Annotation of XML/HTMLparser.c, revision 1.82
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:
1.75 veillard 15: #include <libxml/xmlversion.h>
1.39 daniel 16: #ifdef LIBXML_HTML_ENABLED
1.1 daniel 17: #include <stdio.h>
1.50 veillard 18: #include <string.h>
1.13 daniel 19: #ifdef HAVE_CTYPE_H
1.1 daniel 20: #include <ctype.h>
1.13 daniel 21: #endif
22: #ifdef HAVE_STDLIB_H
1.1 daniel 23: #include <stdlib.h>
1.13 daniel 24: #endif
25: #ifdef HAVE_SYS_STAT_H
1.1 daniel 26: #include <sys/stat.h>
1.13 daniel 27: #endif
1.1 daniel 28: #ifdef HAVE_FCNTL_H
29: #include <fcntl.h>
30: #endif
31: #ifdef HAVE_UNISTD_H
32: #include <unistd.h>
33: #endif
34: #ifdef HAVE_ZLIB_H
35: #include <zlib.h>
36: #endif
37:
1.39 daniel 38: #include <libxml/xmlmemory.h>
39: #include <libxml/tree.h>
1.75 veillard 40: #include <libxml/parser.h>
41: #include <libxml/parserInternals.h>
42: #include <libxml/xmlerror.h>
1.39 daniel 43: #include <libxml/HTMLparser.h>
44: #include <libxml/entities.h>
45: #include <libxml/encoding.h>
46: #include <libxml/valid.h>
47: #include <libxml/xmlIO.h>
1.5 daniel 48:
49: #define HTML_MAX_NAMELEN 1000
1.53 veillard 50: #define HTML_PARSER_BIG_BUFFER_SIZE 1000
1.31 daniel 51: #define HTML_PARSER_BUFFER_SIZE 100
1.1 daniel 52:
53: /* #define DEBUG */
1.31 daniel 54: /* #define DEBUG_PUSH */
1.1 daniel 55:
56: /************************************************************************
57: * *
58: * Parser stacks related functions and macros *
59: * *
60: ************************************************************************/
61:
62: /*
63: * Generic function for accessing stacks in the Parser Context
64: */
65:
1.30 daniel 66: #define PUSH_AND_POP(scope, type, name) \
67: scope int html##name##Push(htmlParserCtxtPtr ctxt, type value) { \
1.1 daniel 68: if (ctxt->name##Nr >= ctxt->name##Max) { \
69: ctxt->name##Max *= 2; \
1.50 veillard 70: ctxt->name##Tab = (type *) xmlRealloc(ctxt->name##Tab, \
1.1 daniel 71: ctxt->name##Max * sizeof(ctxt->name##Tab[0])); \
72: if (ctxt->name##Tab == NULL) { \
1.81 veillard 73: xmlGenericError(xmlGenericErrorContext, \
74: "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:
1.80 veillard 143: #define SKIP_BLANKS htmlSkipBlankChars(ctxt)
1.53 veillard 144:
145: /* Inported from XML */
146:
147: /* #define CUR (ctxt->token ? ctxt->token : (int) (*ctxt->input->cur)) */
148: #define CUR ((int) (*ctxt->input->cur))
1.80 veillard 149: #define NEXT xmlNextChar(ctxt),ctxt->nbChars++
1.53 veillard 150:
151: #define RAW (ctxt->token ? -1 : (*ctxt->input->cur))
152: #define NXT(val) ctxt->input->cur[(val)]
153: #define CUR_PTR ctxt->input->cur
154:
155:
1.80 veillard 156: #define NEXTL(l) do { \
1.53 veillard 157: if (*(ctxt->input->cur) == '\n') { \
158: ctxt->input->line++; ctxt->input->col = 1; \
159: } else ctxt->input->col++; \
1.80 veillard 160: ctxt->token = 0; ctxt->input->cur += l; ctxt->nbChars++; \
161: } while (0)
1.53 veillard 162:
163: /************
164: \
165: if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
166: if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt);
167: ************/
168:
1.80 veillard 169: #define CUR_CHAR(l) htmlCurrentChar(ctxt, &l)
170: #define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l)
1.53 veillard 171:
172: #define COPY_BUF(l,b,i,v) \
173: if (l == 1) b[i++] = (xmlChar) v; \
1.80 veillard 174: else i += xmlCopyChar(l,&b[i],v)
1.53 veillard 175:
176: /**
177: * htmlCurrentChar:
178: * @ctxt: the HTML parser context
179: * @len: pointer to the length of the char read
180: *
181: * The current char value, if using UTF-8 this may actaully span multiple
182: * bytes in the input buffer. Implement the end of line normalization:
183: * 2.11 End-of-Line Handling
184: * If the encoding is unspecified, in the case we find an ISO-Latin-1
185: * char, then the encoding converter is plugged in automatically.
186: *
187: * Returns the current char value and its lenght
188: */
189:
190: int
191: htmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
192: if (ctxt->instate == XML_PARSER_EOF)
193: return(0);
1.35 daniel 194:
1.53 veillard 195: if (ctxt->token != 0) {
196: *len = 0;
197: return(ctxt->token);
198: }
199: if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
200: /*
201: * We are supposed to handle UTF8, check it's valid
202: * From rfc2044: encoding of the Unicode values on UTF-8:
203: *
204: * UCS-4 range (hex.) UTF-8 octet sequence (binary)
205: * 0000 0000-0000 007F 0xxxxxxx
206: * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
207: * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
208: *
209: * Check for the 0x110000 limit too
210: */
211: const unsigned char *cur = ctxt->input->cur;
212: unsigned char c;
213: unsigned int val;
214:
215: c = *cur;
216: if (c & 0x80) {
217: if (cur[1] == 0)
218: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
219: if ((cur[1] & 0xc0) != 0x80)
220: goto encoding_error;
221: if ((c & 0xe0) == 0xe0) {
222:
223: if (cur[2] == 0)
224: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
225: if ((cur[2] & 0xc0) != 0x80)
226: goto encoding_error;
227: if ((c & 0xf0) == 0xf0) {
228: if (cur[3] == 0)
229: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
230: if (((c & 0xf8) != 0xf0) ||
231: ((cur[3] & 0xc0) != 0x80))
232: goto encoding_error;
233: /* 4-byte code */
234: *len = 4;
235: val = (cur[0] & 0x7) << 18;
236: val |= (cur[1] & 0x3f) << 12;
237: val |= (cur[2] & 0x3f) << 6;
238: val |= cur[3] & 0x3f;
239: } else {
240: /* 3-byte code */
241: *len = 3;
242: val = (cur[0] & 0xf) << 12;
243: val |= (cur[1] & 0x3f) << 6;
244: val |= cur[2] & 0x3f;
245: }
246: } else {
247: /* 2-byte code */
248: *len = 2;
249: val = (cur[0] & 0x1f) << 6;
250: val |= cur[1] & 0x3f;
251: }
252: if (!IS_CHAR(val)) {
1.67 veillard 253: ctxt->errNo = XML_ERR_INVALID_ENCODING;
1.53 veillard 254: if ((ctxt->sax != NULL) &&
255: (ctxt->sax->error != NULL))
256: ctxt->sax->error(ctxt->userData,
257: "Char 0x%X out of allowed range\n", val);
258: ctxt->wellFormed = 0;
259: ctxt->disableSAX = 1;
260: }
261: return(val);
262: } else {
263: /* 1-byte code */
264: *len = 1;
265: return((int) *ctxt->input->cur);
266: }
267: }
268: /*
269: * Assume it's a fixed lenght encoding (1) with
270: * a compatibke encoding for the ASCII set, since
271: * XML constructs only use < 128 chars
272: */
273: *len = 1;
274: if ((int) *ctxt->input->cur < 0x80)
275: return((int) *ctxt->input->cur);
276:
277: /*
278: * Humm this is bad, do an automatic flow conversion
279: */
280: xmlSwitchEncoding(ctxt, XML_CHAR_ENCODING_8859_1);
281: ctxt->charset = XML_CHAR_ENCODING_UTF8;
282: return(xmlCurrentChar(ctxt, len));
283:
284: encoding_error:
285: /*
286: * If we detect an UTF8 error that probably mean that the
287: * input encoding didn't get properly advertized in the
288: * declaration header. Report the error and switch the encoding
289: * to ISO-Latin-1 (if you don't like this policy, just declare the
290: * encoding !)
291: */
1.67 veillard 292: ctxt->errNo = XML_ERR_INVALID_ENCODING;
1.53 veillard 293: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) {
294: ctxt->sax->error(ctxt->userData,
295: "Input is not proper UTF-8, indicate encoding !\n");
296: ctxt->sax->error(ctxt->userData, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
297: ctxt->input->cur[0], ctxt->input->cur[1],
298: ctxt->input->cur[2], ctxt->input->cur[3]);
299: }
300:
301: ctxt->charset = XML_CHAR_ENCODING_8859_1;
302: *len = 1;
303: return((int) *ctxt->input->cur);
304: }
1.35 daniel 305:
306: /**
307: * htmlNextChar:
308: * @ctxt: the HTML parser context
309: *
310: * Skip to the next char input char.
311: */
312:
313: void
314: htmlNextChar(htmlParserCtxtPtr ctxt) {
1.44 daniel 315: if (ctxt->instate == XML_PARSER_EOF)
316: return;
1.35 daniel 317: if ((*ctxt->input->cur == 0) &&
318: (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
319: xmlPopInput(ctxt);
320: } else {
321: if (*(ctxt->input->cur) == '\n') {
322: ctxt->input->line++; ctxt->input->col = 1;
323: } else ctxt->input->col++;
324: ctxt->input->cur++;
325: ctxt->nbChars++;
326: if (*ctxt->input->cur == 0)
327: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
328: }
329: }
1.5 daniel 330:
1.36 daniel 331: /**
332: * htmlSkipBlankChars:
333: * @ctxt: the HTML parser context
334: *
335: * skip all blanks character found at that point in the input streams.
336: *
337: * Returns the number of space chars skipped
338: */
339:
340: int
341: htmlSkipBlankChars(xmlParserCtxtPtr ctxt) {
342: int res = 0;
343:
344: while (IS_BLANK(*(ctxt->input->cur))) {
345: if ((*ctxt->input->cur == 0) &&
346: (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
347: xmlPopInput(ctxt);
348: } else {
349: if (*(ctxt->input->cur) == '\n') {
350: ctxt->input->line++; ctxt->input->col = 1;
351: } else ctxt->input->col++;
352: ctxt->input->cur++;
353: ctxt->nbChars++;
354: if (*ctxt->input->cur == 0)
355: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
356: }
357: res++;
358: }
359: return(res);
360: }
1.1 daniel 361:
362:
1.5 daniel 363:
1.1 daniel 364: /************************************************************************
365: * *
366: * The list of HTML elements and their properties *
367: * *
368: ************************************************************************/
369:
370: /*
371: * Start Tag: 1 means the start tag can be ommited
372: * End Tag: 1 means the end tag can be ommited
373: * 2 means it's forbidden (empty elements)
374: * Depr: this element is deprecated
375: * DTD: 1 means that this element is valid only in the Loose DTD
376: * 2 means that this element is valid only in the Frameset DTD
377: *
378: * Name,Start Tag,End Tag, Empty, Depr., DTD, Description
379: */
380: htmlElemDesc html40ElementTable[] = {
1.26 daniel 381: { "a", 0, 0, 0, 0, 0, "anchor " },
382: { "abbr", 0, 0, 0, 0, 0, "abbreviated form" },
383: { "acronym", 0, 0, 0, 0, 0, "" },
384: { "address", 0, 0, 0, 0, 0, "information on author " },
385: { "applet", 0, 0, 0, 1, 1, "java applet " },
386: { "area", 0, 2, 1, 0, 0, "client-side image map area " },
387: { "b", 0, 0, 0, 0, 0, "bold text style" },
388: { "base", 0, 2, 1, 0, 0, "document base uri " },
389: { "basefont", 0, 2, 1, 1, 1, "base font size " },
390: { "bdo", 0, 0, 0, 0, 0, "i18n bidi over-ride " },
391: { "big", 0, 0, 0, 0, 0, "large text style" },
392: { "blockquote", 0, 0, 0, 0, 0, "long quotation " },
393: { "body", 1, 1, 0, 0, 0, "document body " },
394: { "br", 0, 2, 1, 0, 0, "forced line break " },
395: { "button", 0, 0, 0, 0, 0, "push button " },
396: { "caption", 0, 0, 0, 0, 0, "table caption " },
397: { "center", 0, 0, 0, 1, 1, "shorthand for div align=center " },
398: { "cite", 0, 0, 0, 0, 0, "citation" },
399: { "code", 0, 0, 0, 0, 0, "computer code fragment" },
400: { "col", 0, 2, 1, 0, 0, "table column " },
401: { "colgroup", 0, 1, 0, 0, 0, "table column group " },
402: { "dd", 0, 1, 0, 0, 0, "definition description " },
403: { "del", 0, 0, 0, 0, 0, "deleted text " },
404: { "dfn", 0, 0, 0, 0, 0, "instance definition" },
405: { "dir", 0, 0, 0, 1, 1, "directory list" },
406: { "div", 0, 0, 0, 0, 0, "generic language/style container"},
407: { "dl", 0, 0, 0, 0, 0, "definition list " },
408: { "dt", 0, 1, 0, 0, 0, "definition term " },
409: { "em", 0, 0, 0, 0, 0, "emphasis" },
410: { "fieldset", 0, 0, 0, 0, 0, "form control group " },
411: { "font", 0, 0, 0, 1, 1, "local change to font " },
412: { "form", 0, 0, 0, 0, 0, "interactive form " },
413: { "frame", 0, 2, 1, 0, 2, "subwindow " },
414: { "frameset", 0, 0, 0, 0, 2, "window subdivision" },
415: { "h1", 0, 0, 0, 0, 0, "heading " },
416: { "h2", 0, 0, 0, 0, 0, "heading " },
417: { "h3", 0, 0, 0, 0, 0, "heading " },
418: { "h4", 0, 0, 0, 0, 0, "heading " },
419: { "h5", 0, 0, 0, 0, 0, "heading " },
420: { "h6", 0, 0, 0, 0, 0, "heading " },
421: { "head", 1, 1, 0, 0, 0, "document head " },
422: { "hr", 0, 2, 1, 0, 0, "horizontal rule " },
423: { "html", 1, 1, 0, 0, 0, "document root element " },
424: { "i", 0, 0, 0, 0, 0, "italic text style" },
425: { "iframe", 0, 0, 0, 0, 1, "inline subwindow " },
426: { "img", 0, 2, 1, 0, 0, "embedded image " },
427: { "input", 0, 2, 1, 0, 0, "form control " },
428: { "ins", 0, 0, 0, 0, 0, "inserted text" },
429: { "isindex", 0, 2, 1, 1, 1, "single line prompt " },
430: { "kbd", 0, 0, 0, 0, 0, "text to be entered by the user" },
431: { "label", 0, 0, 0, 0, 0, "form field label text " },
432: { "legend", 0, 0, 0, 0, 0, "fieldset legend " },
433: { "li", 0, 1, 0, 0, 0, "list item " },
434: { "link", 0, 2, 1, 0, 0, "a media-independent link " },
435: { "map", 0, 0, 0, 0, 0, "client-side image map " },
436: { "menu", 0, 0, 0, 1, 1, "menu list " },
437: { "meta", 0, 2, 1, 0, 0, "generic metainformation " },
438: { "noframes", 0, 0, 0, 0, 2, "alternate content container for non frame-based rendering " },
439: { "noscript", 0, 0, 0, 0, 0, "alternate content container for non script-based rendering " },
440: { "object", 0, 0, 0, 0, 0, "generic embedded object " },
441: { "ol", 0, 0, 0, 0, 0, "ordered list " },
442: { "optgroup", 0, 0, 0, 0, 0, "option group " },
443: { "option", 0, 1, 0, 0, 0, "selectable choice " },
444: { "p", 0, 1, 0, 0, 0, "paragraph " },
445: { "param", 0, 2, 1, 0, 0, "named property value " },
446: { "pre", 0, 0, 0, 0, 0, "preformatted text " },
447: { "q", 0, 0, 0, 0, 0, "short inline quotation " },
448: { "s", 0, 0, 0, 1, 1, "strike-through text style" },
449: { "samp", 0, 0, 0, 0, 0, "sample program output, scripts, etc." },
450: { "script", 0, 0, 0, 0, 0, "script statements " },
451: { "select", 0, 0, 0, 0, 0, "option selector " },
452: { "small", 0, 0, 0, 0, 0, "small text style" },
453: { "span", 0, 0, 0, 0, 0, "generic language/style container " },
454: { "strike", 0, 0, 0, 1, 1, "strike-through text" },
455: { "strong", 0, 0, 0, 0, 0, "strong emphasis" },
456: { "style", 0, 0, 0, 0, 0, "style info " },
457: { "sub", 0, 0, 0, 0, 0, "subscript" },
458: { "sup", 0, 0, 0, 0, 0, "superscript " },
459: { "table", 0, 0, 0, 0, 0, " " },
460: { "tbody", 1, 1, 0, 0, 0, "table body " },
461: { "td", 0, 1, 0, 0, 0, "table data cell" },
462: { "textarea", 0, 0, 0, 0, 0, "multi-line text field " },
463: { "tfoot", 0, 1, 0, 0, 0, "table footer " },
464: { "th", 0, 1, 0, 0, 0, "table header cell" },
465: { "thead", 0, 1, 0, 0, 0, "table header " },
466: { "title", 0, 0, 0, 0, 0, "document title " },
467: { "tr", 0, 1, 0, 0, 0, "table row " },
468: { "tt", 0, 0, 0, 0, 0, "teletype or monospaced text style" },
469: { "u", 0, 0, 0, 1, 1, "underlined text style" },
470: { "ul", 0, 0, 0, 0, 0, "unordered list " },
471: { "var", 0, 0, 0, 0, 0, "instance of a variable or program argument" },
1.1 daniel 472: };
473:
474: /*
475: * start tags that imply the end of a current element
476: * any tag of each line implies the end of the current element if the type of
477: * that element is in the same line
478: */
1.8 daniel 479: char *htmlEquEnd[] = {
1.26 daniel 480: "dt", "dd", "li", "option", NULL,
481: "h1", "h2", "h3", "h4", "h5", "h6", NULL,
482: "ol", "menu", "dir", "address", "pre", "listing", "xmp", NULL,
1.1 daniel 483: NULL
484: };
485: /*
486: * acording the HTML DTD, HR should be added to the 2nd line above, as it
487: * is not allowed within a H1, H2, H3, etc. But we should tolerate that case
488: * because many documents contain rules in headings...
489: */
490:
491: /*
492: * start tags that imply the end of current element
493: */
1.8 daniel 494: char *htmlStartClose[] = {
1.26 daniel 495: "form", "form", "p", "hr", "h1", "h2", "h3", "h4", "h5", "h6",
496: "dl", "ul", "ol", "menu", "dir", "address", "pre",
497: "listing", "xmp", "head", NULL,
498: "head", "p", NULL,
499: "title", "p", NULL,
500: "body", "head", "style", "link", "title", "p", NULL,
501: "li", "p", "h1", "h2", "h3", "h4", "h5", "h6", "dl", "address",
502: "pre", "listing", "xmp", "head", "li", NULL,
503: "hr", "p", "head", NULL,
504: "h1", "p", "head", NULL,
505: "h2", "p", "head", NULL,
506: "h3", "p", "head", NULL,
507: "h4", "p", "head", NULL,
508: "h5", "p", "head", NULL,
509: "h6", "p", "head", NULL,
510: "dir", "p", "head", NULL,
511: "address", "p", "head", "ul", NULL,
512: "pre", "p", "head", "ul", NULL,
513: "listing", "p", "head", NULL,
514: "xmp", "p", "head", NULL,
515: "blockquote", "p", "head", NULL,
516: "dl", "p", "dt", "menu", "dir", "address", "pre", "listing",
517: "xmp", "head", NULL,
518: "dt", "p", "menu", "dir", "address", "pre", "listing", "xmp",
519: "head", "dd", NULL,
520: "dd", "p", "menu", "dir", "address", "pre", "listing", "xmp",
521: "head", "dt", NULL,
522: "ul", "p", "head", "ol", "menu", "dir", "address", "pre",
523: "listing", "xmp", NULL,
524: "ol", "p", "head", "ul", NULL,
525: "menu", "p", "head", "ul", NULL,
526: "p", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", NULL,
527: "div", "p", "head", NULL,
528: "noscript", "p", "head", NULL,
529: "center", "font", "b", "i", "p", "head", NULL,
530: "a", "a", NULL,
531: "caption", "p", NULL,
532: "colgroup", "caption", "colgroup", "col", "p", NULL,
533: "col", "caption", "col", "p", NULL,
534: "table", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", "pre",
535: "listing", "xmp", "a", NULL,
536: "th", "th", "td", NULL,
537: "td", "th", "td", "p", NULL,
538: "tr", "th", "td", "tr", "caption", "col", "colgroup", "p", NULL,
539: "thead", "caption", "col", "colgroup", NULL,
540: "tfoot", "th", "td", "tr", "caption", "col", "colgroup", "thead",
541: "tbody", "p", NULL,
542: "tbody", "th", "td", "tr", "caption", "col", "colgroup", "thead",
543: "tfoot", "tbody", "p", NULL,
544: "optgroup", "option", NULL,
1.79 veillard 545: "option", "option", NULL,
1.26 daniel 546: "fieldset", "legend", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6",
547: "pre", "listing", "xmp", "a", NULL,
1.1 daniel 548: NULL
549: };
550:
1.59 veillard 551: /*
552: * The list of HTML elements which are supposed not to have
553: * CDATA content and where a p element will be implied
554: *
555: * TODO: extend that list by reading the HTML SGML DtD on
556: * implied paragraph
557: */
558: static char *htmlNoContentElements[] = {
559: "html",
560: "head",
561: "body",
562: NULL
563: };
564:
1.78 veillard 565: /*
566: * The list of HTML attributes which are of content %Script;
567: * NOTE: when adding ones, check htmlIsScriptAttribute() since
568: * it assumes the name starts with 'on'
569: */
570: static char *htmlScriptAttributes[] = {
571: "onclick",
572: "ondblclick",
573: "onmousedown",
574: "onmouseup",
575: "onmouseover",
576: "onmousemove",
577: "onmouseout",
578: "onkeypress",
579: "onkeydown",
580: "onkeyup",
581: "onload",
582: "onunload",
583: "onfocus",
584: "onblur",
585: "onsubmit",
586: "onrest",
587: "onchange",
588: "onselect"
589: };
590:
591:
1.8 daniel 592: static char** htmlStartCloseIndex[100];
1.1 daniel 593: static int htmlStartCloseIndexinitialized = 0;
594:
595: /************************************************************************
596: * *
597: * functions to handle HTML specific data *
598: * *
599: ************************************************************************/
600:
601: /**
602: * htmlInitAutoClose:
603: *
604: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
1.72 veillard 605: * This is not reentrant. Call xmlInitParser() once before processing in
606: * case of use in multithreaded programs.
1.1 daniel 607: */
608: void
609: htmlInitAutoClose(void) {
610: int index, i = 0;
611:
612: if (htmlStartCloseIndexinitialized) return;
613:
614: for (index = 0;index < 100;index ++) htmlStartCloseIndex[index] = NULL;
615: index = 0;
616: while ((htmlStartClose[i] != NULL) && (index < 100 - 1)) {
617: htmlStartCloseIndex[index++] = &htmlStartClose[i];
618: while (htmlStartClose[i] != NULL) i++;
619: i++;
620: }
1.72 veillard 621: htmlStartCloseIndexinitialized = 1;
1.1 daniel 622: }
623:
624: /**
625: * htmlTagLookup:
1.69 veillard 626: * @tag: The tag name in lowercase
1.1 daniel 627: *
628: * Lookup the HTML tag in the ElementTable
629: *
630: * Returns the related htmlElemDescPtr or NULL if not found.
631: */
632: htmlElemDescPtr
1.14 daniel 633: htmlTagLookup(const xmlChar *tag) {
1.61 veillard 634: int i;
1.1 daniel 635:
636: for (i = 0; i < (sizeof(html40ElementTable) /
637: sizeof(html40ElementTable[0]));i++) {
1.73 veillard 638: if (xmlStrEqual(tag, BAD_CAST html40ElementTable[i].name))
1.1 daniel 639: return(&html40ElementTable[i]);
640: }
641: return(NULL);
642: }
643:
644: /**
645: * htmlCheckAutoClose:
1.50 veillard 646: * @newtag: The new tag name
647: * @oldtag: The old tag name
1.1 daniel 648: *
649: * Checks wether the new tag is one of the registered valid tags for closing old.
650: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
651: *
652: * Returns 0 if no, 1 if yes.
653: */
654: int
1.50 veillard 655: htmlCheckAutoClose(const xmlChar *newtag, const xmlChar *oldtag) {
1.1 daniel 656: int i, index;
1.64 veillard 657: char **close = NULL;
1.1 daniel 658:
659: if (htmlStartCloseIndexinitialized == 0) htmlInitAutoClose();
660:
661: /* inefficient, but not a big deal */
662: for (index = 0; index < 100;index++) {
663: close = htmlStartCloseIndex[index];
664: if (close == NULL) return(0);
1.73 veillard 665: if (xmlStrEqual(BAD_CAST *close, newtag)) break;
1.1 daniel 666: }
667:
668: i = close - htmlStartClose;
669: i++;
670: while (htmlStartClose[i] != NULL) {
1.73 veillard 671: if (xmlStrEqual(BAD_CAST htmlStartClose[i], oldtag)) {
1.1 daniel 672: return(1);
673: }
674: i++;
675: }
676: return(0);
677: }
678:
679: /**
1.50 veillard 680: * htmlAutoCloseOnClose:
681: * @ctxt: an HTML parser context
682: * @newtag: The new tag name
683: *
684: * The HTmL DtD allows an ending tag to implicitely close other tags.
685: */
686: void
687: htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
688: htmlElemDescPtr info;
689: xmlChar *oldname;
690: int i;
691:
692: #ifdef DEBUG
1.81 veillard 693: xmlGenericError(xmlGenericErrorContext,"Close of %s stack: %d elements\n", newtag, ctxt->nameNr);
1.50 veillard 694: for (i = 0;i < ctxt->nameNr;i++)
1.81 veillard 695: xmlGenericError(xmlGenericErrorContext,"%d : %s\n", i, ctxt->nameTab[i]);
1.50 veillard 696: #endif
697:
698: for (i = (ctxt->nameNr - 1);i >= 0;i--) {
1.73 veillard 699: if (xmlStrEqual(newtag, ctxt->nameTab[i])) break;
1.50 veillard 700: }
701: if (i < 0) return;
702:
1.73 veillard 703: while (!xmlStrEqual(newtag, ctxt->name)) {
1.50 veillard 704: info = htmlTagLookup(ctxt->name);
705: if ((info == NULL) || (info->endTag == 1)) {
706: #ifdef DEBUG
1.81 veillard 707: xmlGenericError(xmlGenericErrorContext,"htmlAutoCloseOnClose: %s closes %s\n", newtag, ctxt->name);
1.50 veillard 708: #endif
709: } else {
710: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
711: ctxt->sax->error(ctxt->userData,
712: "Opening and ending tag mismatch: %s and %s\n",
713: newtag, ctxt->name);
714: ctxt->wellFormed = 0;
715: }
716: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
717: ctxt->sax->endElement(ctxt->userData, ctxt->name);
718: oldname = htmlnamePop(ctxt);
719: if (oldname != NULL) {
720: #ifdef DEBUG
1.81 veillard 721: xmlGenericError(xmlGenericErrorContext,"htmlAutoCloseOnClose: popped %s\n", oldname);
1.50 veillard 722: #endif
723: xmlFree(oldname);
724: }
725: }
726: }
727:
728: /**
1.1 daniel 729: * htmlAutoClose:
730: * @ctxt: an HTML parser context
1.50 veillard 731: * @newtag: The new tag name or NULL
1.1 daniel 732: *
733: * The HTmL DtD allows a tag to implicitely close other tags.
734: * The list is kept in htmlStartClose array. This function is
735: * called when a new tag has been detected and generates the
736: * appropriates closes if possible/needed.
1.50 veillard 737: * If newtag is NULL this mean we are at the end of the resource
1.47 daniel 738: * and we should check
1.1 daniel 739: */
740: void
1.50 veillard 741: htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
1.15 daniel 742: xmlChar *oldname;
1.50 veillard 743: while ((newtag != NULL) && (ctxt->name != NULL) &&
744: (htmlCheckAutoClose(newtag, ctxt->name))) {
1.1 daniel 745: #ifdef DEBUG
1.81 veillard 746: xmlGenericError(xmlGenericErrorContext,"htmlAutoClose: %s closes %s\n", newtag, ctxt->name);
1.1 daniel 747: #endif
748: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1.15 daniel 749: ctxt->sax->endElement(ctxt->userData, ctxt->name);
1.24 daniel 750: oldname = htmlnamePop(ctxt);
1.18 daniel 751: if (oldname != NULL) {
752: #ifdef DEBUG
1.81 veillard 753: xmlGenericError(xmlGenericErrorContext,"htmlAutoClose: popped %s\n", oldname);
1.18 daniel 754: #endif
1.17 daniel 755: xmlFree(oldname);
1.18 daniel 756: }
1.1 daniel 757: }
1.50 veillard 758: if (newtag == NULL) {
1.49 daniel 759: htmlAutoCloseOnClose(ctxt, BAD_CAST"head");
760: htmlAutoCloseOnClose(ctxt, BAD_CAST"body");
761: htmlAutoCloseOnClose(ctxt, BAD_CAST"html");
762: }
1.50 veillard 763: while ((newtag == NULL) && (ctxt->name != NULL) &&
1.73 veillard 764: ((xmlStrEqual(ctxt->name, BAD_CAST"head")) ||
765: (xmlStrEqual(ctxt->name, BAD_CAST"body")) ||
766: (xmlStrEqual(ctxt->name, BAD_CAST"html")))) {
1.47 daniel 767: #ifdef DEBUG
1.81 veillard 768: xmlGenericError(xmlGenericErrorContext,"htmlAutoClose: EOF closes %s\n", ctxt->name);
1.47 daniel 769: #endif
770: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
771: ctxt->sax->endElement(ctxt->userData, ctxt->name);
772: oldname = htmlnamePop(ctxt);
773: if (oldname != NULL) {
774: #ifdef DEBUG
1.81 veillard 775: xmlGenericError(xmlGenericErrorContext,"htmlAutoClose: popped %s\n", oldname);
1.47 daniel 776: #endif
777: xmlFree(oldname);
778: }
779: }
780:
1.1 daniel 781: }
782:
783: /**
1.28 daniel 784: * htmlAutoCloseTag:
785: * @doc: the HTML document
786: * @name: The tag name
787: * @elem: the HTML element
788: *
789: * The HTmL DtD allows a tag to implicitely close other tags.
790: * The list is kept in htmlStartClose array. This function checks
791: * if the element or one of it's children would autoclose the
792: * given tag.
793: *
794: * Returns 1 if autoclose, 0 otherwise
795: */
796: int
797: htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) {
798: htmlNodePtr child;
799:
800: if (elem == NULL) return(1);
1.73 veillard 801: if (xmlStrEqual(name, elem->name)) return(0);
1.28 daniel 802: if (htmlCheckAutoClose(elem->name, name)) return(1);
1.37 daniel 803: child = elem->children;
1.28 daniel 804: while (child != NULL) {
805: if (htmlAutoCloseTag(doc, name, child)) return(1);
806: child = child->next;
807: }
808: return(0);
809: }
810:
811: /**
812: * htmlIsAutoClosed:
813: * @doc: the HTML document
814: * @elem: the HTML element
815: *
816: * The HTmL DtD allows a tag to implicitely close other tags.
817: * The list is kept in htmlStartClose array. This function checks
818: * if a tag is autoclosed by one of it's child
819: *
820: * Returns 1 if autoclosed, 0 otherwise
821: */
822: int
823: htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) {
824: htmlNodePtr child;
825:
826: if (elem == NULL) return(1);
1.37 daniel 827: child = elem->children;
1.28 daniel 828: while (child != NULL) {
829: if (htmlAutoCloseTag(doc, elem->name, child)) return(1);
830: child = child->next;
831: }
832: return(0);
833: }
834:
835: /**
1.43 daniel 836: * htmlCheckImplied:
837: * @ctxt: an HTML parser context
1.50 veillard 838: * @newtag: The new tag name
1.43 daniel 839: *
840: * The HTmL DtD allows a tag to exists only implicitely
841: * called when a new tag has been detected and generates the
842: * appropriates implicit tags if missing
843: */
844: void
1.50 veillard 845: htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
1.73 veillard 846: if (xmlStrEqual(newtag, BAD_CAST"html"))
1.43 daniel 847: return;
848: if (ctxt->nameNr <= 0) {
849: #ifdef DEBUG
1.81 veillard 850: xmlGenericError(xmlGenericErrorContext,"Implied element html: pushed html\n");
1.43 daniel 851: #endif
852: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"html"));
853: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
854: ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL);
855: }
1.73 veillard 856: if ((xmlStrEqual(newtag, BAD_CAST"body")) || (xmlStrEqual(newtag, BAD_CAST"head")))
1.43 daniel 857: return;
858: if (ctxt->nameNr <= 1) {
1.73 veillard 859: if ((xmlStrEqual(newtag, BAD_CAST"script")) ||
860: (xmlStrEqual(newtag, BAD_CAST"style")) ||
861: (xmlStrEqual(newtag, BAD_CAST"meta")) ||
862: (xmlStrEqual(newtag, BAD_CAST"link")) ||
863: (xmlStrEqual(newtag, BAD_CAST"title")) ||
864: (xmlStrEqual(newtag, BAD_CAST"base"))) {
1.43 daniel 865: /*
866: * dropped OBJECT ... i you put it first BODY will be
867: * assumed !
868: */
869: #ifdef DEBUG
1.81 veillard 870: xmlGenericError(xmlGenericErrorContext,"Implied element head: pushed head\n");
1.43 daniel 871: #endif
872: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"head"));
873: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
874: ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL);
875: } else {
876: #ifdef DEBUG
1.81 veillard 877: xmlGenericError(xmlGenericErrorContext,"Implied element body: pushed body\n");
1.43 daniel 878: #endif
879: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"body"));
880: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
881: ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL);
882: }
883: }
884: }
885:
1.59 veillard 886: /**
887: * htmlCheckParagraph
888: * @ctxt: an HTML parser context
889: *
890: * Check whether a p element need to be implied before inserting
891: * characters in the current element.
892: *
893: * Returns 1 if a paragraph has been inserted, 0 if not and -1
894: * in case of error.
895: */
896:
897: int
898: htmlCheckParagraph(htmlParserCtxtPtr ctxt) {
899: const xmlChar *tag;
900: int i;
901:
902: if (ctxt == NULL)
903: return(-1);
904: tag = ctxt->name;
905: if (tag == NULL) {
906: htmlAutoClose(ctxt, BAD_CAST"p");
907: htmlCheckImplied(ctxt, BAD_CAST"p");
908: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p"));
909: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
910: ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
911: return(1);
912: }
913: for (i = 0; htmlNoContentElements[i] != NULL; i++) {
1.73 veillard 914: if (xmlStrEqual(tag, BAD_CAST htmlNoContentElements[i])) {
1.59 veillard 915: #ifdef DEBUG
1.81 veillard 916: xmlGenericError(xmlGenericErrorContext,"Implied element paragraph\n");
1.59 veillard 917: #endif
918: htmlAutoClose(ctxt, BAD_CAST"p");
919: htmlCheckImplied(ctxt, BAD_CAST"p");
920: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p"));
921: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
922: ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
923: return(1);
924: }
1.78 veillard 925: }
926: return(0);
927: }
928:
929: /**
930: * htmlIsScriptAttribute:
931: * @name: an attribute name
932: *
933: * Check if an attribute is of content type Script
934: *
935: * Returns 1 is the attribute is a script 0 otherwise
936: */
937: int
938: htmlIsScriptAttribute(const xmlChar *name) {
939: int i;
940:
941: if (name == NULL)
942: return(0);
943: /*
944: * all script attributes start with 'on'
945: */
946: if ((name[0] != 'o') || (name[1] != 'n'))
947: return(0);
948: for (i = 0;
949: i < sizeof(htmlScriptAttributes)/sizeof(htmlScriptAttributes[0]);
950: i++) {
951: if (xmlStrEqual(name, (const xmlChar *) htmlScriptAttributes[i]))
952: return(1);
1.59 veillard 953: }
954: return(0);
955: }
956:
1.1 daniel 957: /************************************************************************
958: * *
959: * The list of HTML predefined entities *
960: * *
961: ************************************************************************/
962:
963:
964: htmlEntityDesc html40EntitiesTable[] = {
965: /*
1.61 veillard 966: * the 4 absolute ones, plus apostrophe.
1.1 daniel 967: */
968: { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
969: { 38, "amp", "ampersand, U+0026 ISOnum" },
1.61 veillard 970: { 39, "apos", "single quote" },
1.1 daniel 971: { 60, "lt", "less-than sign, U+003C ISOnum" },
972: { 62, "gt", "greater-than sign, U+003E ISOnum" },
973:
974: /*
975: * A bunch still in the 128-255 range
976: * Replacing them depend really on the charset used.
977: */
978: { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
979: { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
980: { 162, "cent", "cent sign, U+00A2 ISOnum" },
981: { 163, "pound","pound sign, U+00A3 ISOnum" },
982: { 164, "curren","currency sign, U+00A4 ISOnum" },
983: { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
984: { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
985: { 167, "sect", "section sign, U+00A7 ISOnum" },
986: { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
987: { 169, "copy", "copyright sign, U+00A9 ISOnum" },
988: { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
989: { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
990: { 172, "not", "not sign, U+00AC ISOnum" },
991: { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
992: { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
993: { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
994: { 176, "deg", "degree sign, U+00B0 ISOnum" },
995: { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
996: { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
997: { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
998: { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
999: { 181, "micro","micro sign, U+00B5 ISOnum" },
1000: { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1.7 daniel 1001: { 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1.1 daniel 1002: { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
1003: { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
1004: { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1.7 daniel 1005: { 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1.1 daniel 1006: { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
1007: { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
1008: { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
1009: { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
1010: { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
1011: { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
1012: { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
1013: { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
1014: { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
1015: { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
1016: { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
1017: { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
1018: { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
1019: { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
1020: { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
1021: { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
1022: { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
1023: { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
1024: { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
1025: { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
1026: { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
1027: { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
1028: { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
1029: { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
1030: { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
1031: { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
1032: { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
1033: { 215, "times","multiplication sign, U+00D7 ISOnum" },
1.7 daniel 1034: { 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1.1 daniel 1035: { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
1036: { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
1037: { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
1038: { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
1039: { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
1040: { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
1041: { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
1042: { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
1043: { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
1044: { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
1045: { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
1046: { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
1047: { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
1048: { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
1049: { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
1050: { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
1051: { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
1052: { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
1053: { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
1054: { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
1055: { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
1056: { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
1057: { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
1058: { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
1059: { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
1060: { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
1061: { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
1062: { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
1063: { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
1064: { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
1065: { 247, "divide","division sign, U+00F7 ISOnum" },
1066: { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
1067: { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
1068: { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
1069: { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
1070: { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
1071: { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
1072: { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
1073: { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
1074:
1.61 veillard 1075: { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
1076: { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
1077: { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
1078: { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
1079: { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
1080:
1.1 daniel 1081: /*
1082: * Anything below should really be kept as entities references
1083: */
1084: { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
1085:
1.61 veillard 1086: { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
1087: { 732, "tilde","small tilde, U+02DC ISOdia" },
1088:
1.1 daniel 1089: { 913, "Alpha","greek capital letter alpha, U+0391" },
1090: { 914, "Beta", "greek capital letter beta, U+0392" },
1091: { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
1092: { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
1093: { 917, "Epsilon","greek capital letter epsilon, U+0395" },
1094: { 918, "Zeta", "greek capital letter zeta, U+0396" },
1095: { 919, "Eta", "greek capital letter eta, U+0397" },
1096: { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
1097: { 921, "Iota", "greek capital letter iota, U+0399" },
1098: { 922, "Kappa","greek capital letter kappa, U+039A" },
1099: { 923, "Lambda""greek capital letter lambda, U+039B ISOgrk3" },
1100: { 924, "Mu", "greek capital letter mu, U+039C" },
1101: { 925, "Nu", "greek capital letter nu, U+039D" },
1102: { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
1103: { 927, "Omicron","greek capital letter omicron, U+039F" },
1104: { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
1105: { 929, "Rho", "greek capital letter rho, U+03A1" },
1106: { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
1107: { 932, "Tau", "greek capital letter tau, U+03A4" },
1108: { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
1109: { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
1110: { 935, "Chi", "greek capital letter chi, U+03A7" },
1111: { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
1112: { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
1113:
1114: { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
1115: { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
1116: { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
1117: { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
1118: { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
1119: { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
1120: { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
1121: { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
1122: { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
1123: { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
1124: { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
1125: { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
1126: { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
1127: { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
1128: { 959, "omicron","greek small letter omicron, U+03BF NEW" },
1129: { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
1130: { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
1131: { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
1132: { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
1133: { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
1134: { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
1135: { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
1136: { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
1137: { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
1138: { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
1139: { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
1140: { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
1141: { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
1142:
1.61 veillard 1143: { 8194, "ensp", "en space, U+2002 ISOpub" },
1144: { 8195, "emsp", "em space, U+2003 ISOpub" },
1145: { 8201, "thinsp","thin space, U+2009 ISOpub" },
1146: { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
1147: { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
1148: { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
1149: { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
1150: { 8211, "ndash","en dash, U+2013 ISOpub" },
1151: { 8212, "mdash","em dash, U+2014 ISOpub" },
1152: { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
1153: { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
1154: { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
1155: { 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
1156: { 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
1157: { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
1158: { 8224, "dagger","dagger, U+2020 ISOpub" },
1159: { 8225, "Dagger","double dagger, U+2021 ISOpub" },
1160:
1.1 daniel 1161: { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
1162: { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
1.61 veillard 1163:
1164: { 8240, "permil","per mille sign, U+2030 ISOtech" },
1165:
1.1 daniel 1166: { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
1167: { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
1.61 veillard 1168:
1169: { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
1170: { 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
1171:
1.1 daniel 1172: { 8254, "oline","overline = spacing overscore, U+203E NEW" },
1173: { 8260, "frasl","fraction slash, U+2044 NEW" },
1174:
1.61 veillard 1175: { 8364, "euro", "euro sign, U+20AC NEW" },
1176:
1177: { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
1.7 daniel 1178: { 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
1.1 daniel 1179: { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
1180: { 8482, "trade","trade mark sign, U+2122 ISOnum" },
1181: { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
1182: { 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
1183: { 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
1184: { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
1185: { 8595, "darr", "downwards arrow, U+2193 ISOnum" },
1186: { 8596, "harr", "left right arrow, U+2194 ISOamsa" },
1187: { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
1188: { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
1189: { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
1190: { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
1191: { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
1192: { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
1193:
1194: { 8704, "forall","for all, U+2200 ISOtech" },
1195: { 8706, "part", "partial differential, U+2202 ISOtech" },
1196: { 8707, "exist","there exists, U+2203 ISOtech" },
1197: { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
1198: { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
1199: { 8712, "isin", "element of, U+2208 ISOtech" },
1200: { 8713, "notin","not an element of, U+2209 ISOtech" },
1201: { 8715, "ni", "contains as member, U+220B ISOtech" },
1202: { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
1203: { 8721, "sum", "n-ary sumation, U+2211 ISOamsb" },
1204: { 8722, "minus","minus sign, U+2212 ISOtech" },
1205: { 8727, "lowast","asterisk operator, U+2217 ISOtech" },
1206: { 8730, "radic","square root = radical sign, U+221A ISOtech" },
1207: { 8733, "prop", "proportional to, U+221D ISOtech" },
1208: { 8734, "infin","infinity, U+221E ISOtech" },
1209: { 8736, "ang", "angle, U+2220 ISOamso" },
1210: { 8743, "and", "logical and = wedge, U+2227 ISOtech" },
1211: { 8744, "or", "logical or = vee, U+2228 ISOtech" },
1212: { 8745, "cap", "intersection = cap, U+2229 ISOtech" },
1213: { 8746, "cup", "union = cup, U+222A ISOtech" },
1214: { 8747, "int", "integral, U+222B ISOtech" },
1215: { 8756, "there4","therefore, U+2234 ISOtech" },
1216: { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
1217: { 8773, "cong", "approximately equal to, U+2245 ISOtech" },
1218: { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
1219: { 8800, "ne", "not equal to, U+2260 ISOtech" },
1220: { 8801, "equiv","identical to, U+2261 ISOtech" },
1221: { 8804, "le", "less-than or equal to, U+2264 ISOtech" },
1222: { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
1223: { 8834, "sub", "subset of, U+2282 ISOtech" },
1224: { 8835, "sup", "superset of, U+2283 ISOtech" },
1225: { 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
1226: { 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
1227: { 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
1228: { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
1229: { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
1230: { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
1231: { 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
1232: { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
1233: { 8969, "rceil","right ceiling, U+2309 ISOamsc" },
1234: { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
1235: { 8971, "rfloor","right floor, U+230B ISOamsc" },
1236: { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
1237: { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
1238: { 9674, "loz", "lozenge, U+25CA ISOpub" },
1239:
1240: { 9824, "spades","black spade suit, U+2660 ISOpub" },
1241: { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
1242: { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
1243: { 9830, "diams","black diamond suit, U+2666 ISOpub" },
1244:
1245: };
1246:
1247: /************************************************************************
1248: * *
1249: * Commodity functions to handle entities *
1250: * *
1251: ************************************************************************/
1252:
1253: /*
1254: * Macro used to grow the current buffer.
1255: */
1256: #define growBuffer(buffer) { \
1257: buffer##_size *= 2; \
1.14 daniel 1258: buffer = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
1.1 daniel 1259: if (buffer == NULL) { \
1260: perror("realloc failed"); \
1.33 daniel 1261: return(NULL); \
1.1 daniel 1262: } \
1263: }
1264:
1265: /**
1266: * htmlEntityLookup:
1267: * @name: the entity name
1268: *
1269: * Lookup the given entity in EntitiesTable
1270: *
1271: * TODO: the linear scan is really ugly, an hash table is really needed.
1272: *
1273: * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1274: */
1275: htmlEntityDescPtr
1.14 daniel 1276: htmlEntityLookup(const xmlChar *name) {
1.1 daniel 1277: int i;
1278:
1279: for (i = 0;i < (sizeof(html40EntitiesTable)/
1280: sizeof(html40EntitiesTable[0]));i++) {
1.73 veillard 1281: if (xmlStrEqual(name, BAD_CAST html40EntitiesTable[i].name)) {
1.1 daniel 1282: #ifdef DEBUG
1.81 veillard 1283: xmlGenericError(xmlGenericErrorContext,"Found entity %s\n", name);
1.1 daniel 1284: #endif
1285: return(&html40EntitiesTable[i]);
1286: }
1287: }
1288: return(NULL);
1289: }
1290:
1.53 veillard 1291: /**
1.61 veillard 1292: * htmlEntityValueLookup:
1293: * @value: the entity's unicode value
1294: *
1295: * Lookup the given entity in EntitiesTable
1296: *
1297: * TODO: the linear scan is really ugly, an hash table is really needed.
1298: *
1299: * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1300: */
1301: htmlEntityDescPtr
1302: htmlEntityValueLookup(int value) {
1303: int i;
1304: #ifdef DEBUG
1305: int lv = 0;
1306: #endif
1307:
1308: for (i = 0;i < (sizeof(html40EntitiesTable)/
1309: sizeof(html40EntitiesTable[0]));i++) {
1.71 veillard 1310: if ((unsigned int) html40EntitiesTable[i].value >= value) {
1311: if ((unsigned int) html40EntitiesTable[i].value > value)
1.61 veillard 1312: break;
1313: #ifdef DEBUG
1.81 veillard 1314: xmlGenericError(xmlGenericErrorContext,"Found entity %s\n", html40EntitiesTable[i].name);
1.61 veillard 1315: #endif
1316: return(&html40EntitiesTable[i]);
1317: }
1318: #ifdef DEBUG
1319: if (lv > html40EntitiesTable[i].value) {
1.81 veillard 1320: xmlGenericError(xmlGenericErrorContext,
1321: "html40EntitiesTable[] is not sorted (%d > %d)!\n",
1.61 veillard 1322: lv, html40EntitiesTable[i].value);
1323: }
1324: lv = html40EntitiesTable[i].value;
1325: #endif
1326: }
1327: return(NULL);
1328: }
1329:
1330: /**
1.53 veillard 1331: * UTF8ToHtml:
1332: * @out: a pointer to an array of bytes to store the result
1333: * @outlen: the length of @out
1334: * @in: a pointer to an array of UTF-8 chars
1335: * @inlen: the length of @in
1336: *
1337: * Take a block of UTF-8 chars in and try to convert it to an ASCII
1338: * plus HTML entities block of chars out.
1339: *
1340: * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1341: * The value of @inlen after return is the number of octets consumed
1342: * as the return value is positive, else unpredictiable.
1343: * The value of @outlen after return is the number of octets consumed.
1344: */
1345: int
1346: UTF8ToHtml(unsigned char* out, int *outlen,
1347: const unsigned char* in, int *inlen) {
1348: const unsigned char* processed = in;
1349: const unsigned char* outend;
1350: const unsigned char* outstart = out;
1351: const unsigned char* instart = in;
1352: const unsigned char* inend;
1353: unsigned int c, d;
1354: int trailing;
1355:
1356: if (in == NULL) {
1357: /*
1358: * initialization nothing to do
1359: */
1360: *outlen = 0;
1361: *inlen = 0;
1362: return(0);
1363: }
1364: inend = in + (*inlen);
1365: outend = out + (*outlen);
1366: while (in < inend) {
1367: d = *in++;
1368: if (d < 0x80) { c= d; trailing= 0; }
1369: else if (d < 0xC0) {
1370: /* trailing byte in leading position */
1371: *outlen = out - outstart;
1372: *inlen = processed - instart;
1373: return(-2);
1374: } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
1375: else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
1376: else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
1377: else {
1378: /* no chance for this in Ascii */
1379: *outlen = out - outstart;
1380: *inlen = processed - instart;
1381: return(-2);
1382: }
1383:
1384: if (inend - in < trailing) {
1385: break;
1386: }
1387:
1388: for ( ; trailing; trailing--) {
1389: if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
1390: break;
1391: c <<= 6;
1392: c |= d & 0x3F;
1393: }
1394:
1395: /* assertion: c is a single UTF-4 value */
1396: if (c < 0x80) {
1.62 veillard 1397: if (out + 1 >= outend)
1.53 veillard 1398: break;
1399: *out++ = c;
1400: } else {
1.61 veillard 1401: int len;
1402: htmlEntityDescPtr ent;
1403:
1.53 veillard 1404: /*
1405: * Try to lookup a predefined HTML entity for it
1406: */
1407:
1.61 veillard 1408: ent = htmlEntityValueLookup(c);
1409: if (ent == NULL) {
1410: /* no chance for this in Ascii */
1411: *outlen = out - outstart;
1412: *inlen = processed - instart;
1413: return(-2);
1.53 veillard 1414: }
1.61 veillard 1415: len = strlen(ent->name);
1.62 veillard 1416: if (out + 2 + len >= outend)
1.53 veillard 1417: break;
1418: *out++ = '&';
1.61 veillard 1419: memcpy(out, ent->name, len);
1420: out += len;
1.53 veillard 1421: *out++ = ';';
1422: }
1423: processed = in;
1424: }
1425: *outlen = out - outstart;
1426: *inlen = processed - instart;
1427: return(0);
1428: }
1429:
1.62 veillard 1430: /**
1431: * htmlEncodeEntities:
1432: * @out: a pointer to an array of bytes to store the result
1433: * @outlen: the length of @out
1434: * @in: a pointer to an array of UTF-8 chars
1435: * @inlen: the length of @in
1436: * @quoteChar: the quote character to escape (' or ") or zero.
1437: *
1438: * Take a block of UTF-8 chars in and try to convert it to an ASCII
1439: * plus HTML entities block of chars out.
1440: *
1441: * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1442: * The value of @inlen after return is the number of octets consumed
1443: * as the return value is positive, else unpredictiable.
1444: * The value of @outlen after return is the number of octets consumed.
1445: */
1446: int
1447: htmlEncodeEntities(unsigned char* out, int *outlen,
1448: const unsigned char* in, int *inlen, int quoteChar) {
1449: const unsigned char* processed = in;
1450: const unsigned char* outend = out + (*outlen);
1451: const unsigned char* outstart = out;
1452: const unsigned char* instart = in;
1453: const unsigned char* inend = in + (*inlen);
1454: unsigned int c, d;
1455: int trailing;
1456:
1457: while (in < inend) {
1458: d = *in++;
1459: if (d < 0x80) { c= d; trailing= 0; }
1460: else if (d < 0xC0) {
1461: /* trailing byte in leading position */
1462: *outlen = out - outstart;
1463: *inlen = processed - instart;
1464: return(-2);
1465: } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
1466: else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
1467: else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
1468: else {
1469: /* no chance for this in Ascii */
1470: *outlen = out - outstart;
1471: *inlen = processed - instart;
1472: return(-2);
1473: }
1474:
1475: if (inend - in < trailing)
1476: break;
1477:
1478: while (trailing--) {
1479: if (((d= *in++) & 0xC0) != 0x80) {
1480: *outlen = out - outstart;
1481: *inlen = processed - instart;
1482: return(-2);
1483: }
1484: c <<= 6;
1485: c |= d & 0x3F;
1486: }
1487:
1488: /* assertion: c is a single UTF-4 value */
1489: if (c < 0x80 && c != quoteChar && c != '&' && c != '<' && c != '>') {
1490: if (out >= outend)
1491: break;
1492: *out++ = c;
1493: } else {
1494: htmlEntityDescPtr ent;
1495: const char *cp;
1496: char nbuf[16];
1497: int len;
1498:
1499: /*
1500: * Try to lookup a predefined HTML entity for it
1501: */
1502: ent = htmlEntityValueLookup(c);
1503: if (ent == NULL) {
1504: sprintf(nbuf, "#%u", c);
1505: cp = nbuf;
1506: }
1507: else
1508: cp = ent->name;
1509: len = strlen(cp);
1510: if (out + 2 + len > outend)
1511: break;
1512: *out++ = '&';
1513: memcpy(out, cp, len);
1514: out += len;
1515: *out++ = ';';
1516: }
1517: processed = in;
1518: }
1519: *outlen = out - outstart;
1520: *inlen = processed - instart;
1521: return(0);
1522: }
1.1 daniel 1523:
1524: /**
1525: * htmlDecodeEntities:
1526: * @ctxt: the parser context
1527: * @len: the len to decode (in bytes !), -1 for no size limit
1.14 daniel 1528: * @end: an end marker xmlChar, 0 if none
1529: * @end2: an end marker xmlChar, 0 if none
1530: * @end3: an end marker xmlChar, 0 if none
1.1 daniel 1531: *
1532: * Subtitute the HTML entities by their value
1533: *
1.19 daniel 1534: * DEPRECATED !!!!
1.1 daniel 1535: *
1536: * Returns A newly allocated string with the substitution done. The caller
1537: * must deallocate it !
1538: */
1.14 daniel 1539: xmlChar *
1.1 daniel 1540: htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
1.14 daniel 1541: xmlChar end, xmlChar end2, xmlChar end3) {
1.53 veillard 1542: xmlChar *name = NULL;
1.14 daniel 1543: xmlChar *buffer = NULL;
1.53 veillard 1544: unsigned int buffer_size = 0;
1545: unsigned int nbchars = 0;
1.1 daniel 1546: htmlEntityDescPtr ent;
1547: unsigned int max = (unsigned int) len;
1.53 veillard 1548: int c,l;
1549:
1550: if (ctxt->depth > 40) {
1.67 veillard 1551: ctxt->errNo = XML_ERR_ENTITY_LOOP;
1.53 veillard 1552: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1553: ctxt->sax->error(ctxt->userData,
1554: "Detected entity reference loop\n");
1555: ctxt->wellFormed = 0;
1556: ctxt->disableSAX = 1;
1557: return(NULL);
1558: }
1.1 daniel 1559:
1560: /*
1561: * allocate a translation buffer.
1562: */
1.31 daniel 1563: buffer_size = HTML_PARSER_BIG_BUFFER_SIZE;
1.14 daniel 1564: buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
1.1 daniel 1565: if (buffer == NULL) {
1.53 veillard 1566: perror("xmlDecodeEntities: malloc failed");
1.1 daniel 1567: return(NULL);
1568: }
1569:
1570: /*
1571: * Ok loop until we reach one of the ending char or a size limit.
1572: */
1.53 veillard 1573: c = CUR_CHAR(l);
1574: while ((nbchars < max) && (c != end) &&
1575: (c != end2) && (c != end3)) {
1576:
1577: if (c == 0) break;
1578: if (((c == '&') && (ctxt->token != '&')) && (NXT(1) == '#')) {
1579: int val = htmlParseCharRef(ctxt);
1580: COPY_BUF(0,buffer,nbchars,val);
1581: NEXTL(l);
1582: } else if ((c == '&') && (ctxt->token != '&')) {
1583: ent = htmlParseEntityRef(ctxt, &name);
1584: if (name != NULL) {
1585: if (ent != NULL) {
1586: int val = ent->value;
1587: COPY_BUF(0,buffer,nbchars,val);
1588: NEXTL(l);
1589: } else {
1590: const xmlChar *cur = name;
1.1 daniel 1591:
1.53 veillard 1592: buffer[nbchars++] = '&';
1593: if (nbchars > buffer_size - HTML_PARSER_BUFFER_SIZE) {
1594: growBuffer(buffer);
1595: }
1596: while (*cur != 0) {
1597: buffer[nbchars++] = *cur++;
1.1 daniel 1598: }
1.53 veillard 1599: buffer[nbchars++] = ';';
1.1 daniel 1600: }
1601: }
1602: } else {
1.53 veillard 1603: COPY_BUF(l,buffer,nbchars,c);
1604: NEXTL(l);
1605: if (nbchars > buffer_size - HTML_PARSER_BUFFER_SIZE) {
1606: growBuffer(buffer);
1.1 daniel 1607: }
1608: }
1.53 veillard 1609: c = CUR_CHAR(l);
1.1 daniel 1610: }
1.53 veillard 1611: buffer[nbchars++] = 0;
1.1 daniel 1612: return(buffer);
1613: }
1614:
1.31 daniel 1615: /************************************************************************
1616: * *
1617: * Commodity functions to handle streams *
1618: * *
1619: ************************************************************************/
1620:
1621: /**
1622: * htmlFreeInputStream:
1623: * @input: an htmlParserInputPtr
1624: *
1625: * Free up an input stream.
1626: */
1627: void
1628: htmlFreeInputStream(htmlParserInputPtr input) {
1629: if (input == NULL) return;
1630:
1631: if (input->filename != NULL) xmlFree((char *) input->filename);
1632: if (input->directory != NULL) xmlFree((char *) input->directory);
1633: if ((input->free != NULL) && (input->base != NULL))
1634: input->free((xmlChar *) input->base);
1635: if (input->buf != NULL)
1636: xmlFreeParserInputBuffer(input->buf);
1637: memset(input, -1, sizeof(htmlParserInput));
1638: xmlFree(input);
1639: }
1640:
1641: /**
1642: * htmlNewInputStream:
1643: * @ctxt: an HTML parser context
1644: *
1645: * Create a new input stream structure
1646: * Returns the new input stream or NULL
1647: */
1648: htmlParserInputPtr
1649: htmlNewInputStream(htmlParserCtxtPtr ctxt) {
1650: htmlParserInputPtr input;
1651:
1652: input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1653: if (input == NULL) {
1654: ctxt->errNo = XML_ERR_NO_MEMORY;
1655: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1656: ctxt->sax->error(ctxt->userData,
1657: "malloc: couldn't allocate a new input stream\n");
1658: return(NULL);
1659: }
1.51 veillard 1660: memset(input, 0, sizeof(htmlParserInput));
1.31 daniel 1661: input->filename = NULL;
1662: input->directory = NULL;
1663: input->base = NULL;
1664: input->cur = NULL;
1665: input->buf = NULL;
1666: input->line = 1;
1667: input->col = 1;
1668: input->buf = NULL;
1669: input->free = NULL;
1.51 veillard 1670: input->version = NULL;
1.31 daniel 1671: input->consumed = 0;
1672: input->length = 0;
1673: return(input);
1674: }
1675:
1.1 daniel 1676:
1677: /************************************************************************
1678: * *
1679: * Commodity functions, cleanup needed ? *
1680: * *
1681: ************************************************************************/
1682:
1683: /**
1684: * areBlanks:
1685: * @ctxt: an HTML parser context
1.14 daniel 1686: * @str: a xmlChar *
1.1 daniel 1687: * @len: the size of @str
1688: *
1689: * Is this a sequence of blank chars that one can ignore ?
1690: *
1691: * Returns 1 if ignorable 0 otherwise.
1692: */
1693:
1.14 daniel 1694: static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
1.1 daniel 1695: int i;
1696: xmlNodePtr lastChild;
1697:
1698: for (i = 0;i < len;i++)
1699: if (!(IS_BLANK(str[i]))) return(0);
1700:
1.48 daniel 1701: if (CUR == 0) return(1);
1.1 daniel 1702: if (CUR != '<') return(0);
1.62 veillard 1703: if (ctxt->name == NULL)
1704: return(1);
1.73 veillard 1705: if (xmlStrEqual(ctxt->name, BAD_CAST"html"))
1.63 veillard 1706: return(1);
1.73 veillard 1707: if (xmlStrEqual(ctxt->name, BAD_CAST"head"))
1.62 veillard 1708: return(1);
1.73 veillard 1709: if (xmlStrEqual(ctxt->name, BAD_CAST"body"))
1.62 veillard 1710: return(1);
1.1 daniel 1711: if (ctxt->node == NULL) return(0);
1712: lastChild = xmlGetLastChild(ctxt->node);
1713: if (lastChild == NULL) {
1714: if (ctxt->node->content != NULL) return(0);
1.82 ! veillard 1715: } else if (xmlNodeIsText(lastChild)) {
1.1 daniel 1716: return(0);
1.82 ! veillard 1717: } else if (xmlStrEqual(lastChild->name, BAD_CAST"b")) {
! 1718: return(0);
! 1719: } else if (xmlStrEqual(lastChild->name, BAD_CAST"bold")) {
! 1720: return(0);
! 1721: } else if (xmlStrEqual(lastChild->name, BAD_CAST"em")) {
! 1722: return(0);
! 1723: }
1.1 daniel 1724: return(1);
1725: }
1726:
1727: /**
1728: * htmlHandleEntity:
1729: * @ctxt: an HTML parser context
1730: * @entity: an XML entity pointer.
1731: *
1732: * Default handling of an HTML entity, call the parser with the
1733: * substitution string
1734: */
1735:
1736: void
1737: htmlHandleEntity(htmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1738: int len;
1739:
1740: if (entity->content == NULL) {
1741: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1742: ctxt->sax->error(ctxt->userData, "htmlHandleEntity %s: content == NULL\n",
1743: entity->name);
1744: ctxt->wellFormed = 0;
1745: return;
1746: }
1747: len = xmlStrlen(entity->content);
1748:
1749: /*
1750: * Just handle the content as a set of chars.
1751: */
1.59 veillard 1752: htmlCheckParagraph(ctxt);
1.1 daniel 1753: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1754: ctxt->sax->characters(ctxt->userData, entity->content, len);
1755:
1756: }
1757:
1758: /**
1.59 veillard 1759: * htmlNewDocNoDtD:
1.1 daniel 1760: * @URI: URI for the dtd, or NULL
1761: * @ExternalID: the external ID of the DTD, or NULL
1762: *
1.59 veillard 1763: * Returns a new document, do not intialize the DTD if not provided
1.1 daniel 1764: */
1765: htmlDocPtr
1.59 veillard 1766: htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) {
1.1 daniel 1767: xmlDocPtr cur;
1768:
1769: /*
1770: * Allocate a new document and fill the fields.
1771: */
1.11 daniel 1772: cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
1.1 daniel 1773: if (cur == NULL) {
1.81 veillard 1774: xmlGenericError(xmlGenericErrorContext,
1775: "xmlNewDoc : malloc failed\n");
1.1 daniel 1776: return(NULL);
1777: }
1.10 daniel 1778: memset(cur, 0, sizeof(xmlDoc));
1.1 daniel 1779:
1.20 daniel 1780: cur->type = XML_HTML_DOCUMENT_NODE;
1.1 daniel 1781: cur->version = NULL;
1782: cur->intSubset = NULL;
1.59 veillard 1783: if ((ExternalID != NULL) ||
1784: (URI != NULL))
1.28 daniel 1785: xmlCreateIntSubset(cur, BAD_CAST "HTML", ExternalID, URI);
1.41 daniel 1786: cur->doc = cur;
1.1 daniel 1787: cur->name = NULL;
1.37 daniel 1788: cur->children = NULL;
1.1 daniel 1789: cur->extSubset = NULL;
1790: cur->oldNs = NULL;
1791: cur->encoding = NULL;
1792: cur->standalone = 1;
1793: cur->compression = 0;
1.12 daniel 1794: cur->ids = NULL;
1795: cur->refs = NULL;
1.1 daniel 1796: #ifndef XML_WITHOUT_CORBA
1797: cur->_private = NULL;
1798: #endif
1799: return(cur);
1800: }
1801:
1.59 veillard 1802: /**
1803: * htmlNewDoc:
1804: * @URI: URI for the dtd, or NULL
1805: * @ExternalID: the external ID of the DTD, or NULL
1806: *
1807: * Returns a new document
1808: */
1809: htmlDocPtr
1810: htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
1811: if ((URI == NULL) && (ExternalID == NULL))
1812: return(htmlNewDocNoDtD(
1813: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
1814: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd"));
1815:
1816: return(htmlNewDocNoDtD(URI, ExternalID));
1817: }
1818:
1.1 daniel 1819:
1820: /************************************************************************
1821: * *
1822: * The parser itself *
1823: * Relates to http://www.w3.org/TR/html40 *
1824: * *
1825: ************************************************************************/
1826:
1827: /************************************************************************
1828: * *
1829: * The parser itself *
1830: * *
1831: ************************************************************************/
1832:
1833: /**
1834: * htmlParseHTMLName:
1835: * @ctxt: an HTML parser context
1836: *
1.26 daniel 1837: * parse an HTML tag or attribute name, note that we convert it to lowercase
1.1 daniel 1838: * since HTML names are not case-sensitive.
1839: *
1840: * Returns the Tag Name parsed or NULL
1841: */
1842:
1.14 daniel 1843: xmlChar *
1.1 daniel 1844: htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1845: xmlChar *ret = NULL;
1.1 daniel 1846: int i = 0;
1.31 daniel 1847: xmlChar loc[HTML_PARSER_BUFFER_SIZE];
1.1 daniel 1848:
1849: if (!IS_LETTER(CUR) && (CUR != '_') &&
1850: (CUR != ':')) return(NULL);
1851:
1.31 daniel 1852: while ((i < HTML_PARSER_BUFFER_SIZE) &&
1.45 daniel 1853: ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1.76 veillard 1854: (CUR == ':') || (CUR == '-') || (CUR == '_'))) {
1.26 daniel 1855: if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20;
1.1 daniel 1856: else loc[i] = CUR;
1857: i++;
1858:
1859: NEXT;
1860: }
1861:
1862: ret = xmlStrndup(loc, i);
1863:
1864: return(ret);
1865: }
1866:
1867: /**
1868: * htmlParseName:
1869: * @ctxt: an HTML parser context
1870: *
1871: * parse an HTML name, this routine is case sensistive.
1872: *
1873: * Returns the Name parsed or NULL
1874: */
1875:
1.14 daniel 1876: xmlChar *
1.1 daniel 1877: htmlParseName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1878: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1879: int len = 0;
1.1 daniel 1880:
1.5 daniel 1881: GROW;
1882: if (!IS_LETTER(CUR) && (CUR != '_')) {
1883: return(NULL);
1884: }
1.1 daniel 1885:
1886: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1887: (CUR == '.') || (CUR == '-') ||
1888: (CUR == '_') || (CUR == ':') ||
1889: (IS_COMBINING(CUR)) ||
1.5 daniel 1890: (IS_EXTENDER(CUR))) {
1891: buf[len++] = CUR;
1.1 daniel 1892: NEXT;
1.5 daniel 1893: if (len >= HTML_MAX_NAMELEN) {
1.81 veillard 1894: xmlGenericError(xmlGenericErrorContext,
1.5 daniel 1895: "htmlParseName: reached HTML_MAX_NAMELEN limit\n");
1896: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1897: (CUR == '.') || (CUR == '-') ||
1898: (CUR == '_') || (CUR == ':') ||
1899: (IS_COMBINING(CUR)) ||
1900: (IS_EXTENDER(CUR)))
1901: NEXT;
1902: break;
1903: }
1904: }
1905: return(xmlStrndup(buf, len));
1.1 daniel 1906: }
1907:
1908: /**
1909: * htmlParseHTMLAttribute:
1910: * @ctxt: an HTML parser context
1.19 daniel 1911: * @stop: a char stop value
1.1 daniel 1912: *
1.19 daniel 1913: * parse an HTML attribute value till the stop (quote), if
1914: * stop is 0 then it stops at the first space
1.1 daniel 1915: *
1.19 daniel 1916: * Returns the attribute parsed or NULL
1.1 daniel 1917: */
1918:
1.14 daniel 1919: xmlChar *
1.19 daniel 1920: htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) {
1.32 daniel 1921: xmlChar *buffer = NULL;
1922: int buffer_size = 0;
1923: xmlChar *out = NULL;
1924: xmlChar *name = NULL;
1925:
1926: xmlChar *cur = NULL;
1927: htmlEntityDescPtr ent;
1928:
1929: /*
1930: * allocate a translation buffer.
1931: */
1.77 veillard 1932: buffer_size = HTML_PARSER_BUFFER_SIZE;
1.32 daniel 1933: buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
1934: if (buffer == NULL) {
1935: perror("htmlParseHTMLAttribute: malloc failed");
1936: return(NULL);
1937: }
1938: out = buffer;
1939:
1940: /*
1941: * Ok loop until we reach one of the ending chars
1942: */
1943: while ((CUR != 0) && (CUR != stop) && (CUR != '>')) {
1944: if ((stop == 0) && (IS_BLANK(CUR))) break;
1945: if (CUR == '&') {
1946: if (NXT(1) == '#') {
1.52 veillard 1947: unsigned int c;
1948: int bits;
1949:
1950: c = htmlParseCharRef(ctxt);
1951: if (c < 0x80)
1952: { *out++ = c; bits= -6; }
1953: else if (c < 0x800)
1954: { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
1955: else if (c < 0x10000)
1956: { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
1957: else
1958: { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
1959:
1960: for ( ; bits >= 0; bits-= 6) {
1961: *out++ = ((c >> bits) & 0x3F) | 0x80;
1962: }
1.32 daniel 1963: } else {
1964: ent = htmlParseEntityRef(ctxt, &name);
1965: if (name == NULL) {
1966: *out++ = '&';
1967: if (out - buffer > buffer_size - 100) {
1968: int index = out - buffer;
1969:
1970: growBuffer(buffer);
1971: out = &buffer[index];
1972: }
1.52 veillard 1973: } else if (ent == NULL) {
1.32 daniel 1974: *out++ = '&';
1975: cur = name;
1976: while (*cur != 0) {
1977: if (out - buffer > buffer_size - 100) {
1978: int index = out - buffer;
1979:
1980: growBuffer(buffer);
1981: out = &buffer[index];
1982: }
1983: *out++ = *cur++;
1984: }
1985: xmlFree(name);
1986: } else {
1.52 veillard 1987: unsigned int c;
1988: int bits;
1989:
1.32 daniel 1990: if (out - buffer > buffer_size - 100) {
1991: int index = out - buffer;
1992:
1993: growBuffer(buffer);
1994: out = &buffer[index];
1995: }
1.52 veillard 1996: c = (xmlChar)ent->value;
1997: if (c < 0x80)
1998: { *out++ = c; bits= -6; }
1999: else if (c < 0x800)
2000: { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2001: else if (c < 0x10000)
2002: { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2003: else
2004: { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2005:
2006: for ( ; bits >= 0; bits-= 6) {
2007: *out++ = ((c >> bits) & 0x3F) | 0x80;
2008: }
1.32 daniel 2009: xmlFree(name);
2010: }
2011: }
2012: } else {
1.52 veillard 2013: unsigned int c;
1.68 veillard 2014: int bits, l;
1.52 veillard 2015:
1.32 daniel 2016: if (out - buffer > buffer_size - 100) {
1.52 veillard 2017: int index = out - buffer;
2018:
2019: growBuffer(buffer);
2020: out = &buffer[index];
2021: }
1.68 veillard 2022: c = CUR_CHAR(l);
1.52 veillard 2023: if (c < 0x80)
2024: { *out++ = c; bits= -6; }
2025: else if (c < 0x800)
2026: { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2027: else if (c < 0x10000)
2028: { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2029: else
2030: { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2031:
2032: for ( ; bits >= 0; bits-= 6) {
2033: *out++ = ((c >> bits) & 0x3F) | 0x80;
1.32 daniel 2034: }
2035: NEXT;
2036: }
2037: }
2038: *out++ = 0;
2039: return(buffer);
1.1 daniel 2040: }
2041:
2042: /**
2043: * htmlParseNmtoken:
2044: * @ctxt: an HTML parser context
2045: *
2046: * parse an HTML Nmtoken.
2047: *
2048: * Returns the Nmtoken parsed or NULL
2049: */
2050:
1.14 daniel 2051: xmlChar *
1.1 daniel 2052: htmlParseNmtoken(htmlParserCtxtPtr ctxt) {
1.14 daniel 2053: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 2054: int len = 0;
1.1 daniel 2055:
1.5 daniel 2056: GROW;
1.1 daniel 2057: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
2058: (CUR == '.') || (CUR == '-') ||
2059: (CUR == '_') || (CUR == ':') ||
2060: (IS_COMBINING(CUR)) ||
1.5 daniel 2061: (IS_EXTENDER(CUR))) {
2062: buf[len++] = CUR;
1.1 daniel 2063: NEXT;
1.5 daniel 2064: if (len >= HTML_MAX_NAMELEN) {
1.81 veillard 2065: xmlGenericError(xmlGenericErrorContext,
1.5 daniel 2066: "htmlParseNmtoken: reached HTML_MAX_NAMELEN limit\n");
2067: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
2068: (CUR == '.') || (CUR == '-') ||
2069: (CUR == '_') || (CUR == ':') ||
2070: (IS_COMBINING(CUR)) ||
2071: (IS_EXTENDER(CUR)))
2072: NEXT;
2073: break;
2074: }
2075: }
2076: return(xmlStrndup(buf, len));
1.1 daniel 2077: }
2078:
2079: /**
2080: * htmlParseEntityRef:
2081: * @ctxt: an HTML parser context
2082: * @str: location to store the entity name
2083: *
2084: * parse an HTML ENTITY references
2085: *
2086: * [68] EntityRef ::= '&' Name ';'
2087: *
2088: * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
2089: * if non-NULL *str will have to be freed by the caller.
2090: */
2091: htmlEntityDescPtr
1.14 daniel 2092: htmlParseEntityRef(htmlParserCtxtPtr ctxt, xmlChar **str) {
2093: xmlChar *name;
1.1 daniel 2094: htmlEntityDescPtr ent = NULL;
2095: *str = NULL;
2096:
2097: if (CUR == '&') {
2098: NEXT;
2099: name = htmlParseName(ctxt);
2100: if (name == NULL) {
2101: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2102: ctxt->sax->error(ctxt->userData, "htmlParseEntityRef: no name\n");
2103: ctxt->wellFormed = 0;
2104: } else {
1.5 daniel 2105: GROW;
1.1 daniel 2106: if (CUR == ';') {
2107: *str = name;
2108:
2109: /*
2110: * Lookup the entity in the table.
2111: */
2112: ent = htmlEntityLookup(name);
1.32 daniel 2113: if (ent != NULL) /* OK that's ugly !!! */
2114: NEXT;
1.1 daniel 2115: } else {
2116: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2117: ctxt->sax->error(ctxt->userData,
2118: "htmlParseEntityRef: expecting ';'\n");
1.32 daniel 2119: *str = name;
1.1 daniel 2120: }
2121: }
2122: }
2123: return(ent);
2124: }
2125:
2126: /**
2127: * htmlParseAttValue:
2128: * @ctxt: an HTML parser context
2129: *
2130: * parse a value for an attribute
2131: * Note: the parser won't do substitution of entities here, this
2132: * will be handled later in xmlStringGetNodeList, unless it was
2133: * asked for ctxt->replaceEntities != 0
2134: *
2135: * Returns the AttValue parsed or NULL.
2136: */
2137:
1.14 daniel 2138: xmlChar *
1.1 daniel 2139: htmlParseAttValue(htmlParserCtxtPtr ctxt) {
1.14 daniel 2140: xmlChar *ret = NULL;
1.1 daniel 2141:
2142: if (CUR == '"') {
2143: NEXT;
1.19 daniel 2144: ret = htmlParseHTMLAttribute(ctxt, '"');
1.1 daniel 2145: if (CUR != '"') {
2146: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2147: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
2148: ctxt->wellFormed = 0;
2149: } else
2150: NEXT;
2151: } else if (CUR == '\'') {
2152: NEXT;
1.19 daniel 2153: ret = htmlParseHTMLAttribute(ctxt, '\'');
1.1 daniel 2154: if (CUR != '\'') {
2155: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2156: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
2157: ctxt->wellFormed = 0;
2158: } else
2159: NEXT;
2160: } else {
2161: /*
2162: * That's an HTMLism, the attribute value may not be quoted
2163: */
1.19 daniel 2164: ret = htmlParseHTMLAttribute(ctxt, 0);
1.1 daniel 2165: if (ret == NULL) {
2166: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2167: ctxt->sax->error(ctxt->userData, "AttValue: no value found\n");
2168: ctxt->wellFormed = 0;
2169: }
2170: }
2171: return(ret);
2172: }
2173:
2174: /**
2175: * htmlParseSystemLiteral:
2176: * @ctxt: an HTML parser context
2177: *
2178: * parse an HTML Literal
2179: *
2180: * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
2181: *
2182: * Returns the SystemLiteral parsed or NULL
2183: */
2184:
1.14 daniel 2185: xmlChar *
1.1 daniel 2186: htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 2187: const xmlChar *q;
2188: xmlChar *ret = NULL;
1.1 daniel 2189:
2190: if (CUR == '"') {
2191: NEXT;
2192: q = CUR_PTR;
2193: while ((IS_CHAR(CUR)) && (CUR != '"'))
2194: NEXT;
2195: if (!IS_CHAR(CUR)) {
2196: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2197: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
2198: ctxt->wellFormed = 0;
2199: } else {
2200: ret = xmlStrndup(q, CUR_PTR - q);
2201: NEXT;
2202: }
2203: } else if (CUR == '\'') {
2204: NEXT;
2205: q = CUR_PTR;
2206: while ((IS_CHAR(CUR)) && (CUR != '\''))
2207: NEXT;
2208: if (!IS_CHAR(CUR)) {
2209: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2210: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
2211: ctxt->wellFormed = 0;
2212: } else {
2213: ret = xmlStrndup(q, CUR_PTR - q);
2214: NEXT;
2215: }
2216: } else {
2217: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.38 daniel 2218: ctxt->sax->error(ctxt->userData,
2219: "SystemLiteral \" or ' expected\n");
1.1 daniel 2220: ctxt->wellFormed = 0;
2221: }
2222:
2223: return(ret);
2224: }
2225:
2226: /**
2227: * htmlParsePubidLiteral:
2228: * @ctxt: an HTML parser context
2229: *
2230: * parse an HTML public literal
2231: *
2232: * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
2233: *
2234: * Returns the PubidLiteral parsed or NULL.
2235: */
2236:
1.14 daniel 2237: xmlChar *
1.1 daniel 2238: htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 2239: const xmlChar *q;
2240: xmlChar *ret = NULL;
1.1 daniel 2241: /*
2242: * Name ::= (Letter | '_') (NameChar)*
2243: */
2244: if (CUR == '"') {
2245: NEXT;
2246: q = CUR_PTR;
2247: while (IS_PUBIDCHAR(CUR)) NEXT;
2248: if (CUR != '"') {
2249: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2250: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
2251: ctxt->wellFormed = 0;
2252: } else {
2253: ret = xmlStrndup(q, CUR_PTR - q);
2254: NEXT;
2255: }
2256: } else if (CUR == '\'') {
2257: NEXT;
2258: q = CUR_PTR;
2259: while ((IS_LETTER(CUR)) && (CUR != '\''))
2260: NEXT;
2261: if (!IS_LETTER(CUR)) {
2262: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2263: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
2264: ctxt->wellFormed = 0;
2265: } else {
2266: ret = xmlStrndup(q, CUR_PTR - q);
2267: NEXT;
2268: }
2269: } else {
2270: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2271: ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n");
2272: ctxt->wellFormed = 0;
2273: }
2274:
2275: return(ret);
2276: }
2277:
2278: /**
1.77 veillard 2279: * htmlParseScript:
2280: * @ctxt: an HTML parser context
2281: *
2282: * parse the content of an HTML SCRIPT or STYLE element
2283: * http://www.w3.org/TR/html4/sgml/dtd.html#Script
2284: * http://www.w3.org/TR/html4/sgml/dtd.html#StyleSheet
2285: * http://www.w3.org/TR/html4/types.html#type-script
2286: * http://www.w3.org/TR/html4/types.html#h-6.15
2287: * http://www.w3.org/TR/html4/appendix/notes.html#h-B.3.2.1
2288: *
2289: * Script data ( %Script; in the DTD) can be the content of the SCRIPT
2290: * element and the value of intrinsic event attributes. User agents must
2291: * not evaluate script data as HTML markup but instead must pass it on as
2292: * data to a script engine.
2293: * NOTES:
2294: * - The content is passed like CDATA
2295: * - the attributes for style and scripting "onXXX" are also described
2296: * as CDATA but SGML allows entities references in attributes so their
2297: * processing is identical as other attributes
2298: */
2299: void
2300: htmlParseScript(htmlParserCtxtPtr ctxt) {
2301: xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 1];
2302: int nbchar = 0;
2303: xmlChar cur;
2304:
2305: SHRINK;
2306: cur = CUR;
2307: while (IS_CHAR(cur)) {
2308: if ((cur == '<') && (NXT(1) == '/')) {
2309: /*
2310: * One should break here, the specification is clear:
2311: * Authors should therefore escape "</" within the content.
2312: * Escape mechanisms are specific to each scripting or
2313: * style sheet language.
2314: */
2315: if (((NXT(2) >= 'A') && (NXT(2) <= 'Z')) ||
2316: ((NXT(2) >= 'a') && (NXT(2) <= 'z')))
2317: break; /* while */
2318: }
2319: buf[nbchar++] = cur;
2320: if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2321: if (ctxt->sax->cdataBlock!= NULL) {
2322: /*
2323: * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
2324: */
2325: ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
2326: }
2327: nbchar = 0;
2328: }
2329: NEXT;
2330: cur = CUR;
2331: }
2332: if ((nbchar != 0) && (ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2333: if (ctxt->sax->cdataBlock!= NULL) {
2334: /*
2335: * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
2336: */
2337: ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
2338: }
2339: }
2340: }
2341:
2342:
2343: /**
1.1 daniel 2344: * htmlParseCharData:
2345: * @ctxt: an HTML parser context
2346: * @cdata: int indicating whether we are within a CDATA section
2347: *
2348: * parse a CharData section.
2349: * if we are within a CDATA section ']]>' marks an end of section.
2350: *
2351: * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
2352: */
2353:
2354: void
2355: htmlParseCharData(htmlParserCtxtPtr ctxt, int cdata) {
1.53 veillard 2356: xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
2357: int nbchar = 0;
2358: int cur, l;
2359:
2360: SHRINK;
2361: cur = CUR_CHAR(l);
2362: while (((cur != '<') || (ctxt->token == '<')) &&
2363: ((cur != '&') || (ctxt->token == '&')) &&
2364: (IS_CHAR(cur))) {
2365: COPY_BUF(l,buf,nbchar,cur);
2366: if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2367: /*
2368: * Ok the segment is to be consumed as chars.
2369: */
2370: if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2371: if (areBlanks(ctxt, buf, nbchar)) {
2372: if (ctxt->sax->ignorableWhitespace != NULL)
2373: ctxt->sax->ignorableWhitespace(ctxt->userData,
2374: buf, nbchar);
2375: } else {
1.59 veillard 2376: htmlCheckParagraph(ctxt);
1.53 veillard 2377: if (ctxt->sax->characters != NULL)
2378: ctxt->sax->characters(ctxt->userData, buf, nbchar);
2379: }
1.1 daniel 2380: }
1.53 veillard 2381: nbchar = 0;
1.1 daniel 2382: }
1.53 veillard 2383: NEXTL(l);
2384: cur = CUR_CHAR(l);
2385: }
2386: if (nbchar != 0) {
2387: /*
2388: * Ok the segment is to be consumed as chars.
2389: */
2390: if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2391: if (areBlanks(ctxt, buf, nbchar)) {
2392: if (ctxt->sax->ignorableWhitespace != NULL)
2393: ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar);
2394: } else {
1.59 veillard 2395: htmlCheckParagraph(ctxt);
1.53 veillard 2396: if (ctxt->sax->characters != NULL)
2397: ctxt->sax->characters(ctxt->userData, buf, nbchar);
1.25 daniel 2398: }
2399: }
1.1 daniel 2400: }
2401: }
2402:
2403: /**
2404: * htmlParseExternalID:
2405: * @ctxt: an HTML parser context
1.14 daniel 2406: * @publicID: a xmlChar** receiving PubidLiteral
1.1 daniel 2407: * @strict: indicate whether we should restrict parsing to only
2408: * production [75], see NOTE below
2409: *
2410: * Parse an External ID or a Public ID
2411: *
2412: * NOTE: Productions [75] and [83] interract badly since [75] can generate
2413: * 'PUBLIC' S PubidLiteral S SystemLiteral
2414: *
2415: * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
2416: * | 'PUBLIC' S PubidLiteral S SystemLiteral
2417: *
2418: * [83] PublicID ::= 'PUBLIC' S PubidLiteral
2419: *
2420: * Returns the function returns SystemLiteral and in the second
2421: * case publicID receives PubidLiteral, is strict is off
2422: * it is possible to return NULL and have publicID set.
2423: */
2424:
1.14 daniel 2425: xmlChar *
2426: htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID, int strict) {
2427: xmlChar *URI = NULL;
1.1 daniel 2428:
2429: if ((UPPER == 'S') && (UPP(1) == 'Y') &&
2430: (UPP(2) == 'S') && (UPP(3) == 'T') &&
2431: (UPP(4) == 'E') && (UPP(5) == 'M')) {
2432: SKIP(6);
2433: if (!IS_BLANK(CUR)) {
2434: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2435: ctxt->sax->error(ctxt->userData,
2436: "Space required after 'SYSTEM'\n");
2437: ctxt->wellFormed = 0;
2438: }
2439: SKIP_BLANKS;
2440: URI = htmlParseSystemLiteral(ctxt);
2441: if (URI == NULL) {
2442: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2443: ctxt->sax->error(ctxt->userData,
2444: "htmlParseExternalID: SYSTEM, no URI\n");
2445: ctxt->wellFormed = 0;
2446: }
2447: } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
2448: (UPP(2) == 'B') && (UPP(3) == 'L') &&
2449: (UPP(4) == 'I') && (UPP(5) == 'C')) {
2450: SKIP(6);
2451: if (!IS_BLANK(CUR)) {
2452: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2453: ctxt->sax->error(ctxt->userData,
2454: "Space required after 'PUBLIC'\n");
2455: ctxt->wellFormed = 0;
2456: }
2457: SKIP_BLANKS;
2458: *publicID = htmlParsePubidLiteral(ctxt);
2459: if (*publicID == NULL) {
2460: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2461: ctxt->sax->error(ctxt->userData,
2462: "htmlParseExternalID: PUBLIC, no Public Identifier\n");
2463: ctxt->wellFormed = 0;
2464: }
1.5 daniel 2465: SKIP_BLANKS;
2466: if ((CUR == '"') || (CUR == '\'')) {
2467: URI = htmlParseSystemLiteral(ctxt);
1.1 daniel 2468: }
2469: }
2470: return(URI);
2471: }
2472:
2473: /**
2474: * htmlParseComment:
2475: * @ctxt: an HTML parser context
2476: *
2477: * Parse an XML (SGML) comment <!-- .... -->
2478: *
2479: * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
2480: */
2481: void
1.31 daniel 2482: htmlParseComment(htmlParserCtxtPtr ctxt) {
1.25 daniel 2483: xmlChar *buf = NULL;
1.56 veillard 2484: int len;
1.31 daniel 2485: int size = HTML_PARSER_BUFFER_SIZE;
1.56 veillard 2486: int q, ql;
2487: int r, rl;
2488: int cur, l;
2489: xmlParserInputState state;
1.1 daniel 2490:
2491: /*
2492: * Check that there is a comment right here.
2493: */
1.56 veillard 2494: if ((RAW != '<') || (NXT(1) != '!') ||
1.1 daniel 2495: (NXT(2) != '-') || (NXT(3) != '-')) return;
2496:
1.56 veillard 2497: state = ctxt->instate;
2498: ctxt->instate = XML_PARSER_COMMENT;
2499: SHRINK;
2500: SKIP(4);
1.25 daniel 2501: buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar));
2502: if (buf == NULL) {
1.81 veillard 2503: xmlGenericError(xmlGenericErrorContext,
2504: "malloc of %d byte failed\n", size);
1.56 veillard 2505: ctxt->instate = state;
1.25 daniel 2506: return;
2507: }
1.56 veillard 2508: q = CUR_CHAR(ql);
2509: NEXTL(ql);
2510: r = CUR_CHAR(rl);
2511: NEXTL(rl);
2512: cur = CUR_CHAR(l);
2513: len = 0;
2514: while (IS_CHAR(cur) &&
2515: ((cur != '>') ||
2516: (r != '-') || (q != '-'))) {
2517: if (len + 5 >= size) {
1.25 daniel 2518: size *= 2;
1.50 veillard 2519: buf = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
1.25 daniel 2520: if (buf == NULL) {
1.81 veillard 2521: xmlGenericError(xmlGenericErrorContext,
2522: "realloc of %d byte failed\n", size);
1.56 veillard 2523: ctxt->instate = state;
1.25 daniel 2524: return;
2525: }
2526: }
1.56 veillard 2527: COPY_BUF(ql,buf,len,q);
1.25 daniel 2528: q = r;
1.56 veillard 2529: ql = rl;
2530: r = cur;
2531: rl = l;
2532: NEXTL(l);
2533: cur = CUR_CHAR(l);
2534: if (cur == 0) {
2535: SHRINK;
2536: GROW;
2537: cur = CUR_CHAR(l);
2538: }
1.1 daniel 2539: }
1.56 veillard 2540: buf[len] = 0;
2541: if (!IS_CHAR(cur)) {
1.67 veillard 2542: ctxt->errNo = XML_ERR_COMMENT_NOT_FINISHED;
1.1 daniel 2543: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.56 veillard 2544: ctxt->sax->error(ctxt->userData,
2545: "Comment not terminated \n<!--%.50s\n", buf);
1.1 daniel 2546: ctxt->wellFormed = 0;
1.56 veillard 2547: xmlFree(buf);
1.1 daniel 2548: } else {
2549: NEXT;
1.56 veillard 2550: if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) &&
2551: (!ctxt->disableSAX))
1.31 daniel 2552: ctxt->sax->comment(ctxt->userData, buf);
1.56 veillard 2553: xmlFree(buf);
1.1 daniel 2554: }
1.56 veillard 2555: ctxt->instate = state;
1.1 daniel 2556: }
2557:
2558: /**
2559: * htmlParseCharRef:
2560: * @ctxt: an HTML parser context
2561: *
2562: * parse Reference declarations
2563: *
2564: * [66] CharRef ::= '&#' [0-9]+ ';' |
2565: * '&#x' [0-9a-fA-F]+ ';'
2566: *
2567: * Returns the value parsed (as an int)
2568: */
2569: int
2570: htmlParseCharRef(htmlParserCtxtPtr ctxt) {
2571: int val = 0;
2572:
2573: if ((CUR == '&') && (NXT(1) == '#') &&
2574: (NXT(2) == 'x')) {
2575: SKIP(3);
2576: while (CUR != ';') {
2577: if ((CUR >= '0') && (CUR <= '9'))
2578: val = val * 16 + (CUR - '0');
2579: else if ((CUR >= 'a') && (CUR <= 'f'))
2580: val = val * 16 + (CUR - 'a') + 10;
2581: else if ((CUR >= 'A') && (CUR <= 'F'))
2582: val = val * 16 + (CUR - 'A') + 10;
2583: else {
2584: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2585: ctxt->sax->error(ctxt->userData,
2586: "htmlParseCharRef: invalid hexadecimal value\n");
2587: ctxt->wellFormed = 0;
2588: val = 0;
2589: break;
2590: }
2591: NEXT;
2592: }
2593: if (CUR == ';')
2594: NEXT;
2595: } else if ((CUR == '&') && (NXT(1) == '#')) {
2596: SKIP(2);
2597: while (CUR != ';') {
2598: if ((CUR >= '0') && (CUR <= '9'))
2599: val = val * 10 + (CUR - '0');
2600: else {
2601: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2602: ctxt->sax->error(ctxt->userData,
2603: "htmlParseCharRef: invalid decimal value\n");
2604: ctxt->wellFormed = 0;
2605: val = 0;
2606: break;
2607: }
2608: NEXT;
2609: }
2610: if (CUR == ';')
2611: NEXT;
2612: } else {
2613: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2614: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid value\n");
2615: ctxt->wellFormed = 0;
2616: }
2617: /*
2618: * Check the value IS_CHAR ...
2619: */
2620: if (IS_CHAR(val)) {
2621: return(val);
2622: } else {
2623: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.14 daniel 2624: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid xmlChar value %d\n",
1.1 daniel 2625: val);
2626: ctxt->wellFormed = 0;
2627: }
2628: return(0);
2629: }
2630:
2631:
2632: /**
2633: * htmlParseDocTypeDecl :
2634: * @ctxt: an HTML parser context
2635: *
2636: * parse a DOCTYPE declaration
2637: *
2638: * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
2639: * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
2640: */
2641:
2642: void
2643: htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
1.14 daniel 2644: xmlChar *name;
2645: xmlChar *ExternalID = NULL;
2646: xmlChar *URI = NULL;
1.1 daniel 2647:
2648: /*
2649: * We know that '<!DOCTYPE' has been detected.
2650: */
2651: SKIP(9);
2652:
2653: SKIP_BLANKS;
2654:
2655: /*
2656: * Parse the DOCTYPE name.
2657: */
2658: name = htmlParseName(ctxt);
2659: if (name == NULL) {
2660: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2661: ctxt->sax->error(ctxt->userData, "htmlParseDocTypeDecl : no DOCTYPE name !\n");
2662: ctxt->wellFormed = 0;
2663: }
2664: /*
2665: * Check that upper(name) == "HTML" !!!!!!!!!!!!!
2666: */
2667:
2668: SKIP_BLANKS;
2669:
2670: /*
2671: * Check for SystemID and ExternalID
2672: */
1.5 daniel 2673: URI = htmlParseExternalID(ctxt, &ExternalID, 0);
1.1 daniel 2674: SKIP_BLANKS;
2675:
2676: /*
2677: * We should be at the end of the DOCTYPE declaration.
2678: */
2679: if (CUR != '>') {
2680: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2681: ctxt->sax->error(ctxt->userData, "DOCTYPE unproperly terminated\n");
2682: ctxt->wellFormed = 0;
2683: /* We shouldn't try to resynchronize ... */
2684: }
2685: NEXT;
2686:
2687: /*
1.46 daniel 2688: * Create or update the document accordingly to the DOCTYPE
1.1 daniel 2689: */
1.46 daniel 2690: if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) &&
2691: (!ctxt->disableSAX))
2692: ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI);
1.1 daniel 2693:
2694: /*
2695: * Cleanup, since we don't use all those identifiers
2696: */
1.11 daniel 2697: if (URI != NULL) xmlFree(URI);
2698: if (ExternalID != NULL) xmlFree(ExternalID);
2699: if (name != NULL) xmlFree(name);
1.1 daniel 2700: }
2701:
2702: /**
2703: * htmlParseAttribute:
2704: * @ctxt: an HTML parser context
1.14 daniel 2705: * @value: a xmlChar ** used to store the value of the attribute
1.1 daniel 2706: *
2707: * parse an attribute
2708: *
2709: * [41] Attribute ::= Name Eq AttValue
2710: *
2711: * [25] Eq ::= S? '=' S?
2712: *
2713: * With namespace:
2714: *
2715: * [NS 11] Attribute ::= QName Eq AttValue
2716: *
2717: * Also the case QName == xmlns:??? is handled independently as a namespace
2718: * definition.
2719: *
2720: * Returns the attribute name, and the value in *value.
2721: */
2722:
1.14 daniel 2723: xmlChar *
2724: htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
1.31 daniel 2725: xmlChar *name, *val = NULL;
1.1 daniel 2726:
2727: *value = NULL;
1.74 veillard 2728: name = htmlParseHTMLName(ctxt);
1.1 daniel 2729: if (name == NULL) {
2730: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2731: ctxt->sax->error(ctxt->userData, "error parsing attribute name\n");
2732: ctxt->wellFormed = 0;
2733: return(NULL);
2734: }
2735:
2736: /*
2737: * read the value
2738: */
2739: SKIP_BLANKS;
2740: if (CUR == '=') {
2741: NEXT;
2742: SKIP_BLANKS;
2743: val = htmlParseAttValue(ctxt);
1.42 daniel 2744: /******
1.1 daniel 2745: } else {
1.42 daniel 2746: * TODO : some attribute must have values, some may not
1.1 daniel 2747: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.31 daniel 2748: ctxt->sax->warning(ctxt->userData,
1.42 daniel 2749: "No value for attribute %s\n", name); */
1.1 daniel 2750: }
2751:
2752: *value = val;
2753: return(name);
2754: }
2755:
2756: /**
1.47 daniel 2757: * htmlCheckEncoding:
2758: * @ctxt: an HTML parser context
2759: * @attvalue: the attribute value
2760: *
2761: * Checks an http-equiv attribute from a Meta tag to detect
2762: * the encoding
2763: * If a new encoding is detected the parser is switched to decode
2764: * it and pass UTF8
2765: */
2766: void
2767: htmlCheckEncoding(htmlParserCtxtPtr ctxt, const xmlChar *attvalue) {
2768: const xmlChar *encoding;
2769:
2770: if ((ctxt == NULL) || (attvalue == NULL))
2771: return;
2772:
1.69 veillard 2773: encoding = xmlStrcasestr(attvalue, BAD_CAST"charset=");
1.47 daniel 2774: if (encoding != NULL) {
2775: encoding += 8;
2776: } else {
1.69 veillard 2777: encoding = xmlStrcasestr(attvalue, BAD_CAST"charset =");
1.47 daniel 2778: if (encoding != NULL)
2779: encoding += 9;
2780: }
2781: if (encoding != NULL) {
2782: xmlCharEncoding enc;
2783: xmlCharEncodingHandlerPtr handler;
2784:
2785: while ((*encoding == ' ') || (*encoding == '\t')) encoding++;
2786:
2787: if (ctxt->input->encoding != NULL)
2788: xmlFree((xmlChar *) ctxt->input->encoding);
2789: ctxt->input->encoding = xmlStrdup(encoding);
2790:
2791: enc = xmlParseCharEncoding((const char *) encoding);
2792: /*
2793: * registered set of known encodings
2794: */
2795: if (enc != XML_CHAR_ENCODING_ERROR) {
2796: xmlSwitchEncoding(ctxt, enc);
1.53 veillard 2797: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1.47 daniel 2798: } else {
2799: /*
2800: * fallback for unknown encodings
2801: */
2802: handler = xmlFindCharEncodingHandler((const char *) encoding);
2803: if (handler != NULL) {
2804: xmlSwitchToEncoding(ctxt, handler);
1.54 veillard 2805: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1.47 daniel 2806: } else {
2807: ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
2808: }
2809: }
1.54 veillard 2810:
2811: if ((ctxt->input->buf != NULL) &&
2812: (ctxt->input->buf->encoder != NULL) &&
2813: (ctxt->input->buf->raw != NULL) &&
2814: (ctxt->input->buf->buffer != NULL)) {
2815: int nbchars;
1.56 veillard 2816: int processed;
1.54 veillard 2817:
2818: /*
2819: * convert as much as possible to the parser reading buffer.
2820: */
1.56 veillard 2821: processed = ctxt->input->cur - ctxt->input->base;
2822: xmlBufferShrink(ctxt->input->buf->buffer, processed);
1.54 veillard 2823: nbchars = xmlCharEncInFunc(ctxt->input->buf->encoder,
2824: ctxt->input->buf->buffer,
2825: ctxt->input->buf->raw);
2826: if (nbchars < 0) {
1.67 veillard 2827: ctxt->errNo = XML_ERR_INVALID_ENCODING;
1.54 veillard 2828: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2829: ctxt->sax->error(ctxt->userData,
2830: "htmlCheckEncoding: encoder error\n");
2831: }
1.56 veillard 2832: ctxt->input->base =
2833: ctxt->input->cur = ctxt->input->buf->buffer->content;
1.54 veillard 2834: }
1.47 daniel 2835: }
2836: }
2837:
2838: /**
2839: * htmlCheckMeta:
2840: * @ctxt: an HTML parser context
2841: * @atts: the attributes values
2842: *
2843: * Checks an attributes from a Meta tag
2844: */
2845: void
2846: htmlCheckMeta(htmlParserCtxtPtr ctxt, const xmlChar **atts) {
2847: int i;
2848: const xmlChar *att, *value;
2849: int http = 0;
2850: const xmlChar *content = NULL;
2851:
2852: if ((ctxt == NULL) || (atts == NULL))
2853: return;
2854:
2855: i = 0;
2856: att = atts[i++];
2857: while (att != NULL) {
2858: value = atts[i++];
1.69 veillard 2859: if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"http-equiv"))
2860: && (!xmlStrcasecmp(value, BAD_CAST"Content-Type")))
1.47 daniel 2861: http = 1;
1.69 veillard 2862: else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"content")))
1.47 daniel 2863: content = value;
2864: att = atts[i++];
2865: }
2866: if ((http) && (content != NULL))
2867: htmlCheckEncoding(ctxt, content);
2868:
2869: }
2870:
2871: /**
1.1 daniel 2872: * htmlParseStartTag:
2873: * @ctxt: an HTML parser context
2874: *
2875: * parse a start of tag either for rule element or
2876: * EmptyElement. In both case we don't parse the tag closing chars.
2877: *
2878: * [40] STag ::= '<' Name (S Attribute)* S? '>'
2879: *
2880: * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
2881: *
2882: * With namespace:
2883: *
2884: * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
2885: *
2886: * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
2887: *
2888: */
2889:
1.18 daniel 2890: void
1.1 daniel 2891: htmlParseStartTag(htmlParserCtxtPtr ctxt) {
1.14 daniel 2892: xmlChar *name;
2893: xmlChar *attname;
2894: xmlChar *attvalue;
2895: const xmlChar **atts = NULL;
1.1 daniel 2896: int nbatts = 0;
2897: int maxatts = 0;
1.47 daniel 2898: int meta = 0;
1.1 daniel 2899: int i;
2900:
1.18 daniel 2901: if (CUR != '<') return;
1.1 daniel 2902: NEXT;
2903:
1.19 daniel 2904: GROW;
1.1 daniel 2905: name = htmlParseHTMLName(ctxt);
2906: if (name == NULL) {
2907: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2908: ctxt->sax->error(ctxt->userData,
2909: "htmlParseStartTag: invalid element name\n");
2910: ctxt->wellFormed = 0;
1.80 veillard 2911: /* Dump the bogus tag like browsers do */
2912: while ((IS_CHAR(CUR)) && (CUR != '>'))
2913: NEXT;
1.18 daniel 2914: return;
1.1 daniel 2915: }
1.73 veillard 2916: if (xmlStrEqual(name, BAD_CAST"meta"))
1.47 daniel 2917: meta = 1;
1.1 daniel 2918:
2919: /*
2920: * Check for auto-closure of HTML elements.
2921: */
2922: htmlAutoClose(ctxt, name);
1.43 daniel 2923:
2924: /*
2925: * Check for implied HTML elements.
2926: */
2927: htmlCheckImplied(ctxt, name);
1.1 daniel 2928:
2929: /*
2930: * Now parse the attributes, it ends up with the ending
2931: *
2932: * (S Attribute)* S?
2933: */
2934: SKIP_BLANKS;
2935: while ((IS_CHAR(CUR)) &&
2936: (CUR != '>') &&
2937: ((CUR != '/') || (NXT(1) != '>'))) {
1.26 daniel 2938: long cons = ctxt->nbChars;
1.1 daniel 2939:
1.19 daniel 2940: GROW;
1.1 daniel 2941: attname = htmlParseAttribute(ctxt, &attvalue);
1.31 daniel 2942: if (attname != NULL) {
1.47 daniel 2943:
1.1 daniel 2944: /*
2945: * Well formedness requires at most one declaration of an attribute
2946: */
2947: for (i = 0; i < nbatts;i += 2) {
1.73 veillard 2948: if (xmlStrEqual(atts[i], attname)) {
1.1 daniel 2949: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.19 daniel 2950: ctxt->sax->error(ctxt->userData,
2951: "Attribute %s redefined\n",
2952: attname);
1.1 daniel 2953: ctxt->wellFormed = 0;
1.11 daniel 2954: xmlFree(attname);
1.31 daniel 2955: if (attvalue != NULL)
2956: xmlFree(attvalue);
1.19 daniel 2957: goto failed;
1.1 daniel 2958: }
2959: }
2960:
2961: /*
2962: * Add the pair to atts
2963: */
2964: if (atts == NULL) {
2965: maxatts = 10;
1.14 daniel 2966: atts = (const xmlChar **) xmlMalloc(maxatts * sizeof(xmlChar *));
1.1 daniel 2967: if (atts == NULL) {
1.81 veillard 2968: xmlGenericError(xmlGenericErrorContext,
2969: "malloc of %ld byte failed\n",
1.14 daniel 2970: maxatts * (long)sizeof(xmlChar *));
1.18 daniel 2971: if (name != NULL) xmlFree(name);
2972: return;
1.1 daniel 2973: }
1.23 daniel 2974: } else if (nbatts + 4 > maxatts) {
1.1 daniel 2975: maxatts *= 2;
1.71 veillard 2976: atts = (const xmlChar **) xmlRealloc((void *) atts,
2977: maxatts * sizeof(xmlChar *));
1.1 daniel 2978: if (atts == NULL) {
1.81 veillard 2979: xmlGenericError(xmlGenericErrorContext,
2980: "realloc of %ld byte failed\n",
1.14 daniel 2981: maxatts * (long)sizeof(xmlChar *));
1.18 daniel 2982: if (name != NULL) xmlFree(name);
2983: return;
1.1 daniel 2984: }
2985: }
2986: atts[nbatts++] = attname;
2987: atts[nbatts++] = attvalue;
2988: atts[nbatts] = NULL;
2989: atts[nbatts + 1] = NULL;
1.80 veillard 2990: }
2991: else {
2992: /* Dump the bogus attribute string up to the next blank or
2993: * the end of the tag. */
2994: while ((IS_CHAR(CUR)) && !(IS_BLANK(CUR)) && (CUR != '>')
2995: && ((CUR != '/') || (NXT(1) != '>')))
2996: NEXT;
1.1 daniel 2997: }
2998:
1.19 daniel 2999: failed:
1.1 daniel 3000: SKIP_BLANKS;
1.26 daniel 3001: if (cons == ctxt->nbChars) {
1.1 daniel 3002: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3003: ctxt->sax->error(ctxt->userData,
3004: "htmlParseStartTag: problem parsing attributes\n");
3005: ctxt->wellFormed = 0;
3006: break;
3007: }
3008: }
3009:
3010: /*
1.47 daniel 3011: * Handle specific association to the META tag
3012: */
3013: if (meta)
3014: htmlCheckMeta(ctxt, atts);
3015:
3016: /*
1.1 daniel 3017: * SAX: Start of Element !
3018: */
1.15 daniel 3019: htmlnamePush(ctxt, xmlStrdup(name));
1.18 daniel 3020: #ifdef DEBUG
1.81 veillard 3021: xmlGenericError(xmlGenericErrorContext,"Start of element %s: pushed %s\n", name, ctxt->name);
1.18 daniel 3022: #endif
1.1 daniel 3023: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
3024: ctxt->sax->startElement(ctxt->userData, name, atts);
3025:
3026: if (atts != NULL) {
1.31 daniel 3027: for (i = 0;i < nbatts;i++) {
3028: if (atts[i] != NULL)
3029: xmlFree((xmlChar *) atts[i]);
3030: }
1.45 daniel 3031: xmlFree((void *) atts);
1.1 daniel 3032: }
1.18 daniel 3033: if (name != NULL) xmlFree(name);
1.1 daniel 3034: }
3035:
3036: /**
3037: * htmlParseEndTag:
3038: * @ctxt: an HTML parser context
3039: *
3040: * parse an end of tag
3041: *
3042: * [42] ETag ::= '</' Name S? '>'
3043: *
3044: * With namespace
3045: *
3046: * [NS 9] ETag ::= '</' QName S? '>'
3047: */
3048:
3049: void
1.18 daniel 3050: htmlParseEndTag(htmlParserCtxtPtr ctxt) {
1.14 daniel 3051: xmlChar *name;
1.15 daniel 3052: xmlChar *oldname;
1.1 daniel 3053: int i;
3054:
3055: if ((CUR != '<') || (NXT(1) != '/')) {
3056: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3057: ctxt->sax->error(ctxt->userData, "htmlParseEndTag: '</' not found\n");
3058: ctxt->wellFormed = 0;
3059: return;
3060: }
3061: SKIP(2);
3062:
3063: name = htmlParseHTMLName(ctxt);
1.24 daniel 3064: if (name == NULL) return;
1.1 daniel 3065:
3066: /*
3067: * We should definitely be at the ending "S? '>'" part
3068: */
3069: SKIP_BLANKS;
3070: if ((!IS_CHAR(CUR)) || (CUR != '>')) {
3071: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3072: ctxt->sax->error(ctxt->userData, "End tag : expected '>'\n");
3073: ctxt->wellFormed = 0;
3074: } else
3075: NEXT;
3076:
3077: /*
1.18 daniel 3078: * If the name read is not one of the element in the parsing stack
3079: * then return, it's just an error.
1.1 daniel 3080: */
1.18 daniel 3081: for (i = (ctxt->nameNr - 1);i >= 0;i--) {
1.73 veillard 3082: if (xmlStrEqual(name, ctxt->nameTab[i])) break;
1.1 daniel 3083: }
3084: if (i < 0) {
3085: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.18 daniel 3086: ctxt->sax->error(ctxt->userData,
3087: "Unexpected end tag : %s\n", name);
1.11 daniel 3088: xmlFree(name);
1.1 daniel 3089: ctxt->wellFormed = 0;
3090: return;
3091: }
3092:
1.18 daniel 3093:
1.1 daniel 3094: /*
3095: * Check for auto-closure of HTML elements.
3096: */
1.18 daniel 3097:
1.1 daniel 3098: htmlAutoCloseOnClose(ctxt, name);
3099:
3100: /*
3101: * Well formedness constraints, opening and closing must match.
3102: * With the exception that the autoclose may have popped stuff out
3103: * of the stack.
3104: */
1.73 veillard 3105: if (!xmlStrEqual(name, ctxt->name)) {
1.18 daniel 3106: #ifdef DEBUG
1.81 veillard 3107: xmlGenericError(xmlGenericErrorContext,"End of tag %s: expecting %s\n", name, ctxt->name);
1.18 daniel 3108: #endif
1.15 daniel 3109: if ((ctxt->name != NULL) &&
1.73 veillard 3110: (!xmlStrEqual(ctxt->name, name))) {
1.1 daniel 3111: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3112: ctxt->sax->error(ctxt->userData,
3113: "Opening and ending tag mismatch: %s and %s\n",
1.15 daniel 3114: name, ctxt->name);
1.1 daniel 3115: ctxt->wellFormed = 0;
3116: }
3117: }
3118:
3119: /*
3120: * SAX: End of Tag
3121: */
1.15 daniel 3122: oldname = ctxt->name;
1.73 veillard 3123: if ((oldname != NULL) && (xmlStrEqual(oldname, name))) {
1.18 daniel 3124: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3125: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 3126: oldname = htmlnamePop(ctxt);
1.18 daniel 3127: if (oldname != NULL) {
3128: #ifdef DEBUG
1.81 veillard 3129: xmlGenericError(xmlGenericErrorContext,"End of tag %s: popping out %s\n", name, oldname);
1.18 daniel 3130: #endif
3131: xmlFree(oldname);
3132: #ifdef DEBUG
3133: } else {
1.81 veillard 3134: xmlGenericError(xmlGenericErrorContext,"End of tag %s: stack empty !!!\n", name);
1.18 daniel 3135: #endif
3136: }
3137: }
1.1 daniel 3138:
3139: if (name != NULL)
1.11 daniel 3140: xmlFree(name);
1.1 daniel 3141:
3142: return;
3143: }
3144:
3145:
3146: /**
3147: * htmlParseReference:
3148: * @ctxt: an HTML parser context
3149: *
3150: * parse and handle entity references in content,
3151: * this will end-up in a call to character() since this is either a
3152: * CharRef, or a predefined entity.
3153: */
3154: void
3155: htmlParseReference(htmlParserCtxtPtr ctxt) {
3156: htmlEntityDescPtr ent;
1.52 veillard 3157: xmlChar out[6];
1.14 daniel 3158: xmlChar *name;
1.1 daniel 3159: if (CUR != '&') return;
3160:
3161: if (NXT(1) == '#') {
1.52 veillard 3162: unsigned int c;
3163: int bits, i = 0;
3164:
3165: c = htmlParseCharRef(ctxt);
3166: if (c < 0x80) { out[i++]= c; bits= -6; }
3167: else if (c < 0x800) { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3168: else if (c < 0x10000) { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3169: else { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3170:
3171: for ( ; bits >= 0; bits-= 6) {
3172: out[i++]= ((c >> bits) & 0x3F) | 0x80;
3173: }
3174: out[i] = 0;
3175:
1.59 veillard 3176: htmlCheckParagraph(ctxt);
1.1 daniel 3177: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1.52 veillard 3178: ctxt->sax->characters(ctxt->userData, out, i);
1.1 daniel 3179: } else {
3180: ent = htmlParseEntityRef(ctxt, &name);
1.32 daniel 3181: if (name == NULL) {
1.59 veillard 3182: htmlCheckParagraph(ctxt);
1.58 veillard 3183: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3184: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
1.32 daniel 3185: return;
3186: }
1.52 veillard 3187: if ((ent == NULL) || (ent->value <= 0)) {
1.59 veillard 3188: htmlCheckParagraph(ctxt);
1.1 daniel 3189: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
1.8 daniel 3190: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
1.1 daniel 3191: ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
1.32 daniel 3192: /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
1.1 daniel 3193: }
3194: } else {
1.52 veillard 3195: unsigned int c;
3196: int bits, i = 0;
3197:
3198: c = ent->value;
3199: if (c < 0x80)
3200: { out[i++]= c; bits= -6; }
3201: else if (c < 0x800)
3202: { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3203: else if (c < 0x10000)
3204: { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3205: else
3206: { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3207:
3208: for ( ; bits >= 0; bits-= 6) {
3209: out[i++]= ((c >> bits) & 0x3F) | 0x80;
3210: }
3211: out[i] = 0;
3212:
1.59 veillard 3213: htmlCheckParagraph(ctxt);
1.1 daniel 3214: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1.52 veillard 3215: ctxt->sax->characters(ctxt->userData, out, i);
1.1 daniel 3216: }
1.11 daniel 3217: xmlFree(name);
1.1 daniel 3218: }
3219: }
3220:
3221: /**
3222: * htmlParseContent:
3223: * @ctxt: an HTML parser context
3224: * @name: the node name
3225: *
3226: * Parse a content: comment, sub-element, reference or text.
3227: *
3228: */
3229:
3230: void
1.18 daniel 3231: htmlParseContent(htmlParserCtxtPtr ctxt) {
1.15 daniel 3232: xmlChar *currentNode;
1.18 daniel 3233: int depth;
1.1 daniel 3234:
1.26 daniel 3235: currentNode = xmlStrdup(ctxt->name);
1.18 daniel 3236: depth = ctxt->nameNr;
3237: while (1) {
1.26 daniel 3238: long cons = ctxt->nbChars;
1.1 daniel 3239:
1.18 daniel 3240: GROW;
3241: /*
3242: * Our tag or one of it's parent or children is ending.
3243: */
3244: if ((CUR == '<') && (NXT(1) == '/')) {
3245: htmlParseEndTag(ctxt);
1.26 daniel 3246: if (currentNode != NULL) xmlFree(currentNode);
1.18 daniel 3247: return;
3248: }
3249:
3250: /*
3251: * Has this node been popped out during parsing of
3252: * the next element
3253: */
1.73 veillard 3254: if ((!xmlStrEqual(currentNode, ctxt->name)) &&
1.26 daniel 3255: (depth >= ctxt->nameNr)) {
3256: if (currentNode != NULL) xmlFree(currentNode);
3257: return;
3258: }
1.18 daniel 3259:
1.77 veillard 3260: if ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
3261: (xmlStrEqual(currentNode, BAD_CAST"style"))) {
3262: /*
3263: * Handle SCRIPT/STYLE separately
3264: */
3265: htmlParseScript(ctxt);
3266: } else {
3267: /*
3268: * Sometimes DOCTYPE arrives in the middle of the document
3269: */
3270: if ((CUR == '<') && (NXT(1) == '!') &&
3271: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3272: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3273: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3274: (UPP(8) == 'E')) {
3275: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3276: ctxt->sax->error(ctxt->userData,
3277: "Misplaced DOCTYPE declaration\n");
3278: ctxt->wellFormed = 0;
3279: htmlParseDocTypeDecl(ctxt);
3280: }
1.59 veillard 3281:
1.77 veillard 3282: /*
3283: * First case : a comment
3284: */
3285: if ((CUR == '<') && (NXT(1) == '!') &&
3286: (NXT(2) == '-') && (NXT(3) == '-')) {
3287: htmlParseComment(ctxt);
3288: }
1.1 daniel 3289:
1.77 veillard 3290: /*
3291: * Second case : a sub-element.
3292: */
3293: else if (CUR == '<') {
3294: htmlParseElement(ctxt);
3295: }
1.1 daniel 3296:
1.77 veillard 3297: /*
3298: * Third case : a reference. If if has not been resolved,
3299: * parsing returns it's Name, create the node
3300: */
3301: else if (CUR == '&') {
3302: htmlParseReference(ctxt);
3303: }
1.1 daniel 3304:
1.77 veillard 3305: /*
3306: * Fourth : end of the resource
3307: */
3308: else if (CUR == 0) {
3309: htmlAutoClose(ctxt, NULL);
3310: }
1.47 daniel 3311:
1.77 veillard 3312: /*
3313: * Last case, text. Note that References are handled directly.
3314: */
3315: else {
3316: htmlParseCharData(ctxt, 0);
3317: }
1.1 daniel 3318:
1.77 veillard 3319: if (cons == ctxt->nbChars) {
3320: if (ctxt->node != NULL) {
3321: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3322: ctxt->sax->error(ctxt->userData,
3323: "detected an error in element content\n");
3324: ctxt->wellFormed = 0;
3325: }
3326: break;
1.22 daniel 3327: }
1.1 daniel 3328: }
1.5 daniel 3329: GROW;
1.1 daniel 3330: }
1.26 daniel 3331: if (currentNode != NULL) xmlFree(currentNode);
1.1 daniel 3332: }
3333:
3334: /**
3335: * htmlParseElement:
3336: * @ctxt: an HTML parser context
3337: *
3338: * parse an HTML element, this is highly recursive
3339: *
3340: * [39] element ::= EmptyElemTag | STag content ETag
3341: *
3342: * [41] Attribute ::= Name Eq AttValue
3343: */
3344:
3345: void
3346: htmlParseElement(htmlParserCtxtPtr ctxt) {
1.14 daniel 3347: xmlChar *name;
1.16 daniel 3348: xmlChar *currentNode = NULL;
1.1 daniel 3349: htmlElemDescPtr info;
1.10 daniel 3350: htmlParserNodeInfo node_info;
1.31 daniel 3351: xmlChar *oldname;
1.18 daniel 3352: int depth = ctxt->nameNr;
1.1 daniel 3353:
3354: /* Capture start position */
1.10 daniel 3355: if (ctxt->record_info) {
3356: node_info.begin_pos = ctxt->input->consumed +
3357: (CUR_PTR - ctxt->input->base);
3358: node_info.begin_line = ctxt->input->line;
3359: }
1.1 daniel 3360:
1.26 daniel 3361: oldname = xmlStrdup(ctxt->name);
1.18 daniel 3362: htmlParseStartTag(ctxt);
3363: name = ctxt->name;
1.19 daniel 3364: #ifdef DEBUG
3365: if (oldname == NULL)
1.81 veillard 3366: xmlGenericError(xmlGenericErrorContext,
3367: "Start of element %s\n", name);
1.19 daniel 3368: else if (name == NULL)
1.81 veillard 3369: xmlGenericError(xmlGenericErrorContext,
3370: "Start of element failed, was %s\n", oldname);
1.19 daniel 3371: else
1.81 veillard 3372: xmlGenericError(xmlGenericErrorContext,
3373: "Start of element %s, was %s\n", name, oldname);
1.19 daniel 3374: #endif
1.73 veillard 3375: if (((depth == ctxt->nameNr) && (xmlStrEqual(oldname, ctxt->name))) ||
1.18 daniel 3376: (name == NULL)) {
1.19 daniel 3377: if (CUR == '>')
3378: NEXT;
1.26 daniel 3379: if (oldname != NULL)
3380: xmlFree(oldname);
1.1 daniel 3381: return;
3382: }
1.26 daniel 3383: if (oldname != NULL)
3384: xmlFree(oldname);
1.1 daniel 3385:
3386: /*
3387: * Lookup the info for that element.
3388: */
3389: info = htmlTagLookup(name);
3390: if (info == NULL) {
3391: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3392: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
3393: name);
3394: ctxt->wellFormed = 0;
3395: } else if (info->depr) {
3396: /***************************
3397: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
3398: ctxt->sax->warning(ctxt->userData, "Tag %s is deprecated\n",
3399: name);
3400: ***************************/
3401: }
3402:
3403: /*
3404: * Check for an Empty Element labelled the XML/SGML way
3405: */
3406: if ((CUR == '/') && (NXT(1) == '>')) {
3407: SKIP(2);
3408: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3409: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 3410: oldname = htmlnamePop(ctxt);
1.18 daniel 3411: #ifdef DEBUG
1.81 veillard 3412: xmlGenericError(xmlGenericErrorContext,"End of tag the XML way: popping out %s\n", oldname);
1.18 daniel 3413: #endif
1.17 daniel 3414: if (oldname != NULL)
3415: xmlFree(oldname);
1.1 daniel 3416: return;
3417: }
3418:
1.5 daniel 3419: if (CUR == '>') {
3420: NEXT;
3421: } else {
1.1 daniel 3422: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.56 veillard 3423: ctxt->sax->error(ctxt->userData,
3424: "Couldn't find end of Start Tag %s\n",
3425: name);
1.1 daniel 3426: ctxt->wellFormed = 0;
3427:
3428: /*
3429: * end of parsing of this node.
3430: */
1.73 veillard 3431: if (xmlStrEqual(name, ctxt->name)) {
1.18 daniel 3432: nodePop(ctxt);
1.24 daniel 3433: oldname = htmlnamePop(ctxt);
1.18 daniel 3434: #ifdef DEBUG
1.81 veillard 3435: xmlGenericError(xmlGenericErrorContext,"End of start tag problem: popping out %s\n", oldname);
1.18 daniel 3436: #endif
3437: if (oldname != NULL)
3438: xmlFree(oldname);
3439: }
1.10 daniel 3440:
3441: /*
3442: * Capture end position and add node
3443: */
3444: if ( currentNode != NULL && ctxt->record_info ) {
3445: node_info.end_pos = ctxt->input->consumed +
3446: (CUR_PTR - ctxt->input->base);
3447: node_info.end_line = ctxt->input->line;
1.15 daniel 3448: node_info.node = ctxt->node;
1.10 daniel 3449: xmlParserAddNodeInfo(ctxt, &node_info);
3450: }
1.1 daniel 3451: return;
3452: }
3453:
3454: /*
3455: * Check for an Empty Element from DTD definition
3456: */
3457: if ((info != NULL) && (info->empty)) {
3458: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3459: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 3460: oldname = htmlnamePop(ctxt);
1.18 daniel 3461: #ifdef DEBUG
1.81 veillard 3462: xmlGenericError(xmlGenericErrorContext,"End of empty tag %s : popping out %s\n", name, oldname);
1.18 daniel 3463: #endif
1.17 daniel 3464: if (oldname != NULL)
3465: xmlFree(oldname);
1.1 daniel 3466: return;
3467: }
3468:
3469: /*
3470: * Parse the content of the element:
3471: */
1.26 daniel 3472: currentNode = xmlStrdup(ctxt->name);
1.18 daniel 3473: depth = ctxt->nameNr;
3474: while (IS_CHAR(CUR)) {
3475: htmlParseContent(ctxt);
3476: if (ctxt->nameNr < depth) break;
3477: }
1.1 daniel 3478:
3479: if (!IS_CHAR(CUR)) {
1.49 daniel 3480: /************
1.1 daniel 3481: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3482: ctxt->sax->error(ctxt->userData,
1.18 daniel 3483: "Premature end of data in tag %s\n", currentNode);
1.1 daniel 3484: ctxt->wellFormed = 0;
1.49 daniel 3485: *************/
1.1 daniel 3486:
3487: /*
3488: * end of parsing of this node.
3489: */
3490: nodePop(ctxt);
1.24 daniel 3491: oldname = htmlnamePop(ctxt);
1.18 daniel 3492: #ifdef DEBUG
1.81 veillard 3493: xmlGenericError(xmlGenericErrorContext,"Premature end of tag %s : popping out %s\n", name, oldname);
1.18 daniel 3494: #endif
1.17 daniel 3495: if (oldname != NULL)
3496: xmlFree(oldname);
1.26 daniel 3497: if (currentNode != NULL)
3498: xmlFree(currentNode);
1.1 daniel 3499: return;
3500: }
1.10 daniel 3501:
3502: /*
3503: * Capture end position and add node
3504: */
3505: if ( currentNode != NULL && ctxt->record_info ) {
3506: node_info.end_pos = ctxt->input->consumed +
3507: (CUR_PTR - ctxt->input->base);
3508: node_info.end_line = ctxt->input->line;
1.15 daniel 3509: node_info.node = ctxt->node;
1.10 daniel 3510: xmlParserAddNodeInfo(ctxt, &node_info);
3511: }
1.26 daniel 3512: if (currentNode != NULL)
3513: xmlFree(currentNode);
1.1 daniel 3514: }
3515:
3516: /**
3517: * htmlParseDocument :
3518: * @ctxt: an HTML parser context
3519: *
3520: * parse an HTML document (and build a tree if using the standard SAX
3521: * interface).
3522: *
3523: * Returns 0, -1 in case of error. the parser context is augmented
3524: * as a result of the parsing.
3525: */
3526:
3527: int
3528: htmlParseDocument(htmlParserCtxtPtr ctxt) {
1.59 veillard 3529: xmlDtdPtr dtd;
3530:
1.1 daniel 3531: htmlDefaultSAXHandlerInit();
3532: ctxt->html = 1;
3533:
1.5 daniel 3534: GROW;
1.1 daniel 3535: /*
1.9 daniel 3536: * SAX: beginning of the document processing.
1.1 daniel 3537: */
3538: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
3539: ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
3540:
3541: /*
3542: * Wipe out everything which is before the first '<'
3543: */
1.22 daniel 3544: SKIP_BLANKS;
1.1 daniel 3545: if (CUR == 0) {
3546: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3547: ctxt->sax->error(ctxt->userData, "Document is empty\n");
3548: ctxt->wellFormed = 0;
3549: }
3550:
1.40 daniel 3551: if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX))
3552: ctxt->sax->startDocument(ctxt->userData);
3553:
3554:
1.22 daniel 3555: /*
3556: * Parse possible comments before any content
3557: */
3558: while ((CUR == '<') && (NXT(1) == '!') &&
3559: (NXT(2) == '-') && (NXT(3) == '-')) {
1.31 daniel 3560: htmlParseComment(ctxt);
1.22 daniel 3561: SKIP_BLANKS;
3562: }
3563:
1.1 daniel 3564:
3565: /*
3566: * Then possibly doc type declaration(s) and more Misc
3567: * (doctypedecl Misc*)?
3568: */
3569: if ((CUR == '<') && (NXT(1) == '!') &&
3570: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3571: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3572: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3573: (UPP(8) == 'E')) {
3574: htmlParseDocTypeDecl(ctxt);
3575: }
3576: SKIP_BLANKS;
3577:
3578: /*
1.55 veillard 3579: * Parse possible comments before any content
3580: */
3581: while ((CUR == '<') && (NXT(1) == '!') &&
3582: (NXT(2) == '-') && (NXT(3) == '-')) {
3583: htmlParseComment(ctxt);
3584: SKIP_BLANKS;
3585: }
3586:
3587: /*
1.1 daniel 3588: * Time to start parsing the tree itself
3589: */
1.22 daniel 3590: htmlParseContent(ctxt);
1.1 daniel 3591:
3592: /*
1.47 daniel 3593: * autoclose
3594: */
3595: if (CUR == 0)
3596: htmlAutoClose(ctxt, NULL);
3597:
3598:
3599: /*
1.1 daniel 3600: * SAX: end of the document processing.
3601: */
3602: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3603: ctxt->sax->endDocument(ctxt->userData);
1.59 veillard 3604:
3605: if (ctxt->myDoc != NULL) {
3606: dtd = xmlGetIntSubset(ctxt->myDoc);
3607: if (dtd == NULL)
3608: ctxt->myDoc->intSubset =
3609: xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML",
3610: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
3611: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
3612: }
1.1 daniel 3613: if (! ctxt->wellFormed) return(-1);
3614: return(0);
3615: }
3616:
3617:
1.30 daniel 3618: /************************************************************************
3619: * *
3620: * Parser contexts handling *
3621: * *
3622: ************************************************************************/
1.1 daniel 3623:
3624: /**
3625: * xmlInitParserCtxt:
3626: * @ctxt: an HTML parser context
3627: *
3628: * Initialize a parser context
3629: */
3630:
3631: void
3632: htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
3633: {
3634: htmlSAXHandler *sax;
3635:
1.21 daniel 3636: if (ctxt == NULL) return;
3637: memset(ctxt, 0, sizeof(htmlParserCtxt));
3638:
1.11 daniel 3639: sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
1.1 daniel 3640: if (sax == NULL) {
1.81 veillard 3641: xmlGenericError(xmlGenericErrorContext,
3642: "htmlInitParserCtxt: out of memory\n");
1.1 daniel 3643: }
1.68 veillard 3644: else
3645: memset(sax, 0, sizeof(htmlSAXHandler));
1.1 daniel 3646:
3647: /* Allocate the Input stack */
1.19 daniel 3648: ctxt->inputTab = (htmlParserInputPtr *)
3649: xmlMalloc(5 * sizeof(htmlParserInputPtr));
3650: if (ctxt->inputTab == NULL) {
1.81 veillard 3651: xmlGenericError(xmlGenericErrorContext,
3652: "htmlInitParserCtxt: out of memory\n");
1.65 veillard 3653: ctxt->inputNr = 0;
3654: ctxt->inputMax = 0;
3655: ctxt->input = NULL;
3656: return;
1.19 daniel 3657: }
1.1 daniel 3658: ctxt->inputNr = 0;
3659: ctxt->inputMax = 5;
3660: ctxt->input = NULL;
3661: ctxt->version = NULL;
3662: ctxt->encoding = NULL;
3663: ctxt->standalone = -1;
1.30 daniel 3664: ctxt->instate = XML_PARSER_START;
1.1 daniel 3665:
3666: /* Allocate the Node stack */
1.11 daniel 3667: ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
1.65 veillard 3668: if (ctxt->nodeTab == NULL) {
1.81 veillard 3669: xmlGenericError(xmlGenericErrorContext,
3670: "htmlInitParserCtxt: out of memory\n");
1.65 veillard 3671: ctxt->nodeNr = 0;
3672: ctxt->nodeMax = 0;
3673: ctxt->node = NULL;
3674: ctxt->inputNr = 0;
3675: ctxt->inputMax = 0;
3676: ctxt->input = NULL;
3677: return;
3678: }
1.1 daniel 3679: ctxt->nodeNr = 0;
3680: ctxt->nodeMax = 10;
3681: ctxt->node = NULL;
3682:
1.15 daniel 3683: /* Allocate the Name stack */
3684: ctxt->nameTab = (xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
1.65 veillard 3685: if (ctxt->nameTab == NULL) {
1.81 veillard 3686: xmlGenericError(xmlGenericErrorContext,
3687: "htmlInitParserCtxt: out of memory\n");
1.65 veillard 3688: ctxt->nameNr = 0;
3689: ctxt->nameMax = 10;
3690: ctxt->name = NULL;
3691: ctxt->nodeNr = 0;
3692: ctxt->nodeMax = 0;
3693: ctxt->node = NULL;
3694: ctxt->inputNr = 0;
3695: ctxt->inputMax = 0;
3696: ctxt->input = NULL;
3697: return;
3698: }
1.15 daniel 3699: ctxt->nameNr = 0;
3700: ctxt->nameMax = 10;
3701: ctxt->name = NULL;
3702:
1.1 daniel 3703: if (sax == NULL) ctxt->sax = &htmlDefaultSAXHandler;
3704: else {
3705: ctxt->sax = sax;
3706: memcpy(sax, &htmlDefaultSAXHandler, sizeof(htmlSAXHandler));
3707: }
3708: ctxt->userData = ctxt;
3709: ctxt->myDoc = NULL;
3710: ctxt->wellFormed = 1;
3711: ctxt->replaceEntities = 0;
3712: ctxt->html = 1;
3713: ctxt->record_info = 0;
1.21 daniel 3714: ctxt->validate = 0;
1.26 daniel 3715: ctxt->nbChars = 0;
1.30 daniel 3716: ctxt->checkIndex = 0;
1.1 daniel 3717: xmlInitNodeInfoSeq(&ctxt->node_seq);
3718: }
3719:
3720: /**
3721: * htmlFreeParserCtxt:
3722: * @ctxt: an HTML parser context
3723: *
3724: * Free all the memory used by a parser context. However the parsed
3725: * document in ctxt->myDoc is not freed.
3726: */
3727:
3728: void
3729: htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
3730: {
1.47 daniel 3731: xmlFreeParserCtxt(ctxt);
1.1 daniel 3732: }
3733:
3734: /**
3735: * htmlCreateDocParserCtxt :
1.14 daniel 3736: * @cur: a pointer to an array of xmlChar
1.1 daniel 3737: * @encoding: a free form C string describing the HTML document encoding, or NULL
3738: *
3739: * Create a parser context for an HTML document.
3740: *
3741: * Returns the new parser context or NULL
3742: */
3743: htmlParserCtxtPtr
1.14 daniel 3744: htmlCreateDocParserCtxt(xmlChar *cur, const char *encoding) {
1.1 daniel 3745: htmlParserCtxtPtr ctxt;
3746: htmlParserInputPtr input;
3747: /* htmlCharEncoding enc; */
3748:
1.11 daniel 3749: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 3750: if (ctxt == NULL) {
3751: perror("malloc");
3752: return(NULL);
3753: }
3754: htmlInitParserCtxt(ctxt);
1.11 daniel 3755: input = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 3756: if (input == NULL) {
3757: perror("malloc");
1.11 daniel 3758: xmlFree(ctxt);
1.1 daniel 3759: return(NULL);
3760: }
1.19 daniel 3761: memset(input, 0, sizeof(htmlParserInput));
1.1 daniel 3762:
3763: input->line = 1;
3764: input->col = 1;
3765: input->base = cur;
3766: input->cur = cur;
3767:
3768: inputPush(ctxt, input);
3769: return(ctxt);
3770: }
3771:
1.31 daniel 3772: /************************************************************************
3773: * *
3774: * Progressive parsing interfaces *
3775: * *
3776: ************************************************************************/
3777:
3778: /**
3779: * htmlParseLookupSequence:
3780: * @ctxt: an HTML parser context
3781: * @first: the first char to lookup
3782: * @next: the next char to lookup or zero
3783: * @third: the next char to lookup or zero
3784: *
3785: * Try to find if a sequence (first, next, third) or just (first next) or
3786: * (first) is available in the input stream.
3787: * This function has a side effect of (possibly) incrementing ctxt->checkIndex
3788: * to avoid rescanning sequences of bytes, it DOES change the state of the
3789: * parser, do not use liberally.
3790: * This is basically similar to xmlParseLookupSequence()
3791: *
3792: * Returns the index to the current parsing point if the full sequence
3793: * is available, -1 otherwise.
3794: */
3795: int
3796: htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first,
3797: xmlChar next, xmlChar third) {
3798: int base, len;
3799: htmlParserInputPtr in;
3800: const xmlChar *buf;
3801:
3802: in = ctxt->input;
3803: if (in == NULL) return(-1);
3804: base = in->cur - in->base;
3805: if (base < 0) return(-1);
3806: if (ctxt->checkIndex > base)
3807: base = ctxt->checkIndex;
3808: if (in->buf == NULL) {
3809: buf = in->base;
3810: len = in->length;
3811: } else {
3812: buf = in->buf->buffer->content;
3813: len = in->buf->buffer->use;
3814: }
3815: /* take into account the sequence length */
3816: if (third) len -= 2;
3817: else if (next) len --;
3818: for (;base < len;base++) {
3819: if (buf[base] == first) {
3820: if (third != 0) {
3821: if ((buf[base + 1] != next) ||
3822: (buf[base + 2] != third)) continue;
3823: } else if (next != 0) {
3824: if (buf[base + 1] != next) continue;
3825: }
3826: ctxt->checkIndex = 0;
3827: #ifdef DEBUG_PUSH
3828: if (next == 0)
1.81 veillard 3829: xmlGenericError(xmlGenericErrorContext,
3830: "HPP: lookup '%c' found at %d\n",
1.31 daniel 3831: first, base);
3832: else if (third == 0)
1.81 veillard 3833: xmlGenericError(xmlGenericErrorContext,
3834: "HPP: lookup '%c%c' found at %d\n",
1.31 daniel 3835: first, next, base);
3836: else
1.81 veillard 3837: xmlGenericError(xmlGenericErrorContext,
3838: "HPP: lookup '%c%c%c' found at %d\n",
1.31 daniel 3839: first, next, third, base);
3840: #endif
3841: return(base - (in->cur - in->base));
3842: }
3843: }
3844: ctxt->checkIndex = base;
3845: #ifdef DEBUG_PUSH
3846: if (next == 0)
1.81 veillard 3847: xmlGenericError(xmlGenericErrorContext,
3848: "HPP: lookup '%c' failed\n", first);
1.31 daniel 3849: else if (third == 0)
1.81 veillard 3850: xmlGenericError(xmlGenericErrorContext,
3851: "HPP: lookup '%c%c' failed\n", first, next);
1.31 daniel 3852: else
1.81 veillard 3853: xmlGenericError(xmlGenericErrorContext,
3854: "HPP: lookup '%c%c%c' failed\n", first, next, third);
1.31 daniel 3855: #endif
3856: return(-1);
3857: }
3858:
3859: /**
1.32 daniel 3860: * htmlParseTryOrFinish:
1.31 daniel 3861: * @ctxt: an HTML parser context
1.32 daniel 3862: * @terminate: last chunk indicator
1.31 daniel 3863: *
3864: * Try to progress on parsing
3865: *
3866: * Returns zero if no parsing was possible
3867: */
3868: int
1.32 daniel 3869: htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
1.31 daniel 3870: int ret = 0;
3871: htmlParserInputPtr in;
1.47 daniel 3872: int avail = 0;
1.31 daniel 3873: xmlChar cur, next;
3874:
3875: #ifdef DEBUG_PUSH
3876: switch (ctxt->instate) {
3877: case XML_PARSER_EOF:
1.81 veillard 3878: xmlGenericError(xmlGenericErrorContext,
3879: "HPP: try EOF\n"); break;
1.31 daniel 3880: case XML_PARSER_START:
1.81 veillard 3881: xmlGenericError(xmlGenericErrorContext,
3882: "HPP: try START\n"); break;
1.31 daniel 3883: case XML_PARSER_MISC:
1.81 veillard 3884: xmlGenericError(xmlGenericErrorContext,
3885: "HPP: try MISC\n");break;
1.31 daniel 3886: case XML_PARSER_COMMENT:
1.81 veillard 3887: xmlGenericError(xmlGenericErrorContext,
3888: "HPP: try COMMENT\n");break;
1.31 daniel 3889: case XML_PARSER_PROLOG:
1.81 veillard 3890: xmlGenericError(xmlGenericErrorContext,
3891: "HPP: try PROLOG\n");break;
1.31 daniel 3892: case XML_PARSER_START_TAG:
1.81 veillard 3893: xmlGenericError(xmlGenericErrorContext,
3894: "HPP: try START_TAG\n");break;
1.31 daniel 3895: case XML_PARSER_CONTENT:
1.81 veillard 3896: xmlGenericError(xmlGenericErrorContext,
3897: "HPP: try CONTENT\n");break;
1.31 daniel 3898: case XML_PARSER_CDATA_SECTION:
1.81 veillard 3899: xmlGenericError(xmlGenericErrorContext,
3900: "HPP: try CDATA_SECTION\n");break;
1.31 daniel 3901: case XML_PARSER_END_TAG:
1.81 veillard 3902: xmlGenericError(xmlGenericErrorContext,
3903: "HPP: try END_TAG\n");break;
1.31 daniel 3904: case XML_PARSER_ENTITY_DECL:
1.81 veillard 3905: xmlGenericError(xmlGenericErrorContext,
3906: "HPP: try ENTITY_DECL\n");break;
1.31 daniel 3907: case XML_PARSER_ENTITY_VALUE:
1.81 veillard 3908: xmlGenericError(xmlGenericErrorContext,
3909: "HPP: try ENTITY_VALUE\n");break;
1.31 daniel 3910: case XML_PARSER_ATTRIBUTE_VALUE:
1.81 veillard 3911: xmlGenericError(xmlGenericErrorContext,
3912: "HPP: try ATTRIBUTE_VALUE\n");break;
1.31 daniel 3913: case XML_PARSER_DTD:
1.81 veillard 3914: xmlGenericError(xmlGenericErrorContext,
3915: "HPP: try DTD\n");break;
1.31 daniel 3916: case XML_PARSER_EPILOG:
1.81 veillard 3917: xmlGenericError(xmlGenericErrorContext,
3918: "HPP: try EPILOG\n");break;
1.31 daniel 3919: case XML_PARSER_PI:
1.81 veillard 3920: xmlGenericError(xmlGenericErrorContext,
3921: "HPP: try PI\n");break;
1.77 veillard 3922: case XML_PARSER_SYSTEM_LITERAL:
1.81 veillard 3923: xmlGenericError(xmlGenericErrorContext,
3924: "HPP: try SYSTEM_LITERAL\n");break;
1.31 daniel 3925: }
3926: #endif
3927:
3928: while (1) {
3929:
3930: in = ctxt->input;
3931: if (in == NULL) break;
3932: if (in->buf == NULL)
3933: avail = in->length - (in->cur - in->base);
3934: else
3935: avail = in->buf->buffer->use - (in->cur - in->base);
1.47 daniel 3936: if ((avail == 0) && (terminate)) {
3937: htmlAutoClose(ctxt, NULL);
1.54 veillard 3938: if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
3939: /*
3940: * SAX: end of the document processing.
3941: */
1.47 daniel 3942: ctxt->instate = XML_PARSER_EOF;
1.54 veillard 3943: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3944: ctxt->sax->endDocument(ctxt->userData);
3945: }
1.47 daniel 3946: }
1.31 daniel 3947: if (avail < 1)
3948: goto done;
3949: switch (ctxt->instate) {
3950: case XML_PARSER_EOF:
3951: /*
3952: * Document parsing is done !
3953: */
3954: goto done;
3955: case XML_PARSER_START:
3956: /*
3957: * Very first chars read from the document flow.
3958: */
3959: cur = in->cur[0];
3960: if (IS_BLANK(cur)) {
3961: SKIP_BLANKS;
3962: if (in->buf == NULL)
3963: avail = in->length - (in->cur - in->base);
3964: else
3965: avail = in->buf->buffer->use - (in->cur - in->base);
3966: }
3967: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
3968: ctxt->sax->setDocumentLocator(ctxt->userData,
3969: &xmlDefaultSAXLocator);
1.46 daniel 3970: if ((ctxt->sax) && (ctxt->sax->startDocument) &&
3971: (!ctxt->disableSAX))
3972: ctxt->sax->startDocument(ctxt->userData);
3973:
1.31 daniel 3974: cur = in->cur[0];
3975: next = in->cur[1];
3976: if ((cur == '<') && (next == '!') &&
3977: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3978: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3979: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3980: (UPP(8) == 'E')) {
1.32 daniel 3981: if ((!terminate) &&
3982: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3983: goto done;
3984: #ifdef DEBUG_PUSH
1.81 veillard 3985: xmlGenericError(xmlGenericErrorContext,
3986: "HPP: Parsing internal subset\n");
1.31 daniel 3987: #endif
3988: htmlParseDocTypeDecl(ctxt);
3989: ctxt->instate = XML_PARSER_PROLOG;
3990: #ifdef DEBUG_PUSH
1.81 veillard 3991: xmlGenericError(xmlGenericErrorContext,
3992: "HPP: entering PROLOG\n");
1.31 daniel 3993: #endif
3994: } else {
3995: ctxt->instate = XML_PARSER_MISC;
3996: }
3997: #ifdef DEBUG_PUSH
1.81 veillard 3998: xmlGenericError(xmlGenericErrorContext,
3999: "HPP: entering MISC\n");
1.31 daniel 4000: #endif
4001: break;
4002: case XML_PARSER_MISC:
4003: SKIP_BLANKS;
4004: if (in->buf == NULL)
4005: avail = in->length - (in->cur - in->base);
4006: else
4007: avail = in->buf->buffer->use - (in->cur - in->base);
4008: if (avail < 2)
4009: goto done;
4010: cur = in->cur[0];
4011: next = in->cur[1];
4012: if ((cur == '<') && (next == '!') &&
4013: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 4014: if ((!terminate) &&
4015: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 4016: goto done;
4017: #ifdef DEBUG_PUSH
1.81 veillard 4018: xmlGenericError(xmlGenericErrorContext,
4019: "HPP: Parsing Comment\n");
1.31 daniel 4020: #endif
4021: htmlParseComment(ctxt);
4022: ctxt->instate = XML_PARSER_MISC;
4023: } else if ((cur == '<') && (next == '!') &&
4024: (UPP(2) == 'D') && (UPP(3) == 'O') &&
4025: (UPP(4) == 'C') && (UPP(5) == 'T') &&
4026: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4027: (UPP(8) == 'E')) {
1.32 daniel 4028: if ((!terminate) &&
4029: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 4030: goto done;
4031: #ifdef DEBUG_PUSH
1.81 veillard 4032: xmlGenericError(xmlGenericErrorContext,
4033: "HPP: Parsing internal subset\n");
1.31 daniel 4034: #endif
4035: htmlParseDocTypeDecl(ctxt);
4036: ctxt->instate = XML_PARSER_PROLOG;
4037: #ifdef DEBUG_PUSH
1.81 veillard 4038: xmlGenericError(xmlGenericErrorContext,
4039: "HPP: entering PROLOG\n");
1.31 daniel 4040: #endif
4041: } else if ((cur == '<') && (next == '!') &&
4042: (avail < 9)) {
4043: goto done;
4044: } else {
4045: ctxt->instate = XML_PARSER_START_TAG;
4046: #ifdef DEBUG_PUSH
1.81 veillard 4047: xmlGenericError(xmlGenericErrorContext,
4048: "HPP: entering START_TAG\n");
1.31 daniel 4049: #endif
4050: }
4051: break;
4052: case XML_PARSER_PROLOG:
4053: SKIP_BLANKS;
4054: if (in->buf == NULL)
4055: avail = in->length - (in->cur - in->base);
4056: else
4057: avail = in->buf->buffer->use - (in->cur - in->base);
4058: if (avail < 2)
4059: goto done;
4060: cur = in->cur[0];
4061: next = in->cur[1];
4062: if ((cur == '<') && (next == '!') &&
4063: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 4064: if ((!terminate) &&
4065: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 4066: goto done;
4067: #ifdef DEBUG_PUSH
1.81 veillard 4068: xmlGenericError(xmlGenericErrorContext,
4069: "HPP: Parsing Comment\n");
1.31 daniel 4070: #endif
4071: htmlParseComment(ctxt);
4072: ctxt->instate = XML_PARSER_PROLOG;
4073: } else if ((cur == '<') && (next == '!') &&
4074: (avail < 4)) {
4075: goto done;
4076: } else {
4077: ctxt->instate = XML_PARSER_START_TAG;
4078: #ifdef DEBUG_PUSH
1.81 veillard 4079: xmlGenericError(xmlGenericErrorContext,
4080: "HPP: entering START_TAG\n");
1.31 daniel 4081: #endif
4082: }
4083: break;
4084: case XML_PARSER_EPILOG:
4085: if (in->buf == NULL)
4086: avail = in->length - (in->cur - in->base);
4087: else
4088: avail = in->buf->buffer->use - (in->cur - in->base);
1.55 veillard 4089: if (avail < 1)
4090: goto done;
4091: cur = in->cur[0];
4092: if (IS_BLANK(cur)) {
4093: htmlParseCharData(ctxt, 0);
4094: goto done;
4095: }
1.31 daniel 4096: if (avail < 2)
4097: goto done;
4098: next = in->cur[1];
4099: if ((cur == '<') && (next == '!') &&
4100: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 4101: if ((!terminate) &&
4102: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 4103: goto done;
4104: #ifdef DEBUG_PUSH
1.81 veillard 4105: xmlGenericError(xmlGenericErrorContext,
4106: "HPP: Parsing Comment\n");
1.31 daniel 4107: #endif
4108: htmlParseComment(ctxt);
4109: ctxt->instate = XML_PARSER_EPILOG;
4110: } else if ((cur == '<') && (next == '!') &&
4111: (avail < 4)) {
4112: goto done;
4113: } else {
1.67 veillard 4114: ctxt->errNo = XML_ERR_DOCUMENT_END;
1.31 daniel 4115: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4116: ctxt->sax->error(ctxt->userData,
4117: "Extra content at the end of the document\n");
4118: ctxt->wellFormed = 0;
4119: ctxt->instate = XML_PARSER_EOF;
4120: #ifdef DEBUG_PUSH
1.81 veillard 4121: xmlGenericError(xmlGenericErrorContext,
4122: "HPP: entering EOF\n");
1.31 daniel 4123: #endif
4124: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4125: ctxt->sax->endDocument(ctxt->userData);
4126: goto done;
4127: }
4128: break;
4129: case XML_PARSER_START_TAG: {
4130: xmlChar *name, *oldname;
4131: int depth = ctxt->nameNr;
4132: htmlElemDescPtr info;
4133:
4134: if (avail < 2)
4135: goto done;
4136: cur = in->cur[0];
4137: if (cur != '<') {
4138: ctxt->instate = XML_PARSER_CONTENT;
4139: #ifdef DEBUG_PUSH
1.81 veillard 4140: xmlGenericError(xmlGenericErrorContext,
4141: "HPP: entering CONTENT\n");
1.31 daniel 4142: #endif
4143: break;
4144: }
1.32 daniel 4145: if ((!terminate) &&
4146: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 4147: goto done;
4148:
4149: oldname = xmlStrdup(ctxt->name);
4150: htmlParseStartTag(ctxt);
4151: name = ctxt->name;
4152: #ifdef DEBUG
4153: if (oldname == NULL)
1.81 veillard 4154: xmlGenericError(xmlGenericErrorContext,
4155: "Start of element %s\n", name);
1.31 daniel 4156: else if (name == NULL)
1.81 veillard 4157: xmlGenericError(xmlGenericErrorContext,
4158: "Start of element failed, was %s\n",
1.31 daniel 4159: oldname);
4160: else
1.81 veillard 4161: xmlGenericError(xmlGenericErrorContext,
4162: "Start of element %s, was %s\n",
1.31 daniel 4163: name, oldname);
4164: #endif
4165: if (((depth == ctxt->nameNr) &&
1.73 veillard 4166: (xmlStrEqual(oldname, ctxt->name))) ||
1.31 daniel 4167: (name == NULL)) {
4168: if (CUR == '>')
4169: NEXT;
4170: if (oldname != NULL)
4171: xmlFree(oldname);
4172: break;
4173: }
4174: if (oldname != NULL)
4175: xmlFree(oldname);
4176:
4177: /*
4178: * Lookup the info for that element.
4179: */
4180: info = htmlTagLookup(name);
4181: if (info == NULL) {
4182: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4183: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
4184: name);
4185: ctxt->wellFormed = 0;
4186: } else if (info->depr) {
4187: /***************************
4188: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
4189: ctxt->sax->warning(ctxt->userData,
4190: "Tag %s is deprecated\n",
4191: name);
4192: ***************************/
4193: }
4194:
4195: /*
4196: * Check for an Empty Element labelled the XML/SGML way
4197: */
4198: if ((CUR == '/') && (NXT(1) == '>')) {
4199: SKIP(2);
4200: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4201: ctxt->sax->endElement(ctxt->userData, name);
4202: oldname = htmlnamePop(ctxt);
4203: #ifdef DEBUG
1.81 veillard 4204: xmlGenericError(xmlGenericErrorContext,"End of tag the XML way: popping out %s\n",
1.31 daniel 4205: oldname);
4206: #endif
4207: if (oldname != NULL)
4208: xmlFree(oldname);
4209: ctxt->instate = XML_PARSER_CONTENT;
4210: #ifdef DEBUG_PUSH
1.81 veillard 4211: xmlGenericError(xmlGenericErrorContext,
4212: "HPP: entering CONTENT\n");
1.31 daniel 4213: #endif
4214: break;
4215: }
4216:
4217: if (CUR == '>') {
4218: NEXT;
4219: } else {
4220: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4221: ctxt->sax->error(ctxt->userData,
4222: "Couldn't find end of Start Tag %s\n",
4223: name);
4224: ctxt->wellFormed = 0;
4225:
4226: /*
4227: * end of parsing of this node.
4228: */
1.73 veillard 4229: if (xmlStrEqual(name, ctxt->name)) {
1.31 daniel 4230: nodePop(ctxt);
4231: oldname = htmlnamePop(ctxt);
4232: #ifdef DEBUG
1.81 veillard 4233: xmlGenericError(xmlGenericErrorContext,
1.31 daniel 4234: "End of start tag problem: popping out %s\n", oldname);
4235: #endif
4236: if (oldname != NULL)
4237: xmlFree(oldname);
4238: }
4239:
4240: ctxt->instate = XML_PARSER_CONTENT;
4241: #ifdef DEBUG_PUSH
1.81 veillard 4242: xmlGenericError(xmlGenericErrorContext,
4243: "HPP: entering CONTENT\n");
1.31 daniel 4244: #endif
4245: break;
4246: }
4247:
4248: /*
4249: * Check for an Empty Element from DTD definition
4250: */
4251: if ((info != NULL) && (info->empty)) {
4252: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4253: ctxt->sax->endElement(ctxt->userData, name);
4254: oldname = htmlnamePop(ctxt);
4255: #ifdef DEBUG
1.81 veillard 4256: xmlGenericError(xmlGenericErrorContext,"End of empty tag %s : popping out %s\n", name, oldname);
1.31 daniel 4257: #endif
4258: if (oldname != NULL)
4259: xmlFree(oldname);
4260: }
4261: ctxt->instate = XML_PARSER_CONTENT;
4262: #ifdef DEBUG_PUSH
1.81 veillard 4263: xmlGenericError(xmlGenericErrorContext,
4264: "HPP: entering CONTENT\n");
1.31 daniel 4265: #endif
4266: break;
4267: }
1.56 veillard 4268: case XML_PARSER_CONTENT: {
4269: long cons;
1.31 daniel 4270: /*
4271: * Handle preparsed entities and charRef
4272: */
4273: if (ctxt->token != 0) {
1.47 daniel 4274: xmlChar chr[2] = { 0 , 0 } ;
1.31 daniel 4275:
1.47 daniel 4276: chr[0] = (xmlChar) ctxt->token;
1.59 veillard 4277: htmlCheckParagraph(ctxt);
1.31 daniel 4278: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1.47 daniel 4279: ctxt->sax->characters(ctxt->userData, chr, 1);
1.31 daniel 4280: ctxt->token = 0;
4281: ctxt->checkIndex = 0;
4282: }
1.47 daniel 4283: if ((avail == 1) && (terminate)) {
4284: cur = in->cur[0];
4285: if ((cur != '<') && (cur != '&')) {
1.48 daniel 4286: if (ctxt->sax != NULL) {
4287: if (IS_BLANK(cur)) {
4288: if (ctxt->sax->ignorableWhitespace != NULL)
4289: ctxt->sax->ignorableWhitespace(
4290: ctxt->userData, &cur, 1);
4291: } else {
1.59 veillard 4292: htmlCheckParagraph(ctxt);
1.48 daniel 4293: if (ctxt->sax->characters != NULL)
4294: ctxt->sax->characters(
4295: ctxt->userData, &cur, 1);
4296: }
4297: }
1.47 daniel 4298: ctxt->token = 0;
4299: ctxt->checkIndex = 0;
4300: NEXT;
4301: }
4302: break;
4303: }
1.31 daniel 4304: if (avail < 2)
4305: goto done;
4306: cur = in->cur[0];
4307: next = in->cur[1];
1.56 veillard 4308: cons = ctxt->nbChars;
1.77 veillard 4309: if ((xmlStrEqual(ctxt->name, BAD_CAST"script")) ||
4310: (xmlStrEqual(ctxt->name, BAD_CAST"style"))) {
4311: /*
4312: * Handle SCRIPT/STYLE separately
4313: */
1.59 veillard 4314: if ((!terminate) &&
1.77 veillard 4315: (htmlParseLookupSequence(ctxt, '<', '/', 0) < 0))
1.31 daniel 4316: goto done;
1.77 veillard 4317: htmlParseScript(ctxt);
4318: if ((cur == '<') && (next == '/')) {
4319: ctxt->instate = XML_PARSER_END_TAG;
4320: ctxt->checkIndex = 0;
1.31 daniel 4321: #ifdef DEBUG_PUSH
1.81 veillard 4322: xmlGenericError(xmlGenericErrorContext,
4323: "HPP: entering END_TAG\n");
1.31 daniel 4324: #endif
1.77 veillard 4325: break;
4326: }
4327: } else {
4328: /*
4329: * Sometimes DOCTYPE arrives in the middle of the document
4330: */
4331: if ((cur == '<') && (next == '!') &&
4332: (UPP(2) == 'D') && (UPP(3) == 'O') &&
4333: (UPP(4) == 'C') && (UPP(5) == 'T') &&
4334: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4335: (UPP(8) == 'E')) {
4336: if ((!terminate) &&
4337: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
4338: goto done;
4339: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4340: ctxt->sax->error(ctxt->userData,
4341: "Misplaced DOCTYPE declaration\n");
4342: ctxt->wellFormed = 0;
4343: htmlParseDocTypeDecl(ctxt);
4344: } else if ((cur == '<') && (next == '!') &&
4345: (in->cur[2] == '-') && (in->cur[3] == '-')) {
4346: if ((!terminate) &&
4347: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
4348: goto done;
1.31 daniel 4349: #ifdef DEBUG_PUSH
1.81 veillard 4350: xmlGenericError(xmlGenericErrorContext,
4351: "HPP: Parsing Comment\n");
1.31 daniel 4352: #endif
1.77 veillard 4353: htmlParseComment(ctxt);
4354: ctxt->instate = XML_PARSER_CONTENT;
4355: } else if ((cur == '<') && (next == '!') && (avail < 4)) {
4356: goto done;
4357: } else if ((cur == '<') && (next == '/')) {
4358: ctxt->instate = XML_PARSER_END_TAG;
4359: ctxt->checkIndex = 0;
1.31 daniel 4360: #ifdef DEBUG_PUSH
1.81 veillard 4361: xmlGenericError(xmlGenericErrorContext,
4362: "HPP: entering END_TAG\n");
1.31 daniel 4363: #endif
1.77 veillard 4364: break;
4365: } else if (cur == '<') {
4366: ctxt->instate = XML_PARSER_START_TAG;
4367: ctxt->checkIndex = 0;
1.31 daniel 4368: #ifdef DEBUG_PUSH
1.81 veillard 4369: xmlGenericError(xmlGenericErrorContext,
4370: "HPP: entering START_TAG\n");
1.31 daniel 4371: #endif
1.77 veillard 4372: break;
4373: } else if (cur == '&') {
1.32 daniel 4374: if ((!terminate) &&
1.77 veillard 4375: (htmlParseLookupSequence(ctxt, ';', 0, 0) < 0))
1.31 daniel 4376: goto done;
4377: #ifdef DEBUG_PUSH
1.81 veillard 4378: xmlGenericError(xmlGenericErrorContext,
4379: "HPP: Parsing Reference\n");
1.77 veillard 4380: #endif
4381: /* TODO: check generation of subtrees if noent !!! */
4382: htmlParseReference(ctxt);
4383: } else {
4384: /* TODO Avoid the extra copy, handle directly !!!!!! */
4385: /*
4386: * Goal of the following test is :
4387: * - minimize calls to the SAX 'character' callback
4388: * when they are mergeable
4389: */
4390: if ((ctxt->inputNr == 1) &&
4391: (avail < HTML_PARSER_BIG_BUFFER_SIZE)) {
4392: if ((!terminate) &&
4393: (htmlParseLookupSequence(ctxt, '<', 0, 0) < 0))
4394: goto done;
4395: }
4396: ctxt->checkIndex = 0;
4397: #ifdef DEBUG_PUSH
1.81 veillard 4398: xmlGenericError(xmlGenericErrorContext,
4399: "HPP: Parsing char data\n");
1.31 daniel 4400: #endif
1.77 veillard 4401: htmlParseCharData(ctxt, 0);
4402: }
1.31 daniel 4403: }
1.56 veillard 4404: if (cons == ctxt->nbChars) {
4405: if (ctxt->node != NULL) {
4406: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4407: ctxt->sax->error(ctxt->userData,
4408: "detected an error in element content\n");
4409: ctxt->wellFormed = 0;
4410: }
1.70 veillard 4411: NEXT;
1.56 veillard 4412: break;
4413: }
4414:
1.31 daniel 4415: break;
1.56 veillard 4416: }
1.31 daniel 4417: case XML_PARSER_END_TAG:
4418: if (avail < 2)
4419: goto done;
1.32 daniel 4420: if ((!terminate) &&
4421: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 4422: goto done;
4423: htmlParseEndTag(ctxt);
4424: if (ctxt->nameNr == 0) {
4425: ctxt->instate = XML_PARSER_EPILOG;
4426: } else {
4427: ctxt->instate = XML_PARSER_CONTENT;
4428: }
4429: ctxt->checkIndex = 0;
4430: #ifdef DEBUG_PUSH
1.81 veillard 4431: xmlGenericError(xmlGenericErrorContext,
4432: "HPP: entering CONTENT\n");
1.31 daniel 4433: #endif
4434: break;
4435: case XML_PARSER_CDATA_SECTION:
1.81 veillard 4436: xmlGenericError(xmlGenericErrorContext,
4437: "HPP: internal error, state == CDATA\n");
1.31 daniel 4438: ctxt->instate = XML_PARSER_CONTENT;
4439: ctxt->checkIndex = 0;
4440: #ifdef DEBUG_PUSH
1.81 veillard 4441: xmlGenericError(xmlGenericErrorContext,
4442: "HPP: entering CONTENT\n");
1.31 daniel 4443: #endif
4444: break;
4445: case XML_PARSER_DTD:
1.81 veillard 4446: xmlGenericError(xmlGenericErrorContext,
4447: "HPP: internal error, state == DTD\n");
1.31 daniel 4448: ctxt->instate = XML_PARSER_CONTENT;
4449: ctxt->checkIndex = 0;
4450: #ifdef DEBUG_PUSH
1.81 veillard 4451: xmlGenericError(xmlGenericErrorContext,
4452: "HPP: entering CONTENT\n");
1.31 daniel 4453: #endif
4454: break;
4455: case XML_PARSER_COMMENT:
1.81 veillard 4456: xmlGenericError(xmlGenericErrorContext,
4457: "HPP: internal error, state == COMMENT\n");
1.31 daniel 4458: ctxt->instate = XML_PARSER_CONTENT;
4459: ctxt->checkIndex = 0;
4460: #ifdef DEBUG_PUSH
1.81 veillard 4461: xmlGenericError(xmlGenericErrorContext,
4462: "HPP: entering CONTENT\n");
1.31 daniel 4463: #endif
4464: break;
4465: case XML_PARSER_PI:
1.81 veillard 4466: xmlGenericError(xmlGenericErrorContext,
4467: "HPP: internal error, state == PI\n");
1.31 daniel 4468: ctxt->instate = XML_PARSER_CONTENT;
4469: ctxt->checkIndex = 0;
4470: #ifdef DEBUG_PUSH
1.81 veillard 4471: xmlGenericError(xmlGenericErrorContext,
4472: "HPP: entering CONTENT\n");
1.31 daniel 4473: #endif
4474: break;
4475: case XML_PARSER_ENTITY_DECL:
1.81 veillard 4476: xmlGenericError(xmlGenericErrorContext,
4477: "HPP: internal error, state == ENTITY_DECL\n");
1.31 daniel 4478: ctxt->instate = XML_PARSER_CONTENT;
4479: ctxt->checkIndex = 0;
4480: #ifdef DEBUG_PUSH
1.81 veillard 4481: xmlGenericError(xmlGenericErrorContext,
4482: "HPP: entering CONTENT\n");
1.31 daniel 4483: #endif
4484: break;
4485: case XML_PARSER_ENTITY_VALUE:
1.81 veillard 4486: xmlGenericError(xmlGenericErrorContext,
4487: "HPP: internal error, state == ENTITY_VALUE\n");
1.31 daniel 4488: ctxt->instate = XML_PARSER_CONTENT;
4489: ctxt->checkIndex = 0;
4490: #ifdef DEBUG_PUSH
1.81 veillard 4491: xmlGenericError(xmlGenericErrorContext,
4492: "HPP: entering DTD\n");
1.31 daniel 4493: #endif
4494: break;
4495: case XML_PARSER_ATTRIBUTE_VALUE:
1.81 veillard 4496: xmlGenericError(xmlGenericErrorContext,
4497: "HPP: internal error, state == ATTRIBUTE_VALUE\n");
1.31 daniel 4498: ctxt->instate = XML_PARSER_START_TAG;
4499: ctxt->checkIndex = 0;
4500: #ifdef DEBUG_PUSH
1.81 veillard 4501: xmlGenericError(xmlGenericErrorContext,
4502: "HPP: entering START_TAG\n");
1.53 veillard 4503: #endif
4504: break;
4505: case XML_PARSER_SYSTEM_LITERAL:
1.81 veillard 4506: xmlGenericError(xmlGenericErrorContext,
4507: "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n");
1.53 veillard 4508: ctxt->instate = XML_PARSER_CONTENT;
4509: ctxt->checkIndex = 0;
4510: #ifdef DEBUG_PUSH
1.81 veillard 4511: xmlGenericError(xmlGenericErrorContext,
4512: "HPP: entering CONTENT\n");
1.31 daniel 4513: #endif
4514: break;
4515: }
4516: }
4517: done:
1.47 daniel 4518: if ((avail == 0) && (terminate)) {
4519: htmlAutoClose(ctxt, NULL);
1.54 veillard 4520: if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
4521: /*
4522: * SAX: end of the document processing.
4523: */
1.47 daniel 4524: ctxt->instate = XML_PARSER_EOF;
1.54 veillard 4525: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4526: ctxt->sax->endDocument(ctxt->userData);
4527: }
1.59 veillard 4528: }
4529: if ((ctxt->myDoc != NULL) &&
4530: ((terminate) || (ctxt->instate == XML_PARSER_EOF) ||
4531: (ctxt->instate == XML_PARSER_EPILOG))) {
4532: xmlDtdPtr dtd;
4533: dtd = xmlGetIntSubset(ctxt->myDoc);
4534: if (dtd == NULL)
4535: ctxt->myDoc->intSubset =
4536: xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML",
4537: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
4538: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
1.47 daniel 4539: }
1.31 daniel 4540: #ifdef DEBUG_PUSH
1.81 veillard 4541: xmlGenericError(xmlGenericErrorContext, "HPP: done %d\n", ret);
1.31 daniel 4542: #endif
4543: return(ret);
4544: }
4545:
4546: /**
1.32 daniel 4547: * htmlParseTry:
4548: * @ctxt: an HTML parser context
4549: *
4550: * Try to progress on parsing
4551: *
4552: * Returns zero if no parsing was possible
4553: */
4554: int
4555: htmlParseTry(htmlParserCtxtPtr ctxt) {
4556: return(htmlParseTryOrFinish(ctxt, 0));
4557: }
4558:
4559: /**
1.31 daniel 4560: * htmlParseChunk:
4561: * @ctxt: an XML parser context
4562: * @chunk: an char array
4563: * @size: the size in byte of the chunk
4564: * @terminate: last chunk indicator
4565: *
4566: * Parse a Chunk of memory
4567: *
4568: * Returns zero if no error, the xmlParserErrors otherwise.
4569: */
4570: int
4571: htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
4572: int terminate) {
4573: if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
4574: (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) {
4575: int base = ctxt->input->base - ctxt->input->buf->buffer->content;
4576: int cur = ctxt->input->cur - ctxt->input->base;
4577:
4578: xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
4579: ctxt->input->base = ctxt->input->buf->buffer->content + base;
4580: ctxt->input->cur = ctxt->input->base + cur;
4581: #ifdef DEBUG_PUSH
1.81 veillard 4582: xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
1.31 daniel 4583: #endif
4584:
1.34 daniel 4585: if ((terminate) || (ctxt->input->buf->buffer->use > 80))
4586: htmlParseTryOrFinish(ctxt, terminate);
1.60 veillard 4587: } else if (ctxt->instate != XML_PARSER_EOF) {
4588: xmlParserInputBufferPush(ctxt->input->buf, 0, "");
1.32 daniel 4589: htmlParseTryOrFinish(ctxt, terminate);
1.60 veillard 4590: }
1.31 daniel 4591: if (terminate) {
4592: if ((ctxt->instate != XML_PARSER_EOF) &&
4593: (ctxt->instate != XML_PARSER_EPILOG) &&
4594: (ctxt->instate != XML_PARSER_MISC)) {
1.67 veillard 4595: ctxt->errNo = XML_ERR_DOCUMENT_END;
1.31 daniel 4596: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
4597: ctxt->sax->error(ctxt->userData,
4598: "Extra content at the end of the document\n");
4599: ctxt->wellFormed = 0;
4600: }
4601: if (ctxt->instate != XML_PARSER_EOF) {
4602: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4603: ctxt->sax->endDocument(ctxt->userData);
4604: }
4605: ctxt->instate = XML_PARSER_EOF;
4606: }
4607: return((xmlParserErrors) ctxt->errNo);
4608: }
4609:
4610: /************************************************************************
4611: * *
4612: * User entry points *
4613: * *
4614: ************************************************************************/
4615:
4616: /**
4617: * htmlCreatePushParserCtxt :
4618: * @sax: a SAX handler
4619: * @user_data: The user data returned on SAX callbacks
4620: * @chunk: a pointer to an array of chars
4621: * @size: number of chars in the array
4622: * @filename: an optional file name or URI
4623: * @enc: an optional encoding
4624: *
4625: * Create a parser context for using the HTML parser in push mode
4626: * To allow content encoding detection, @size should be >= 4
4627: * The value of @filename is used for fetching external entities
4628: * and error/warning reports.
4629: *
4630: * Returns the new parser context or NULL
4631: */
4632: htmlParserCtxtPtr
4633: htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data,
4634: const char *chunk, int size, const char *filename,
4635: xmlCharEncoding enc) {
4636: htmlParserCtxtPtr ctxt;
4637: htmlParserInputPtr inputStream;
4638: xmlParserInputBufferPtr buf;
4639:
4640: buf = xmlAllocParserInputBuffer(enc);
4641: if (buf == NULL) return(NULL);
4642:
4643: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
4644: if (ctxt == NULL) {
4645: xmlFree(buf);
4646: return(NULL);
4647: }
4648: memset(ctxt, 0, sizeof(htmlParserCtxt));
4649: htmlInitParserCtxt(ctxt);
4650: if (sax != NULL) {
4651: if (ctxt->sax != &htmlDefaultSAXHandler)
4652: xmlFree(ctxt->sax);
4653: ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler));
4654: if (ctxt->sax == NULL) {
4655: xmlFree(buf);
4656: xmlFree(ctxt);
4657: return(NULL);
4658: }
4659: memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler));
4660: if (user_data != NULL)
4661: ctxt->userData = user_data;
4662: }
4663: if (filename == NULL) {
4664: ctxt->directory = NULL;
4665: } else {
4666: ctxt->directory = xmlParserGetDirectory(filename);
4667: }
4668:
4669: inputStream = htmlNewInputStream(ctxt);
4670: if (inputStream == NULL) {
4671: xmlFreeParserCtxt(ctxt);
4672: return(NULL);
4673: }
4674:
4675: if (filename == NULL)
4676: inputStream->filename = NULL;
4677: else
4678: inputStream->filename = xmlMemStrdup(filename);
4679: inputStream->buf = buf;
4680: inputStream->base = inputStream->buf->buffer->content;
4681: inputStream->cur = inputStream->buf->buffer->content;
4682:
4683: inputPush(ctxt, inputStream);
4684:
4685: if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
4686: (ctxt->input->buf != NULL)) {
4687: xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
4688: #ifdef DEBUG_PUSH
1.81 veillard 4689: xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
1.31 daniel 4690: #endif
4691: }
4692:
4693: return(ctxt);
4694: }
1.1 daniel 4695:
4696: /**
4697: * htmlSAXParseDoc :
1.14 daniel 4698: * @cur: a pointer to an array of xmlChar
1.1 daniel 4699: * @encoding: a free form C string describing the HTML document encoding, or NULL
4700: * @sax: the SAX handler block
4701: * @userData: if using SAX, this pointer will be provided on callbacks.
4702: *
4703: * parse an HTML in-memory document and build a tree.
4704: * It use the given SAX function block to handle the parsing callback.
4705: * If sax is NULL, fallback to the default DOM tree building routines.
4706: *
4707: * Returns the resulting document tree
4708: */
4709:
4710: htmlDocPtr
1.14 daniel 4711: htmlSAXParseDoc(xmlChar *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) {
1.1 daniel 4712: htmlDocPtr ret;
4713: htmlParserCtxtPtr ctxt;
4714:
4715: if (cur == NULL) return(NULL);
4716:
4717:
4718: ctxt = htmlCreateDocParserCtxt(cur, encoding);
4719: if (ctxt == NULL) return(NULL);
4720: if (sax != NULL) {
4721: ctxt->sax = sax;
4722: ctxt->userData = userData;
4723: }
4724:
4725: htmlParseDocument(ctxt);
4726: ret = ctxt->myDoc;
4727: if (sax != NULL) {
4728: ctxt->sax = NULL;
4729: ctxt->userData = NULL;
4730: }
4731: htmlFreeParserCtxt(ctxt);
4732:
4733: return(ret);
4734: }
4735:
4736: /**
4737: * htmlParseDoc :
1.14 daniel 4738: * @cur: a pointer to an array of xmlChar
1.1 daniel 4739: * @encoding: a free form C string describing the HTML document encoding, or NULL
4740: *
4741: * parse an HTML in-memory document and build a tree.
4742: *
4743: * Returns the resulting document tree
4744: */
4745:
4746: htmlDocPtr
1.14 daniel 4747: htmlParseDoc(xmlChar *cur, const char *encoding) {
1.1 daniel 4748: return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
4749: }
4750:
4751:
4752: /**
4753: * htmlCreateFileParserCtxt :
4754: * @filename: the filename
4755: * @encoding: a free form C string describing the HTML document encoding, or NULL
4756: *
4757: * Create a parser context for a file content.
4758: * Automatic support for ZLIB/Compress compressed document is provided
4759: * by default if found at compile-time.
4760: *
4761: * Returns the new parser context or NULL
4762: */
4763: htmlParserCtxtPtr
4764: htmlCreateFileParserCtxt(const char *filename, const char *encoding)
4765: {
4766: htmlParserCtxtPtr ctxt;
4767: htmlParserInputPtr inputStream;
1.5 daniel 4768: xmlParserInputBufferPtr buf;
1.1 daniel 4769: /* htmlCharEncoding enc; */
4770:
1.5 daniel 4771: buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
4772: if (buf == NULL) return(NULL);
1.1 daniel 4773:
1.11 daniel 4774: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 4775: if (ctxt == NULL) {
4776: perror("malloc");
4777: return(NULL);
4778: }
1.19 daniel 4779: memset(ctxt, 0, sizeof(htmlParserCtxt));
1.1 daniel 4780: htmlInitParserCtxt(ctxt);
1.11 daniel 4781: inputStream = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 4782: if (inputStream == NULL) {
4783: perror("malloc");
1.11 daniel 4784: xmlFree(ctxt);
1.1 daniel 4785: return(NULL);
4786: }
1.19 daniel 4787: memset(inputStream, 0, sizeof(htmlParserInput));
1.1 daniel 4788:
1.11 daniel 4789: inputStream->filename = xmlMemStrdup(filename);
1.1 daniel 4790: inputStream->line = 1;
4791: inputStream->col = 1;
1.5 daniel 4792: inputStream->buf = buf;
1.21 daniel 4793: inputStream->directory = NULL;
1.1 daniel 4794:
1.5 daniel 4795: inputStream->base = inputStream->buf->buffer->content;
4796: inputStream->cur = inputStream->buf->buffer->content;
4797: inputStream->free = NULL;
1.1 daniel 4798:
4799: inputPush(ctxt, inputStream);
4800: return(ctxt);
4801: }
4802:
4803: /**
4804: * htmlSAXParseFile :
4805: * @filename: the filename
4806: * @encoding: a free form C string describing the HTML document encoding, or NULL
4807: * @sax: the SAX handler block
4808: * @userData: if using SAX, this pointer will be provided on callbacks.
4809: *
4810: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
4811: * compressed document is provided by default if found at compile-time.
4812: * It use the given SAX function block to handle the parsing callback.
4813: * If sax is NULL, fallback to the default DOM tree building routines.
4814: *
4815: * Returns the resulting document tree
4816: */
4817:
4818: htmlDocPtr
4819: htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
4820: void *userData) {
4821: htmlDocPtr ret;
4822: htmlParserCtxtPtr ctxt;
1.57 veillard 4823: htmlSAXHandlerPtr oldsax = NULL;
1.1 daniel 4824:
4825: ctxt = htmlCreateFileParserCtxt(filename, encoding);
4826: if (ctxt == NULL) return(NULL);
4827: if (sax != NULL) {
1.55 veillard 4828: oldsax = ctxt->sax;
1.1 daniel 4829: ctxt->sax = sax;
4830: ctxt->userData = userData;
4831: }
4832:
4833: htmlParseDocument(ctxt);
4834:
4835: ret = ctxt->myDoc;
4836: if (sax != NULL) {
1.55 veillard 4837: ctxt->sax = oldsax;
1.1 daniel 4838: ctxt->userData = NULL;
4839: }
4840: htmlFreeParserCtxt(ctxt);
4841:
4842: return(ret);
4843: }
4844:
4845: /**
4846: * htmlParseFile :
4847: * @filename: the filename
4848: * @encoding: a free form C string describing the HTML document encoding, or NULL
4849: *
4850: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
4851: * compressed document is provided by default if found at compile-time.
4852: *
4853: * Returns the resulting document tree
4854: */
4855:
4856: htmlDocPtr
4857: htmlParseFile(const char *filename, const char *encoding) {
4858: return(htmlSAXParseFile(filename, encoding, NULL, NULL));
4859: }
1.39 daniel 4860:
4861: #endif /* LIBXML_HTML_ENABLED */
Webmaster
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to HTMLparser.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Public] / XML