= d & 0x3F;
1434: }
1435:
1436: /* assertion: c is a single UTF-4 value */
1437: if (c < 0x80 && c != quoteChar && c != '&' && c != '<' && c != '>') {
1438: if (out >= outend)
1439: break;
1440: *out++ = c;
1441: } else {
1442: htmlEntityDescPtr ent;
1443: const char *cp;
1444: char nbuf[16];
1445: int len;
1446:
1447: /*
1448: * Try to lookup a predefined HTML entity for it
1449: */
1450: ent = htmlEntityValueLookup(c);
1451: if (ent == NULL) {
1452: sprintf(nbuf, "#%u", c);
1453: cp = nbuf;
1454: }
1455: else
1456: cp = ent->name;
1457: len = strlen(cp);
1458: if (out + 2 + len > outend)
1459: break;
1460: *out++ = '&';
1461: memcpy(out, cp, len);
1462: out += len;
1463: *out++ = ';';
1464: }
1465: processed = in;
1466: }
1467: *outlen = out - outstart;
1468: *inlen = processed - instart;
1469: return(0);
1470: }
1.1 daniel 1471:
1472: /**
1473: * htmlDecodeEntities:
1474: * @ctxt: the parser context
1475: * @len: the len to decode (in bytes !), -1 for no size limit
1.14 daniel 1476: * @end: an end marker xmlChar, 0 if none
1477: * @end2: an end marker xmlChar, 0 if none
1478: * @end3: an end marker xmlChar, 0 if none
1.1 daniel 1479: *
1480: * Subtitute the HTML entities by their value
1481: *
1.19 daniel 1482: * DEPRECATED !!!!
1.1 daniel 1483: *
1484: * Returns A newly allocated string with the substitution done. The caller
1485: * must deallocate it !
1486: */
1.14 daniel 1487: xmlChar *
1.1 daniel 1488: htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
1.14 daniel 1489: xmlChar end, xmlChar end2, xmlChar end3) {
1.53 veillard 1490: xmlChar *name = NULL;
1.14 daniel 1491: xmlChar *buffer = NULL;
1.53 veillard 1492: unsigned int buffer_size = 0;
1493: unsigned int nbchars = 0;
1.1 daniel 1494: htmlEntityDescPtr ent;
1495: unsigned int max = (unsigned int) len;
1.53 veillard 1496: int c,l;
1497:
1498: if (ctxt->depth > 40) {
1.67 veillard 1499: ctxt->errNo = XML_ERR_ENTITY_LOOP;
1.53 veillard 1500: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1501: ctxt->sax->error(ctxt->userData,
1502: "Detected entity reference loop\n");
1503: ctxt->wellFormed = 0;
1504: ctxt->disableSAX = 1;
1505: return(NULL);
1506: }
1.1 daniel 1507:
1508: /*
1509: * allocate a translation buffer.
1510: */
1.31 daniel 1511: buffer_size = HTML_PARSER_BIG_BUFFER_SIZE;
1.14 daniel 1512: buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
1.1 daniel 1513: if (buffer == NULL) {
1.53 veillard 1514: perror("xmlDecodeEntities: malloc failed");
1.1 daniel 1515: return(NULL);
1516: }
1517:
1518: /*
1519: * Ok loop until we reach one of the ending char or a size limit.
1520: */
1.53 veillard 1521: c = CUR_CHAR(l);
1522: while ((nbchars < max) && (c != end) &&
1523: (c != end2) && (c != end3)) {
1524:
1525: if (c == 0) break;
1526: if (((c == '&') && (ctxt->token != '&')) && (NXT(1) == '#')) {
1527: int val = htmlParseCharRef(ctxt);
1528: COPY_BUF(0,buffer,nbchars,val);
1529: NEXTL(l);
1530: } else if ((c == '&') && (ctxt->token != '&')) {
1531: ent = htmlParseEntityRef(ctxt, &name);
1532: if (name != NULL) {
1533: if (ent != NULL) {
1534: int val = ent->value;
1535: COPY_BUF(0,buffer,nbchars,val);
1536: NEXTL(l);
1537: } else {
1538: const xmlChar *cur = name;
1.1 daniel 1539:
1.53 veillard 1540: buffer[nbchars++] = '&';
1541: if (nbchars > buffer_size - HTML_PARSER_BUFFER_SIZE) {
1542: growBuffer(buffer);
1543: }
1544: while (*cur != 0) {
1545: buffer[nbchars++] = *cur++;
1.1 daniel 1546: }
1.53 veillard 1547: buffer[nbchars++] = ';';
1.1 daniel 1548: }
1549: }
1550: } else {
1.53 veillard 1551: COPY_BUF(l,buffer,nbchars,c);
1552: NEXTL(l);
1553: if (nbchars > buffer_size - HTML_PARSER_BUFFER_SIZE) {
1554: growBuffer(buffer);
1.1 daniel 1555: }
1556: }
1.53 veillard 1557: c = CUR_CHAR(l);
1.1 daniel 1558: }
1.53 veillard 1559: buffer[nbchars++] = 0;
1.1 daniel 1560: return(buffer);
1561: }
1562:
1.31 daniel 1563: /************************************************************************
1564: * *
1565: * Commodity functions to handle streams *
1566: * *
1567: ************************************************************************/
1568:
1569: /**
1570: * htmlFreeInputStream:
1571: * @input: an htmlParserInputPtr
1572: *
1573: * Free up an input stream.
1574: */
1575: void
1576: htmlFreeInputStream(htmlParserInputPtr input) {
1577: if (input == NULL) return;
1578:
1579: if (input->filename != NULL) xmlFree((char *) input->filename);
1580: if (input->directory != NULL) xmlFree((char *) input->directory);
1581: if ((input->free != NULL) && (input->base != NULL))
1582: input->free((xmlChar *) input->base);
1583: if (input->buf != NULL)
1584: xmlFreeParserInputBuffer(input->buf);
1585: memset(input, -1, sizeof(htmlParserInput));
1586: xmlFree(input);
1587: }
1588:
1589: /**
1590: * htmlNewInputStream:
1591: * @ctxt: an HTML parser context
1592: *
1593: * Create a new input stream structure
1594: * Returns the new input stream or NULL
1595: */
1596: htmlParserInputPtr
1597: htmlNewInputStream(htmlParserCtxtPtr ctxt) {
1598: htmlParserInputPtr input;
1599:
1600: input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1601: if (input == NULL) {
1602: ctxt->errNo = XML_ERR_NO_MEMORY;
1603: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1604: ctxt->sax->error(ctxt->userData,
1605: "malloc: couldn't allocate a new input stream\n");
1606: return(NULL);
1607: }
1.51 veillard 1608: memset(input, 0, sizeof(htmlParserInput));
1.31 daniel 1609: input->filename = NULL;
1610: input->directory = NULL;
1611: input->base = NULL;
1612: input->cur = NULL;
1613: input->buf = NULL;
1614: input->line = 1;
1615: input->col = 1;
1616: input->buf = NULL;
1617: input->free = NULL;
1.51 veillard 1618: input->version = NULL;
1.31 daniel 1619: input->consumed = 0;
1620: input->length = 0;
1621: return(input);
1622: }
1623:
1.1 daniel 1624:
1625: /************************************************************************
1626: * *
1627: * Commodity functions, cleanup needed ? *
1628: * *
1629: ************************************************************************/
1630:
1631: /**
1632: * areBlanks:
1633: * @ctxt: an HTML parser context
1.14 daniel 1634: * @str: a xmlChar *
1.1 daniel 1635: * @len: the size of @str
1636: *
1637: * Is this a sequence of blank chars that one can ignore ?
1638: *
1639: * Returns 1 if ignorable 0 otherwise.
1640: */
1641:
1.14 daniel 1642: static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
1.1 daniel 1643: int i;
1644: xmlNodePtr lastChild;
1645:
1646: for (i = 0;i < len;i++)
1647: if (!(IS_BLANK(str[i]))) return(0);
1648:
1.48 daniel 1649: if (CUR == 0) return(1);
1.1 daniel 1650: if (CUR != '<') return(0);
1.62 veillard 1651: if (ctxt->name == NULL)
1652: return(1);
1.73 ! veillard 1653: if (xmlStrEqual(ctxt->name, BAD_CAST"html"))
1.63 veillard 1654: return(1);
1.73 ! veillard 1655: if (xmlStrEqual(ctxt->name, BAD_CAST"head"))
1.62 veillard 1656: return(1);
1.73 ! veillard 1657: if (xmlStrEqual(ctxt->name, BAD_CAST"body"))
1.62 veillard 1658: return(1);
1.1 daniel 1659: if (ctxt->node == NULL) return(0);
1660: lastChild = xmlGetLastChild(ctxt->node);
1661: if (lastChild == NULL) {
1662: if (ctxt->node->content != NULL) return(0);
1663: } else if (xmlNodeIsText(lastChild))
1664: return(0);
1665: return(1);
1666: }
1667:
1668: /**
1669: * htmlHandleEntity:
1670: * @ctxt: an HTML parser context
1671: * @entity: an XML entity pointer.
1672: *
1673: * Default handling of an HTML entity, call the parser with the
1674: * substitution string
1675: */
1676:
1677: void
1678: htmlHandleEntity(htmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1679: int len;
1680:
1681: if (entity->content == NULL) {
1682: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1683: ctxt->sax->error(ctxt->userData, "htmlHandleEntity %s: content == NULL\n",
1684: entity->name);
1685: ctxt->wellFormed = 0;
1686: return;
1687: }
1688: len = xmlStrlen(entity->content);
1689:
1690: /*
1691: * Just handle the content as a set of chars.
1692: */
1.59 veillard 1693: htmlCheckParagraph(ctxt);
1.1 daniel 1694: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1695: ctxt->sax->characters(ctxt->userData, entity->content, len);
1696:
1697: }
1698:
1699: /**
1.59 veillard 1700: * htmlNewDocNoDtD:
1.1 daniel 1701: * @URI: URI for the dtd, or NULL
1702: * @ExternalID: the external ID of the DTD, or NULL
1703: *
1.59 veillard 1704: * Returns a new document, do not intialize the DTD if not provided
1.1 daniel 1705: */
1706: htmlDocPtr
1.59 veillard 1707: htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) {
1.1 daniel 1708: xmlDocPtr cur;
1709:
1710: /*
1711: * Allocate a new document and fill the fields.
1712: */
1.11 daniel 1713: cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
1.1 daniel 1714: if (cur == NULL) {
1715: fprintf(stderr, "xmlNewDoc : malloc failed\n");
1716: return(NULL);
1717: }
1.10 daniel 1718: memset(cur, 0, sizeof(xmlDoc));
1.1 daniel 1719:
1.20 daniel 1720: cur->type = XML_HTML_DOCUMENT_NODE;
1.1 daniel 1721: cur->version = NULL;
1722: cur->intSubset = NULL;
1.59 veillard 1723: if ((ExternalID != NULL) ||
1724: (URI != NULL))
1.28 daniel 1725: xmlCreateIntSubset(cur, BAD_CAST "HTML", ExternalID, URI);
1.41 daniel 1726: cur->doc = cur;
1.1 daniel 1727: cur->name = NULL;
1.37 daniel 1728: cur->children = NULL;
1.1 daniel 1729: cur->extSubset = NULL;
1730: cur->oldNs = NULL;
1731: cur->encoding = NULL;
1732: cur->standalone = 1;
1733: cur->compression = 0;
1.12 daniel 1734: cur->ids = NULL;
1735: cur->refs = NULL;
1.1 daniel 1736: #ifndef XML_WITHOUT_CORBA
1737: cur->_private = NULL;
1738: #endif
1739: return(cur);
1740: }
1741:
1.59 veillard 1742: /**
1743: * htmlNewDoc:
1744: * @URI: URI for the dtd, or NULL
1745: * @ExternalID: the external ID of the DTD, or NULL
1746: *
1747: * Returns a new document
1748: */
1749: htmlDocPtr
1750: htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
1751: if ((URI == NULL) && (ExternalID == NULL))
1752: return(htmlNewDocNoDtD(
1753: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
1754: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd"));
1755:
1756: return(htmlNewDocNoDtD(URI, ExternalID));
1757: }
1758:
1.1 daniel 1759:
1760: /************************************************************************
1761: * *
1762: * The parser itself *
1763: * Relates to http://www.w3.org/TR/html40 *
1764: * *
1765: ************************************************************************/
1766:
1767: /************************************************************************
1768: * *
1769: * The parser itself *
1770: * *
1771: ************************************************************************/
1772:
1773: /**
1774: * htmlParseHTMLName:
1775: * @ctxt: an HTML parser context
1776: *
1.26 daniel 1777: * parse an HTML tag or attribute name, note that we convert it to lowercase
1.1 daniel 1778: * since HTML names are not case-sensitive.
1779: *
1780: * Returns the Tag Name parsed or NULL
1781: */
1782:
1.14 daniel 1783: xmlChar *
1.1 daniel 1784: htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1785: xmlChar *ret = NULL;
1.1 daniel 1786: int i = 0;
1.31 daniel 1787: xmlChar loc[HTML_PARSER_BUFFER_SIZE];
1.1 daniel 1788:
1789: if (!IS_LETTER(CUR) && (CUR != '_') &&
1790: (CUR != ':')) return(NULL);
1791:
1.31 daniel 1792: while ((i < HTML_PARSER_BUFFER_SIZE) &&
1.45 daniel 1793: ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1794: (CUR == ':') || (CUR == '_'))) {
1.26 daniel 1795: if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20;
1.1 daniel 1796: else loc[i] = CUR;
1797: i++;
1798:
1799: NEXT;
1800: }
1801:
1802: ret = xmlStrndup(loc, i);
1803:
1804: return(ret);
1805: }
1806:
1807: /**
1808: * htmlParseName:
1809: * @ctxt: an HTML parser context
1810: *
1811: * parse an HTML name, this routine is case sensistive.
1812: *
1813: * Returns the Name parsed or NULL
1814: */
1815:
1.14 daniel 1816: xmlChar *
1.1 daniel 1817: htmlParseName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1818: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1819: int len = 0;
1.1 daniel 1820:
1.5 daniel 1821: GROW;
1822: if (!IS_LETTER(CUR) && (CUR != '_')) {
1823: return(NULL);
1824: }
1.1 daniel 1825:
1826: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1827: (CUR == '.') || (CUR == '-') ||
1828: (CUR == '_') || (CUR == ':') ||
1829: (IS_COMBINING(CUR)) ||
1.5 daniel 1830: (IS_EXTENDER(CUR))) {
1831: buf[len++] = CUR;
1.1 daniel 1832: NEXT;
1.5 daniel 1833: if (len >= HTML_MAX_NAMELEN) {
1834: fprintf(stderr,
1835: "htmlParseName: reached HTML_MAX_NAMELEN limit\n");
1836: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1837: (CUR == '.') || (CUR == '-') ||
1838: (CUR == '_') || (CUR == ':') ||
1839: (IS_COMBINING(CUR)) ||
1840: (IS_EXTENDER(CUR)))
1841: NEXT;
1842: break;
1843: }
1844: }
1845: return(xmlStrndup(buf, len));
1.1 daniel 1846: }
1847:
1848: /**
1849: * htmlParseHTMLAttribute:
1850: * @ctxt: an HTML parser context
1.19 daniel 1851: * @stop: a char stop value
1.1 daniel 1852: *
1.19 daniel 1853: * parse an HTML attribute value till the stop (quote), if
1854: * stop is 0 then it stops at the first space
1.1 daniel 1855: *
1.19 daniel 1856: * Returns the attribute parsed or NULL
1.1 daniel 1857: */
1858:
1.14 daniel 1859: xmlChar *
1.19 daniel 1860: htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) {
1.32 daniel 1861: #if 0
1.14 daniel 1862: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1863: int len = 0;
1.1 daniel 1864:
1.5 daniel 1865: GROW;
1.19 daniel 1866: while ((CUR != 0) && (CUR != stop) && (CUR != '>')) {
1867: if ((stop == 0) && (IS_BLANK(CUR))) break;
1.5 daniel 1868: buf[len++] = CUR;
1.1 daniel 1869: NEXT;
1.5 daniel 1870: if (len >= HTML_MAX_NAMELEN) {
1871: fprintf(stderr,
1872: "htmlParseHTMLAttribute: reached HTML_MAX_NAMELEN limit\n");
1873: while ((!IS_BLANK(CUR)) && (CUR != '<') &&
1.19 daniel 1874: (CUR != '>') &&
1.5 daniel 1875: (CUR != '\'') && (CUR != '"'))
1876: NEXT;
1877: break;
1878: }
1879: }
1880: return(xmlStrndup(buf, len));
1.32 daniel 1881: #else
1882: xmlChar *buffer = NULL;
1883: int buffer_size = 0;
1884: xmlChar *out = NULL;
1885: xmlChar *name = NULL;
1886:
1887: xmlChar *cur = NULL;
1888: htmlEntityDescPtr ent;
1889:
1890: /*
1891: * allocate a translation buffer.
1892: */
1893: buffer_size = HTML_PARSER_BIG_BUFFER_SIZE;
1894: buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
1895: if (buffer == NULL) {
1896: perror("htmlParseHTMLAttribute: malloc failed");
1897: return(NULL);
1898: }
1899: out = buffer;
1900:
1901: /*
1902: * Ok loop until we reach one of the ending chars
1903: */
1904: while ((CUR != 0) && (CUR != stop) && (CUR != '>')) {
1905: if ((stop == 0) && (IS_BLANK(CUR))) break;
1906: if (CUR == '&') {
1907: if (NXT(1) == '#') {
1.52 veillard 1908: unsigned int c;
1909: int bits;
1910:
1911: c = htmlParseCharRef(ctxt);
1912: if (c < 0x80)
1913: { *out++ = c; bits= -6; }
1914: else if (c < 0x800)
1915: { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
1916: else if (c < 0x10000)
1917: { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
1918: else
1919: { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
1920:
1921: for ( ; bits >= 0; bits-= 6) {
1922: *out++ = ((c >> bits) & 0x3F) | 0x80;
1923: }
1.32 daniel 1924: } else {
1925: ent = htmlParseEntityRef(ctxt, &name);
1926: if (name == NULL) {
1927: *out++ = '&';
1928: if (out - buffer > buffer_size - 100) {
1929: int index = out - buffer;
1930:
1931: growBuffer(buffer);
1932: out = &buffer[index];
1933: }
1.52 veillard 1934: } else if (ent == NULL) {
1.32 daniel 1935: *out++ = '&';
1936: cur = name;
1937: while (*cur != 0) {
1938: if (out - buffer > buffer_size - 100) {
1939: int index = out - buffer;
1940:
1941: growBuffer(buffer);
1942: out = &buffer[index];
1943: }
1944: *out++ = *cur++;
1945: }
1946: xmlFree(name);
1947: } else {
1.52 veillard 1948: unsigned int c;
1949: int bits;
1950:
1.32 daniel 1951: if (out - buffer > buffer_size - 100) {
1952: int index = out - buffer;
1953:
1954: growBuffer(buffer);
1955: out = &buffer[index];
1956: }
1.52 veillard 1957: c = (xmlChar)ent->value;
1958: if (c < 0x80)
1959: { *out++ = c; bits= -6; }
1960: else if (c < 0x800)
1961: { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
1962: else if (c < 0x10000)
1963: { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
1964: else
1965: { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
1966:
1967: for ( ; bits >= 0; bits-= 6) {
1968: *out++ = ((c >> bits) & 0x3F) | 0x80;
1969: }
1.32 daniel 1970: xmlFree(name);
1971: }
1972: }
1973: } else {
1.52 veillard 1974: unsigned int c;
1.68 veillard 1975: int bits, l;
1.52 veillard 1976:
1.32 daniel 1977: if (out - buffer > buffer_size - 100) {
1.52 veillard 1978: int index = out - buffer;
1979:
1980: growBuffer(buffer);
1981: out = &buffer[index];
1982: }
1.68 veillard 1983: c = CUR_CHAR(l);
1.52 veillard 1984: if (c < 0x80)
1985: { *out++ = c; bits= -6; }
1986: else if (c < 0x800)
1987: { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
1988: else if (c < 0x10000)
1989: { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
1990: else
1991: { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
1992:
1993: for ( ; bits >= 0; bits-= 6) {
1994: *out++ = ((c >> bits) & 0x3F) | 0x80;
1.32 daniel 1995: }
1996: NEXT;
1997: }
1998: }
1999: *out++ = 0;
2000: return(buffer);
2001: #endif
1.1 daniel 2002: }
2003:
2004: /**
2005: * htmlParseNmtoken:
2006: * @ctxt: an HTML parser context
2007: *
2008: * parse an HTML Nmtoken.
2009: *
2010: * Returns the Nmtoken parsed or NULL
2011: */
2012:
1.14 daniel 2013: xmlChar *
1.1 daniel 2014: htmlParseNmtoken(htmlParserCtxtPtr ctxt) {
1.14 daniel 2015: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 2016: int len = 0;
1.1 daniel 2017:
1.5 daniel 2018: GROW;
1.1 daniel 2019: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
2020: (CUR == '.') || (CUR == '-') ||
2021: (CUR == '_') || (CUR == ':') ||
2022: (IS_COMBINING(CUR)) ||
1.5 daniel 2023: (IS_EXTENDER(CUR))) {
2024: buf[len++] = CUR;
1.1 daniel 2025: NEXT;
1.5 daniel 2026: if (len >= HTML_MAX_NAMELEN) {
2027: fprintf(stderr,
2028: "htmlParseNmtoken: reached HTML_MAX_NAMELEN limit\n");
2029: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
2030: (CUR == '.') || (CUR == '-') ||
2031: (CUR == '_') || (CUR == ':') ||
2032: (IS_COMBINING(CUR)) ||
2033: (IS_EXTENDER(CUR)))
2034: NEXT;
2035: break;
2036: }
2037: }
2038: return(xmlStrndup(buf, len));
1.1 daniel 2039: }
2040:
2041: /**
2042: * htmlParseEntityRef:
2043: * @ctxt: an HTML parser context
2044: * @str: location to store the entity name
2045: *
2046: * parse an HTML ENTITY references
2047: *
2048: * [68] EntityRef ::= '&' Name ';'
2049: *
2050: * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
2051: * if non-NULL *str will have to be freed by the caller.
2052: */
2053: htmlEntityDescPtr
1.14 daniel 2054: htmlParseEntityRef(htmlParserCtxtPtr ctxt, xmlChar **str) {
2055: xmlChar *name;
1.1 daniel 2056: htmlEntityDescPtr ent = NULL;
2057: *str = NULL;
2058:
2059: if (CUR == '&') {
2060: NEXT;
2061: name = htmlParseName(ctxt);
2062: if (name == NULL) {
2063: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2064: ctxt->sax->error(ctxt->userData, "htmlParseEntityRef: no name\n");
2065: ctxt->wellFormed = 0;
2066: } else {
1.5 daniel 2067: GROW;
1.1 daniel 2068: if (CUR == ';') {
2069: *str = name;
2070:
2071: /*
2072: * Lookup the entity in the table.
2073: */
2074: ent = htmlEntityLookup(name);
1.32 daniel 2075: if (ent != NULL) /* OK that's ugly !!! */
2076: NEXT;
1.1 daniel 2077: } else {
2078: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2079: ctxt->sax->error(ctxt->userData,
2080: "htmlParseEntityRef: expecting ';'\n");
1.32 daniel 2081: *str = name;
1.1 daniel 2082: }
2083: }
2084: }
2085: return(ent);
2086: }
2087:
2088: /**
2089: * htmlParseAttValue:
2090: * @ctxt: an HTML parser context
2091: *
2092: * parse a value for an attribute
2093: * Note: the parser won't do substitution of entities here, this
2094: * will be handled later in xmlStringGetNodeList, unless it was
2095: * asked for ctxt->replaceEntities != 0
2096: *
2097: * Returns the AttValue parsed or NULL.
2098: */
2099:
1.14 daniel 2100: xmlChar *
1.1 daniel 2101: htmlParseAttValue(htmlParserCtxtPtr ctxt) {
1.14 daniel 2102: xmlChar *ret = NULL;
1.1 daniel 2103:
2104: if (CUR == '"') {
2105: NEXT;
1.19 daniel 2106: ret = htmlParseHTMLAttribute(ctxt, '"');
1.1 daniel 2107: if (CUR != '"') {
2108: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2109: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
2110: ctxt->wellFormed = 0;
2111: } else
2112: NEXT;
2113: } else if (CUR == '\'') {
2114: NEXT;
1.19 daniel 2115: ret = htmlParseHTMLAttribute(ctxt, '\'');
1.1 daniel 2116: if (CUR != '\'') {
2117: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2118: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
2119: ctxt->wellFormed = 0;
2120: } else
2121: NEXT;
2122: } else {
2123: /*
2124: * That's an HTMLism, the attribute value may not be quoted
2125: */
1.19 daniel 2126: ret = htmlParseHTMLAttribute(ctxt, 0);
1.1 daniel 2127: if (ret == NULL) {
2128: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2129: ctxt->sax->error(ctxt->userData, "AttValue: no value found\n");
2130: ctxt->wellFormed = 0;
2131: }
2132: }
2133: return(ret);
2134: }
2135:
2136: /**
2137: * htmlParseSystemLiteral:
2138: * @ctxt: an HTML parser context
2139: *
2140: * parse an HTML Literal
2141: *
2142: * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
2143: *
2144: * Returns the SystemLiteral parsed or NULL
2145: */
2146:
1.14 daniel 2147: xmlChar *
1.1 daniel 2148: htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 2149: const xmlChar *q;
2150: xmlChar *ret = NULL;
1.1 daniel 2151:
2152: if (CUR == '"') {
2153: NEXT;
2154: q = CUR_PTR;
2155: while ((IS_CHAR(CUR)) && (CUR != '"'))
2156: NEXT;
2157: if (!IS_CHAR(CUR)) {
2158: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2159: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
2160: ctxt->wellFormed = 0;
2161: } else {
2162: ret = xmlStrndup(q, CUR_PTR - q);
2163: NEXT;
2164: }
2165: } else if (CUR == '\'') {
2166: NEXT;
2167: q = CUR_PTR;
2168: while ((IS_CHAR(CUR)) && (CUR != '\''))
2169: NEXT;
2170: if (!IS_CHAR(CUR)) {
2171: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2172: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
2173: ctxt->wellFormed = 0;
2174: } else {
2175: ret = xmlStrndup(q, CUR_PTR - q);
2176: NEXT;
2177: }
2178: } else {
2179: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.38 daniel 2180: ctxt->sax->error(ctxt->userData,
2181: "SystemLiteral \" or ' expected\n");
1.1 daniel 2182: ctxt->wellFormed = 0;
2183: }
2184:
2185: return(ret);
2186: }
2187:
2188: /**
2189: * htmlParsePubidLiteral:
2190: * @ctxt: an HTML parser context
2191: *
2192: * parse an HTML public literal
2193: *
2194: * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
2195: *
2196: * Returns the PubidLiteral parsed or NULL.
2197: */
2198:
1.14 daniel 2199: xmlChar *
1.1 daniel 2200: htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 2201: const xmlChar *q;
2202: xmlChar *ret = NULL;
1.1 daniel 2203: /*
2204: * Name ::= (Letter | '_') (NameChar)*
2205: */
2206: if (CUR == '"') {
2207: NEXT;
2208: q = CUR_PTR;
2209: while (IS_PUBIDCHAR(CUR)) NEXT;
2210: if (CUR != '"') {
2211: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2212: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
2213: ctxt->wellFormed = 0;
2214: } else {
2215: ret = xmlStrndup(q, CUR_PTR - q);
2216: NEXT;
2217: }
2218: } else if (CUR == '\'') {
2219: NEXT;
2220: q = CUR_PTR;
2221: while ((IS_LETTER(CUR)) && (CUR != '\''))
2222: NEXT;
2223: if (!IS_LETTER(CUR)) {
2224: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2225: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
2226: ctxt->wellFormed = 0;
2227: } else {
2228: ret = xmlStrndup(q, CUR_PTR - q);
2229: NEXT;
2230: }
2231: } else {
2232: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2233: ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n");
2234: ctxt->wellFormed = 0;
2235: }
2236:
2237: return(ret);
2238: }
2239:
2240: /**
2241: * htmlParseCharData:
2242: * @ctxt: an HTML parser context
2243: * @cdata: int indicating whether we are within a CDATA section
2244: *
2245: * parse a CharData section.
2246: * if we are within a CDATA section ']]>' marks an end of section.
2247: *
2248: * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
2249: */
2250:
2251: void
2252: htmlParseCharData(htmlParserCtxtPtr ctxt, int cdata) {
1.53 veillard 2253: xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
2254: int nbchar = 0;
2255: int cur, l;
2256:
2257: SHRINK;
2258: cur = CUR_CHAR(l);
2259: while (((cur != '<') || (ctxt->token == '<')) &&
2260: ((cur != '&') || (ctxt->token == '&')) &&
2261: (IS_CHAR(cur))) {
2262: COPY_BUF(l,buf,nbchar,cur);
2263: if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2264: /*
2265: * Ok the segment is to be consumed as chars.
2266: */
2267: if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2268: if (areBlanks(ctxt, buf, nbchar)) {
2269: if (ctxt->sax->ignorableWhitespace != NULL)
2270: ctxt->sax->ignorableWhitespace(ctxt->userData,
2271: buf, nbchar);
2272: } else {
1.59 veillard 2273: htmlCheckParagraph(ctxt);
1.53 veillard 2274: if (ctxt->sax->characters != NULL)
2275: ctxt->sax->characters(ctxt->userData, buf, nbchar);
2276: }
1.1 daniel 2277: }
1.53 veillard 2278: nbchar = 0;
1.1 daniel 2279: }
1.53 veillard 2280: NEXTL(l);
2281: cur = CUR_CHAR(l);
2282: }
2283: if (nbchar != 0) {
2284: /*
2285: * Ok the segment is to be consumed as chars.
2286: */
2287: if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2288: if (areBlanks(ctxt, buf, nbchar)) {
2289: if (ctxt->sax->ignorableWhitespace != NULL)
2290: ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar);
2291: } else {
1.59 veillard 2292: htmlCheckParagraph(ctxt);
1.53 veillard 2293: if (ctxt->sax->characters != NULL)
2294: ctxt->sax->characters(ctxt->userData, buf, nbchar);
1.25 daniel 2295: }
2296: }
1.1 daniel 2297: }
2298: }
2299:
2300: /**
2301: * htmlParseExternalID:
2302: * @ctxt: an HTML parser context
1.14 daniel 2303: * @publicID: a xmlChar** receiving PubidLiteral
1.1 daniel 2304: * @strict: indicate whether we should restrict parsing to only
2305: * production [75], see NOTE below
2306: *
2307: * Parse an External ID or a Public ID
2308: *
2309: * NOTE: Productions [75] and [83] interract badly since [75] can generate
2310: * 'PUBLIC' S PubidLiteral S SystemLiteral
2311: *
2312: * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
2313: * | 'PUBLIC' S PubidLiteral S SystemLiteral
2314: *
2315: * [83] PublicID ::= 'PUBLIC' S PubidLiteral
2316: *
2317: * Returns the function returns SystemLiteral and in the second
2318: * case publicID receives PubidLiteral, is strict is off
2319: * it is possible to return NULL and have publicID set.
2320: */
2321:
1.14 daniel 2322: xmlChar *
2323: htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID, int strict) {
2324: xmlChar *URI = NULL;
1.1 daniel 2325:
2326: if ((UPPER == 'S') && (UPP(1) == 'Y') &&
2327: (UPP(2) == 'S') && (UPP(3) == 'T') &&
2328: (UPP(4) == 'E') && (UPP(5) == 'M')) {
2329: SKIP(6);
2330: if (!IS_BLANK(CUR)) {
2331: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2332: ctxt->sax->error(ctxt->userData,
2333: "Space required after 'SYSTEM'\n");
2334: ctxt->wellFormed = 0;
2335: }
2336: SKIP_BLANKS;
2337: URI = htmlParseSystemLiteral(ctxt);
2338: if (URI == NULL) {
2339: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2340: ctxt->sax->error(ctxt->userData,
2341: "htmlParseExternalID: SYSTEM, no URI\n");
2342: ctxt->wellFormed = 0;
2343: }
2344: } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
2345: (UPP(2) == 'B') && (UPP(3) == 'L') &&
2346: (UPP(4) == 'I') && (UPP(5) == 'C')) {
2347: SKIP(6);
2348: if (!IS_BLANK(CUR)) {
2349: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2350: ctxt->sax->error(ctxt->userData,
2351: "Space required after 'PUBLIC'\n");
2352: ctxt->wellFormed = 0;
2353: }
2354: SKIP_BLANKS;
2355: *publicID = htmlParsePubidLiteral(ctxt);
2356: if (*publicID == NULL) {
2357: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2358: ctxt->sax->error(ctxt->userData,
2359: "htmlParseExternalID: PUBLIC, no Public Identifier\n");
2360: ctxt->wellFormed = 0;
2361: }
1.5 daniel 2362: SKIP_BLANKS;
2363: if ((CUR == '"') || (CUR == '\'')) {
2364: URI = htmlParseSystemLiteral(ctxt);
1.1 daniel 2365: }
2366: }
2367: return(URI);
2368: }
2369:
2370: /**
2371: * htmlParseComment:
2372: * @ctxt: an HTML parser context
2373: *
2374: * Parse an XML (SGML) comment <!-- .... -->
2375: *
2376: * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
2377: */
2378: void
1.31 daniel 2379: htmlParseComment(htmlParserCtxtPtr ctxt) {
1.25 daniel 2380: xmlChar *buf = NULL;
1.56 veillard 2381: int len;
1.31 daniel 2382: int size = HTML_PARSER_BUFFER_SIZE;
1.56 veillard 2383: int q, ql;
2384: int r, rl;
2385: int cur, l;
2386: xmlParserInputState state;
1.1 daniel 2387:
2388: /*
2389: * Check that there is a comment right here.
2390: */
1.56 veillard 2391: if ((RAW != '<') || (NXT(1) != '!') ||
1.1 daniel 2392: (NXT(2) != '-') || (NXT(3) != '-')) return;
2393:
1.56 veillard 2394: state = ctxt->instate;
2395: ctxt->instate = XML_PARSER_COMMENT;
2396: SHRINK;
2397: SKIP(4);
1.25 daniel 2398: buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar));
2399: if (buf == NULL) {
2400: fprintf(stderr, "malloc of %d byte failed\n", size);
1.56 veillard 2401: ctxt->instate = state;
1.25 daniel 2402: return;
2403: }
1.56 veillard 2404: q = CUR_CHAR(ql);
2405: NEXTL(ql);
2406: r = CUR_CHAR(rl);
2407: NEXTL(rl);
2408: cur = CUR_CHAR(l);
2409: len = 0;
2410: while (IS_CHAR(cur) &&
2411: ((cur != '>') ||
2412: (r != '-') || (q != '-'))) {
2413: if (len + 5 >= size) {
1.25 daniel 2414: size *= 2;
1.50 veillard 2415: buf = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
1.25 daniel 2416: if (buf == NULL) {
2417: fprintf(stderr, "realloc of %d byte failed\n", size);
1.56 veillard 2418: ctxt->instate = state;
1.25 daniel 2419: return;
2420: }
2421: }
1.56 veillard 2422: COPY_BUF(ql,buf,len,q);
1.25 daniel 2423: q = r;
1.56 veillard 2424: ql = rl;
2425: r = cur;
2426: rl = l;
2427: NEXTL(l);
2428: cur = CUR_CHAR(l);
2429: if (cur == 0) {
2430: SHRINK;
2431: GROW;
2432: cur = CUR_CHAR(l);
2433: }
1.1 daniel 2434: }
1.56 veillard 2435: buf[len] = 0;
2436: if (!IS_CHAR(cur)) {
1.67 veillard 2437: ctxt->errNo = XML_ERR_COMMENT_NOT_FINISHED;
1.1 daniel 2438: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.56 veillard 2439: ctxt->sax->error(ctxt->userData,
2440: "Comment not terminated \n<!--%.50s\n", buf);
1.1 daniel 2441: ctxt->wellFormed = 0;
1.56 veillard 2442: xmlFree(buf);
1.1 daniel 2443: } else {
2444: NEXT;
1.56 veillard 2445: if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) &&
2446: (!ctxt->disableSAX))
1.31 daniel 2447: ctxt->sax->comment(ctxt->userData, buf);
1.56 veillard 2448: xmlFree(buf);
1.1 daniel 2449: }
1.56 veillard 2450: ctxt->instate = state;
1.1 daniel 2451: }
2452:
2453: /**
2454: * htmlParseCharRef:
2455: * @ctxt: an HTML parser context
2456: *
2457: * parse Reference declarations
2458: *
2459: * [66] CharRef ::= '&#' [0-9]+ ';' |
2460: * '&#x' [0-9a-fA-F]+ ';'
2461: *
2462: * Returns the value parsed (as an int)
2463: */
2464: int
2465: htmlParseCharRef(htmlParserCtxtPtr ctxt) {
2466: int val = 0;
2467:
2468: if ((CUR == '&') && (NXT(1) == '#') &&
2469: (NXT(2) == 'x')) {
2470: SKIP(3);
2471: while (CUR != ';') {
2472: if ((CUR >= '0') && (CUR <= '9'))
2473: val = val * 16 + (CUR - '0');
2474: else if ((CUR >= 'a') && (CUR <= 'f'))
2475: val = val * 16 + (CUR - 'a') + 10;
2476: else if ((CUR >= 'A') && (CUR <= 'F'))
2477: val = val * 16 + (CUR - 'A') + 10;
2478: else {
2479: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2480: ctxt->sax->error(ctxt->userData,
2481: "htmlParseCharRef: invalid hexadecimal value\n");
2482: ctxt->wellFormed = 0;
2483: val = 0;
2484: break;
2485: }
2486: NEXT;
2487: }
2488: if (CUR == ';')
2489: NEXT;
2490: } else if ((CUR == '&') && (NXT(1) == '#')) {
2491: SKIP(2);
2492: while (CUR != ';') {
2493: if ((CUR >= '0') && (CUR <= '9'))
2494: val = val * 10 + (CUR - '0');
2495: else {
2496: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2497: ctxt->sax->error(ctxt->userData,
2498: "htmlParseCharRef: invalid decimal value\n");
2499: ctxt->wellFormed = 0;
2500: val = 0;
2501: break;
2502: }
2503: NEXT;
2504: }
2505: if (CUR == ';')
2506: NEXT;
2507: } else {
2508: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2509: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid value\n");
2510: ctxt->wellFormed = 0;
2511: }
2512: /*
2513: * Check the value IS_CHAR ...
2514: */
2515: if (IS_CHAR(val)) {
2516: return(val);
2517: } else {
2518: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.14 daniel 2519: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid xmlChar value %d\n",
1.1 daniel 2520: val);
2521: ctxt->wellFormed = 0;
2522: }
2523: return(0);
2524: }
2525:
2526:
2527: /**
2528: * htmlParseDocTypeDecl :
2529: * @ctxt: an HTML parser context
2530: *
2531: * parse a DOCTYPE declaration
2532: *
2533: * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
2534: * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
2535: */
2536:
2537: void
2538: htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
1.14 daniel 2539: xmlChar *name;
2540: xmlChar *ExternalID = NULL;
2541: xmlChar *URI = NULL;
1.1 daniel 2542:
2543: /*
2544: * We know that '<!DOCTYPE' has been detected.
2545: */
2546: SKIP(9);
2547:
2548: SKIP_BLANKS;
2549:
2550: /*
2551: * Parse the DOCTYPE name.
2552: */
2553: name = htmlParseName(ctxt);
2554: if (name == NULL) {
2555: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2556: ctxt->sax->error(ctxt->userData, "htmlParseDocTypeDecl : no DOCTYPE name !\n");
2557: ctxt->wellFormed = 0;
2558: }
2559: /*
2560: * Check that upper(name) == "HTML" !!!!!!!!!!!!!
2561: */
2562:
2563: SKIP_BLANKS;
2564:
2565: /*
2566: * Check for SystemID and ExternalID
2567: */
1.5 daniel 2568: URI = htmlParseExternalID(ctxt, &ExternalID, 0);
1.1 daniel 2569: SKIP_BLANKS;
2570:
2571: /*
2572: * We should be at the end of the DOCTYPE declaration.
2573: */
2574: if (CUR != '>') {
2575: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2576: ctxt->sax->error(ctxt->userData, "DOCTYPE unproperly terminated\n");
2577: ctxt->wellFormed = 0;
2578: /* We shouldn't try to resynchronize ... */
2579: }
2580: NEXT;
2581:
2582: /*
1.46 daniel 2583: * Create or update the document accordingly to the DOCTYPE
1.1 daniel 2584: */
1.46 daniel 2585: if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) &&
2586: (!ctxt->disableSAX))
2587: ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI);
1.1 daniel 2588:
2589: /*
2590: * Cleanup, since we don't use all those identifiers
2591: */
1.11 daniel 2592: if (URI != NULL) xmlFree(URI);
2593: if (ExternalID != NULL) xmlFree(ExternalID);
2594: if (name != NULL) xmlFree(name);
1.1 daniel 2595: }
2596:
2597: /**
2598: * htmlParseAttribute:
2599: * @ctxt: an HTML parser context
1.14 daniel 2600: * @value: a xmlChar ** used to store the value of the attribute
1.1 daniel 2601: *
2602: * parse an attribute
2603: *
2604: * [41] Attribute ::= Name Eq AttValue
2605: *
2606: * [25] Eq ::= S? '=' S?
2607: *
2608: * With namespace:
2609: *
2610: * [NS 11] Attribute ::= QName Eq AttValue
2611: *
2612: * Also the case QName == xmlns:??? is handled independently as a namespace
2613: * definition.
2614: *
2615: * Returns the attribute name, and the value in *value.
2616: */
2617:
1.14 daniel 2618: xmlChar *
2619: htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
1.31 daniel 2620: xmlChar *name, *val = NULL;
1.1 daniel 2621:
2622: *value = NULL;
2623: name = htmlParseName(ctxt);
2624: if (name == NULL) {
2625: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2626: ctxt->sax->error(ctxt->userData, "error parsing attribute name\n");
2627: ctxt->wellFormed = 0;
2628: return(NULL);
2629: }
2630:
2631: /*
2632: * read the value
2633: */
2634: SKIP_BLANKS;
2635: if (CUR == '=') {
2636: NEXT;
2637: SKIP_BLANKS;
2638: val = htmlParseAttValue(ctxt);
1.42 daniel 2639: /******
1.1 daniel 2640: } else {
1.42 daniel 2641: * TODO : some attribute must have values, some may not
1.1 daniel 2642: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.31 daniel 2643: ctxt->sax->warning(ctxt->userData,
1.42 daniel 2644: "No value for attribute %s\n", name); */
1.1 daniel 2645: }
2646:
2647: *value = val;
2648: return(name);
2649: }
2650:
2651: /**
1.47 daniel 2652: * htmlCheckEncoding:
2653: * @ctxt: an HTML parser context
2654: * @attvalue: the attribute value
2655: *
2656: * Checks an http-equiv attribute from a Meta tag to detect
2657: * the encoding
2658: * If a new encoding is detected the parser is switched to decode
2659: * it and pass UTF8
2660: */
2661: void
2662: htmlCheckEncoding(htmlParserCtxtPtr ctxt, const xmlChar *attvalue) {
2663: const xmlChar *encoding;
2664:
2665: if ((ctxt == NULL) || (attvalue == NULL))
2666: return;
2667:
1.69 veillard 2668: encoding = xmlStrcasestr(attvalue, BAD_CAST"charset=");
1.47 daniel 2669: if (encoding != NULL) {
2670: encoding += 8;
2671: } else {
1.69 veillard 2672: encoding = xmlStrcasestr(attvalue, BAD_CAST"charset =");
1.47 daniel 2673: if (encoding != NULL)
2674: encoding += 9;
2675: }
2676: if (encoding != NULL) {
2677: xmlCharEncoding enc;
2678: xmlCharEncodingHandlerPtr handler;
2679:
2680: while ((*encoding == ' ') || (*encoding == '\t')) encoding++;
2681:
2682: if (ctxt->input->encoding != NULL)
2683: xmlFree((xmlChar *) ctxt->input->encoding);
2684: ctxt->input->encoding = xmlStrdup(encoding);
2685:
2686: enc = xmlParseCharEncoding((const char *) encoding);
2687: /*
2688: * registered set of known encodings
2689: */
2690: if (enc != XML_CHAR_ENCODING_ERROR) {
2691: xmlSwitchEncoding(ctxt, enc);
1.53 veillard 2692: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1.47 daniel 2693: } else {
2694: /*
2695: * fallback for unknown encodings
2696: */
2697: handler = xmlFindCharEncodingHandler((const char *) encoding);
2698: if (handler != NULL) {
2699: xmlSwitchToEncoding(ctxt, handler);
1.54 veillard 2700: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1.47 daniel 2701: } else {
2702: ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
2703: }
2704: }
1.54 veillard 2705:
2706: if ((ctxt->input->buf != NULL) &&
2707: (ctxt->input->buf->encoder != NULL) &&
2708: (ctxt->input->buf->raw != NULL) &&
2709: (ctxt->input->buf->buffer != NULL)) {
2710: int nbchars;
1.56 veillard 2711: int processed;
1.54 veillard 2712:
2713: /*
2714: * convert as much as possible to the parser reading buffer.
2715: */
1.56 veillard 2716: processed = ctxt->input->cur - ctxt->input->base;
2717: xmlBufferShrink(ctxt->input->buf->buffer, processed);
1.54 veillard 2718: nbchars = xmlCharEncInFunc(ctxt->input->buf->encoder,
2719: ctxt->input->buf->buffer,
2720: ctxt->input->buf->raw);
2721: if (nbchars < 0) {
1.67 veillard 2722: ctxt->errNo = XML_ERR_INVALID_ENCODING;
1.54 veillard 2723: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2724: ctxt->sax->error(ctxt->userData,
2725: "htmlCheckEncoding: encoder error\n");
2726: }
1.56 veillard 2727: ctxt->input->base =
2728: ctxt->input->cur = ctxt->input->buf->buffer->content;
1.54 veillard 2729: }
1.47 daniel 2730: }
2731: }
2732:
2733: /**
2734: * htmlCheckMeta:
2735: * @ctxt: an HTML parser context
2736: * @atts: the attributes values
2737: *
2738: * Checks an attributes from a Meta tag
2739: */
2740: void
2741: htmlCheckMeta(htmlParserCtxtPtr ctxt, const xmlChar **atts) {
2742: int i;
2743: const xmlChar *att, *value;
2744: int http = 0;
2745: const xmlChar *content = NULL;
2746:
2747: if ((ctxt == NULL) || (atts == NULL))
2748: return;
2749:
2750: i = 0;
2751: att = atts[i++];
2752: while (att != NULL) {
2753: value = atts[i++];
1.69 veillard 2754: if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"http-equiv"))
2755: && (!xmlStrcasecmp(value, BAD_CAST"Content-Type")))
1.47 daniel 2756: http = 1;
1.69 veillard 2757: else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"content")))
1.47 daniel 2758: content = value;
2759: att = atts[i++];
2760: }
2761: if ((http) && (content != NULL))
2762: htmlCheckEncoding(ctxt, content);
2763:
2764: }
2765:
2766: /**
1.1 daniel 2767: * htmlParseStartTag:
2768: * @ctxt: an HTML parser context
2769: *
2770: * parse a start of tag either for rule element or
2771: * EmptyElement. In both case we don't parse the tag closing chars.
2772: *
2773: * [40] STag ::= '<' Name (S Attribute)* S? '>'
2774: *
2775: * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
2776: *
2777: * With namespace:
2778: *
2779: * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
2780: *
2781: * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
2782: *
2783: */
2784:
1.18 daniel 2785: void
1.1 daniel 2786: htmlParseStartTag(htmlParserCtxtPtr ctxt) {
1.14 daniel 2787: xmlChar *name;
2788: xmlChar *attname;
2789: xmlChar *attvalue;
2790: const xmlChar **atts = NULL;
1.1 daniel 2791: int nbatts = 0;
2792: int maxatts = 0;
1.47 daniel 2793: int meta = 0;
1.1 daniel 2794: int i;
2795:
1.18 daniel 2796: if (CUR != '<') return;
1.1 daniel 2797: NEXT;
2798:
1.19 daniel 2799: GROW;
1.1 daniel 2800: name = htmlParseHTMLName(ctxt);
2801: if (name == NULL) {
2802: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2803: ctxt->sax->error(ctxt->userData,
2804: "htmlParseStartTag: invalid element name\n");
2805: ctxt->wellFormed = 0;
1.18 daniel 2806: return;
1.1 daniel 2807: }
1.73 ! veillard 2808: if (xmlStrEqual(name, BAD_CAST"meta"))
1.47 daniel 2809: meta = 1;
1.1 daniel 2810:
2811: /*
2812: * Check for auto-closure of HTML elements.
2813: */
2814: htmlAutoClose(ctxt, name);
1.43 daniel 2815:
2816: /*
2817: * Check for implied HTML elements.
2818: */
2819: htmlCheckImplied(ctxt, name);
1.1 daniel 2820:
2821: /*
2822: * Now parse the attributes, it ends up with the ending
2823: *
2824: * (S Attribute)* S?
2825: */
2826: SKIP_BLANKS;
2827: while ((IS_CHAR(CUR)) &&
2828: (CUR != '>') &&
2829: ((CUR != '/') || (NXT(1) != '>'))) {
1.26 daniel 2830: long cons = ctxt->nbChars;
1.1 daniel 2831:
1.19 daniel 2832: GROW;
1.1 daniel 2833: attname = htmlParseAttribute(ctxt, &attvalue);
1.31 daniel 2834: if (attname != NULL) {
1.47 daniel 2835:
1.1 daniel 2836: /*
2837: * Well formedness requires at most one declaration of an attribute
2838: */
2839: for (i = 0; i < nbatts;i += 2) {
1.73 ! veillard 2840: if (xmlStrEqual(atts[i], attname)) {
1.1 daniel 2841: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.19 daniel 2842: ctxt->sax->error(ctxt->userData,
2843: "Attribute %s redefined\n",
2844: attname);
1.1 daniel 2845: ctxt->wellFormed = 0;
1.11 daniel 2846: xmlFree(attname);
1.31 daniel 2847: if (attvalue != NULL)
2848: xmlFree(attvalue);
1.19 daniel 2849: goto failed;
1.1 daniel 2850: }
2851: }
2852:
2853: /*
2854: * Add the pair to atts
2855: */
2856: if (atts == NULL) {
2857: maxatts = 10;
1.14 daniel 2858: atts = (const xmlChar **) xmlMalloc(maxatts * sizeof(xmlChar *));
1.1 daniel 2859: if (atts == NULL) {
2860: fprintf(stderr, "malloc of %ld byte failed\n",
1.14 daniel 2861: maxatts * (long)sizeof(xmlChar *));
1.18 daniel 2862: if (name != NULL) xmlFree(name);
2863: return;
1.1 daniel 2864: }
1.23 daniel 2865: } else if (nbatts + 4 > maxatts) {
1.1 daniel 2866: maxatts *= 2;
1.71 veillard 2867: atts = (const xmlChar **) xmlRealloc((void *) atts,
2868: maxatts * sizeof(xmlChar *));
1.1 daniel 2869: if (atts == NULL) {
2870: fprintf(stderr, "realloc of %ld byte failed\n",
1.14 daniel 2871: maxatts * (long)sizeof(xmlChar *));
1.18 daniel 2872: if (name != NULL) xmlFree(name);
2873: return;
1.1 daniel 2874: }
2875: }
2876: atts[nbatts++] = attname;
2877: atts[nbatts++] = attvalue;
2878: atts[nbatts] = NULL;
2879: atts[nbatts + 1] = NULL;
2880: }
2881:
1.19 daniel 2882: failed:
1.1 daniel 2883: SKIP_BLANKS;
1.26 daniel 2884: if (cons == ctxt->nbChars) {
1.1 daniel 2885: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2886: ctxt->sax->error(ctxt->userData,
2887: "htmlParseStartTag: problem parsing attributes\n");
2888: ctxt->wellFormed = 0;
2889: break;
2890: }
2891: }
2892:
2893: /*
1.47 daniel 2894: * Handle specific association to the META tag
2895: */
2896: if (meta)
2897: htmlCheckMeta(ctxt, atts);
2898:
2899: /*
1.1 daniel 2900: * SAX: Start of Element !
2901: */
1.15 daniel 2902: htmlnamePush(ctxt, xmlStrdup(name));
1.18 daniel 2903: #ifdef DEBUG
2904: fprintf(stderr,"Start of element %s: pushed %s\n", name, ctxt->name);
2905: #endif
1.1 daniel 2906: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
2907: ctxt->sax->startElement(ctxt->userData, name, atts);
2908:
2909: if (atts != NULL) {
1.31 daniel 2910: for (i = 0;i < nbatts;i++) {
2911: if (atts[i] != NULL)
2912: xmlFree((xmlChar *) atts[i]);
2913: }
1.45 daniel 2914: xmlFree((void *) atts);
1.1 daniel 2915: }
1.18 daniel 2916: if (name != NULL) xmlFree(name);
1.1 daniel 2917: }
2918:
2919: /**
2920: * htmlParseEndTag:
2921: * @ctxt: an HTML parser context
2922: *
2923: * parse an end of tag
2924: *
2925: * [42] ETag ::= '</' Name S? '>'
2926: *
2927: * With namespace
2928: *
2929: * [NS 9] ETag ::= '</' QName S? '>'
2930: */
2931:
2932: void
1.18 daniel 2933: htmlParseEndTag(htmlParserCtxtPtr ctxt) {
1.14 daniel 2934: xmlChar *name;
1.15 daniel 2935: xmlChar *oldname;
1.1 daniel 2936: int i;
2937:
2938: if ((CUR != '<') || (NXT(1) != '/')) {
2939: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2940: ctxt->sax->error(ctxt->userData, "htmlParseEndTag: '</' not found\n");
2941: ctxt->wellFormed = 0;
2942: return;
2943: }
2944: SKIP(2);
2945:
2946: name = htmlParseHTMLName(ctxt);
1.24 daniel 2947: if (name == NULL) return;
1.1 daniel 2948:
2949: /*
2950: * We should definitely be at the ending "S? '>'" part
2951: */
2952: SKIP_BLANKS;
2953: if ((!IS_CHAR(CUR)) || (CUR != '>')) {
2954: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2955: ctxt->sax->error(ctxt->userData, "End tag : expected '>'\n");
2956: ctxt->wellFormed = 0;
2957: } else
2958: NEXT;
2959:
2960: /*
1.18 daniel 2961: * If the name read is not one of the element in the parsing stack
2962: * then return, it's just an error.
1.1 daniel 2963: */
1.18 daniel 2964: for (i = (ctxt->nameNr - 1);i >= 0;i--) {
1.73 ! veillard 2965: if (xmlStrEqual(name, ctxt->nameTab[i])) break;
1.1 daniel 2966: }
2967: if (i < 0) {
2968: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.18 daniel 2969: ctxt->sax->error(ctxt->userData,
2970: "Unexpected end tag : %s\n", name);
1.11 daniel 2971: xmlFree(name);
1.1 daniel 2972: ctxt->wellFormed = 0;
2973: return;
2974: }
2975:
1.18 daniel 2976:
1.1 daniel 2977: /*
2978: * Check for auto-closure of HTML elements.
2979: */
1.18 daniel 2980:
1.1 daniel 2981: htmlAutoCloseOnClose(ctxt, name);
2982:
2983: /*
2984: * Well formedness constraints, opening and closing must match.
2985: * With the exception that the autoclose may have popped stuff out
2986: * of the stack.
2987: */
1.73 ! veillard 2988: if (!xmlStrEqual(name, ctxt->name)) {
1.18 daniel 2989: #ifdef DEBUG
2990: fprintf(stderr,"End of tag %s: expecting %s\n", name, ctxt->name);
2991: #endif
1.15 daniel 2992: if ((ctxt->name != NULL) &&
1.73 ! veillard 2993: (!xmlStrEqual(ctxt->name, name))) {
1.1 daniel 2994: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2995: ctxt->sax->error(ctxt->userData,
2996: "Opening and ending tag mismatch: %s and %s\n",
1.15 daniel 2997: name, ctxt->name);
1.1 daniel 2998: ctxt->wellFormed = 0;
2999: }
3000: }
3001:
3002: /*
3003: * SAX: End of Tag
3004: */
1.15 daniel 3005: oldname = ctxt->name;
1.73 ! veillard 3006: if ((oldname != NULL) && (xmlStrEqual(oldname, name))) {
1.18 daniel 3007: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3008: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 3009: oldname = htmlnamePop(ctxt);
1.18 daniel 3010: if (oldname != NULL) {
3011: #ifdef DEBUG
3012: fprintf(stderr,"End of tag %s: popping out %s\n", name, oldname);
3013: #endif
3014: xmlFree(oldname);
3015: #ifdef DEBUG
3016: } else {
3017: fprintf(stderr,"End of tag %s: stack empty !!!\n", name);
3018: #endif
3019: }
3020: }
1.1 daniel 3021:
3022: if (name != NULL)
1.11 daniel 3023: xmlFree(name);
1.1 daniel 3024:
3025: return;
3026: }
3027:
3028:
3029: /**
3030: * htmlParseReference:
3031: * @ctxt: an HTML parser context
3032: *
3033: * parse and handle entity references in content,
3034: * this will end-up in a call to character() since this is either a
3035: * CharRef, or a predefined entity.
3036: */
3037: void
3038: htmlParseReference(htmlParserCtxtPtr ctxt) {
3039: htmlEntityDescPtr ent;
1.52 veillard 3040: xmlChar out[6];
1.14 daniel 3041: xmlChar *name;
1.1 daniel 3042: if (CUR != '&') return;
3043:
3044: if (NXT(1) == '#') {
1.52 veillard 3045: unsigned int c;
3046: int bits, i = 0;
3047:
3048: c = htmlParseCharRef(ctxt);
3049: if (c < 0x80) { out[i++]= c; bits= -6; }
3050: else if (c < 0x800) { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3051: else if (c < 0x10000) { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3052: else { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3053:
3054: for ( ; bits >= 0; bits-= 6) {
3055: out[i++]= ((c >> bits) & 0x3F) | 0x80;
3056: }
3057: out[i] = 0;
3058:
1.59 veillard 3059: htmlCheckParagraph(ctxt);
1.1 daniel 3060: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1.52 veillard 3061: ctxt->sax->characters(ctxt->userData, out, i);
1.1 daniel 3062: } else {
3063: ent = htmlParseEntityRef(ctxt, &name);
1.32 daniel 3064: if (name == NULL) {
1.59 veillard 3065: htmlCheckParagraph(ctxt);
1.58 veillard 3066: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3067: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
1.32 daniel 3068: return;
3069: }
1.52 veillard 3070: if ((ent == NULL) || (ent->value <= 0)) {
1.59 veillard 3071: htmlCheckParagraph(ctxt);
1.1 daniel 3072: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
1.8 daniel 3073: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
1.1 daniel 3074: ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
1.32 daniel 3075: /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
1.1 daniel 3076: }
3077: } else {
1.52 veillard 3078: unsigned int c;
3079: int bits, i = 0;
3080:
3081: c = ent->value;
3082: if (c < 0x80)
3083: { out[i++]= c; bits= -6; }
3084: else if (c < 0x800)
3085: { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3086: else if (c < 0x10000)
3087: { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3088: else
3089: { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3090:
3091: for ( ; bits >= 0; bits-= 6) {
3092: out[i++]= ((c >> bits) & 0x3F) | 0x80;
3093: }
3094: out[i] = 0;
3095:
1.59 veillard 3096: htmlCheckParagraph(ctxt);
1.1 daniel 3097: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1.52 veillard 3098: ctxt->sax->characters(ctxt->userData, out, i);
1.1 daniel 3099: }
1.11 daniel 3100: xmlFree(name);
1.1 daniel 3101: }
3102: }
3103:
3104: /**
3105: * htmlParseContent:
3106: * @ctxt: an HTML parser context
3107: * @name: the node name
3108: *
3109: * Parse a content: comment, sub-element, reference or text.
3110: *
3111: */
3112:
3113: void
1.18 daniel 3114: htmlParseContent(htmlParserCtxtPtr ctxt) {
1.15 daniel 3115: xmlChar *currentNode;
1.18 daniel 3116: int depth;
1.1 daniel 3117:
1.26 daniel 3118: currentNode = xmlStrdup(ctxt->name);
1.18 daniel 3119: depth = ctxt->nameNr;
3120: while (1) {
1.26 daniel 3121: long cons = ctxt->nbChars;
1.1 daniel 3122:
1.18 daniel 3123: GROW;
3124: /*
3125: * Our tag or one of it's parent or children is ending.
3126: */
3127: if ((CUR == '<') && (NXT(1) == '/')) {
3128: htmlParseEndTag(ctxt);
1.26 daniel 3129: if (currentNode != NULL) xmlFree(currentNode);
1.18 daniel 3130: return;
3131: }
3132:
3133: /*
3134: * Has this node been popped out during parsing of
3135: * the next element
3136: */
1.73 ! veillard 3137: if ((!xmlStrEqual(currentNode, ctxt->name)) &&
1.26 daniel 3138: (depth >= ctxt->nameNr)) {
3139: if (currentNode != NULL) xmlFree(currentNode);
3140: return;
3141: }
1.18 daniel 3142:
1.1 daniel 3143: /*
1.59 veillard 3144: * Sometimes DOCTYPE arrives in the middle of the document
3145: */
3146: if ((CUR == '<') && (NXT(1) == '!') &&
3147: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3148: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3149: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3150: (UPP(8) == 'E')) {
3151: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3152: ctxt->sax->error(ctxt->userData,
3153: "Misplaced DOCTYPE declaration\n");
3154: ctxt->wellFormed = 0;
3155: htmlParseDocTypeDecl(ctxt);
3156: }
3157:
3158: /*
1.1 daniel 3159: * First case : a comment
3160: */
3161: if ((CUR == '<') && (NXT(1) == '!') &&
3162: (NXT(2) == '-') && (NXT(3) == '-')) {
1.31 daniel 3163: htmlParseComment(ctxt);
1.1 daniel 3164: }
3165:
3166: /*
3167: * Second case : a sub-element.
3168: */
3169: else if (CUR == '<') {
3170: htmlParseElement(ctxt);
3171: }
3172:
3173: /*
3174: * Third case : a reference. If if has not been resolved,
3175: * parsing returns it's Name, create the node
3176: */
3177: else if (CUR == '&') {
3178: htmlParseReference(ctxt);
3179: }
3180:
3181: /*
1.47 daniel 3182: * Fourth : end of the resource
3183: */
3184: else if (CUR == 0) {
3185: htmlAutoClose(ctxt, NULL);
3186: }
3187:
3188: /*
1.1 daniel 3189: * Last case, text. Note that References are handled directly.
3190: */
3191: else {
3192: htmlParseCharData(ctxt, 0);
3193: }
3194:
1.26 daniel 3195: if (cons == ctxt->nbChars) {
1.22 daniel 3196: if (ctxt->node != NULL) {
3197: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3198: ctxt->sax->error(ctxt->userData,
3199: "detected an error in element content\n");
3200: ctxt->wellFormed = 0;
3201: }
1.1 daniel 3202: break;
3203: }
1.17 daniel 3204:
1.5 daniel 3205: GROW;
1.1 daniel 3206: }
1.26 daniel 3207: if (currentNode != NULL) xmlFree(currentNode);
1.1 daniel 3208: }
3209:
3210: /**
3211: * htmlParseElement:
3212: * @ctxt: an HTML parser context
3213: *
3214: * parse an HTML element, this is highly recursive
3215: *
3216: * [39] element ::= EmptyElemTag | STag content ETag
3217: *
3218: * [41] Attribute ::= Name Eq AttValue
3219: */
3220:
3221: void
3222: htmlParseElement(htmlParserCtxtPtr ctxt) {
1.14 daniel 3223: xmlChar *name;
1.16 daniel 3224: xmlChar *currentNode = NULL;
1.1 daniel 3225: htmlElemDescPtr info;
1.10 daniel 3226: htmlParserNodeInfo node_info;
1.31 daniel 3227: xmlChar *oldname;
1.18 daniel 3228: int depth = ctxt->nameNr;
1.1 daniel 3229:
3230: /* Capture start position */
1.10 daniel 3231: if (ctxt->record_info) {
3232: node_info.begin_pos = ctxt->input->consumed +
3233: (CUR_PTR - ctxt->input->base);
3234: node_info.begin_line = ctxt->input->line;
3235: }
1.1 daniel 3236:
1.26 daniel 3237: oldname = xmlStrdup(ctxt->name);
1.18 daniel 3238: htmlParseStartTag(ctxt);
3239: name = ctxt->name;
1.19 daniel 3240: #ifdef DEBUG
3241: if (oldname == NULL)
3242: fprintf(stderr, "Start of element %s\n", name);
3243: else if (name == NULL)
3244: fprintf(stderr, "Start of element failed, was %s\n", oldname);
3245: else
3246: fprintf(stderr, "Start of element %s, was %s\n", name, oldname);
3247: #endif
1.73 ! veillard 3248: if (((depth == ctxt->nameNr) && (xmlStrEqual(oldname, ctxt->name))) ||
1.18 daniel 3249: (name == NULL)) {
1.19 daniel 3250: if (CUR == '>')
3251: NEXT;
1.26 daniel 3252: if (oldname != NULL)
3253: xmlFree(oldname);
1.1 daniel 3254: return;
3255: }
1.26 daniel 3256: if (oldname != NULL)
3257: xmlFree(oldname);
1.1 daniel 3258:
3259: /*
3260: * Lookup the info for that element.
3261: */
3262: info = htmlTagLookup(name);
3263: if (info == NULL) {
3264: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3265: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
3266: name);
3267: ctxt->wellFormed = 0;
3268: } else if (info->depr) {
3269: /***************************
3270: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
3271: ctxt->sax->warning(ctxt->userData, "Tag %s is deprecated\n",
3272: name);
3273: ***************************/
3274: }
3275:
3276: /*
3277: * Check for an Empty Element labelled the XML/SGML way
3278: */
3279: if ((CUR == '/') && (NXT(1) == '>')) {
3280: SKIP(2);
3281: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3282: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 3283: oldname = htmlnamePop(ctxt);
1.18 daniel 3284: #ifdef DEBUG
3285: fprintf(stderr,"End of tag the XML way: popping out %s\n", oldname);
3286: #endif
1.17 daniel 3287: if (oldname != NULL)
3288: xmlFree(oldname);
1.1 daniel 3289: return;
3290: }
3291:
1.5 daniel 3292: if (CUR == '>') {
3293: NEXT;
3294: } else {
1.1 daniel 3295: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.56 veillard 3296: ctxt->sax->error(ctxt->userData,
3297: "Couldn't find end of Start Tag %s\n",
3298: name);
1.1 daniel 3299: ctxt->wellFormed = 0;
3300:
3301: /*
3302: * end of parsing of this node.
3303: */
1.73 ! veillard 3304: if (xmlStrEqual(name, ctxt->name)) {
1.18 daniel 3305: nodePop(ctxt);
1.24 daniel 3306: oldname = htmlnamePop(ctxt);
1.18 daniel 3307: #ifdef DEBUG
3308: fprintf(stderr,"End of start tag problem: popping out %s\n", oldname);
3309: #endif
3310: if (oldname != NULL)
3311: xmlFree(oldname);
3312: }
1.10 daniel 3313:
3314: /*
3315: * Capture end position and add node
3316: */
3317: if ( currentNode != NULL && ctxt->record_info ) {
3318: node_info.end_pos = ctxt->input->consumed +
3319: (CUR_PTR - ctxt->input->base);
3320: node_info.end_line = ctxt->input->line;
1.15 daniel 3321: node_info.node = ctxt->node;
1.10 daniel 3322: xmlParserAddNodeInfo(ctxt, &node_info);
3323: }
1.1 daniel 3324: return;
3325: }
3326:
3327: /*
3328: * Check for an Empty Element from DTD definition
3329: */
3330: if ((info != NULL) && (info->empty)) {
3331: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3332: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 3333: oldname = htmlnamePop(ctxt);
1.18 daniel 3334: #ifdef DEBUG
3335: fprintf(stderr,"End of empty tag %s : popping out %s\n", name, oldname);
3336: #endif
1.17 daniel 3337: if (oldname != NULL)
3338: xmlFree(oldname);
1.1 daniel 3339: return;
3340: }
3341:
3342: /*
3343: * Parse the content of the element:
3344: */
1.26 daniel 3345: currentNode = xmlStrdup(ctxt->name);
1.18 daniel 3346: depth = ctxt->nameNr;
3347: while (IS_CHAR(CUR)) {
3348: htmlParseContent(ctxt);
3349: if (ctxt->nameNr < depth) break;
3350: }
1.1 daniel 3351:
3352: if (!IS_CHAR(CUR)) {
1.49 daniel 3353: /************
1.1 daniel 3354: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3355: ctxt->sax->error(ctxt->userData,
1.18 daniel 3356: "Premature end of data in tag %s\n", currentNode);
1.1 daniel 3357: ctxt->wellFormed = 0;
1.49 daniel 3358: *************/
1.1 daniel 3359:
3360: /*
3361: * end of parsing of this node.
3362: */
3363: nodePop(ctxt);
1.24 daniel 3364: oldname = htmlnamePop(ctxt);
1.18 daniel 3365: #ifdef DEBUG
3366: fprintf(stderr,"Premature end of tag %s : popping out %s\n", name, oldname);
3367: #endif
1.17 daniel 3368: if (oldname != NULL)
3369: xmlFree(oldname);
1.26 daniel 3370: if (currentNode != NULL)
3371: xmlFree(currentNode);
1.1 daniel 3372: return;
3373: }
1.10 daniel 3374:
3375: /*
3376: * Capture end position and add node
3377: */
3378: if ( currentNode != NULL && ctxt->record_info ) {
3379: node_info.end_pos = ctxt->input->consumed +
3380: (CUR_PTR - ctxt->input->base);
3381: node_info.end_line = ctxt->input->line;
1.15 daniel 3382: node_info.node = ctxt->node;
1.10 daniel 3383: xmlParserAddNodeInfo(ctxt, &node_info);
3384: }
1.26 daniel 3385: if (currentNode != NULL)
3386: xmlFree(currentNode);
1.1 daniel 3387: }
3388:
3389: /**
3390: * htmlParseDocument :
3391: * @ctxt: an HTML parser context
3392: *
3393: * parse an HTML document (and build a tree if using the standard SAX
3394: * interface).
3395: *
3396: * Returns 0, -1 in case of error. the parser context is augmented
3397: * as a result of the parsing.
3398: */
3399:
3400: int
3401: htmlParseDocument(htmlParserCtxtPtr ctxt) {
1.59 veillard 3402: xmlDtdPtr dtd;
3403:
1.1 daniel 3404: htmlDefaultSAXHandlerInit();
3405: ctxt->html = 1;
3406:
1.5 daniel 3407: GROW;
1.1 daniel 3408: /*
1.9 daniel 3409: * SAX: beginning of the document processing.
1.1 daniel 3410: */
3411: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
3412: ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
3413:
3414: /*
3415: * Wipe out everything which is before the first '<'
3416: */
1.22 daniel 3417: SKIP_BLANKS;
1.1 daniel 3418: if (CUR == 0) {
3419: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3420: ctxt->sax->error(ctxt->userData, "Document is empty\n");
3421: ctxt->wellFormed = 0;
3422: }
3423:
1.40 daniel 3424: if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX))
3425: ctxt->sax->startDocument(ctxt->userData);
3426:
3427:
1.22 daniel 3428: /*
3429: * Parse possible comments before any content
3430: */
3431: while ((CUR == '<') && (NXT(1) == '!') &&
3432: (NXT(2) == '-') && (NXT(3) == '-')) {
1.31 daniel 3433: htmlParseComment(ctxt);
1.22 daniel 3434: SKIP_BLANKS;
3435: }
3436:
1.1 daniel 3437:
3438: /*
3439: * Then possibly doc type declaration(s) and more Misc
3440: * (doctypedecl Misc*)?
3441: */
3442: if ((CUR == '<') && (NXT(1) == '!') &&
3443: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3444: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3445: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3446: (UPP(8) == 'E')) {
3447: htmlParseDocTypeDecl(ctxt);
3448: }
3449: SKIP_BLANKS;
3450:
3451: /*
1.55 veillard 3452: * Parse possible comments before any content
3453: */
3454: while ((CUR == '<') && (NXT(1) == '!') &&
3455: (NXT(2) == '-') && (NXT(3) == '-')) {
3456: htmlParseComment(ctxt);
3457: SKIP_BLANKS;
3458: }
3459:
3460: /*
1.1 daniel 3461: * Time to start parsing the tree itself
3462: */
1.22 daniel 3463: htmlParseContent(ctxt);
1.1 daniel 3464:
3465: /*
1.47 daniel 3466: * autoclose
3467: */
3468: if (CUR == 0)
3469: htmlAutoClose(ctxt, NULL);
3470:
3471:
3472: /*
1.1 daniel 3473: * SAX: end of the document processing.
3474: */
3475: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3476: ctxt->sax->endDocument(ctxt->userData);
1.59 veillard 3477:
3478: if (ctxt->myDoc != NULL) {
3479: dtd = xmlGetIntSubset(ctxt->myDoc);
3480: if (dtd == NULL)
3481: ctxt->myDoc->intSubset =
3482: xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML",
3483: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
3484: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
3485: }
1.1 daniel 3486: if (! ctxt->wellFormed) return(-1);
3487: return(0);
3488: }
3489:
3490:
1.30 daniel 3491: /************************************************************************
3492: * *
3493: * Parser contexts handling *
3494: * *
3495: ************************************************************************/
1.1 daniel 3496:
3497: /**
3498: * xmlInitParserCtxt:
3499: * @ctxt: an HTML parser context
3500: *
3501: * Initialize a parser context
3502: */
3503:
3504: void
3505: htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
3506: {
3507: htmlSAXHandler *sax;
3508:
1.21 daniel 3509: if (ctxt == NULL) return;
3510: memset(ctxt, 0, sizeof(htmlParserCtxt));
3511:
1.11 daniel 3512: sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
1.1 daniel 3513: if (sax == NULL) {
3514: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
3515: }
1.68 veillard 3516: else
3517: memset(sax, 0, sizeof(htmlSAXHandler));
1.1 daniel 3518:
3519: /* Allocate the Input stack */
1.19 daniel 3520: ctxt->inputTab = (htmlParserInputPtr *)
3521: xmlMalloc(5 * sizeof(htmlParserInputPtr));
3522: if (ctxt->inputTab == NULL) {
3523: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
1.65 veillard 3524: ctxt->inputNr = 0;
3525: ctxt->inputMax = 0;
3526: ctxt->input = NULL;
3527: return;
1.19 daniel 3528: }
1.1 daniel 3529: ctxt->inputNr = 0;
3530: ctxt->inputMax = 5;
3531: ctxt->input = NULL;
3532: ctxt->version = NULL;
3533: ctxt->encoding = NULL;
3534: ctxt->standalone = -1;
1.30 daniel 3535: ctxt->instate = XML_PARSER_START;
1.1 daniel 3536:
3537: /* Allocate the Node stack */
1.11 daniel 3538: ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
1.65 veillard 3539: if (ctxt->nodeTab == NULL) {
3540: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
3541: ctxt->nodeNr = 0;
3542: ctxt->nodeMax = 0;
3543: ctxt->node = NULL;
3544: ctxt->inputNr = 0;
3545: ctxt->inputMax = 0;
3546: ctxt->input = NULL;
3547: return;
3548: }
1.1 daniel 3549: ctxt->nodeNr = 0;
3550: ctxt->nodeMax = 10;
3551: ctxt->node = NULL;
3552:
1.15 daniel 3553: /* Allocate the Name stack */
3554: ctxt->nameTab = (xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
1.65 veillard 3555: if (ctxt->nameTab == NULL) {
3556: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
3557: ctxt->nameNr = 0;
3558: ctxt->nameMax = 10;
3559: ctxt->name = NULL;
3560: ctxt->nodeNr = 0;
3561: ctxt->nodeMax = 0;
3562: ctxt->node = NULL;
3563: ctxt->inputNr = 0;
3564: ctxt->inputMax = 0;
3565: ctxt->input = NULL;
3566: return;
3567: }
1.15 daniel 3568: ctxt->nameNr = 0;
3569: ctxt->nameMax = 10;
3570: ctxt->name = NULL;
3571:
1.1 daniel 3572: if (sax == NULL) ctxt->sax = &htmlDefaultSAXHandler;
3573: else {
3574: ctxt->sax = sax;
3575: memcpy(sax, &htmlDefaultSAXHandler, sizeof(htmlSAXHandler));
3576: }
3577: ctxt->userData = ctxt;
3578: ctxt->myDoc = NULL;
3579: ctxt->wellFormed = 1;
3580: ctxt->replaceEntities = 0;
3581: ctxt->html = 1;
3582: ctxt->record_info = 0;
1.21 daniel 3583: ctxt->validate = 0;
1.26 daniel 3584: ctxt->nbChars = 0;
1.30 daniel 3585: ctxt->checkIndex = 0;
1.1 daniel 3586: xmlInitNodeInfoSeq(&ctxt->node_seq);
3587: }
3588:
3589: /**
3590: * htmlFreeParserCtxt:
3591: * @ctxt: an HTML parser context
3592: *
3593: * Free all the memory used by a parser context. However the parsed
3594: * document in ctxt->myDoc is not freed.
3595: */
3596:
3597: void
3598: htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
3599: {
1.47 daniel 3600: xmlFreeParserCtxt(ctxt);
1.1 daniel 3601: }
3602:
3603: /**
3604: * htmlCreateDocParserCtxt :
1.14 daniel 3605: * @cur: a pointer to an array of xmlChar
1.1 daniel 3606: * @encoding: a free form C string describing the HTML document encoding, or NULL
3607: *
3608: * Create a parser context for an HTML document.
3609: *
3610: * Returns the new parser context or NULL
3611: */
3612: htmlParserCtxtPtr
1.14 daniel 3613: htmlCreateDocParserCtxt(xmlChar *cur, const char *encoding) {
1.1 daniel 3614: htmlParserCtxtPtr ctxt;
3615: htmlParserInputPtr input;
3616: /* htmlCharEncoding enc; */
3617:
1.11 daniel 3618: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 3619: if (ctxt == NULL) {
3620: perror("malloc");
3621: return(NULL);
3622: }
3623: htmlInitParserCtxt(ctxt);
1.11 daniel 3624: input = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 3625: if (input == NULL) {
3626: perror("malloc");
1.11 daniel 3627: xmlFree(ctxt);
1.1 daniel 3628: return(NULL);
3629: }
1.19 daniel 3630: memset(input, 0, sizeof(htmlParserInput));
1.1 daniel 3631:
3632: input->line = 1;
3633: input->col = 1;
3634: input->base = cur;
3635: input->cur = cur;
3636:
3637: inputPush(ctxt, input);
3638: return(ctxt);
3639: }
3640:
1.31 daniel 3641: /************************************************************************
3642: * *
3643: * Progressive parsing interfaces *
3644: * *
3645: ************************************************************************/
3646:
3647: /**
3648: * htmlParseLookupSequence:
3649: * @ctxt: an HTML parser context
3650: * @first: the first char to lookup
3651: * @next: the next char to lookup or zero
3652: * @third: the next char to lookup or zero
3653: *
3654: * Try to find if a sequence (first, next, third) or just (first next) or
3655: * (first) is available in the input stream.
3656: * This function has a side effect of (possibly) incrementing ctxt->checkIndex
3657: * to avoid rescanning sequences of bytes, it DOES change the state of the
3658: * parser, do not use liberally.
3659: * This is basically similar to xmlParseLookupSequence()
3660: *
3661: * Returns the index to the current parsing point if the full sequence
3662: * is available, -1 otherwise.
3663: */
3664: int
3665: htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first,
3666: xmlChar next, xmlChar third) {
3667: int base, len;
3668: htmlParserInputPtr in;
3669: const xmlChar *buf;
3670:
3671: in = ctxt->input;
3672: if (in == NULL) return(-1);
3673: base = in->cur - in->base;
3674: if (base < 0) return(-1);
3675: if (ctxt->checkIndex > base)
3676: base = ctxt->checkIndex;
3677: if (in->buf == NULL) {
3678: buf = in->base;
3679: len = in->length;
3680: } else {
3681: buf = in->buf->buffer->content;
3682: len = in->buf->buffer->use;
3683: }
3684: /* take into account the sequence length */
3685: if (third) len -= 2;
3686: else if (next) len --;
3687: for (;base < len;base++) {
3688: if (buf[base] == first) {
3689: if (third != 0) {
3690: if ((buf[base + 1] != next) ||
3691: (buf[base + 2] != third)) continue;
3692: } else if (next != 0) {
3693: if (buf[base + 1] != next) continue;
3694: }
3695: ctxt->checkIndex = 0;
3696: #ifdef DEBUG_PUSH
3697: if (next == 0)
3698: fprintf(stderr, "HPP: lookup '%c' found at %d\n",
3699: first, base);
3700: else if (third == 0)
3701: fprintf(stderr, "HPP: lookup '%c%c' found at %d\n",
3702: first, next, base);
3703: else
3704: fprintf(stderr, "HPP: lookup '%c%c%c' found at %d\n",
3705: first, next, third, base);
3706: #endif
3707: return(base - (in->cur - in->base));
3708: }
3709: }
3710: ctxt->checkIndex = base;
3711: #ifdef DEBUG_PUSH
3712: if (next == 0)
3713: fprintf(stderr, "HPP: lookup '%c' failed\n", first);
3714: else if (third == 0)
3715: fprintf(stderr, "HPP: lookup '%c%c' failed\n", first, next);
3716: else
3717: fprintf(stderr, "HPP: lookup '%c%c%c' failed\n", first, next, third);
3718: #endif
3719: return(-1);
3720: }
3721:
3722: /**
1.32 daniel 3723: * htmlParseTryOrFinish:
1.31 daniel 3724: * @ctxt: an HTML parser context
1.32 daniel 3725: * @terminate: last chunk indicator
1.31 daniel 3726: *
3727: * Try to progress on parsing
3728: *
3729: * Returns zero if no parsing was possible
3730: */
3731: int
1.32 daniel 3732: htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
1.31 daniel 3733: int ret = 0;
3734: htmlParserInputPtr in;
1.47 daniel 3735: int avail = 0;
1.31 daniel 3736: xmlChar cur, next;
3737:
3738: #ifdef DEBUG_PUSH
3739: switch (ctxt->instate) {
3740: case XML_PARSER_EOF:
3741: fprintf(stderr, "HPP: try EOF\n"); break;
3742: case XML_PARSER_START:
3743: fprintf(stderr, "HPP: try START\n"); break;
3744: case XML_PARSER_MISC:
3745: fprintf(stderr, "HPP: try MISC\n");break;
3746: case XML_PARSER_COMMENT:
3747: fprintf(stderr, "HPP: try COMMENT\n");break;
3748: case XML_PARSER_PROLOG:
3749: fprintf(stderr, "HPP: try PROLOG\n");break;
3750: case XML_PARSER_START_TAG:
3751: fprintf(stderr, "HPP: try START_TAG\n");break;
3752: case XML_PARSER_CONTENT:
3753: fprintf(stderr, "HPP: try CONTENT\n");break;
3754: case XML_PARSER_CDATA_SECTION:
3755: fprintf(stderr, "HPP: try CDATA_SECTION\n");break;
3756: case XML_PARSER_END_TAG:
3757: fprintf(stderr, "HPP: try END_TAG\n");break;
3758: case XML_PARSER_ENTITY_DECL:
3759: fprintf(stderr, "HPP: try ENTITY_DECL\n");break;
3760: case XML_PARSER_ENTITY_VALUE:
3761: fprintf(stderr, "HPP: try ENTITY_VALUE\n");break;
3762: case XML_PARSER_ATTRIBUTE_VALUE:
3763: fprintf(stderr, "HPP: try ATTRIBUTE_VALUE\n");break;
3764: case XML_PARSER_DTD:
3765: fprintf(stderr, "HPP: try DTD\n");break;
3766: case XML_PARSER_EPILOG:
3767: fprintf(stderr, "HPP: try EPILOG\n");break;
3768: case XML_PARSER_PI:
3769: fprintf(stderr, "HPP: try PI\n");break;
3770: }
3771: #endif
3772:
3773: while (1) {
3774:
3775: in = ctxt->input;
3776: if (in == NULL) break;
3777: if (in->buf == NULL)
3778: avail = in->length - (in->cur - in->base);
3779: else
3780: avail = in->buf->buffer->use - (in->cur - in->base);
1.47 daniel 3781: if ((avail == 0) && (terminate)) {
3782: htmlAutoClose(ctxt, NULL);
1.54 veillard 3783: if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
3784: /*
3785: * SAX: end of the document processing.
3786: */
1.47 daniel 3787: ctxt->instate = XML_PARSER_EOF;
1.54 veillard 3788: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3789: ctxt->sax->endDocument(ctxt->userData);
3790: }
1.47 daniel 3791: }
1.31 daniel 3792: if (avail < 1)
3793: goto done;
3794: switch (ctxt->instate) {
3795: case XML_PARSER_EOF:
3796: /*
3797: * Document parsing is done !
3798: */
3799: goto done;
3800: case XML_PARSER_START:
3801: /*
3802: * Very first chars read from the document flow.
3803: */
3804: cur = in->cur[0];
3805: if (IS_BLANK(cur)) {
3806: SKIP_BLANKS;
3807: if (in->buf == NULL)
3808: avail = in->length - (in->cur - in->base);
3809: else
3810: avail = in->buf->buffer->use - (in->cur - in->base);
3811: }
3812: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
3813: ctxt->sax->setDocumentLocator(ctxt->userData,
3814: &xmlDefaultSAXLocator);
1.46 daniel 3815: if ((ctxt->sax) && (ctxt->sax->startDocument) &&
3816: (!ctxt->disableSAX))
3817: ctxt->sax->startDocument(ctxt->userData);
3818:
1.31 daniel 3819: cur = in->cur[0];
3820: next = in->cur[1];
3821: if ((cur == '<') && (next == '!') &&
3822: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3823: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3824: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3825: (UPP(8) == 'E')) {
1.32 daniel 3826: if ((!terminate) &&
3827: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3828: goto done;
3829: #ifdef DEBUG_PUSH
3830: fprintf(stderr, "HPP: Parsing internal subset\n");
3831: #endif
3832: htmlParseDocTypeDecl(ctxt);
3833: ctxt->instate = XML_PARSER_PROLOG;
3834: #ifdef DEBUG_PUSH
3835: fprintf(stderr, "HPP: entering PROLOG\n");
3836: #endif
3837: } else {
3838: ctxt->instate = XML_PARSER_MISC;
3839: }
3840: #ifdef DEBUG_PUSH
3841: fprintf(stderr, "HPP: entering MISC\n");
3842: #endif
3843: break;
3844: case XML_PARSER_MISC:
3845: SKIP_BLANKS;
3846: if (in->buf == NULL)
3847: avail = in->length - (in->cur - in->base);
3848: else
3849: avail = in->buf->buffer->use - (in->cur - in->base);
3850: if (avail < 2)
3851: goto done;
3852: cur = in->cur[0];
3853: next = in->cur[1];
3854: if ((cur == '<') && (next == '!') &&
3855: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3856: if ((!terminate) &&
3857: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3858: goto done;
3859: #ifdef DEBUG_PUSH
3860: fprintf(stderr, "HPP: Parsing Comment\n");
3861: #endif
3862: htmlParseComment(ctxt);
3863: ctxt->instate = XML_PARSER_MISC;
3864: } else if ((cur == '<') && (next == '!') &&
3865: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3866: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3867: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3868: (UPP(8) == 'E')) {
1.32 daniel 3869: if ((!terminate) &&
3870: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3871: goto done;
3872: #ifdef DEBUG_PUSH
3873: fprintf(stderr, "HPP: Parsing internal subset\n");
3874: #endif
3875: htmlParseDocTypeDecl(ctxt);
3876: ctxt->instate = XML_PARSER_PROLOG;
3877: #ifdef DEBUG_PUSH
3878: fprintf(stderr, "HPP: entering PROLOG\n");
3879: #endif
3880: } else if ((cur == '<') && (next == '!') &&
3881: (avail < 9)) {
3882: goto done;
3883: } else {
3884: ctxt->instate = XML_PARSER_START_TAG;
3885: #ifdef DEBUG_PUSH
3886: fprintf(stderr, "HPP: entering START_TAG\n");
3887: #endif
3888: }
3889: break;
3890: case XML_PARSER_PROLOG:
3891: SKIP_BLANKS;
3892: if (in->buf == NULL)
3893: avail = in->length - (in->cur - in->base);
3894: else
3895: avail = in->buf->buffer->use - (in->cur - in->base);
3896: if (avail < 2)
3897: goto done;
3898: cur = in->cur[0];
3899: next = in->cur[1];
3900: if ((cur == '<') && (next == '!') &&
3901: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3902: if ((!terminate) &&
3903: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3904: goto done;
3905: #ifdef DEBUG_PUSH
3906: fprintf(stderr, "HPP: Parsing Comment\n");
3907: #endif
3908: htmlParseComment(ctxt);
3909: ctxt->instate = XML_PARSER_PROLOG;
3910: } else if ((cur == '<') && (next == '!') &&
3911: (avail < 4)) {
3912: goto done;
3913: } else {
3914: ctxt->instate = XML_PARSER_START_TAG;
3915: #ifdef DEBUG_PUSH
3916: fprintf(stderr, "HPP: entering START_TAG\n");
3917: #endif
3918: }
3919: break;
3920: case XML_PARSER_EPILOG:
3921: if (in->buf == NULL)
3922: avail = in->length - (in->cur - in->base);
3923: else
3924: avail = in->buf->buffer->use - (in->cur - in->base);
1.55 veillard 3925: if (avail < 1)
3926: goto done;
3927: cur = in->cur[0];
3928: if (IS_BLANK(cur)) {
3929: htmlParseCharData(ctxt, 0);
3930: goto done;
3931: }
1.31 daniel 3932: if (avail < 2)
3933: goto done;
3934: next = in->cur[1];
3935: if ((cur == '<') && (next == '!') &&
3936: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3937: if ((!terminate) &&
3938: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3939: goto done;
3940: #ifdef DEBUG_PUSH
3941: fprintf(stderr, "HPP: Parsing Comment\n");
3942: #endif
3943: htmlParseComment(ctxt);
3944: ctxt->instate = XML_PARSER_EPILOG;
3945: } else if ((cur == '<') && (next == '!') &&
3946: (avail < 4)) {
3947: goto done;
3948: } else {
1.67 veillard 3949: ctxt->errNo = XML_ERR_DOCUMENT_END;
1.31 daniel 3950: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3951: ctxt->sax->error(ctxt->userData,
3952: "Extra content at the end of the document\n");
3953: ctxt->wellFormed = 0;
3954: ctxt->instate = XML_PARSER_EOF;
3955: #ifdef DEBUG_PUSH
3956: fprintf(stderr, "HPP: entering EOF\n");
3957: #endif
3958: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3959: ctxt->sax->endDocument(ctxt->userData);
3960: goto done;
3961: }
3962: break;
3963: case XML_PARSER_START_TAG: {
3964: xmlChar *name, *oldname;
3965: int depth = ctxt->nameNr;
3966: htmlElemDescPtr info;
3967:
3968: if (avail < 2)
3969: goto done;
3970: cur = in->cur[0];
3971: if (cur != '<') {
3972: ctxt->instate = XML_PARSER_CONTENT;
3973: #ifdef DEBUG_PUSH
3974: fprintf(stderr, "HPP: entering CONTENT\n");
3975: #endif
3976: break;
3977: }
1.32 daniel 3978: if ((!terminate) &&
3979: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3980: goto done;
3981:
3982: oldname = xmlStrdup(ctxt->name);
3983: htmlParseStartTag(ctxt);
3984: name = ctxt->name;
3985: #ifdef DEBUG
3986: if (oldname == NULL)
3987: fprintf(stderr, "Start of element %s\n", name);
3988: else if (name == NULL)
3989: fprintf(stderr, "Start of element failed, was %s\n",
3990: oldname);
3991: else
3992: fprintf(stderr, "Start of element %s, was %s\n",
3993: name, oldname);
3994: #endif
3995: if (((depth == ctxt->nameNr) &&
1.73 ! veillard 3996: (xmlStrEqual(oldname, ctxt->name))) ||
1.31 daniel 3997: (name == NULL)) {
3998: if (CUR == '>')
3999: NEXT;
4000: if (oldname != NULL)
4001: xmlFree(oldname);
4002: break;
4003: }
4004: if (oldname != NULL)
4005: xmlFree(oldname);
4006:
4007: /*
4008: * Lookup the info for that element.
4009: */
4010: info = htmlTagLookup(name);
4011: if (info == NULL) {
4012: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4013: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
4014: name);
4015: ctxt->wellFormed = 0;
4016: } else if (info->depr) {
4017: /***************************
4018: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
4019: ctxt->sax->warning(ctxt->userData,
4020: "Tag %s is deprecated\n",
4021: name);
4022: ***************************/
4023: }
4024:
4025: /*
4026: * Check for an Empty Element labelled the XML/SGML way
4027: */
4028: if ((CUR == '/') && (NXT(1) == '>')) {
4029: SKIP(2);
4030: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4031: ctxt->sax->endElement(ctxt->userData, name);
4032: oldname = htmlnamePop(ctxt);
4033: #ifdef DEBUG
4034: fprintf(stderr,"End of tag the XML way: popping out %s\n",
4035: oldname);
4036: #endif
4037: if (oldname != NULL)
4038: xmlFree(oldname);
4039: ctxt->instate = XML_PARSER_CONTENT;
4040: #ifdef DEBUG_PUSH
4041: fprintf(stderr, "HPP: entering CONTENT\n");
4042: #endif
4043: break;
4044: }
4045:
4046: if (CUR == '>') {
4047: NEXT;
4048: } else {
4049: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4050: ctxt->sax->error(ctxt->userData,
4051: "Couldn't find end of Start Tag %s\n",
4052: name);
4053: ctxt->wellFormed = 0;
4054:
4055: /*
4056: * end of parsing of this node.
4057: */
1.73 ! veillard 4058: if (xmlStrEqual(name, ctxt->name)) {
1.31 daniel 4059: nodePop(ctxt);
4060: oldname = htmlnamePop(ctxt);
4061: #ifdef DEBUG
4062: fprintf(stderr,
4063: "End of start tag problem: popping out %s\n", oldname);
4064: #endif
4065: if (oldname != NULL)
4066: xmlFree(oldname);
4067: }
4068:
4069: ctxt->instate = XML_PARSER_CONTENT;
4070: #ifdef DEBUG_PUSH
4071: fprintf(stderr, "HPP: entering CONTENT\n");
4072: #endif
4073: break;
4074: }
4075:
4076: /*
4077: * Check for an Empty Element from DTD definition
4078: */
4079: if ((info != NULL) && (info->empty)) {
4080: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4081: ctxt->sax->endElement(ctxt->userData, name);
4082: oldname = htmlnamePop(ctxt);
4083: #ifdef DEBUG
4084: fprintf(stderr,"End of empty tag %s : popping out %s\n", name, oldname);
4085: #endif
4086: if (oldname != NULL)
4087: xmlFree(oldname);
4088: }
4089: ctxt->instate = XML_PARSER_CONTENT;
4090: #ifdef DEBUG_PUSH
4091: fprintf(stderr, "HPP: entering CONTENT\n");
4092: #endif
4093: break;
4094: }
1.56 veillard 4095: case XML_PARSER_CONTENT: {
4096: long cons;
1.31 daniel 4097: /*
4098: * Handle preparsed entities and charRef
4099: */
4100: if (ctxt->token != 0) {
1.47 daniel 4101: xmlChar chr[2] = { 0 , 0 } ;
1.31 daniel 4102:
1.47 daniel 4103: chr[0] = (xmlChar) ctxt->token;
1.59 veillard 4104: htmlCheckParagraph(ctxt);
1.31 daniel 4105: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1.47 daniel 4106: ctxt->sax->characters(ctxt->userData, chr, 1);
1.31 daniel 4107: ctxt->token = 0;
4108: ctxt->checkIndex = 0;
4109: }
1.47 daniel 4110: if ((avail == 1) && (terminate)) {
4111: cur = in->cur[0];
4112: if ((cur != '<') && (cur != '&')) {
1.48 daniel 4113: if (ctxt->sax != NULL) {
4114: if (IS_BLANK(cur)) {
4115: if (ctxt->sax->ignorableWhitespace != NULL)
4116: ctxt->sax->ignorableWhitespace(
4117: ctxt->userData, &cur, 1);
4118: } else {
1.59 veillard 4119: htmlCheckParagraph(ctxt);
1.48 daniel 4120: if (ctxt->sax->characters != NULL)
4121: ctxt->sax->characters(
4122: ctxt->userData, &cur, 1);
4123: }
4124: }
1.47 daniel 4125: ctxt->token = 0;
4126: ctxt->checkIndex = 0;
4127: NEXT;
4128: }
4129: break;
4130: }
1.31 daniel 4131: if (avail < 2)
4132: goto done;
4133: cur = in->cur[0];
4134: next = in->cur[1];
1.56 veillard 4135: cons = ctxt->nbChars;
1.59 veillard 4136: /*
4137: * Sometimes DOCTYPE arrives in the middle of the document
4138: */
4139: if ((cur == '<') && (next == '!') &&
4140: (UPP(2) == 'D') && (UPP(3) == 'O') &&
4141: (UPP(4) == 'C') && (UPP(5) == 'T') &&
4142: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4143: (UPP(8) == 'E')) {
4144: if ((!terminate) &&
4145: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
4146: goto done;
4147: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4148: ctxt->sax->error(ctxt->userData,
4149: "Misplaced DOCTYPE declaration\n");
4150: ctxt->wellFormed = 0;
4151: htmlParseDocTypeDecl(ctxt);
4152: } else if ((cur == '<') && (next == '!') &&
1.31 daniel 4153: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 4154: if ((!terminate) &&
4155: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 4156: goto done;
4157: #ifdef DEBUG_PUSH
4158: fprintf(stderr, "HPP: Parsing Comment\n");
4159: #endif
4160: htmlParseComment(ctxt);
4161: ctxt->instate = XML_PARSER_CONTENT;
4162: } else if ((cur == '<') && (next == '!') && (avail < 4)) {
4163: goto done;
4164: } else if ((cur == '<') && (next == '/')) {
4165: ctxt->instate = XML_PARSER_END_TAG;
4166: ctxt->checkIndex = 0;
4167: #ifdef DEBUG_PUSH
4168: fprintf(stderr, "HPP: entering END_TAG\n");
4169: #endif
4170: break;
4171: } else if (cur == '<') {
4172: ctxt->instate = XML_PARSER_START_TAG;
4173: ctxt->checkIndex = 0;
4174: #ifdef DEBUG_PUSH
4175: fprintf(stderr, "HPP: entering START_TAG\n");
4176: #endif
4177: break;
4178: } else if (cur == '&') {
1.32 daniel 4179: if ((!terminate) &&
4180: (htmlParseLookupSequence(ctxt, ';', 0, 0) < 0))
1.31 daniel 4181: goto done;
4182: #ifdef DEBUG_PUSH
4183: fprintf(stderr, "HPP: Parsing Reference\n");
4184: #endif
4185: /* TODO: check generation of subtrees if noent !!! */
4186: htmlParseReference(ctxt);
4187: } else {
4188: /* TODO Avoid the extra copy, handle directly !!!!!! */
4189: /*
4190: * Goal of the following test is :
4191: * - minimize calls to the SAX 'character' callback
4192: * when they are mergeable
4193: */
4194: if ((ctxt->inputNr == 1) &&
4195: (avail < HTML_PARSER_BIG_BUFFER_SIZE)) {
1.32 daniel 4196: if ((!terminate) &&
4197: (htmlParseLookupSequence(ctxt, '<', 0, 0) < 0))
1.31 daniel 4198: goto done;
4199: }
4200: ctxt->checkIndex = 0;
4201: #ifdef DEBUG_PUSH
4202: fprintf(stderr, "HPP: Parsing char data\n");
4203: #endif
4204: htmlParseCharData(ctxt, 0);
4205: }
1.56 veillard 4206: if (cons == ctxt->nbChars) {
4207: if (ctxt->node != NULL) {
4208: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4209: ctxt->sax->error(ctxt->userData,
4210: "detected an error in element content\n");
4211: ctxt->wellFormed = 0;
4212: }
1.70 veillard 4213: NEXT;
1.56 veillard 4214: break;
4215: }
4216:
1.31 daniel 4217: break;
1.56 veillard 4218: }
1.31 daniel 4219: case XML_PARSER_END_TAG:
4220: if (avail < 2)
4221: goto done;
1.32 daniel 4222: if ((!terminate) &&
4223: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 4224: goto done;
4225: htmlParseEndTag(ctxt);
4226: if (ctxt->nameNr == 0) {
4227: ctxt->instate = XML_PARSER_EPILOG;
4228: } else {
4229: ctxt->instate = XML_PARSER_CONTENT;
4230: }
4231: ctxt->checkIndex = 0;
4232: #ifdef DEBUG_PUSH
4233: fprintf(stderr, "HPP: entering CONTENT\n");
4234: #endif
4235: break;
4236: case XML_PARSER_CDATA_SECTION:
4237: fprintf(stderr, "HPP: internal error, state == CDATA\n");
4238: ctxt->instate = XML_PARSER_CONTENT;
4239: ctxt->checkIndex = 0;
4240: #ifdef DEBUG_PUSH
4241: fprintf(stderr, "HPP: entering CONTENT\n");
4242: #endif
4243: break;
4244: case XML_PARSER_DTD:
4245: fprintf(stderr, "HPP: internal error, state == DTD\n");
4246: ctxt->instate = XML_PARSER_CONTENT;
4247: ctxt->checkIndex = 0;
4248: #ifdef DEBUG_PUSH
4249: fprintf(stderr, "HPP: entering CONTENT\n");
4250: #endif
4251: break;
4252: case XML_PARSER_COMMENT:
4253: fprintf(stderr, "HPP: internal error, state == COMMENT\n");
4254: ctxt->instate = XML_PARSER_CONTENT;
4255: ctxt->checkIndex = 0;
4256: #ifdef DEBUG_PUSH
4257: fprintf(stderr, "HPP: entering CONTENT\n");
4258: #endif
4259: break;
4260: case XML_PARSER_PI:
4261: fprintf(stderr, "HPP: internal error, state == PI\n");
4262: ctxt->instate = XML_PARSER_CONTENT;
4263: ctxt->checkIndex = 0;
4264: #ifdef DEBUG_PUSH
4265: fprintf(stderr, "HPP: entering CONTENT\n");
4266: #endif
4267: break;
4268: case XML_PARSER_ENTITY_DECL:
4269: fprintf(stderr, "HPP: internal error, state == ENTITY_DECL\n");
4270: ctxt->instate = XML_PARSER_CONTENT;
4271: ctxt->checkIndex = 0;
4272: #ifdef DEBUG_PUSH
4273: fprintf(stderr, "HPP: entering CONTENT\n");
4274: #endif
4275: break;
4276: case XML_PARSER_ENTITY_VALUE:
4277: fprintf(stderr, "HPP: internal error, state == ENTITY_VALUE\n");
4278: ctxt->instate = XML_PARSER_CONTENT;
4279: ctxt->checkIndex = 0;
4280: #ifdef DEBUG_PUSH
4281: fprintf(stderr, "HPP: entering DTD\n");
4282: #endif
4283: break;
4284: case XML_PARSER_ATTRIBUTE_VALUE:
4285: fprintf(stderr, "HPP: internal error, state == ATTRIBUTE_VALUE\n");
4286: ctxt->instate = XML_PARSER_START_TAG;
4287: ctxt->checkIndex = 0;
4288: #ifdef DEBUG_PUSH
4289: fprintf(stderr, "HPP: entering START_TAG\n");
1.53 veillard 4290: #endif
4291: break;
4292: case XML_PARSER_SYSTEM_LITERAL:
4293: fprintf(stderr, "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n");
4294: ctxt->instate = XML_PARSER_CONTENT;
4295: ctxt->checkIndex = 0;
4296: #ifdef DEBUG_PUSH
4297: fprintf(stderr, "HPP: entering CONTENT\n");
1.31 daniel 4298: #endif
4299: break;
4300: }
4301: }
4302: done:
1.47 daniel 4303: if ((avail == 0) && (terminate)) {
4304: htmlAutoClose(ctxt, NULL);
1.54 veillard 4305: if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
4306: /*
4307: * SAX: end of the document processing.
4308: */
1.47 daniel 4309: ctxt->instate = XML_PARSER_EOF;
1.54 veillard 4310: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4311: ctxt->sax->endDocument(ctxt->userData);
4312: }
1.59 veillard 4313: }
4314: if ((ctxt->myDoc != NULL) &&
4315: ((terminate) || (ctxt->instate == XML_PARSER_EOF) ||
4316: (ctxt->instate == XML_PARSER_EPILOG))) {
4317: xmlDtdPtr dtd;
4318: dtd = xmlGetIntSubset(ctxt->myDoc);
4319: if (dtd == NULL)
4320: ctxt->myDoc->intSubset =
4321: xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML",
4322: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
4323: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
1.47 daniel 4324: }
1.31 daniel 4325: #ifdef DEBUG_PUSH
4326: fprintf(stderr, "HPP: done %d\n", ret);
4327: #endif
4328: return(ret);
4329: }
4330:
4331: /**
1.32 daniel 4332: * htmlParseTry:
4333: * @ctxt: an HTML parser context
4334: *
4335: * Try to progress on parsing
4336: *
4337: * Returns zero if no parsing was possible
4338: */
4339: int
4340: htmlParseTry(htmlParserCtxtPtr ctxt) {
4341: return(htmlParseTryOrFinish(ctxt, 0));
4342: }
4343:
4344: /**
1.31 daniel 4345: * htmlParseChunk:
4346: * @ctxt: an XML parser context
4347: * @chunk: an char array
4348: * @size: the size in byte of the chunk
4349: * @terminate: last chunk indicator
4350: *
4351: * Parse a Chunk of memory
4352: *
4353: * Returns zero if no error, the xmlParserErrors otherwise.
4354: */
4355: int
4356: htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
4357: int terminate) {
4358: if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
4359: (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) {
4360: int base = ctxt->input->base - ctxt->input->buf->buffer->content;
4361: int cur = ctxt->input->cur - ctxt->input->base;
4362:
4363: xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
4364: ctxt->input->base = ctxt->input->buf->buffer->content + base;
4365: ctxt->input->cur = ctxt->input->base + cur;
4366: #ifdef DEBUG_PUSH
4367: fprintf(stderr, "HPP: pushed %d\n", size);
4368: #endif
4369:
1.34 daniel 4370: if ((terminate) || (ctxt->input->buf->buffer->use > 80))
4371: htmlParseTryOrFinish(ctxt, terminate);
1.60 veillard 4372: } else if (ctxt->instate != XML_PARSER_EOF) {
4373: xmlParserInputBufferPush(ctxt->input->buf, 0, "");
1.32 daniel 4374: htmlParseTryOrFinish(ctxt, terminate);
1.60 veillard 4375: }
1.31 daniel 4376: if (terminate) {
4377: if ((ctxt->instate != XML_PARSER_EOF) &&
4378: (ctxt->instate != XML_PARSER_EPILOG) &&
4379: (ctxt->instate != XML_PARSER_MISC)) {
1.67 veillard 4380: ctxt->errNo = XML_ERR_DOCUMENT_END;
1.31 daniel 4381: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4382: ctxt->sax->error(ctxt->userData,
4383: "Extra content at the end of the document\n");
4384: ctxt->wellFormed = 0;
4385: }
4386: if (ctxt->instate != XML_PARSER_EOF) {
4387: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4388: ctxt->sax->endDocument(ctxt->userData);
4389: }
4390: ctxt->instate = XML_PARSER_EOF;
4391: }
4392: return((xmlParserErrors) ctxt->errNo);
4393: }
4394:
4395: /************************************************************************
4396: * *
4397: * User entry points *
4398: * *
4399: ************************************************************************/
4400:
4401: /**
4402: * htmlCreatePushParserCtxt :
4403: * @sax: a SAX handler
4404: * @user_data: The user data returned on SAX callbacks
4405: * @chunk: a pointer to an array of chars
4406: * @size: number of chars in the array
4407: * @filename: an optional file name or URI
4408: * @enc: an optional encoding
4409: *
4410: * Create a parser context for using the HTML parser in push mode
4411: * To allow content encoding detection, @size should be >= 4
4412: * The value of @filename is used for fetching external entities
4413: * and error/warning reports.
4414: *
4415: * Returns the new parser context or NULL
4416: */
4417: htmlParserCtxtPtr
4418: htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data,
4419: const char *chunk, int size, const char *filename,
4420: xmlCharEncoding enc) {
4421: htmlParserCtxtPtr ctxt;
4422: htmlParserInputPtr inputStream;
4423: xmlParserInputBufferPtr buf;
4424:
4425: buf = xmlAllocParserInputBuffer(enc);
4426: if (buf == NULL) return(NULL);
4427:
4428: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
4429: if (ctxt == NULL) {
4430: xmlFree(buf);
4431: return(NULL);
4432: }
4433: memset(ctxt, 0, sizeof(htmlParserCtxt));
4434: htmlInitParserCtxt(ctxt);
4435: if (sax != NULL) {
4436: if (ctxt->sax != &htmlDefaultSAXHandler)
4437: xmlFree(ctxt->sax);
4438: ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler));
4439: if (ctxt->sax == NULL) {
4440: xmlFree(buf);
4441: xmlFree(ctxt);
4442: return(NULL);
4443: }
4444: memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler));
4445: if (user_data != NULL)
4446: ctxt->userData = user_data;
4447: }
4448: if (filename == NULL) {
4449: ctxt->directory = NULL;
4450: } else {
4451: ctxt->directory = xmlParserGetDirectory(filename);
4452: }
4453:
4454: inputStream = htmlNewInputStream(ctxt);
4455: if (inputStream == NULL) {
4456: xmlFreeParserCtxt(ctxt);
4457: return(NULL);
4458: }
4459:
4460: if (filename == NULL)
4461: inputStream->filename = NULL;
4462: else
4463: inputStream->filename = xmlMemStrdup(filename);
4464: inputStream->buf = buf;
4465: inputStream->base = inputStream->buf->buffer->content;
4466: inputStream->cur = inputStream->buf->buffer->content;
4467:
4468: inputPush(ctxt, inputStream);
4469:
4470: if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
4471: (ctxt->input->buf != NULL)) {
4472: xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
4473: #ifdef DEBUG_PUSH
4474: fprintf(stderr, "HPP: pushed %d\n", size);
4475: #endif
4476: }
4477:
4478: return(ctxt);
4479: }
1.1 daniel 4480:
4481: /**
4482: * htmlSAXParseDoc :
1.14 daniel 4483: * @cur: a pointer to an array of xmlChar
1.1 daniel 4484: * @encoding: a free form C string describing the HTML document encoding, or NULL
4485: * @sax: the SAX handler block
4486: * @userData: if using SAX, this pointer will be provided on callbacks.
4487: *
4488: * parse an HTML in-memory document and build a tree.
4489: * It use the given SAX function block to handle the parsing callback.
4490: * If sax is NULL, fallback to the default DOM tree building routines.
4491: *
4492: * Returns the resulting document tree
4493: */
4494:
4495: htmlDocPtr
1.14 daniel 4496: htmlSAXParseDoc(xmlChar *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) {
1.1 daniel 4497: htmlDocPtr ret;
4498: htmlParserCtxtPtr ctxt;
4499:
4500: if (cur == NULL) return(NULL);
4501:
4502:
4503: ctxt = htmlCreateDocParserCtxt(cur, encoding);
4504: if (ctxt == NULL) return(NULL);
4505: if (sax != NULL) {
4506: ctxt->sax = sax;
4507: ctxt->userData = userData;
4508: }
4509:
4510: htmlParseDocument(ctxt);
4511: ret = ctxt->myDoc;
4512: if (sax != NULL) {
4513: ctxt->sax = NULL;
4514: ctxt->userData = NULL;
4515: }
4516: htmlFreeParserCtxt(ctxt);
4517:
4518: return(ret);
4519: }
4520:
4521: /**
4522: * htmlParseDoc :
1.14 daniel 4523: * @cur: a pointer to an array of xmlChar
1.1 daniel 4524: * @encoding: a free form C string describing the HTML document encoding, or NULL
4525: *
4526: * parse an HTML in-memory document and build a tree.
4527: *
4528: * Returns the resulting document tree
4529: */
4530:
4531: htmlDocPtr
1.14 daniel 4532: htmlParseDoc(xmlChar *cur, const char *encoding) {
1.1 daniel 4533: return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
4534: }
4535:
4536:
4537: /**
4538: * htmlCreateFileParserCtxt :
4539: * @filename: the filename
4540: * @encoding: a free form C string describing the HTML document encoding, or NULL
4541: *
4542: * Create a parser context for a file content.
4543: * Automatic support for ZLIB/Compress compressed document is provided
4544: * by default if found at compile-time.
4545: *
4546: * Returns the new parser context or NULL
4547: */
4548: htmlParserCtxtPtr
4549: htmlCreateFileParserCtxt(const char *filename, const char *encoding)
4550: {
4551: htmlParserCtxtPtr ctxt;
4552: htmlParserInputPtr inputStream;
1.5 daniel 4553: xmlParserInputBufferPtr buf;
1.1 daniel 4554: /* htmlCharEncoding enc; */
4555:
1.5 daniel 4556: buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
4557: if (buf == NULL) return(NULL);
1.1 daniel 4558:
1.11 daniel 4559: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 4560: if (ctxt == NULL) {
4561: perror("malloc");
4562: return(NULL);
4563: }
1.19 daniel 4564: memset(ctxt, 0, sizeof(htmlParserCtxt));
1.1 daniel 4565: htmlInitParserCtxt(ctxt);
1.11 daniel 4566: inputStream = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 4567: if (inputStream == NULL) {
4568: perror("malloc");
1.11 daniel 4569: xmlFree(ctxt);
1.1 daniel 4570: return(NULL);
4571: }
1.19 daniel 4572: memset(inputStream, 0, sizeof(htmlParserInput));
1.1 daniel 4573:
1.11 daniel 4574: inputStream->filename = xmlMemStrdup(filename);
1.1 daniel 4575: inputStream->line = 1;
4576: inputStream->col = 1;
1.5 daniel 4577: inputStream->buf = buf;
1.21 daniel 4578: inputStream->directory = NULL;
1.1 daniel 4579:
1.5 daniel 4580: inputStream->base = inputStream->buf->buffer->content;
4581: inputStream->cur = inputStream->buf->buffer->content;
4582: inputStream->free = NULL;
1.1 daniel 4583:
4584: inputPush(ctxt, inputStream);
4585: return(ctxt);
4586: }
4587:
4588: /**
4589: * htmlSAXParseFile :
4590: * @filename: the filename
4591: * @encoding: a free form C string describing the HTML document encoding, or NULL
4592: * @sax: the SAX handler block
4593: * @userData: if using SAX, this pointer will be provided on callbacks.
4594: *
4595: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
4596: * compressed document is provided by default if found at compile-time.
4597: * It use the given SAX function block to handle the parsing callback.
4598: * If sax is NULL, fallback to the default DOM tree building routines.
4599: *
4600: * Returns the resulting document tree
4601: */
4602:
4603: htmlDocPtr
4604: htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
4605: void *userData) {
4606: htmlDocPtr ret;
4607: htmlParserCtxtPtr ctxt;
1.57 veillard 4608: htmlSAXHandlerPtr oldsax = NULL;
1.1 daniel 4609:
4610: ctxt = htmlCreateFileParserCtxt(filename, encoding);
4611: if (ctxt == NULL) return(NULL);
4612: if (sax != NULL) {
1.55 veillard 4613: oldsax = ctxt->sax;
1.1 daniel 4614: ctxt->sax = sax;
4615: ctxt->userData = userData;
4616: }
4617:
4618: htmlParseDocument(ctxt);
4619:
4620: ret = ctxt->myDoc;
4621: if (sax != NULL) {
1.55 veillard 4622: ctxt->sax = oldsax;
1.1 daniel 4623: ctxt->userData = NULL;
4624: }
4625: htmlFreeParserCtxt(ctxt);
4626:
4627: return(ret);
4628: }
4629:
4630: /**
4631: * htmlParseFile :
4632: * @filename: the filename
4633: * @encoding: a free form C string describing the HTML document encoding, or NULL
4634: *
4635: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
4636: * compressed document is provided by default if found at compile-time.
4637: *
4638: * Returns the resulting document tree
4639: */
4640:
4641: htmlDocPtr
4642: htmlParseFile(const char *filename, const char *encoding) {
4643: return(htmlSAXParseFile(filename, encoding, NULL, NULL));
4644: }
1.39 daniel 4645:
4646: #endif /* LIBXML_HTML_ENABLED */
Webmaster