Annotation of XML/HTMLparser.c, revision 1.46
1.1 daniel 1: /*
2: * HTMLparser.c : an HTML 4.0 non-verifying parser
3: *
4: * See Copyright for the status of this software.
5: *
6: * Daniel.Veillard@w3.org
7: */
8:
9: #ifdef WIN32
1.29 daniel 10: #include "win32config.h"
1.1 daniel 11: #else
1.13 daniel 12: #include "config.h"
1.1 daniel 13: #endif
1.13 daniel 14:
1.39 daniel 15: #include "xmlversion.h"
16: #ifdef LIBXML_HTML_ENABLED
17:
1.1 daniel 18: #include <stdio.h>
1.13 daniel 19: #include <string.h> /* for memset() only */
20: #ifdef HAVE_CTYPE_H
1.1 daniel 21: #include <ctype.h>
1.13 daniel 22: #endif
23: #ifdef HAVE_STDLIB_H
1.1 daniel 24: #include <stdlib.h>
1.13 daniel 25: #endif
26: #ifdef HAVE_SYS_STAT_H
1.1 daniel 27: #include <sys/stat.h>
1.13 daniel 28: #endif
1.1 daniel 29: #ifdef HAVE_FCNTL_H
30: #include <fcntl.h>
31: #endif
32: #ifdef HAVE_UNISTD_H
33: #include <unistd.h>
34: #endif
35: #ifdef HAVE_ZLIB_H
36: #include <zlib.h>
37: #endif
38:
1.39 daniel 39: #include <libxml/xmlmemory.h>
40: #include <libxml/tree.h>
41: #include <libxml/HTMLparser.h>
42: #include <libxml/entities.h>
43: #include <libxml/encoding.h>
44: #include <libxml/valid.h>
45: #include <libxml/parserInternals.h>
46: #include <libxml/xmlIO.h>
1.31 daniel 47: #include "xml-error.h"
1.5 daniel 48:
49: #define HTML_MAX_NAMELEN 1000
50: #define INPUT_CHUNK 50
1.31 daniel 51: #define HTML_PARSER_BIG_BUFFER_SIZE 1024
52: #define HTML_PARSER_BUFFER_SIZE 100
1.1 daniel 53:
54: /* #define DEBUG */
1.31 daniel 55: /* #define DEBUG_PUSH */
1.1 daniel 56:
57: /************************************************************************
58: * *
59: * Parser stacks related functions and macros *
60: * *
61: ************************************************************************/
62:
63: /*
64: * Generic function for accessing stacks in the Parser Context
65: */
66:
1.30 daniel 67: #define PUSH_AND_POP(scope, type, name) \
68: scope int html##name##Push(htmlParserCtxtPtr ctxt, type value) { \
1.1 daniel 69: if (ctxt->name##Nr >= ctxt->name##Max) { \
70: ctxt->name##Max *= 2; \
1.11 daniel 71: ctxt->name##Tab = (void *) xmlRealloc(ctxt->name##Tab, \
1.1 daniel 72: ctxt->name##Max * sizeof(ctxt->name##Tab[0])); \
73: if (ctxt->name##Tab == NULL) { \
74: fprintf(stderr, "realloc failed !\n"); \
1.33 daniel 75: return(0); \
1.1 daniel 76: } \
77: } \
78: ctxt->name##Tab[ctxt->name##Nr] = value; \
79: ctxt->name = value; \
80: return(ctxt->name##Nr++); \
81: } \
1.30 daniel 82: scope type html##name##Pop(htmlParserCtxtPtr ctxt) { \
1.1 daniel 83: type ret; \
1.18 daniel 84: if (ctxt->name##Nr < 0) return(0); \
1.1 daniel 85: ctxt->name##Nr--; \
1.18 daniel 86: if (ctxt->name##Nr < 0) return(0); \
1.1 daniel 87: if (ctxt->name##Nr > 0) \
88: ctxt->name = ctxt->name##Tab[ctxt->name##Nr - 1]; \
89: else \
90: ctxt->name = NULL; \
91: ret = ctxt->name##Tab[ctxt->name##Nr]; \
92: ctxt->name##Tab[ctxt->name##Nr] = 0; \
93: return(ret); \
94: } \
95:
1.30 daniel 96: PUSH_AND_POP(extern, xmlNodePtr, node)
97: PUSH_AND_POP(extern, xmlChar*, name)
1.1 daniel 98:
99: /*
100: * Macros for accessing the content. Those should be used only by the parser,
101: * and not exported.
102: *
103: * Dirty macros, i.e. one need to make assumption on the context to use them
104: *
1.14 daniel 105: * CUR_PTR return the current pointer to the xmlChar to be parsed.
106: * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled
1.1 daniel 107: * in ISO-Latin or UTF-8, and the current 16 bit value if compiled
108: * in UNICODE mode. This should be used internally by the parser
109: * only to compare to ASCII values otherwise it would break when
110: * running with UTF-8 encoding.
1.14 daniel 111: * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
1.1 daniel 112: * to compare on ASCII based substring.
1.14 daniel 113: * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR
1.1 daniel 114: * it should be used only to compare on ASCII based substring.
1.14 daniel 115: * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
1.1 daniel 116: * strings within the parser.
117: *
118: * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
119: *
120: * CURRENT Returns the current char value, with the full decoding of
121: * UTF-8 if we are using this mode. It returns an int.
122: * NEXT Skip to the next character, this does the proper decoding
123: * in UTF-8 mode. It also pop-up unfinished entities on the fly.
124: * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly
125: */
126:
1.36 daniel 127: #define CUR ((int) (*ctxt->input->cur))
128:
1.1 daniel 129: #define UPPER (toupper(*ctxt->input->cur))
1.36 daniel 130:
1.26 daniel 131: #define SKIP(val) ctxt->nbChars += (val),ctxt->input->cur += (val)
1.36 daniel 132:
1.1 daniel 133: #define NXT(val) ctxt->input->cur[(val)]
1.36 daniel 134:
1.1 daniel 135: #define UPP(val) (toupper(ctxt->input->cur[(val)]))
1.36 daniel 136:
1.1 daniel 137: #define CUR_PTR ctxt->input->cur
1.36 daniel 138:
1.5 daniel 139: #define SHRINK xmlParserInputShrink(ctxt->input)
1.36 daniel 140:
1.5 daniel 141: #define GROW xmlParserInputGrow(ctxt->input, INPUT_CHUNK)
1.1 daniel 142:
1.36 daniel 143: #define CURRENT ((int) (*ctxt->input->cur))
1.1 daniel 144:
1.36 daniel 145: #define NEXT htmlNextChar(ctxt);
1.35 daniel 146:
1.36 daniel 147: #define SKIP_BLANKS htmlSkipBlankChars(ctxt);
1.35 daniel 148:
149: /**
150: * htmlNextChar:
151: * @ctxt: the HTML parser context
152: *
153: * Skip to the next char input char.
154: */
155:
156: void
157: htmlNextChar(htmlParserCtxtPtr ctxt) {
1.44 daniel 158: if (ctxt->instate == XML_PARSER_EOF)
159: return;
1.35 daniel 160: if ((*ctxt->input->cur == 0) &&
161: (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
162: xmlPopInput(ctxt);
163: } else {
164: if (*(ctxt->input->cur) == '\n') {
165: ctxt->input->line++; ctxt->input->col = 1;
166: } else ctxt->input->col++;
167: ctxt->input->cur++;
168: ctxt->nbChars++;
169: if (*ctxt->input->cur == 0)
170: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
171: }
172: }
1.5 daniel 173:
1.36 daniel 174: /**
175: * htmlSkipBlankChars:
176: * @ctxt: the HTML parser context
177: *
178: * skip all blanks character found at that point in the input streams.
179: *
180: * Returns the number of space chars skipped
181: */
182:
183: int
184: htmlSkipBlankChars(xmlParserCtxtPtr ctxt) {
185: int res = 0;
186:
187: while (IS_BLANK(*(ctxt->input->cur))) {
188: if ((*ctxt->input->cur == 0) &&
189: (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
190: xmlPopInput(ctxt);
191: } else {
192: if (*(ctxt->input->cur) == '\n') {
193: ctxt->input->line++; ctxt->input->col = 1;
194: } else ctxt->input->col++;
195: ctxt->input->cur++;
196: ctxt->nbChars++;
197: if (*ctxt->input->cur == 0)
198: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
199: }
200: res++;
201: }
202: return(res);
203: }
1.1 daniel 204:
205:
1.5 daniel 206:
1.1 daniel 207: /************************************************************************
208: * *
209: * The list of HTML elements and their properties *
210: * *
211: ************************************************************************/
212:
213: /*
214: * Start Tag: 1 means the start tag can be ommited
215: * End Tag: 1 means the end tag can be ommited
216: * 2 means it's forbidden (empty elements)
217: * Depr: this element is deprecated
218: * DTD: 1 means that this element is valid only in the Loose DTD
219: * 2 means that this element is valid only in the Frameset DTD
220: *
221: * Name,Start Tag,End Tag, Empty, Depr., DTD, Description
222: */
223: htmlElemDesc html40ElementTable[] = {
1.26 daniel 224: { "a", 0, 0, 0, 0, 0, "anchor " },
225: { "abbr", 0, 0, 0, 0, 0, "abbreviated form" },
226: { "acronym", 0, 0, 0, 0, 0, "" },
227: { "address", 0, 0, 0, 0, 0, "information on author " },
228: { "applet", 0, 0, 0, 1, 1, "java applet " },
229: { "area", 0, 2, 1, 0, 0, "client-side image map area " },
230: { "b", 0, 0, 0, 0, 0, "bold text style" },
231: { "base", 0, 2, 1, 0, 0, "document base uri " },
232: { "basefont", 0, 2, 1, 1, 1, "base font size " },
233: { "bdo", 0, 0, 0, 0, 0, "i18n bidi over-ride " },
234: { "big", 0, 0, 0, 0, 0, "large text style" },
235: { "blockquote", 0, 0, 0, 0, 0, "long quotation " },
236: { "body", 1, 1, 0, 0, 0, "document body " },
237: { "br", 0, 2, 1, 0, 0, "forced line break " },
238: { "button", 0, 0, 0, 0, 0, "push button " },
239: { "caption", 0, 0, 0, 0, 0, "table caption " },
240: { "center", 0, 0, 0, 1, 1, "shorthand for div align=center " },
241: { "cite", 0, 0, 0, 0, 0, "citation" },
242: { "code", 0, 0, 0, 0, 0, "computer code fragment" },
243: { "col", 0, 2, 1, 0, 0, "table column " },
244: { "colgroup", 0, 1, 0, 0, 0, "table column group " },
245: { "dd", 0, 1, 0, 0, 0, "definition description " },
246: { "del", 0, 0, 0, 0, 0, "deleted text " },
247: { "dfn", 0, 0, 0, 0, 0, "instance definition" },
248: { "dir", 0, 0, 0, 1, 1, "directory list" },
249: { "div", 0, 0, 0, 0, 0, "generic language/style container"},
250: { "dl", 0, 0, 0, 0, 0, "definition list " },
251: { "dt", 0, 1, 0, 0, 0, "definition term " },
252: { "em", 0, 0, 0, 0, 0, "emphasis" },
253: { "fieldset", 0, 0, 0, 0, 0, "form control group " },
254: { "font", 0, 0, 0, 1, 1, "local change to font " },
255: { "form", 0, 0, 0, 0, 0, "interactive form " },
256: { "frame", 0, 2, 1, 0, 2, "subwindow " },
257: { "frameset", 0, 0, 0, 0, 2, "window subdivision" },
258: { "h1", 0, 0, 0, 0, 0, "heading " },
259: { "h2", 0, 0, 0, 0, 0, "heading " },
260: { "h3", 0, 0, 0, 0, 0, "heading " },
261: { "h4", 0, 0, 0, 0, 0, "heading " },
262: { "h5", 0, 0, 0, 0, 0, "heading " },
263: { "h6", 0, 0, 0, 0, 0, "heading " },
264: { "head", 1, 1, 0, 0, 0, "document head " },
265: { "hr", 0, 2, 1, 0, 0, "horizontal rule " },
266: { "html", 1, 1, 0, 0, 0, "document root element " },
267: { "i", 0, 0, 0, 0, 0, "italic text style" },
268: { "iframe", 0, 0, 0, 0, 1, "inline subwindow " },
269: { "img", 0, 2, 1, 0, 0, "embedded image " },
270: { "input", 0, 2, 1, 0, 0, "form control " },
271: { "ins", 0, 0, 0, 0, 0, "inserted text" },
272: { "isindex", 0, 2, 1, 1, 1, "single line prompt " },
273: { "kbd", 0, 0, 0, 0, 0, "text to be entered by the user" },
274: { "label", 0, 0, 0, 0, 0, "form field label text " },
275: { "legend", 0, 0, 0, 0, 0, "fieldset legend " },
276: { "li", 0, 1, 0, 0, 0, "list item " },
277: { "link", 0, 2, 1, 0, 0, "a media-independent link " },
278: { "map", 0, 0, 0, 0, 0, "client-side image map " },
279: { "menu", 0, 0, 0, 1, 1, "menu list " },
280: { "meta", 0, 2, 1, 0, 0, "generic metainformation " },
281: { "noframes", 0, 0, 0, 0, 2, "alternate content container for non frame-based rendering " },
282: { "noscript", 0, 0, 0, 0, 0, "alternate content container for non script-based rendering " },
283: { "object", 0, 0, 0, 0, 0, "generic embedded object " },
284: { "ol", 0, 0, 0, 0, 0, "ordered list " },
285: { "optgroup", 0, 0, 0, 0, 0, "option group " },
286: { "option", 0, 1, 0, 0, 0, "selectable choice " },
287: { "p", 0, 1, 0, 0, 0, "paragraph " },
288: { "param", 0, 2, 1, 0, 0, "named property value " },
289: { "pre", 0, 0, 0, 0, 0, "preformatted text " },
290: { "q", 0, 0, 0, 0, 0, "short inline quotation " },
291: { "s", 0, 0, 0, 1, 1, "strike-through text style" },
292: { "samp", 0, 0, 0, 0, 0, "sample program output, scripts, etc." },
293: { "script", 0, 0, 0, 0, 0, "script statements " },
294: { "select", 0, 0, 0, 0, 0, "option selector " },
295: { "small", 0, 0, 0, 0, 0, "small text style" },
296: { "span", 0, 0, 0, 0, 0, "generic language/style container " },
297: { "strike", 0, 0, 0, 1, 1, "strike-through text" },
298: { "strong", 0, 0, 0, 0, 0, "strong emphasis" },
299: { "style", 0, 0, 0, 0, 0, "style info " },
300: { "sub", 0, 0, 0, 0, 0, "subscript" },
301: { "sup", 0, 0, 0, 0, 0, "superscript " },
302: { "table", 0, 0, 0, 0, 0, " " },
303: { "tbody", 1, 1, 0, 0, 0, "table body " },
304: { "td", 0, 1, 0, 0, 0, "table data cell" },
305: { "textarea", 0, 0, 0, 0, 0, "multi-line text field " },
306: { "tfoot", 0, 1, 0, 0, 0, "table footer " },
307: { "th", 0, 1, 0, 0, 0, "table header cell" },
308: { "thead", 0, 1, 0, 0, 0, "table header " },
309: { "title", 0, 0, 0, 0, 0, "document title " },
310: { "tr", 0, 1, 0, 0, 0, "table row " },
311: { "tt", 0, 0, 0, 0, 0, "teletype or monospaced text style" },
312: { "u", 0, 0, 0, 1, 1, "underlined text style" },
313: { "ul", 0, 0, 0, 0, 0, "unordered list " },
314: { "var", 0, 0, 0, 0, 0, "instance of a variable or program argument" },
1.1 daniel 315: };
316:
317: /*
318: * start tags that imply the end of a current element
319: * any tag of each line implies the end of the current element if the type of
320: * that element is in the same line
321: */
1.8 daniel 322: char *htmlEquEnd[] = {
1.26 daniel 323: "dt", "dd", "li", "option", NULL,
324: "h1", "h2", "h3", "h4", "h5", "h6", NULL,
325: "ol", "menu", "dir", "address", "pre", "listing", "xmp", NULL,
1.1 daniel 326: NULL
327: };
328: /*
329: * acording the HTML DTD, HR should be added to the 2nd line above, as it
330: * is not allowed within a H1, H2, H3, etc. But we should tolerate that case
331: * because many documents contain rules in headings...
332: */
333:
334: /*
335: * start tags that imply the end of current element
336: */
1.8 daniel 337: char *htmlStartClose[] = {
1.26 daniel 338: "form", "form", "p", "hr", "h1", "h2", "h3", "h4", "h5", "h6",
339: "dl", "ul", "ol", "menu", "dir", "address", "pre",
340: "listing", "xmp", "head", NULL,
341: "head", "p", NULL,
342: "title", "p", NULL,
343: "body", "head", "style", "link", "title", "p", NULL,
344: "li", "p", "h1", "h2", "h3", "h4", "h5", "h6", "dl", "address",
345: "pre", "listing", "xmp", "head", "li", NULL,
346: "hr", "p", "head", NULL,
347: "h1", "p", "head", NULL,
348: "h2", "p", "head", NULL,
349: "h3", "p", "head", NULL,
350: "h4", "p", "head", NULL,
351: "h5", "p", "head", NULL,
352: "h6", "p", "head", NULL,
353: "dir", "p", "head", NULL,
354: "address", "p", "head", "ul", NULL,
355: "pre", "p", "head", "ul", NULL,
356: "listing", "p", "head", NULL,
357: "xmp", "p", "head", NULL,
358: "blockquote", "p", "head", NULL,
359: "dl", "p", "dt", "menu", "dir", "address", "pre", "listing",
360: "xmp", "head", NULL,
361: "dt", "p", "menu", "dir", "address", "pre", "listing", "xmp",
362: "head", "dd", NULL,
363: "dd", "p", "menu", "dir", "address", "pre", "listing", "xmp",
364: "head", "dt", NULL,
365: "ul", "p", "head", "ol", "menu", "dir", "address", "pre",
366: "listing", "xmp", NULL,
367: "ol", "p", "head", "ul", NULL,
368: "menu", "p", "head", "ul", NULL,
369: "p", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", NULL,
370: "div", "p", "head", NULL,
371: "noscript", "p", "head", NULL,
372: "center", "font", "b", "i", "p", "head", NULL,
373: "a", "a", NULL,
374: "caption", "p", NULL,
375: "colgroup", "caption", "colgroup", "col", "p", NULL,
376: "col", "caption", "col", "p", NULL,
377: "table", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", "pre",
378: "listing", "xmp", "a", NULL,
379: "th", "th", "td", NULL,
380: "td", "th", "td", "p", NULL,
381: "tr", "th", "td", "tr", "caption", "col", "colgroup", "p", NULL,
382: "thead", "caption", "col", "colgroup", NULL,
383: "tfoot", "th", "td", "tr", "caption", "col", "colgroup", "thead",
384: "tbody", "p", NULL,
385: "tbody", "th", "td", "tr", "caption", "col", "colgroup", "thead",
386: "tfoot", "tbody", "p", NULL,
387: "optgroup", "option", NULL,
388: "fieldset", "legend", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6",
389: "pre", "listing", "xmp", "a", NULL,
1.1 daniel 390: NULL
391: };
392:
1.43 daniel 393:
1.8 daniel 394: static char** htmlStartCloseIndex[100];
1.1 daniel 395: static int htmlStartCloseIndexinitialized = 0;
396:
397: /************************************************************************
398: * *
399: * functions to handle HTML specific data *
400: * *
401: ************************************************************************/
402:
403: /**
404: * htmlInitAutoClose:
405: *
406: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
407: *
408: */
409: void
410: htmlInitAutoClose(void) {
411: int index, i = 0;
412:
413: if (htmlStartCloseIndexinitialized) return;
414:
415: for (index = 0;index < 100;index ++) htmlStartCloseIndex[index] = NULL;
416: index = 0;
417: while ((htmlStartClose[i] != NULL) && (index < 100 - 1)) {
418: htmlStartCloseIndex[index++] = &htmlStartClose[i];
419: while (htmlStartClose[i] != NULL) i++;
420: i++;
421: }
422: }
423:
424: /**
425: * htmlTagLookup:
426: * @tag: The tag name
427: *
428: * Lookup the HTML tag in the ElementTable
429: *
430: * Returns the related htmlElemDescPtr or NULL if not found.
431: */
432: htmlElemDescPtr
1.14 daniel 433: htmlTagLookup(const xmlChar *tag) {
1.1 daniel 434: int i = 0;
435:
436: for (i = 0; i < (sizeof(html40ElementTable) /
437: sizeof(html40ElementTable[0]));i++) {
1.8 daniel 438: if (!xmlStrcmp(tag, BAD_CAST html40ElementTable[i].name))
1.1 daniel 439: return(&html40ElementTable[i]);
440: }
441: return(NULL);
442: }
443:
444: /**
445: * htmlCheckAutoClose:
446: * @new: The new tag name
447: * @old: The old tag name
448: *
449: * Checks wether the new tag is one of the registered valid tags for closing old.
450: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
451: *
452: * Returns 0 if no, 1 if yes.
453: */
454: int
1.14 daniel 455: htmlCheckAutoClose(const xmlChar *new, const xmlChar *old) {
1.1 daniel 456: int i, index;
1.8 daniel 457: char **close;
1.1 daniel 458:
459: if (htmlStartCloseIndexinitialized == 0) htmlInitAutoClose();
460:
461: /* inefficient, but not a big deal */
462: for (index = 0; index < 100;index++) {
463: close = htmlStartCloseIndex[index];
464: if (close == NULL) return(0);
1.8 daniel 465: if (!xmlStrcmp(BAD_CAST *close, new)) break;
1.1 daniel 466: }
467:
468: i = close - htmlStartClose;
469: i++;
470: while (htmlStartClose[i] != NULL) {
1.8 daniel 471: if (!xmlStrcmp(BAD_CAST htmlStartClose[i], old)) {
1.1 daniel 472: return(1);
473: }
474: i++;
475: }
476: return(0);
477: }
478:
479: /**
480: * htmlAutoClose:
481: * @ctxt: an HTML parser context
482: * @new: The new tag name
483: *
484: * The HTmL DtD allows a tag to implicitely close other tags.
485: * The list is kept in htmlStartClose array. This function is
486: * called when a new tag has been detected and generates the
487: * appropriates closes if possible/needed.
488: */
489: void
1.14 daniel 490: htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar *new) {
1.15 daniel 491: xmlChar *oldname;
492: while ((ctxt->name != NULL) &&
493: (htmlCheckAutoClose(new, ctxt->name))) {
1.1 daniel 494: #ifdef DEBUG
1.18 daniel 495: fprintf(stderr,"htmlAutoClose: %s closes %s\n", new, ctxt->name);
1.1 daniel 496: #endif
497: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1.15 daniel 498: ctxt->sax->endElement(ctxt->userData, ctxt->name);
1.24 daniel 499: oldname = htmlnamePop(ctxt);
1.18 daniel 500: if (oldname != NULL) {
501: #ifdef DEBUG
502: fprintf(stderr,"htmlAutoClose: popped %s\n", oldname);
503: #endif
1.17 daniel 504: xmlFree(oldname);
1.18 daniel 505: }
1.1 daniel 506: }
507: }
508:
509: /**
1.28 daniel 510: * htmlAutoCloseTag:
511: * @doc: the HTML document
512: * @name: The tag name
513: * @elem: the HTML element
514: *
515: * The HTmL DtD allows a tag to implicitely close other tags.
516: * The list is kept in htmlStartClose array. This function checks
517: * if the element or one of it's children would autoclose the
518: * given tag.
519: *
520: * Returns 1 if autoclose, 0 otherwise
521: */
522: int
523: htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) {
524: htmlNodePtr child;
525:
526: if (elem == NULL) return(1);
527: if (!xmlStrcmp(name, elem->name)) return(0);
528: if (htmlCheckAutoClose(elem->name, name)) return(1);
1.37 daniel 529: child = elem->children;
1.28 daniel 530: while (child != NULL) {
531: if (htmlAutoCloseTag(doc, name, child)) return(1);
532: child = child->next;
533: }
534: return(0);
535: }
536:
537: /**
538: * htmlIsAutoClosed:
539: * @doc: the HTML document
540: * @elem: the HTML element
541: *
542: * The HTmL DtD allows a tag to implicitely close other tags.
543: * The list is kept in htmlStartClose array. This function checks
544: * if a tag is autoclosed by one of it's child
545: *
546: * Returns 1 if autoclosed, 0 otherwise
547: */
548: int
549: htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) {
550: htmlNodePtr child;
551:
552: if (elem == NULL) return(1);
1.37 daniel 553: child = elem->children;
1.28 daniel 554: while (child != NULL) {
555: if (htmlAutoCloseTag(doc, elem->name, child)) return(1);
556: child = child->next;
557: }
558: return(0);
559: }
560:
561: /**
1.1 daniel 562: * htmlAutoCloseOnClose:
563: * @ctxt: an HTML parser context
564: * @new: The new tag name
565: *
566: * The HTmL DtD allows an ending tag to implicitely close other tags.
567: */
568: void
1.14 daniel 569: htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar *new) {
1.1 daniel 570: htmlElemDescPtr info;
1.15 daniel 571: xmlChar *oldname;
1.18 daniel 572: int i;
1.1 daniel 573:
1.18 daniel 574: #ifdef DEBUG
575: fprintf(stderr,"Close of %s stack: %d elements\n", new, ctxt->nameNr);
576: for (i = 0;i < ctxt->nameNr;i++)
577: fprintf(stderr,"%d : %s\n", i, ctxt->nameTab[i]);
578: #endif
579:
580: for (i = (ctxt->nameNr - 1);i >= 0;i--) {
581: if (!xmlStrcmp(new, ctxt->nameTab[i])) break;
582: }
583: if (i < 0) return;
584:
585: while (xmlStrcmp(new, ctxt->name)) {
1.15 daniel 586: info = htmlTagLookup(ctxt->name);
1.1 daniel 587: if ((info == NULL) || (info->endTag == 1)) {
588: #ifdef DEBUG
1.18 daniel 589: fprintf(stderr,"htmlAutoCloseOnClose: %s closes %s\n", new, ctxt->name);
590: #endif
591: } else {
592: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
593: ctxt->sax->error(ctxt->userData,
594: "Opening and ending tag mismatch: %s and %s\n",
595: new, ctxt->name);
596: ctxt->wellFormed = 0;
597: }
598: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
599: ctxt->sax->endElement(ctxt->userData, ctxt->name);
1.24 daniel 600: oldname = htmlnamePop(ctxt);
1.18 daniel 601: if (oldname != NULL) {
602: #ifdef DEBUG
603: fprintf(stderr,"htmlAutoCloseOnClose: popped %s\n", oldname);
1.1 daniel 604: #endif
1.18 daniel 605: xmlFree(oldname);
606: }
1.1 daniel 607: }
608: }
609:
1.43 daniel 610: /**
611: * htmlCheckImplied:
612: * @ctxt: an HTML parser context
613: * @new: The new tag name
614: *
615: * The HTmL DtD allows a tag to exists only implicitely
616: * called when a new tag has been detected and generates the
617: * appropriates implicit tags if missing
618: */
619: void
620: htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *new) {
1.46 ! daniel 621: if (!xmlStrcmp(new, BAD_CAST"html"))
1.43 daniel 622: return;
623: if (ctxt->nameNr <= 0) {
624: #ifdef DEBUG
625: fprintf(stderr,"Implied element html: pushed html\n");
626: #endif
627: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"html"));
628: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
629: ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL);
630: }
1.46 ! daniel 631: if ((!xmlStrcmp(new, BAD_CAST"body")) || (!xmlStrcmp(new, BAD_CAST"head")))
1.43 daniel 632: return;
633: if (ctxt->nameNr <= 1) {
1.46 ! daniel 634: if ((!xmlStrcmp(new, BAD_CAST"script")) ||
! 635: (!xmlStrcmp(new, BAD_CAST"style")) ||
! 636: (!xmlStrcmp(new, BAD_CAST"meta")) ||
! 637: (!xmlStrcmp(new, BAD_CAST"link")) ||
! 638: (!xmlStrcmp(new, BAD_CAST"title")) ||
! 639: (!xmlStrcmp(new, BAD_CAST"base"))) {
1.43 daniel 640: /*
641: * dropped OBJECT ... i you put it first BODY will be
642: * assumed !
643: */
644: #ifdef DEBUG
645: fprintf(stderr,"Implied element head: pushed head\n");
646: #endif
647: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"head"));
648: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
649: ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL);
650: } else {
651: #ifdef DEBUG
652: fprintf(stderr,"Implied element body: pushed body\n");
653: #endif
654: htmlnamePush(ctxt, xmlStrdup(BAD_CAST"body"));
655: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
656: ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL);
657: }
658: }
659: }
660:
1.1 daniel 661: /************************************************************************
662: * *
663: * The list of HTML predefined entities *
664: * *
665: ************************************************************************/
666:
667:
668: htmlEntityDesc html40EntitiesTable[] = {
669: /*
670: * the 4 absolute ones,
671: */
672: { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
673: { 38, "amp", "ampersand, U+0026 ISOnum" },
674: { 60, "lt", "less-than sign, U+003C ISOnum" },
675: { 62, "gt", "greater-than sign, U+003E ISOnum" },
676:
677: /*
678: * A bunch still in the 128-255 range
679: * Replacing them depend really on the charset used.
680: */
1.28 daniel 681: { 39, "apos", "single quote" },
1.1 daniel 682: { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
683: { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
684: { 162, "cent", "cent sign, U+00A2 ISOnum" },
685: { 163, "pound","pound sign, U+00A3 ISOnum" },
686: { 164, "curren","currency sign, U+00A4 ISOnum" },
687: { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
688: { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
689: { 167, "sect", "section sign, U+00A7 ISOnum" },
690: { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
691: { 169, "copy", "copyright sign, U+00A9 ISOnum" },
692: { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
693: { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
694: { 172, "not", "not sign, U+00AC ISOnum" },
695: { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
696: { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
697: { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
698: { 176, "deg", "degree sign, U+00B0 ISOnum" },
699: { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
700: { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
701: { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
702: { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
703: { 181, "micro","micro sign, U+00B5 ISOnum" },
704: { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1.7 daniel 705: { 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1.1 daniel 706: { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
707: { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
708: { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1.7 daniel 709: { 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1.1 daniel 710: { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
711: { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
712: { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
713: { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
714: { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
715: { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
716: { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
717: { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
718: { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
719: { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
720: { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
721: { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
722: { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
723: { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
724: { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
725: { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
726: { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
727: { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
728: { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
729: { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
730: { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
731: { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
732: { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
733: { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
734: { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
735: { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
736: { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
737: { 215, "times","multiplication sign, U+00D7 ISOnum" },
1.7 daniel 738: { 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1.1 daniel 739: { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
740: { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
741: { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
742: { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
743: { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
744: { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
745: { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
746: { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
747: { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
748: { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
749: { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
750: { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
751: { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
752: { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
753: { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
754: { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
755: { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
756: { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
757: { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
758: { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
759: { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
760: { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
761: { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
762: { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
763: { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
764: { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
765: { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
766: { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
767: { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
768: { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
769: { 247, "divide","division sign, U+00F7 ISOnum" },
770: { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
771: { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
772: { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
773: { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
774: { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
775: { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
776: { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
777: { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
778:
779: /*
780: * Anything below should really be kept as entities references
781: */
782: { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
783:
784: { 913, "Alpha","greek capital letter alpha, U+0391" },
785: { 914, "Beta", "greek capital letter beta, U+0392" },
786: { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
787: { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
788: { 917, "Epsilon","greek capital letter epsilon, U+0395" },
789: { 918, "Zeta", "greek capital letter zeta, U+0396" },
790: { 919, "Eta", "greek capital letter eta, U+0397" },
791: { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
792: { 921, "Iota", "greek capital letter iota, U+0399" },
793: { 922, "Kappa","greek capital letter kappa, U+039A" },
794: { 923, "Lambda""greek capital letter lambda, U+039B ISOgrk3" },
795: { 924, "Mu", "greek capital letter mu, U+039C" },
796: { 925, "Nu", "greek capital letter nu, U+039D" },
797: { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
798: { 927, "Omicron","greek capital letter omicron, U+039F" },
799: { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
800: { 929, "Rho", "greek capital letter rho, U+03A1" },
801: { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
802: { 932, "Tau", "greek capital letter tau, U+03A4" },
803: { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
804: { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
805: { 935, "Chi", "greek capital letter chi, U+03A7" },
806: { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
807: { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
808:
809: { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
810: { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
811: { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
812: { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
813: { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
814: { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
815: { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
816: { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
817: { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
818: { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
819: { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
820: { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
821: { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
822: { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
823: { 959, "omicron","greek small letter omicron, U+03BF NEW" },
824: { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
825: { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
826: { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
827: { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
828: { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
829: { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
830: { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
831: { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
832: { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
833: { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
834: { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
835: { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
836: { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
837:
838: { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
839: { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
840: { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
841: { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
842: { 8254, "oline","overline = spacing overscore, U+203E NEW" },
843: { 8260, "frasl","fraction slash, U+2044 NEW" },
844:
1.7 daniel 845: { 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
1.1 daniel 846: { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
847: { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
848: { 8482, "trade","trade mark sign, U+2122 ISOnum" },
849: { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
850: { 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
851: { 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
852: { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
853: { 8595, "darr", "downwards arrow, U+2193 ISOnum" },
854: { 8596, "harr", "left right arrow, U+2194 ISOamsa" },
855: { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
856: { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
857: { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
858: { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
859: { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
860: { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
861:
862:
863: { 8704, "forall","for all, U+2200 ISOtech" },
864: { 8706, "part", "partial differential, U+2202 ISOtech" },
865: { 8707, "exist","there exists, U+2203 ISOtech" },
866: { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
867: { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
868: { 8712, "isin", "element of, U+2208 ISOtech" },
869: { 8713, "notin","not an element of, U+2209 ISOtech" },
870: { 8715, "ni", "contains as member, U+220B ISOtech" },
871: { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
872: { 8721, "sum", "n-ary sumation, U+2211 ISOamsb" },
873: { 8722, "minus","minus sign, U+2212 ISOtech" },
874: { 8727, "lowast","asterisk operator, U+2217 ISOtech" },
875: { 8730, "radic","square root = radical sign, U+221A ISOtech" },
876: { 8733, "prop", "proportional to, U+221D ISOtech" },
877: { 8734, "infin","infinity, U+221E ISOtech" },
878: { 8736, "ang", "angle, U+2220 ISOamso" },
879: { 8743, "and", "logical and = wedge, U+2227 ISOtech" },
880: { 8744, "or", "logical or = vee, U+2228 ISOtech" },
881: { 8745, "cap", "intersection = cap, U+2229 ISOtech" },
882: { 8746, "cup", "union = cup, U+222A ISOtech" },
883: { 8747, "int", "integral, U+222B ISOtech" },
884: { 8756, "there4","therefore, U+2234 ISOtech" },
885: { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
886: { 8773, "cong", "approximately equal to, U+2245 ISOtech" },
887: { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
888: { 8800, "ne", "not equal to, U+2260 ISOtech" },
889: { 8801, "equiv","identical to, U+2261 ISOtech" },
890: { 8804, "le", "less-than or equal to, U+2264 ISOtech" },
891: { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
892: { 8834, "sub", "subset of, U+2282 ISOtech" },
893: { 8835, "sup", "superset of, U+2283 ISOtech" },
894: { 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
895: { 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
896: { 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
897: { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
898: { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
899: { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
900: { 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
901: { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
902: { 8969, "rceil","right ceiling, U+2309 ISOamsc" },
903: { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
904: { 8971, "rfloor","right floor, U+230B ISOamsc" },
905: { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
906: { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
907: { 9674, "loz", "lozenge, U+25CA ISOpub" },
908:
909: { 9824, "spades","black spade suit, U+2660 ISOpub" },
910: { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
911: { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
912: { 9830, "diams","black diamond suit, U+2666 ISOpub" },
913:
914: { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
915: { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
916: { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
917: { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
918: { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
919: { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
920: { 732, "tilde","small tilde, U+02DC ISOdia" },
921:
922: { 8194, "ensp", "en space, U+2002 ISOpub" },
923: { 8195, "emsp", "em space, U+2003 ISOpub" },
924: { 8201, "thinsp","thin space, U+2009 ISOpub" },
925: { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
926: { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
927: { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
928: { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
929: { 8211, "ndash","en dash, U+2013 ISOpub" },
930: { 8212, "mdash","em dash, U+2014 ISOpub" },
931: { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
932: { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
933: { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
934: { 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
935: { 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
936: { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
937: { 8224, "dagger","dagger, U+2020 ISOpub" },
938: { 8225, "Dagger","double dagger, U+2021 ISOpub" },
939: { 8240, "permil","per mille sign, U+2030 ISOtech" },
940: { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
1.7 daniel 941: { 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
1.1 daniel 942: { 8364, "euro", "euro sign, U+20AC NEW" }
943: };
944:
945: /************************************************************************
946: * *
947: * Commodity functions to handle entities *
948: * *
949: ************************************************************************/
950:
951: /*
952: * Macro used to grow the current buffer.
953: */
954: #define growBuffer(buffer) { \
955: buffer##_size *= 2; \
1.14 daniel 956: buffer = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
1.1 daniel 957: if (buffer == NULL) { \
958: perror("realloc failed"); \
1.33 daniel 959: return(NULL); \
1.1 daniel 960: } \
961: }
962:
963: /**
964: * htmlEntityLookup:
965: * @name: the entity name
966: *
967: * Lookup the given entity in EntitiesTable
968: *
969: * TODO: the linear scan is really ugly, an hash table is really needed.
970: *
971: * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
972: */
973: htmlEntityDescPtr
1.14 daniel 974: htmlEntityLookup(const xmlChar *name) {
1.1 daniel 975: int i;
976:
977: for (i = 0;i < (sizeof(html40EntitiesTable)/
978: sizeof(html40EntitiesTable[0]));i++) {
1.8 daniel 979: if (!xmlStrcmp(name, BAD_CAST html40EntitiesTable[i].name)) {
1.1 daniel 980: #ifdef DEBUG
1.18 daniel 981: fprintf(stderr,"Found entity %s\n", name);
1.1 daniel 982: #endif
983: return(&html40EntitiesTable[i]);
984: }
985: }
986: return(NULL);
987: }
988:
989:
990: /**
991: * htmlDecodeEntities:
992: * @ctxt: the parser context
993: * @len: the len to decode (in bytes !), -1 for no size limit
1.14 daniel 994: * @end: an end marker xmlChar, 0 if none
995: * @end2: an end marker xmlChar, 0 if none
996: * @end3: an end marker xmlChar, 0 if none
1.1 daniel 997: *
998: * Subtitute the HTML entities by their value
999: *
1.19 daniel 1000: * DEPRECATED !!!!
1.1 daniel 1001: *
1002: * Returns A newly allocated string with the substitution done. The caller
1003: * must deallocate it !
1004: */
1.14 daniel 1005: xmlChar *
1.1 daniel 1006: htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
1.14 daniel 1007: xmlChar end, xmlChar end2, xmlChar end3) {
1008: xmlChar *buffer = NULL;
1.1 daniel 1009: int buffer_size = 0;
1.14 daniel 1010: xmlChar *out = NULL;
1011: xmlChar *name = NULL;
1.1 daniel 1012:
1.14 daniel 1013: xmlChar *cur = NULL;
1.1 daniel 1014: htmlEntityDescPtr ent;
1.5 daniel 1015: int nbchars = 0;
1.1 daniel 1016: unsigned int max = (unsigned int) len;
1017:
1018: /*
1019: * allocate a translation buffer.
1020: */
1.31 daniel 1021: buffer_size = HTML_PARSER_BIG_BUFFER_SIZE;
1.14 daniel 1022: buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
1.1 daniel 1023: if (buffer == NULL) {
1024: perror("htmlDecodeEntities: malloc failed");
1025: return(NULL);
1026: }
1027: out = buffer;
1028:
1029: /*
1030: * Ok loop until we reach one of the ending char or a size limit.
1031: */
1.45 daniel 1032: while ((nbchars < (int) max) && (CUR != end) &&
1.1 daniel 1033: (CUR != end2) && (CUR != end3)) {
1034:
1035: if (CUR == '&') {
1036: if (NXT(1) == '#') {
1037: int val = htmlParseCharRef(ctxt);
1.8 daniel 1038: /* invalid for UTF-8 variable encoding !!!!! */
1.1 daniel 1039: *out++ = val;
1.5 daniel 1040: nbchars += 3; /* !!!! */
1.1 daniel 1041: } else {
1042: ent = htmlParseEntityRef(ctxt, &name);
1043: if (name != NULL) {
1044: if ((ent == NULL) || (ent->value <= 0) ||
1045: (ent->value >= 255)) {
1046: *out++ = '&';
1047: cur = name;
1048: while (*cur != 0) {
1049: if (out - buffer > buffer_size - 100) {
1050: int index = out - buffer;
1051:
1052: growBuffer(buffer);
1053: out = &buffer[index];
1054: }
1055: *out++ = *cur++;
1056: }
1057: *out++ = ';';
1058: } else {
1.8 daniel 1059: /* invalid for UTF-8 variable encoding !!!!! */
1.14 daniel 1060: *out++ = (xmlChar)ent->value;
1.1 daniel 1061: if (out - buffer > buffer_size - 100) {
1062: int index = out - buffer;
1063:
1064: growBuffer(buffer);
1065: out = &buffer[index];
1066: }
1067: }
1.5 daniel 1068: nbchars += 2 + xmlStrlen(name);
1.11 daniel 1069: xmlFree(name);
1.1 daniel 1070: }
1071: }
1072: } else {
1.8 daniel 1073: /* invalid for UTF-8 , use COPY(out); !!!!! */
1.1 daniel 1074: *out++ = CUR;
1.5 daniel 1075: nbchars++;
1.1 daniel 1076: if (out - buffer > buffer_size - 100) {
1077: int index = out - buffer;
1078:
1079: growBuffer(buffer);
1080: out = &buffer[index];
1081: }
1082: NEXT;
1083: }
1084: }
1085: *out++ = 0;
1086: return(buffer);
1087: }
1088:
1089:
1090: /************************************************************************
1091: * *
1092: * Commodity functions to handle encodings *
1093: * *
1094: ************************************************************************/
1095:
1096: /**
1097: * htmlSwitchEncoding:
1098: * @ctxt: the parser context
1099: * @len: the len of @cur
1100: *
1101: * change the input functions when discovering the character encoding
1102: * of a given entity.
1103: *
1104: */
1105: void
1106: htmlSwitchEncoding(htmlParserCtxtPtr ctxt, xmlCharEncoding enc)
1107: {
1108: switch (enc) {
1109: case XML_CHAR_ENCODING_ERROR:
1110: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1111: ctxt->sax->error(ctxt->userData, "encoding unknown\n");
1112: ctxt->wellFormed = 0;
1113: break;
1114: case XML_CHAR_ENCODING_NONE:
1115: /* let's assume it's UTF-8 without the XML decl */
1116: return;
1117: case XML_CHAR_ENCODING_UTF8:
1118: /* default encoding, no conversion should be needed */
1119: return;
1120: case XML_CHAR_ENCODING_UTF16LE:
1121: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1122: ctxt->sax->error(ctxt->userData,
1123: "char encoding UTF16 little endian not supported\n");
1124: break;
1125: case XML_CHAR_ENCODING_UTF16BE:
1126: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1127: ctxt->sax->error(ctxt->userData,
1128: "char encoding UTF16 big endian not supported\n");
1129: break;
1130: case XML_CHAR_ENCODING_UCS4LE:
1131: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1132: ctxt->sax->error(ctxt->userData,
1133: "char encoding USC4 little endian not supported\n");
1134: break;
1135: case XML_CHAR_ENCODING_UCS4BE:
1136: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1137: ctxt->sax->error(ctxt->userData,
1138: "char encoding USC4 big endian not supported\n");
1139: break;
1140: case XML_CHAR_ENCODING_EBCDIC:
1141: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1142: ctxt->sax->error(ctxt->userData,
1143: "char encoding EBCDIC not supported\n");
1144: break;
1145: case XML_CHAR_ENCODING_UCS4_2143:
1146: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1147: ctxt->sax->error(ctxt->userData,
1148: "char encoding UCS4 2143 not supported\n");
1149: break;
1150: case XML_CHAR_ENCODING_UCS4_3412:
1151: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1152: ctxt->sax->error(ctxt->userData,
1153: "char encoding UCS4 3412 not supported\n");
1154: break;
1155: case XML_CHAR_ENCODING_UCS2:
1156: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1157: ctxt->sax->error(ctxt->userData,
1158: "char encoding UCS2 not supported\n");
1159: break;
1160: case XML_CHAR_ENCODING_8859_1:
1161: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1162: ctxt->sax->error(ctxt->userData,
1163: "char encoding ISO_8859_1 ISO Latin 1 not supported\n");
1164: break;
1165: case XML_CHAR_ENCODING_8859_2:
1166: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1167: ctxt->sax->error(ctxt->userData,
1168: "char encoding ISO_8859_2 ISO Latin 2 not supported\n");
1169: break;
1170: case XML_CHAR_ENCODING_8859_3:
1171: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1172: ctxt->sax->error(ctxt->userData,
1173: "char encoding ISO_8859_3 not supported\n");
1174: break;
1175: case XML_CHAR_ENCODING_8859_4:
1176: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1177: ctxt->sax->error(ctxt->userData,
1178: "char encoding ISO_8859_4 not supported\n");
1179: break;
1180: case XML_CHAR_ENCODING_8859_5:
1181: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1182: ctxt->sax->error(ctxt->userData,
1183: "char encoding ISO_8859_5 not supported\n");
1184: break;
1185: case XML_CHAR_ENCODING_8859_6:
1186: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1187: ctxt->sax->error(ctxt->userData,
1188: "char encoding ISO_8859_6 not supported\n");
1189: break;
1190: case XML_CHAR_ENCODING_8859_7:
1191: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1192: ctxt->sax->error(ctxt->userData,
1193: "char encoding ISO_8859_7 not supported\n");
1194: break;
1195: case XML_CHAR_ENCODING_8859_8:
1196: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1197: ctxt->sax->error(ctxt->userData,
1198: "char encoding ISO_8859_8 not supported\n");
1199: break;
1200: case XML_CHAR_ENCODING_8859_9:
1201: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1202: ctxt->sax->error(ctxt->userData,
1203: "char encoding ISO_8859_9 not supported\n");
1204: break;
1205: case XML_CHAR_ENCODING_2022_JP:
1206: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1207: ctxt->sax->error(ctxt->userData,
1208: "char encoding ISO-2022-JPnot supported\n");
1209: break;
1210: case XML_CHAR_ENCODING_SHIFT_JIS:
1211: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1212: ctxt->sax->error(ctxt->userData,
1213: "char encoding Shift_JISnot supported\n");
1214: break;
1215: case XML_CHAR_ENCODING_EUC_JP:
1216: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1217: ctxt->sax->error(ctxt->userData,
1218: "char encoding EUC-JPnot supported\n");
1219: break;
1220: }
1221: }
1222:
1.31 daniel 1223: /************************************************************************
1224: * *
1225: * Commodity functions to handle streams *
1226: * *
1227: ************************************************************************/
1228:
1229: /**
1230: * htmlFreeInputStream:
1231: * @input: an htmlParserInputPtr
1232: *
1233: * Free up an input stream.
1234: */
1235: void
1236: htmlFreeInputStream(htmlParserInputPtr input) {
1237: if (input == NULL) return;
1238:
1239: if (input->filename != NULL) xmlFree((char *) input->filename);
1240: if (input->directory != NULL) xmlFree((char *) input->directory);
1241: if ((input->free != NULL) && (input->base != NULL))
1242: input->free((xmlChar *) input->base);
1243: if (input->buf != NULL)
1244: xmlFreeParserInputBuffer(input->buf);
1245: memset(input, -1, sizeof(htmlParserInput));
1246: xmlFree(input);
1247: }
1248:
1249: /**
1250: * htmlNewInputStream:
1251: * @ctxt: an HTML parser context
1252: *
1253: * Create a new input stream structure
1254: * Returns the new input stream or NULL
1255: */
1256: htmlParserInputPtr
1257: htmlNewInputStream(htmlParserCtxtPtr ctxt) {
1258: htmlParserInputPtr input;
1259:
1260: input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1261: if (input == NULL) {
1262: ctxt->errNo = XML_ERR_NO_MEMORY;
1263: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1264: ctxt->sax->error(ctxt->userData,
1265: "malloc: couldn't allocate a new input stream\n");
1266: ctxt->errNo = XML_ERR_NO_MEMORY;
1267: return(NULL);
1268: }
1269: input->filename = NULL;
1270: input->directory = NULL;
1271: input->base = NULL;
1272: input->cur = NULL;
1273: input->buf = NULL;
1274: input->line = 1;
1275: input->col = 1;
1276: input->buf = NULL;
1277: input->free = NULL;
1278: input->consumed = 0;
1279: input->length = 0;
1280: return(input);
1281: }
1282:
1.1 daniel 1283:
1284: /************************************************************************
1285: * *
1286: * Commodity functions, cleanup needed ? *
1287: * *
1288: ************************************************************************/
1289:
1290: /**
1291: * areBlanks:
1292: * @ctxt: an HTML parser context
1.14 daniel 1293: * @str: a xmlChar *
1.1 daniel 1294: * @len: the size of @str
1295: *
1296: * Is this a sequence of blank chars that one can ignore ?
1297: *
1298: * Returns 1 if ignorable 0 otherwise.
1299: */
1300:
1.14 daniel 1301: static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
1.1 daniel 1302: int i;
1303: xmlNodePtr lastChild;
1304:
1305: for (i = 0;i < len;i++)
1306: if (!(IS_BLANK(str[i]))) return(0);
1307:
1308: if (CUR != '<') return(0);
1309: if (ctxt->node == NULL) return(0);
1310: lastChild = xmlGetLastChild(ctxt->node);
1311: if (lastChild == NULL) {
1312: if (ctxt->node->content != NULL) return(0);
1313: } else if (xmlNodeIsText(lastChild))
1314: return(0);
1315: return(1);
1316: }
1317:
1318: /**
1319: * htmlHandleEntity:
1320: * @ctxt: an HTML parser context
1321: * @entity: an XML entity pointer.
1322: *
1323: * Default handling of an HTML entity, call the parser with the
1324: * substitution string
1325: */
1326:
1327: void
1328: htmlHandleEntity(htmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1329: int len;
1330:
1331: if (entity->content == NULL) {
1332: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1333: ctxt->sax->error(ctxt->userData, "htmlHandleEntity %s: content == NULL\n",
1334: entity->name);
1335: ctxt->wellFormed = 0;
1336: return;
1337: }
1338: len = xmlStrlen(entity->content);
1339:
1340: /*
1341: * Just handle the content as a set of chars.
1342: */
1343: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1344: ctxt->sax->characters(ctxt->userData, entity->content, len);
1345:
1346: }
1347:
1348: /**
1349: * htmlNewDoc:
1350: * @URI: URI for the dtd, or NULL
1351: * @ExternalID: the external ID of the DTD, or NULL
1352: *
1353: * Returns a new document
1354: */
1355: htmlDocPtr
1.14 daniel 1356: htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
1.1 daniel 1357: xmlDocPtr cur;
1358:
1359: /*
1360: * Allocate a new document and fill the fields.
1361: */
1.11 daniel 1362: cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
1.1 daniel 1363: if (cur == NULL) {
1364: fprintf(stderr, "xmlNewDoc : malloc failed\n");
1365: return(NULL);
1366: }
1.10 daniel 1367: memset(cur, 0, sizeof(xmlDoc));
1.1 daniel 1368:
1.20 daniel 1369: cur->type = XML_HTML_DOCUMENT_NODE;
1.1 daniel 1370: cur->version = NULL;
1371: cur->intSubset = NULL;
1.28 daniel 1372: if ((ExternalID == NULL) &&
1373: (URI == NULL))
1374: xmlCreateIntSubset(cur, BAD_CAST "HTML",
1375: BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
1376: BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
1377: else
1378: xmlCreateIntSubset(cur, BAD_CAST "HTML", ExternalID, URI);
1.41 daniel 1379: cur->doc = cur;
1.1 daniel 1380: cur->name = NULL;
1.37 daniel 1381: cur->children = NULL;
1.1 daniel 1382: cur->extSubset = NULL;
1383: cur->oldNs = NULL;
1384: cur->encoding = NULL;
1385: cur->standalone = 1;
1386: cur->compression = 0;
1.12 daniel 1387: cur->ids = NULL;
1388: cur->refs = NULL;
1.1 daniel 1389: #ifndef XML_WITHOUT_CORBA
1390: cur->_private = NULL;
1391: #endif
1392: return(cur);
1393: }
1394:
1395:
1396: /************************************************************************
1397: * *
1398: * The parser itself *
1399: * Relates to http://www.w3.org/TR/html40 *
1400: * *
1401: ************************************************************************/
1402:
1403: /************************************************************************
1404: * *
1405: * The parser itself *
1406: * *
1407: ************************************************************************/
1408:
1409: /**
1410: * htmlParseHTMLName:
1411: * @ctxt: an HTML parser context
1412: *
1.26 daniel 1413: * parse an HTML tag or attribute name, note that we convert it to lowercase
1.1 daniel 1414: * since HTML names are not case-sensitive.
1415: *
1416: * Returns the Tag Name parsed or NULL
1417: */
1418:
1.14 daniel 1419: xmlChar *
1.1 daniel 1420: htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1421: xmlChar *ret = NULL;
1.1 daniel 1422: int i = 0;
1.31 daniel 1423: xmlChar loc[HTML_PARSER_BUFFER_SIZE];
1.1 daniel 1424:
1425: if (!IS_LETTER(CUR) && (CUR != '_') &&
1426: (CUR != ':')) return(NULL);
1427:
1.31 daniel 1428: while ((i < HTML_PARSER_BUFFER_SIZE) &&
1.45 daniel 1429: ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1430: (CUR == ':') || (CUR == '_'))) {
1.26 daniel 1431: if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20;
1.1 daniel 1432: else loc[i] = CUR;
1433: i++;
1434:
1435: NEXT;
1436: }
1437:
1438: ret = xmlStrndup(loc, i);
1439:
1440: return(ret);
1441: }
1442:
1443: /**
1444: * htmlParseName:
1445: * @ctxt: an HTML parser context
1446: *
1447: * parse an HTML name, this routine is case sensistive.
1448: *
1449: * Returns the Name parsed or NULL
1450: */
1451:
1.14 daniel 1452: xmlChar *
1.1 daniel 1453: htmlParseName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1454: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1455: int len = 0;
1.1 daniel 1456:
1.5 daniel 1457: GROW;
1458: if (!IS_LETTER(CUR) && (CUR != '_')) {
1459: return(NULL);
1460: }
1.1 daniel 1461:
1462: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1463: (CUR == '.') || (CUR == '-') ||
1464: (CUR == '_') || (CUR == ':') ||
1465: (IS_COMBINING(CUR)) ||
1.5 daniel 1466: (IS_EXTENDER(CUR))) {
1467: buf[len++] = CUR;
1.1 daniel 1468: NEXT;
1.5 daniel 1469: if (len >= HTML_MAX_NAMELEN) {
1470: fprintf(stderr,
1471: "htmlParseName: reached HTML_MAX_NAMELEN limit\n");
1472: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1473: (CUR == '.') || (CUR == '-') ||
1474: (CUR == '_') || (CUR == ':') ||
1475: (IS_COMBINING(CUR)) ||
1476: (IS_EXTENDER(CUR)))
1477: NEXT;
1478: break;
1479: }
1480: }
1481: return(xmlStrndup(buf, len));
1.1 daniel 1482: }
1483:
1484: /**
1485: * htmlParseHTMLAttribute:
1486: * @ctxt: an HTML parser context
1.19 daniel 1487: * @stop: a char stop value
1.1 daniel 1488: *
1.19 daniel 1489: * parse an HTML attribute value till the stop (quote), if
1490: * stop is 0 then it stops at the first space
1.1 daniel 1491: *
1.19 daniel 1492: * Returns the attribute parsed or NULL
1.1 daniel 1493: */
1494:
1.14 daniel 1495: xmlChar *
1.19 daniel 1496: htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) {
1.32 daniel 1497: #if 0
1.14 daniel 1498: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1499: int len = 0;
1.1 daniel 1500:
1.5 daniel 1501: GROW;
1.19 daniel 1502: while ((CUR != 0) && (CUR != stop) && (CUR != '>')) {
1503: if ((stop == 0) && (IS_BLANK(CUR))) break;
1.5 daniel 1504: buf[len++] = CUR;
1.1 daniel 1505: NEXT;
1.5 daniel 1506: if (len >= HTML_MAX_NAMELEN) {
1507: fprintf(stderr,
1508: "htmlParseHTMLAttribute: reached HTML_MAX_NAMELEN limit\n");
1509: while ((!IS_BLANK(CUR)) && (CUR != '<') &&
1.19 daniel 1510: (CUR != '>') &&
1.5 daniel 1511: (CUR != '\'') && (CUR != '"'))
1512: NEXT;
1513: break;
1514: }
1515: }
1516: return(xmlStrndup(buf, len));
1.32 daniel 1517: #else
1518: xmlChar *buffer = NULL;
1519: int buffer_size = 0;
1520: xmlChar *out = NULL;
1521: xmlChar *name = NULL;
1522:
1523: xmlChar *cur = NULL;
1524: htmlEntityDescPtr ent;
1525:
1526: /*
1527: * allocate a translation buffer.
1528: */
1529: buffer_size = HTML_PARSER_BIG_BUFFER_SIZE;
1530: buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
1531: if (buffer == NULL) {
1532: perror("htmlParseHTMLAttribute: malloc failed");
1533: return(NULL);
1534: }
1535: out = buffer;
1536:
1537: /*
1538: * Ok loop until we reach one of the ending chars
1539: */
1540: while ((CUR != 0) && (CUR != stop) && (CUR != '>')) {
1541: if ((stop == 0) && (IS_BLANK(CUR))) break;
1542: if (CUR == '&') {
1543: if (NXT(1) == '#') {
1544: int val = htmlParseCharRef(ctxt);
1545: *out++ = val;
1546: } else {
1547: ent = htmlParseEntityRef(ctxt, &name);
1548: if (name == NULL) {
1549: *out++ = '&';
1550: if (out - buffer > buffer_size - 100) {
1551: int index = out - buffer;
1552:
1553: growBuffer(buffer);
1554: out = &buffer[index];
1555: }
1556: } else if ((ent == NULL) || (ent->value <= 0) ||
1557: (ent->value >= 255)) {
1558: *out++ = '&';
1559: cur = name;
1560: while (*cur != 0) {
1561: if (out - buffer > buffer_size - 100) {
1562: int index = out - buffer;
1563:
1564: growBuffer(buffer);
1565: out = &buffer[index];
1566: }
1567: *out++ = *cur++;
1568: }
1569: xmlFree(name);
1570: } else {
1571: *out++ = ent->value;
1572: if (out - buffer > buffer_size - 100) {
1573: int index = out - buffer;
1574:
1575: growBuffer(buffer);
1576: out = &buffer[index];
1577: }
1578: xmlFree(name);
1579: }
1580: }
1581: } else {
1582: *out++ = CUR;
1583: if (out - buffer > buffer_size - 100) {
1584: int index = out - buffer;
1585:
1586: growBuffer(buffer);
1587: out = &buffer[index];
1588: }
1589: NEXT;
1590: }
1591: }
1592: *out++ = 0;
1593: return(buffer);
1594: #endif
1.1 daniel 1595: }
1596:
1597: /**
1598: * htmlParseNmtoken:
1599: * @ctxt: an HTML parser context
1600: *
1601: * parse an HTML Nmtoken.
1602: *
1603: * Returns the Nmtoken parsed or NULL
1604: */
1605:
1.14 daniel 1606: xmlChar *
1.1 daniel 1607: htmlParseNmtoken(htmlParserCtxtPtr ctxt) {
1.14 daniel 1608: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1609: int len = 0;
1.1 daniel 1610:
1.5 daniel 1611: GROW;
1.1 daniel 1612: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1613: (CUR == '.') || (CUR == '-') ||
1614: (CUR == '_') || (CUR == ':') ||
1615: (IS_COMBINING(CUR)) ||
1.5 daniel 1616: (IS_EXTENDER(CUR))) {
1617: buf[len++] = CUR;
1.1 daniel 1618: NEXT;
1.5 daniel 1619: if (len >= HTML_MAX_NAMELEN) {
1620: fprintf(stderr,
1621: "htmlParseNmtoken: reached HTML_MAX_NAMELEN limit\n");
1622: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1623: (CUR == '.') || (CUR == '-') ||
1624: (CUR == '_') || (CUR == ':') ||
1625: (IS_COMBINING(CUR)) ||
1626: (IS_EXTENDER(CUR)))
1627: NEXT;
1628: break;
1629: }
1630: }
1631: return(xmlStrndup(buf, len));
1.1 daniel 1632: }
1633:
1634: /**
1635: * htmlParseEntityRef:
1636: * @ctxt: an HTML parser context
1637: * @str: location to store the entity name
1638: *
1639: * parse an HTML ENTITY references
1640: *
1641: * [68] EntityRef ::= '&' Name ';'
1642: *
1643: * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
1644: * if non-NULL *str will have to be freed by the caller.
1645: */
1646: htmlEntityDescPtr
1.14 daniel 1647: htmlParseEntityRef(htmlParserCtxtPtr ctxt, xmlChar **str) {
1648: xmlChar *name;
1.1 daniel 1649: htmlEntityDescPtr ent = NULL;
1650: *str = NULL;
1651:
1652: if (CUR == '&') {
1653: NEXT;
1654: name = htmlParseName(ctxt);
1655: if (name == NULL) {
1656: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1657: ctxt->sax->error(ctxt->userData, "htmlParseEntityRef: no name\n");
1658: ctxt->wellFormed = 0;
1659: } else {
1.5 daniel 1660: GROW;
1.1 daniel 1661: if (CUR == ';') {
1662: *str = name;
1663:
1664: /*
1665: * Lookup the entity in the table.
1666: */
1667: ent = htmlEntityLookup(name);
1.32 daniel 1668: if (ent != NULL) /* OK that's ugly !!! */
1669: NEXT;
1.1 daniel 1670: } else {
1671: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1672: ctxt->sax->error(ctxt->userData,
1673: "htmlParseEntityRef: expecting ';'\n");
1.32 daniel 1674: *str = name;
1.1 daniel 1675: }
1676: }
1677: }
1678: return(ent);
1679: }
1680:
1681: /**
1682: * htmlParseAttValue:
1683: * @ctxt: an HTML parser context
1684: *
1685: * parse a value for an attribute
1686: * Note: the parser won't do substitution of entities here, this
1687: * will be handled later in xmlStringGetNodeList, unless it was
1688: * asked for ctxt->replaceEntities != 0
1689: *
1690: * Returns the AttValue parsed or NULL.
1691: */
1692:
1.14 daniel 1693: xmlChar *
1.1 daniel 1694: htmlParseAttValue(htmlParserCtxtPtr ctxt) {
1.14 daniel 1695: xmlChar *ret = NULL;
1.1 daniel 1696:
1697: if (CUR == '"') {
1698: NEXT;
1.19 daniel 1699: ret = htmlParseHTMLAttribute(ctxt, '"');
1.1 daniel 1700: if (CUR != '"') {
1701: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1702: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
1703: ctxt->wellFormed = 0;
1704: } else
1705: NEXT;
1706: } else if (CUR == '\'') {
1707: NEXT;
1.19 daniel 1708: ret = htmlParseHTMLAttribute(ctxt, '\'');
1.1 daniel 1709: if (CUR != '\'') {
1710: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1711: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
1712: ctxt->wellFormed = 0;
1713: } else
1714: NEXT;
1715: } else {
1716: /*
1717: * That's an HTMLism, the attribute value may not be quoted
1718: */
1.19 daniel 1719: ret = htmlParseHTMLAttribute(ctxt, 0);
1.1 daniel 1720: if (ret == NULL) {
1721: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1722: ctxt->sax->error(ctxt->userData, "AttValue: no value found\n");
1723: ctxt->wellFormed = 0;
1724: }
1725: }
1726: return(ret);
1727: }
1728:
1729: /**
1730: * htmlParseSystemLiteral:
1731: * @ctxt: an HTML parser context
1732: *
1733: * parse an HTML Literal
1734: *
1735: * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
1736: *
1737: * Returns the SystemLiteral parsed or NULL
1738: */
1739:
1.14 daniel 1740: xmlChar *
1.1 daniel 1741: htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 1742: const xmlChar *q;
1743: xmlChar *ret = NULL;
1.1 daniel 1744:
1745: if (CUR == '"') {
1746: NEXT;
1747: q = CUR_PTR;
1748: while ((IS_CHAR(CUR)) && (CUR != '"'))
1749: NEXT;
1750: if (!IS_CHAR(CUR)) {
1751: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1752: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
1753: ctxt->wellFormed = 0;
1754: } else {
1755: ret = xmlStrndup(q, CUR_PTR - q);
1756: NEXT;
1757: }
1758: } else if (CUR == '\'') {
1759: NEXT;
1760: q = CUR_PTR;
1761: while ((IS_CHAR(CUR)) && (CUR != '\''))
1762: NEXT;
1763: if (!IS_CHAR(CUR)) {
1764: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1765: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
1766: ctxt->wellFormed = 0;
1767: } else {
1768: ret = xmlStrndup(q, CUR_PTR - q);
1769: NEXT;
1770: }
1771: } else {
1772: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.38 daniel 1773: ctxt->sax->error(ctxt->userData,
1774: "SystemLiteral \" or ' expected\n");
1.1 daniel 1775: ctxt->wellFormed = 0;
1776: }
1777:
1778: return(ret);
1779: }
1780:
1781: /**
1782: * htmlParsePubidLiteral:
1783: * @ctxt: an HTML parser context
1784: *
1785: * parse an HTML public literal
1786: *
1787: * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
1788: *
1789: * Returns the PubidLiteral parsed or NULL.
1790: */
1791:
1.14 daniel 1792: xmlChar *
1.1 daniel 1793: htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 1794: const xmlChar *q;
1795: xmlChar *ret = NULL;
1.1 daniel 1796: /*
1797: * Name ::= (Letter | '_') (NameChar)*
1798: */
1799: if (CUR == '"') {
1800: NEXT;
1801: q = CUR_PTR;
1802: while (IS_PUBIDCHAR(CUR)) NEXT;
1803: if (CUR != '"') {
1804: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1805: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
1806: ctxt->wellFormed = 0;
1807: } else {
1808: ret = xmlStrndup(q, CUR_PTR - q);
1809: NEXT;
1810: }
1811: } else if (CUR == '\'') {
1812: NEXT;
1813: q = CUR_PTR;
1814: while ((IS_LETTER(CUR)) && (CUR != '\''))
1815: NEXT;
1816: if (!IS_LETTER(CUR)) {
1817: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1818: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
1819: ctxt->wellFormed = 0;
1820: } else {
1821: ret = xmlStrndup(q, CUR_PTR - q);
1822: NEXT;
1823: }
1824: } else {
1825: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1826: ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n");
1827: ctxt->wellFormed = 0;
1828: }
1829:
1830: return(ret);
1831: }
1832:
1833: /**
1834: * htmlParseCharData:
1835: * @ctxt: an HTML parser context
1836: * @cdata: int indicating whether we are within a CDATA section
1837: *
1838: * parse a CharData section.
1839: * if we are within a CDATA section ']]>' marks an end of section.
1840: *
1841: * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
1842: */
1843:
1844: void
1845: htmlParseCharData(htmlParserCtxtPtr ctxt, int cdata) {
1.25 daniel 1846: xmlChar *buf = NULL;
1847: int len = 0;
1.31 daniel 1848: int size = HTML_PARSER_BUFFER_SIZE;
1.25 daniel 1849: xmlChar q;
1850:
1851: buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar));
1852: if (buf == NULL) {
1853: fprintf(stderr, "malloc of %d byte failed\n", size);
1854: return;
1855: }
1.1 daniel 1856:
1.25 daniel 1857: q = CUR;
1858: while ((IS_CHAR(q)) && (q != '<') &&
1859: (q != '&')) {
1860: if ((q == ']') && (NXT(1) == ']') &&
1.1 daniel 1861: (NXT(2) == '>')) {
1862: if (cdata) break;
1863: else {
1864: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1865: ctxt->sax->error(ctxt->userData,
1866: "Sequence ']]>' not allowed in content\n");
1867: ctxt->wellFormed = 0;
1868: }
1869: }
1.25 daniel 1870: if (len + 1 >= size) {
1871: size *= 2;
1872: buf = xmlRealloc(buf, size * sizeof(xmlChar));
1873: if (buf == NULL) {
1874: fprintf(stderr, "realloc of %d byte failed\n", size);
1875: return;
1876: }
1877: }
1878: buf[len++] = q;
1.1 daniel 1879: NEXT;
1.25 daniel 1880: q = CUR;
1881: }
1882: if (len == 0) {
1883: xmlFree(buf);
1884: return;
1.1 daniel 1885: }
1886:
1887: /*
1.25 daniel 1888: * Ok the buffer is to be consumed as chars.
1.1 daniel 1889: */
1890: if (ctxt->sax != NULL) {
1.25 daniel 1891: if (areBlanks(ctxt, buf, len)) {
1.1 daniel 1892: if (ctxt->sax->ignorableWhitespace != NULL)
1.25 daniel 1893: ctxt->sax->ignorableWhitespace(ctxt->userData, buf, len);
1.1 daniel 1894: } else {
1895: if (ctxt->sax->characters != NULL)
1.25 daniel 1896: ctxt->sax->characters(ctxt->userData, buf, len);
1.1 daniel 1897: }
1898: }
1.25 daniel 1899: xmlFree(buf);
1.1 daniel 1900: }
1901:
1902: /**
1903: * htmlParseExternalID:
1904: * @ctxt: an HTML parser context
1.14 daniel 1905: * @publicID: a xmlChar** receiving PubidLiteral
1.1 daniel 1906: * @strict: indicate whether we should restrict parsing to only
1907: * production [75], see NOTE below
1908: *
1909: * Parse an External ID or a Public ID
1910: *
1911: * NOTE: Productions [75] and [83] interract badly since [75] can generate
1912: * 'PUBLIC' S PubidLiteral S SystemLiteral
1913: *
1914: * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
1915: * | 'PUBLIC' S PubidLiteral S SystemLiteral
1916: *
1917: * [83] PublicID ::= 'PUBLIC' S PubidLiteral
1918: *
1919: * Returns the function returns SystemLiteral and in the second
1920: * case publicID receives PubidLiteral, is strict is off
1921: * it is possible to return NULL and have publicID set.
1922: */
1923:
1.14 daniel 1924: xmlChar *
1925: htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID, int strict) {
1926: xmlChar *URI = NULL;
1.1 daniel 1927:
1928: if ((UPPER == 'S') && (UPP(1) == 'Y') &&
1929: (UPP(2) == 'S') && (UPP(3) == 'T') &&
1930: (UPP(4) == 'E') && (UPP(5) == 'M')) {
1931: SKIP(6);
1932: if (!IS_BLANK(CUR)) {
1933: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1934: ctxt->sax->error(ctxt->userData,
1935: "Space required after 'SYSTEM'\n");
1936: ctxt->wellFormed = 0;
1937: }
1938: SKIP_BLANKS;
1939: URI = htmlParseSystemLiteral(ctxt);
1940: if (URI == NULL) {
1941: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1942: ctxt->sax->error(ctxt->userData,
1943: "htmlParseExternalID: SYSTEM, no URI\n");
1944: ctxt->wellFormed = 0;
1945: }
1946: } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
1947: (UPP(2) == 'B') && (UPP(3) == 'L') &&
1948: (UPP(4) == 'I') && (UPP(5) == 'C')) {
1949: SKIP(6);
1950: if (!IS_BLANK(CUR)) {
1951: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1952: ctxt->sax->error(ctxt->userData,
1953: "Space required after 'PUBLIC'\n");
1954: ctxt->wellFormed = 0;
1955: }
1956: SKIP_BLANKS;
1957: *publicID = htmlParsePubidLiteral(ctxt);
1958: if (*publicID == NULL) {
1959: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1960: ctxt->sax->error(ctxt->userData,
1961: "htmlParseExternalID: PUBLIC, no Public Identifier\n");
1962: ctxt->wellFormed = 0;
1963: }
1.5 daniel 1964: SKIP_BLANKS;
1965: if ((CUR == '"') || (CUR == '\'')) {
1966: URI = htmlParseSystemLiteral(ctxt);
1.1 daniel 1967: }
1968: }
1969: return(URI);
1970: }
1971:
1972: /**
1973: * htmlParseComment:
1974: * @ctxt: an HTML parser context
1975: *
1976: * Parse an XML (SGML) comment <!-- .... -->
1977: *
1978: * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
1979: */
1980: void
1.31 daniel 1981: htmlParseComment(htmlParserCtxtPtr ctxt) {
1.25 daniel 1982: xmlChar *buf = NULL;
1983: int len = 0;
1.31 daniel 1984: int size = HTML_PARSER_BUFFER_SIZE;
1.25 daniel 1985: register xmlChar s, r, q;
1.1 daniel 1986:
1987: /*
1988: * Check that there is a comment right here.
1989: */
1990: if ((CUR != '<') || (NXT(1) != '!') ||
1991: (NXT(2) != '-') || (NXT(3) != '-')) return;
1992:
1.25 daniel 1993: buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar));
1994: if (buf == NULL) {
1995: fprintf(stderr, "malloc of %d byte failed\n", size);
1996: return;
1997: }
1998: q = r = '-'; /* 0 or '-' to cover our ass against <!--> and <!---> ? !!! */
1.1 daniel 1999: SKIP(4);
1.25 daniel 2000: s = CUR;
2001:
2002: while (IS_CHAR(s) &&
2003: ((s != '>') || (r != '-') || (q != '-'))) {
2004: if (len + 1 >= size) {
2005: size *= 2;
2006: buf = xmlRealloc(buf, size * sizeof(xmlChar));
2007: if (buf == NULL) {
2008: fprintf(stderr, "realloc of %d byte failed\n", size);
2009: return;
2010: }
2011: }
2012: buf[len++] = s;
2013: NEXT;
2014: q = r;
2015: r = s;
2016: s = CUR;
1.1 daniel 2017: }
1.25 daniel 2018: buf[len - 2] = 0;
2019: if (!IS_CHAR(s)) {
1.1 daniel 2020: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.25 daniel 2021: ctxt->sax->error(ctxt->userData, "Comment not terminated \n<!--%.50s\n", buf);
1.1 daniel 2022: ctxt->wellFormed = 0;
2023: } else {
2024: NEXT;
1.31 daniel 2025: if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL)) {
2026: ctxt->sax->comment(ctxt->userData, buf);
1.1 daniel 2027: }
2028: }
1.25 daniel 2029: xmlFree(buf);
1.1 daniel 2030: }
2031:
2032: /**
2033: * htmlParseCharRef:
2034: * @ctxt: an HTML parser context
2035: *
2036: * parse Reference declarations
2037: *
2038: * [66] CharRef ::= '&#' [0-9]+ ';' |
2039: * '&#x' [0-9a-fA-F]+ ';'
2040: *
2041: * Returns the value parsed (as an int)
2042: */
2043: int
2044: htmlParseCharRef(htmlParserCtxtPtr ctxt) {
2045: int val = 0;
2046:
2047: if ((CUR == '&') && (NXT(1) == '#') &&
2048: (NXT(2) == 'x')) {
2049: SKIP(3);
2050: while (CUR != ';') {
2051: if ((CUR >= '0') && (CUR <= '9'))
2052: val = val * 16 + (CUR - '0');
2053: else if ((CUR >= 'a') && (CUR <= 'f'))
2054: val = val * 16 + (CUR - 'a') + 10;
2055: else if ((CUR >= 'A') && (CUR <= 'F'))
2056: val = val * 16 + (CUR - 'A') + 10;
2057: else {
2058: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2059: ctxt->sax->error(ctxt->userData,
2060: "htmlParseCharRef: invalid hexadecimal value\n");
2061: ctxt->wellFormed = 0;
2062: val = 0;
2063: break;
2064: }
2065: NEXT;
2066: }
2067: if (CUR == ';')
2068: NEXT;
2069: } else if ((CUR == '&') && (NXT(1) == '#')) {
2070: SKIP(2);
2071: while (CUR != ';') {
2072: if ((CUR >= '0') && (CUR <= '9'))
2073: val = val * 10 + (CUR - '0');
2074: else {
2075: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2076: ctxt->sax->error(ctxt->userData,
2077: "htmlParseCharRef: invalid decimal value\n");
2078: ctxt->wellFormed = 0;
2079: val = 0;
2080: break;
2081: }
2082: NEXT;
2083: }
2084: if (CUR == ';')
2085: NEXT;
2086: } else {
2087: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2088: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid value\n");
2089: ctxt->wellFormed = 0;
2090: }
2091: /*
2092: * Check the value IS_CHAR ...
2093: */
2094: if (IS_CHAR(val)) {
2095: return(val);
2096: } else {
2097: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.14 daniel 2098: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid xmlChar value %d\n",
1.1 daniel 2099: val);
2100: ctxt->wellFormed = 0;
2101: }
2102: return(0);
2103: }
2104:
2105:
2106: /**
2107: * htmlParseDocTypeDecl :
2108: * @ctxt: an HTML parser context
2109: *
2110: * parse a DOCTYPE declaration
2111: *
2112: * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
2113: * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
2114: */
2115:
2116: void
2117: htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
1.14 daniel 2118: xmlChar *name;
2119: xmlChar *ExternalID = NULL;
2120: xmlChar *URI = NULL;
1.1 daniel 2121:
2122: /*
2123: * We know that '<!DOCTYPE' has been detected.
2124: */
2125: SKIP(9);
2126:
2127: SKIP_BLANKS;
2128:
2129: /*
2130: * Parse the DOCTYPE name.
2131: */
2132: name = htmlParseName(ctxt);
2133: if (name == NULL) {
2134: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2135: ctxt->sax->error(ctxt->userData, "htmlParseDocTypeDecl : no DOCTYPE name !\n");
2136: ctxt->wellFormed = 0;
2137: }
2138: /*
2139: * Check that upper(name) == "HTML" !!!!!!!!!!!!!
2140: */
2141:
2142: SKIP_BLANKS;
2143:
2144: /*
2145: * Check for SystemID and ExternalID
2146: */
1.5 daniel 2147: URI = htmlParseExternalID(ctxt, &ExternalID, 0);
1.1 daniel 2148: SKIP_BLANKS;
2149:
2150: /*
2151: * We should be at the end of the DOCTYPE declaration.
2152: */
2153: if (CUR != '>') {
2154: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2155: ctxt->sax->error(ctxt->userData, "DOCTYPE unproperly terminated\n");
2156: ctxt->wellFormed = 0;
2157: /* We shouldn't try to resynchronize ... */
2158: }
2159: NEXT;
2160:
2161: /*
1.46 ! daniel 2162: * Create or update the document accordingly to the DOCTYPE
1.1 daniel 2163: */
1.46 ! daniel 2164: if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) &&
! 2165: (!ctxt->disableSAX))
! 2166: ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI);
1.1 daniel 2167:
2168: /*
2169: * Cleanup, since we don't use all those identifiers
2170: */
1.11 daniel 2171: if (URI != NULL) xmlFree(URI);
2172: if (ExternalID != NULL) xmlFree(ExternalID);
2173: if (name != NULL) xmlFree(name);
1.1 daniel 2174: }
2175:
2176: /**
2177: * htmlParseAttribute:
2178: * @ctxt: an HTML parser context
1.14 daniel 2179: * @value: a xmlChar ** used to store the value of the attribute
1.1 daniel 2180: *
2181: * parse an attribute
2182: *
2183: * [41] Attribute ::= Name Eq AttValue
2184: *
2185: * [25] Eq ::= S? '=' S?
2186: *
2187: * With namespace:
2188: *
2189: * [NS 11] Attribute ::= QName Eq AttValue
2190: *
2191: * Also the case QName == xmlns:??? is handled independently as a namespace
2192: * definition.
2193: *
2194: * Returns the attribute name, and the value in *value.
2195: */
2196:
1.14 daniel 2197: xmlChar *
2198: htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
1.31 daniel 2199: xmlChar *name, *val = NULL;
1.1 daniel 2200:
2201: *value = NULL;
2202: name = htmlParseName(ctxt);
2203: if (name == NULL) {
2204: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2205: ctxt->sax->error(ctxt->userData, "error parsing attribute name\n");
2206: ctxt->wellFormed = 0;
2207: return(NULL);
2208: }
2209:
2210: /*
2211: * read the value
2212: */
2213: SKIP_BLANKS;
2214: if (CUR == '=') {
2215: NEXT;
2216: SKIP_BLANKS;
2217: val = htmlParseAttValue(ctxt);
1.42 daniel 2218: /******
1.1 daniel 2219: } else {
1.42 daniel 2220: * TODO : some attribute must have values, some may not
1.1 daniel 2221: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.31 daniel 2222: ctxt->sax->warning(ctxt->userData,
1.42 daniel 2223: "No value for attribute %s\n", name); */
1.1 daniel 2224: }
2225:
2226: *value = val;
2227: return(name);
2228: }
2229:
2230: /**
2231: * htmlParseStartTag:
2232: * @ctxt: an HTML parser context
2233: *
2234: * parse a start of tag either for rule element or
2235: * EmptyElement. In both case we don't parse the tag closing chars.
2236: *
2237: * [40] STag ::= '<' Name (S Attribute)* S? '>'
2238: *
2239: * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
2240: *
2241: * With namespace:
2242: *
2243: * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
2244: *
2245: * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
2246: *
2247: */
2248:
1.18 daniel 2249: void
1.1 daniel 2250: htmlParseStartTag(htmlParserCtxtPtr ctxt) {
1.14 daniel 2251: xmlChar *name;
2252: xmlChar *attname;
2253: xmlChar *attvalue;
2254: const xmlChar **atts = NULL;
1.1 daniel 2255: int nbatts = 0;
2256: int maxatts = 0;
2257: int i;
2258:
1.18 daniel 2259: if (CUR != '<') return;
1.1 daniel 2260: NEXT;
2261:
1.19 daniel 2262: GROW;
1.1 daniel 2263: name = htmlParseHTMLName(ctxt);
2264: if (name == NULL) {
2265: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2266: ctxt->sax->error(ctxt->userData,
2267: "htmlParseStartTag: invalid element name\n");
2268: ctxt->wellFormed = 0;
1.18 daniel 2269: return;
1.1 daniel 2270: }
2271:
2272: /*
2273: * Check for auto-closure of HTML elements.
2274: */
2275: htmlAutoClose(ctxt, name);
1.43 daniel 2276:
2277: /*
2278: * Check for implied HTML elements.
2279: */
2280: htmlCheckImplied(ctxt, name);
1.1 daniel 2281:
2282: /*
2283: * Now parse the attributes, it ends up with the ending
2284: *
2285: * (S Attribute)* S?
2286: */
2287: SKIP_BLANKS;
2288: while ((IS_CHAR(CUR)) &&
2289: (CUR != '>') &&
2290: ((CUR != '/') || (NXT(1) != '>'))) {
1.26 daniel 2291: long cons = ctxt->nbChars;
1.1 daniel 2292:
1.19 daniel 2293: GROW;
1.1 daniel 2294: attname = htmlParseAttribute(ctxt, &attvalue);
1.31 daniel 2295: if (attname != NULL) {
1.1 daniel 2296: /*
2297: * Well formedness requires at most one declaration of an attribute
2298: */
2299: for (i = 0; i < nbatts;i += 2) {
2300: if (!xmlStrcmp(atts[i], attname)) {
2301: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.19 daniel 2302: ctxt->sax->error(ctxt->userData,
2303: "Attribute %s redefined\n",
2304: attname);
1.1 daniel 2305: ctxt->wellFormed = 0;
1.11 daniel 2306: xmlFree(attname);
1.31 daniel 2307: if (attvalue != NULL)
2308: xmlFree(attvalue);
1.19 daniel 2309: goto failed;
1.1 daniel 2310: }
2311: }
2312:
2313: /*
2314: * Add the pair to atts
2315: */
2316: if (atts == NULL) {
2317: maxatts = 10;
1.14 daniel 2318: atts = (const xmlChar **) xmlMalloc(maxatts * sizeof(xmlChar *));
1.1 daniel 2319: if (atts == NULL) {
2320: fprintf(stderr, "malloc of %ld byte failed\n",
1.14 daniel 2321: maxatts * (long)sizeof(xmlChar *));
1.18 daniel 2322: if (name != NULL) xmlFree(name);
2323: return;
1.1 daniel 2324: }
1.23 daniel 2325: } else if (nbatts + 4 > maxatts) {
1.1 daniel 2326: maxatts *= 2;
1.14 daniel 2327: atts = (const xmlChar **) xmlRealloc(atts, maxatts * sizeof(xmlChar *));
1.1 daniel 2328: if (atts == NULL) {
2329: fprintf(stderr, "realloc of %ld byte failed\n",
1.14 daniel 2330: maxatts * (long)sizeof(xmlChar *));
1.18 daniel 2331: if (name != NULL) xmlFree(name);
2332: return;
1.1 daniel 2333: }
2334: }
2335: atts[nbatts++] = attname;
2336: atts[nbatts++] = attvalue;
2337: atts[nbatts] = NULL;
2338: atts[nbatts + 1] = NULL;
2339: }
2340:
1.19 daniel 2341: failed:
1.1 daniel 2342: SKIP_BLANKS;
1.26 daniel 2343: if (cons == ctxt->nbChars) {
1.1 daniel 2344: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2345: ctxt->sax->error(ctxt->userData,
2346: "htmlParseStartTag: problem parsing attributes\n");
2347: ctxt->wellFormed = 0;
2348: break;
2349: }
2350: }
2351:
2352: /*
2353: * SAX: Start of Element !
2354: */
1.15 daniel 2355: htmlnamePush(ctxt, xmlStrdup(name));
1.18 daniel 2356: #ifdef DEBUG
2357: fprintf(stderr,"Start of element %s: pushed %s\n", name, ctxt->name);
2358: #endif
1.1 daniel 2359: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
2360: ctxt->sax->startElement(ctxt->userData, name, atts);
2361:
2362: if (atts != NULL) {
1.31 daniel 2363: for (i = 0;i < nbatts;i++) {
2364: if (atts[i] != NULL)
2365: xmlFree((xmlChar *) atts[i]);
2366: }
1.45 daniel 2367: xmlFree((void *) atts);
1.1 daniel 2368: }
1.18 daniel 2369: if (name != NULL) xmlFree(name);
1.1 daniel 2370: }
2371:
2372: /**
2373: * htmlParseEndTag:
2374: * @ctxt: an HTML parser context
2375: *
2376: * parse an end of tag
2377: *
2378: * [42] ETag ::= '</' Name S? '>'
2379: *
2380: * With namespace
2381: *
2382: * [NS 9] ETag ::= '</' QName S? '>'
2383: */
2384:
2385: void
1.18 daniel 2386: htmlParseEndTag(htmlParserCtxtPtr ctxt) {
1.14 daniel 2387: xmlChar *name;
1.15 daniel 2388: xmlChar *oldname;
1.1 daniel 2389: int i;
2390:
2391: if ((CUR != '<') || (NXT(1) != '/')) {
2392: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2393: ctxt->sax->error(ctxt->userData, "htmlParseEndTag: '</' not found\n");
2394: ctxt->wellFormed = 0;
2395: return;
2396: }
2397: SKIP(2);
2398:
2399: name = htmlParseHTMLName(ctxt);
1.24 daniel 2400: if (name == NULL) return;
1.1 daniel 2401:
2402: /*
2403: * We should definitely be at the ending "S? '>'" part
2404: */
2405: SKIP_BLANKS;
2406: if ((!IS_CHAR(CUR)) || (CUR != '>')) {
2407: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2408: ctxt->sax->error(ctxt->userData, "End tag : expected '>'\n");
2409: ctxt->wellFormed = 0;
2410: } else
2411: NEXT;
2412:
2413: /*
1.18 daniel 2414: * If the name read is not one of the element in the parsing stack
2415: * then return, it's just an error.
1.1 daniel 2416: */
1.18 daniel 2417: for (i = (ctxt->nameNr - 1);i >= 0;i--) {
2418: if (!xmlStrcmp(name, ctxt->nameTab[i])) break;
1.1 daniel 2419: }
2420: if (i < 0) {
2421: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.18 daniel 2422: ctxt->sax->error(ctxt->userData,
2423: "Unexpected end tag : %s\n", name);
1.11 daniel 2424: xmlFree(name);
1.1 daniel 2425: ctxt->wellFormed = 0;
2426: return;
2427: }
2428:
1.18 daniel 2429:
1.1 daniel 2430: /*
2431: * Check for auto-closure of HTML elements.
2432: */
1.18 daniel 2433:
1.1 daniel 2434: htmlAutoCloseOnClose(ctxt, name);
2435:
2436: /*
2437: * Well formedness constraints, opening and closing must match.
2438: * With the exception that the autoclose may have popped stuff out
2439: * of the stack.
2440: */
1.18 daniel 2441: if (xmlStrcmp(name, ctxt->name)) {
2442: #ifdef DEBUG
2443: fprintf(stderr,"End of tag %s: expecting %s\n", name, ctxt->name);
2444: #endif
1.15 daniel 2445: if ((ctxt->name != NULL) &&
2446: (xmlStrcmp(ctxt->name, name))) {
1.1 daniel 2447: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2448: ctxt->sax->error(ctxt->userData,
2449: "Opening and ending tag mismatch: %s and %s\n",
1.15 daniel 2450: name, ctxt->name);
1.1 daniel 2451: ctxt->wellFormed = 0;
2452: }
2453: }
2454:
2455: /*
2456: * SAX: End of Tag
2457: */
1.15 daniel 2458: oldname = ctxt->name;
1.24 daniel 2459: if ((oldname != NULL) && (!xmlStrcmp(oldname, name))) {
1.18 daniel 2460: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
2461: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 2462: oldname = htmlnamePop(ctxt);
1.18 daniel 2463: if (oldname != NULL) {
2464: #ifdef DEBUG
2465: fprintf(stderr,"End of tag %s: popping out %s\n", name, oldname);
2466: #endif
2467: xmlFree(oldname);
2468: #ifdef DEBUG
2469: } else {
2470: fprintf(stderr,"End of tag %s: stack empty !!!\n", name);
2471: #endif
2472: }
2473: }
1.1 daniel 2474:
2475: if (name != NULL)
1.11 daniel 2476: xmlFree(name);
1.1 daniel 2477:
2478: return;
2479: }
2480:
2481:
2482: /**
2483: * htmlParseReference:
2484: * @ctxt: an HTML parser context
2485: *
2486: * parse and handle entity references in content,
2487: * this will end-up in a call to character() since this is either a
2488: * CharRef, or a predefined entity.
2489: */
2490: void
2491: htmlParseReference(htmlParserCtxtPtr ctxt) {
2492: htmlEntityDescPtr ent;
1.14 daniel 2493: xmlChar out[2];
2494: xmlChar *name;
1.1 daniel 2495: int val;
2496: if (CUR != '&') return;
2497:
2498: if (NXT(1) == '#') {
2499: val = htmlParseCharRef(ctxt);
1.8 daniel 2500: /* invalid for UTF-8 variable encoding !!!!! */
1.1 daniel 2501: out[0] = val;
2502: out[1] = 0;
2503: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
2504: ctxt->sax->characters(ctxt->userData, out, 1);
2505: } else {
2506: ent = htmlParseEntityRef(ctxt, &name);
1.32 daniel 2507: if (name == NULL) {
2508: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
2509: return;
2510: }
1.1 daniel 2511: if ((ent == NULL) || (ent->value <= 0) || (ent->value >= 255)) {
2512: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
1.8 daniel 2513: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
1.1 daniel 2514: ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
1.32 daniel 2515: /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
1.1 daniel 2516: }
2517: } else {
1.8 daniel 2518: /* invalid for UTF-8 variable encoding !!!!! */
1.1 daniel 2519: out[0] = ent->value;
2520: out[1] = 0;
2521: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
2522: ctxt->sax->characters(ctxt->userData, out, 1);
2523: }
1.11 daniel 2524: xmlFree(name);
1.1 daniel 2525: }
2526: }
2527:
2528: /**
2529: * htmlParseContent:
2530: * @ctxt: an HTML parser context
2531: * @name: the node name
2532: *
2533: * Parse a content: comment, sub-element, reference or text.
2534: *
2535: */
2536:
2537: void
1.18 daniel 2538: htmlParseContent(htmlParserCtxtPtr ctxt) {
1.15 daniel 2539: xmlChar *currentNode;
1.18 daniel 2540: int depth;
1.1 daniel 2541:
1.26 daniel 2542: currentNode = xmlStrdup(ctxt->name);
1.18 daniel 2543: depth = ctxt->nameNr;
2544: while (1) {
1.26 daniel 2545: long cons = ctxt->nbChars;
1.1 daniel 2546:
1.18 daniel 2547: GROW;
2548: /*
2549: * Our tag or one of it's parent or children is ending.
2550: */
2551: if ((CUR == '<') && (NXT(1) == '/')) {
2552: htmlParseEndTag(ctxt);
1.26 daniel 2553: if (currentNode != NULL) xmlFree(currentNode);
1.18 daniel 2554: return;
2555: }
2556:
2557: /*
2558: * Has this node been popped out during parsing of
2559: * the next element
2560: */
1.26 daniel 2561: if ((xmlStrcmp(currentNode, ctxt->name)) &&
2562: (depth >= ctxt->nameNr)) {
2563: if (currentNode != NULL) xmlFree(currentNode);
2564: return;
2565: }
1.18 daniel 2566:
1.1 daniel 2567: /*
2568: * First case : a comment
2569: */
2570: if ((CUR == '<') && (NXT(1) == '!') &&
2571: (NXT(2) == '-') && (NXT(3) == '-')) {
1.31 daniel 2572: htmlParseComment(ctxt);
1.1 daniel 2573: }
2574:
2575: /*
2576: * Second case : a sub-element.
2577: */
2578: else if (CUR == '<') {
2579: htmlParseElement(ctxt);
2580: }
2581:
2582: /*
2583: * Third case : a reference. If if has not been resolved,
2584: * parsing returns it's Name, create the node
2585: */
2586: else if (CUR == '&') {
2587: htmlParseReference(ctxt);
2588: }
2589:
2590: /*
2591: * Last case, text. Note that References are handled directly.
2592: */
2593: else {
2594: htmlParseCharData(ctxt, 0);
2595: }
2596:
1.26 daniel 2597: if (cons == ctxt->nbChars) {
1.22 daniel 2598: if (ctxt->node != NULL) {
2599: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2600: ctxt->sax->error(ctxt->userData,
2601: "detected an error in element content\n");
2602: ctxt->wellFormed = 0;
2603: }
1.1 daniel 2604: break;
2605: }
1.17 daniel 2606:
1.5 daniel 2607: GROW;
1.1 daniel 2608: }
1.26 daniel 2609: if (currentNode != NULL) xmlFree(currentNode);
1.1 daniel 2610: }
2611:
2612: /**
2613: * htmlParseElement:
2614: * @ctxt: an HTML parser context
2615: *
2616: * parse an HTML element, this is highly recursive
2617: *
2618: * [39] element ::= EmptyElemTag | STag content ETag
2619: *
2620: * [41] Attribute ::= Name Eq AttValue
2621: */
2622:
2623: void
2624: htmlParseElement(htmlParserCtxtPtr ctxt) {
1.14 daniel 2625: const xmlChar *openTag = CUR_PTR;
2626: xmlChar *name;
1.16 daniel 2627: xmlChar *currentNode = NULL;
1.1 daniel 2628: htmlElemDescPtr info;
1.10 daniel 2629: htmlParserNodeInfo node_info;
1.31 daniel 2630: xmlChar *oldname;
1.18 daniel 2631: int depth = ctxt->nameNr;
1.1 daniel 2632:
2633: /* Capture start position */
1.10 daniel 2634: if (ctxt->record_info) {
2635: node_info.begin_pos = ctxt->input->consumed +
2636: (CUR_PTR - ctxt->input->base);
2637: node_info.begin_line = ctxt->input->line;
2638: }
1.1 daniel 2639:
1.26 daniel 2640: oldname = xmlStrdup(ctxt->name);
1.18 daniel 2641: htmlParseStartTag(ctxt);
2642: name = ctxt->name;
1.19 daniel 2643: #ifdef DEBUG
2644: if (oldname == NULL)
2645: fprintf(stderr, "Start of element %s\n", name);
2646: else if (name == NULL)
2647: fprintf(stderr, "Start of element failed, was %s\n", oldname);
2648: else
2649: fprintf(stderr, "Start of element %s, was %s\n", name, oldname);
2650: #endif
1.26 daniel 2651: if (((depth == ctxt->nameNr) && (!xmlStrcmp(oldname, ctxt->name))) ||
1.18 daniel 2652: (name == NULL)) {
1.19 daniel 2653: if (CUR == '>')
2654: NEXT;
1.26 daniel 2655: if (oldname != NULL)
2656: xmlFree(oldname);
1.1 daniel 2657: return;
2658: }
1.26 daniel 2659: if (oldname != NULL)
2660: xmlFree(oldname);
1.1 daniel 2661:
2662: /*
2663: * Lookup the info for that element.
2664: */
2665: info = htmlTagLookup(name);
2666: if (info == NULL) {
2667: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2668: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
2669: name);
2670: ctxt->wellFormed = 0;
2671: } else if (info->depr) {
2672: /***************************
2673: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
2674: ctxt->sax->warning(ctxt->userData, "Tag %s is deprecated\n",
2675: name);
2676: ***************************/
2677: }
2678:
2679: /*
2680: * Check for an Empty Element labelled the XML/SGML way
2681: */
2682: if ((CUR == '/') && (NXT(1) == '>')) {
2683: SKIP(2);
2684: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
2685: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 2686: oldname = htmlnamePop(ctxt);
1.18 daniel 2687: #ifdef DEBUG
2688: fprintf(stderr,"End of tag the XML way: popping out %s\n", oldname);
2689: #endif
1.17 daniel 2690: if (oldname != NULL)
2691: xmlFree(oldname);
1.1 daniel 2692: return;
2693: }
2694:
1.5 daniel 2695: if (CUR == '>') {
2696: NEXT;
2697: } else {
1.1 daniel 2698: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2699: ctxt->sax->error(ctxt->userData, "Couldn't find end of Start Tag\n%.30s\n",
2700: openTag);
2701: ctxt->wellFormed = 0;
2702:
2703: /*
2704: * end of parsing of this node.
2705: */
1.18 daniel 2706: if (!xmlStrcmp(name, ctxt->name)) {
2707: nodePop(ctxt);
1.24 daniel 2708: oldname = htmlnamePop(ctxt);
1.18 daniel 2709: #ifdef DEBUG
2710: fprintf(stderr,"End of start tag problem: popping out %s\n", oldname);
2711: #endif
2712: if (oldname != NULL)
2713: xmlFree(oldname);
2714: }
1.10 daniel 2715:
2716: /*
2717: * Capture end position and add node
2718: */
2719: if ( currentNode != NULL && ctxt->record_info ) {
2720: node_info.end_pos = ctxt->input->consumed +
2721: (CUR_PTR - ctxt->input->base);
2722: node_info.end_line = ctxt->input->line;
1.15 daniel 2723: node_info.node = ctxt->node;
1.10 daniel 2724: xmlParserAddNodeInfo(ctxt, &node_info);
2725: }
1.1 daniel 2726: return;
2727: }
2728:
2729: /*
2730: * Check for an Empty Element from DTD definition
2731: */
2732: if ((info != NULL) && (info->empty)) {
2733: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
2734: ctxt->sax->endElement(ctxt->userData, name);
1.24 daniel 2735: oldname = htmlnamePop(ctxt);
1.18 daniel 2736: #ifdef DEBUG
2737: fprintf(stderr,"End of empty tag %s : popping out %s\n", name, oldname);
2738: #endif
1.17 daniel 2739: if (oldname != NULL)
2740: xmlFree(oldname);
1.1 daniel 2741: return;
2742: }
2743:
2744: /*
2745: * Parse the content of the element:
2746: */
1.26 daniel 2747: currentNode = xmlStrdup(ctxt->name);
1.18 daniel 2748: depth = ctxt->nameNr;
2749: while (IS_CHAR(CUR)) {
2750: htmlParseContent(ctxt);
2751: if (ctxt->nameNr < depth) break;
2752: }
1.1 daniel 2753:
2754: if (!IS_CHAR(CUR)) {
2755: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2756: ctxt->sax->error(ctxt->userData,
1.18 daniel 2757: "Premature end of data in tag %s\n", currentNode);
1.1 daniel 2758: ctxt->wellFormed = 0;
2759:
2760: /*
2761: * end of parsing of this node.
2762: */
2763: nodePop(ctxt);
1.24 daniel 2764: oldname = htmlnamePop(ctxt);
1.18 daniel 2765: #ifdef DEBUG
2766: fprintf(stderr,"Premature end of tag %s : popping out %s\n", name, oldname);
2767: #endif
1.17 daniel 2768: if (oldname != NULL)
2769: xmlFree(oldname);
1.26 daniel 2770: if (currentNode != NULL)
2771: xmlFree(currentNode);
1.1 daniel 2772: return;
2773: }
1.10 daniel 2774:
2775: /*
2776: * Capture end position and add node
2777: */
2778: if ( currentNode != NULL && ctxt->record_info ) {
2779: node_info.end_pos = ctxt->input->consumed +
2780: (CUR_PTR - ctxt->input->base);
2781: node_info.end_line = ctxt->input->line;
1.15 daniel 2782: node_info.node = ctxt->node;
1.10 daniel 2783: xmlParserAddNodeInfo(ctxt, &node_info);
2784: }
1.26 daniel 2785: if (currentNode != NULL)
2786: xmlFree(currentNode);
1.1 daniel 2787: }
2788:
2789: /**
2790: * htmlParseDocument :
2791: * @ctxt: an HTML parser context
2792: *
2793: * parse an HTML document (and build a tree if using the standard SAX
2794: * interface).
2795: *
2796: * Returns 0, -1 in case of error. the parser context is augmented
2797: * as a result of the parsing.
2798: */
2799:
2800: int
2801: htmlParseDocument(htmlParserCtxtPtr ctxt) {
2802: htmlDefaultSAXHandlerInit();
2803: ctxt->html = 1;
2804:
1.5 daniel 2805: GROW;
1.1 daniel 2806: /*
1.9 daniel 2807: * SAX: beginning of the document processing.
1.1 daniel 2808: */
2809: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
2810: ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
2811:
2812: /*
2813: * Wipe out everything which is before the first '<'
2814: */
1.22 daniel 2815: SKIP_BLANKS;
1.1 daniel 2816: if (CUR == 0) {
2817: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2818: ctxt->sax->error(ctxt->userData, "Document is empty\n");
2819: ctxt->wellFormed = 0;
2820: }
2821:
1.40 daniel 2822: if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX))
2823: ctxt->sax->startDocument(ctxt->userData);
2824:
2825:
1.22 daniel 2826: /*
2827: * Parse possible comments before any content
2828: */
2829: while ((CUR == '<') && (NXT(1) == '!') &&
2830: (NXT(2) == '-') && (NXT(3) == '-')) {
1.31 daniel 2831: htmlParseComment(ctxt);
1.22 daniel 2832: SKIP_BLANKS;
2833: }
2834:
1.1 daniel 2835:
2836: /*
2837: * Then possibly doc type declaration(s) and more Misc
2838: * (doctypedecl Misc*)?
2839: */
2840: if ((CUR == '<') && (NXT(1) == '!') &&
2841: (UPP(2) == 'D') && (UPP(3) == 'O') &&
2842: (UPP(4) == 'C') && (UPP(5) == 'T') &&
2843: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
2844: (UPP(8) == 'E')) {
2845: htmlParseDocTypeDecl(ctxt);
2846: }
2847: SKIP_BLANKS;
2848:
2849: /*
2850: * Time to start parsing the tree itself
2851: */
1.22 daniel 2852: htmlParseContent(ctxt);
1.1 daniel 2853:
2854: /*
2855: * SAX: end of the document processing.
2856: */
2857: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
2858: ctxt->sax->endDocument(ctxt->userData);
2859: if (! ctxt->wellFormed) return(-1);
2860: return(0);
2861: }
2862:
2863:
1.30 daniel 2864: /************************************************************************
2865: * *
2866: * Parser contexts handling *
2867: * *
2868: ************************************************************************/
1.1 daniel 2869:
2870: /**
2871: * xmlInitParserCtxt:
2872: * @ctxt: an HTML parser context
2873: *
2874: * Initialize a parser context
2875: */
2876:
2877: void
2878: htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
2879: {
2880: htmlSAXHandler *sax;
2881:
1.21 daniel 2882: if (ctxt == NULL) return;
2883: memset(ctxt, 0, sizeof(htmlParserCtxt));
2884:
1.11 daniel 2885: sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
1.1 daniel 2886: if (sax == NULL) {
2887: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
2888: }
1.19 daniel 2889: memset(sax, 0, sizeof(htmlSAXHandler));
1.1 daniel 2890:
2891: /* Allocate the Input stack */
1.19 daniel 2892: ctxt->inputTab = (htmlParserInputPtr *)
2893: xmlMalloc(5 * sizeof(htmlParserInputPtr));
2894: if (ctxt->inputTab == NULL) {
2895: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
2896: }
1.1 daniel 2897: ctxt->inputNr = 0;
2898: ctxt->inputMax = 5;
2899: ctxt->input = NULL;
2900: ctxt->version = NULL;
2901: ctxt->encoding = NULL;
2902: ctxt->standalone = -1;
1.30 daniel 2903: ctxt->instate = XML_PARSER_START;
1.1 daniel 2904:
2905: /* Allocate the Node stack */
1.11 daniel 2906: ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
1.1 daniel 2907: ctxt->nodeNr = 0;
2908: ctxt->nodeMax = 10;
2909: ctxt->node = NULL;
2910:
1.15 daniel 2911: /* Allocate the Name stack */
2912: ctxt->nameTab = (xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
2913: ctxt->nameNr = 0;
2914: ctxt->nameMax = 10;
2915: ctxt->name = NULL;
2916:
1.1 daniel 2917: if (sax == NULL) ctxt->sax = &htmlDefaultSAXHandler;
2918: else {
2919: ctxt->sax = sax;
2920: memcpy(sax, &htmlDefaultSAXHandler, sizeof(htmlSAXHandler));
2921: }
2922: ctxt->userData = ctxt;
2923: ctxt->myDoc = NULL;
2924: ctxt->wellFormed = 1;
2925: ctxt->replaceEntities = 0;
2926: ctxt->html = 1;
2927: ctxt->record_info = 0;
1.21 daniel 2928: ctxt->validate = 0;
1.26 daniel 2929: ctxt->nbChars = 0;
1.30 daniel 2930: ctxt->checkIndex = 0;
1.1 daniel 2931: xmlInitNodeInfoSeq(&ctxt->node_seq);
2932: }
2933:
2934: /**
2935: * htmlFreeParserCtxt:
2936: * @ctxt: an HTML parser context
2937: *
2938: * Free all the memory used by a parser context. However the parsed
2939: * document in ctxt->myDoc is not freed.
2940: */
2941:
2942: void
2943: htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
2944: {
2945: htmlParserInputPtr input;
1.15 daniel 2946: xmlChar *oldname;
1.1 daniel 2947:
2948: if (ctxt == NULL) return;
2949:
2950: while ((input = inputPop(ctxt)) != NULL) {
2951: xmlFreeInputStream(input);
2952: }
2953:
1.11 daniel 2954: if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab);
1.24 daniel 2955: while ((oldname = htmlnamePop(ctxt)) != NULL) {
2956: xmlFree(oldname);
1.15 daniel 2957: }
2958: if (ctxt->nameTab != NULL) xmlFree(ctxt->nameTab);
1.31 daniel 2959: if (ctxt->directory != NULL) xmlFree(ctxt->directory);
1.11 daniel 2960: if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab);
2961: if (ctxt->version != NULL) xmlFree((char *) ctxt->version);
1.1 daniel 2962: if ((ctxt->sax != NULL) && (ctxt->sax != &htmlDefaultSAXHandler))
1.11 daniel 2963: xmlFree(ctxt->sax);
2964: xmlFree(ctxt);
1.1 daniel 2965: }
2966:
2967: /**
2968: * htmlCreateDocParserCtxt :
1.14 daniel 2969: * @cur: a pointer to an array of xmlChar
1.1 daniel 2970: * @encoding: a free form C string describing the HTML document encoding, or NULL
2971: *
2972: * Create a parser context for an HTML document.
2973: *
2974: * Returns the new parser context or NULL
2975: */
2976: htmlParserCtxtPtr
1.14 daniel 2977: htmlCreateDocParserCtxt(xmlChar *cur, const char *encoding) {
1.1 daniel 2978: htmlParserCtxtPtr ctxt;
2979: htmlParserInputPtr input;
2980: /* htmlCharEncoding enc; */
2981:
1.11 daniel 2982: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 2983: if (ctxt == NULL) {
2984: perror("malloc");
2985: return(NULL);
2986: }
2987: htmlInitParserCtxt(ctxt);
1.11 daniel 2988: input = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 2989: if (input == NULL) {
2990: perror("malloc");
1.11 daniel 2991: xmlFree(ctxt);
1.1 daniel 2992: return(NULL);
2993: }
1.19 daniel 2994: memset(input, 0, sizeof(htmlParserInput));
1.1 daniel 2995:
2996: input->line = 1;
2997: input->col = 1;
2998: input->base = cur;
2999: input->cur = cur;
3000:
3001: inputPush(ctxt, input);
3002: return(ctxt);
3003: }
3004:
1.31 daniel 3005: /************************************************************************
3006: * *
3007: * Progressive parsing interfaces *
3008: * *
3009: ************************************************************************/
3010:
3011: /**
3012: * htmlParseLookupSequence:
3013: * @ctxt: an HTML parser context
3014: * @first: the first char to lookup
3015: * @next: the next char to lookup or zero
3016: * @third: the next char to lookup or zero
3017: *
3018: * Try to find if a sequence (first, next, third) or just (first next) or
3019: * (first) is available in the input stream.
3020: * This function has a side effect of (possibly) incrementing ctxt->checkIndex
3021: * to avoid rescanning sequences of bytes, it DOES change the state of the
3022: * parser, do not use liberally.
3023: * This is basically similar to xmlParseLookupSequence()
3024: *
3025: * Returns the index to the current parsing point if the full sequence
3026: * is available, -1 otherwise.
3027: */
3028: int
3029: htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first,
3030: xmlChar next, xmlChar third) {
3031: int base, len;
3032: htmlParserInputPtr in;
3033: const xmlChar *buf;
3034:
3035: in = ctxt->input;
3036: if (in == NULL) return(-1);
3037: base = in->cur - in->base;
3038: if (base < 0) return(-1);
3039: if (ctxt->checkIndex > base)
3040: base = ctxt->checkIndex;
3041: if (in->buf == NULL) {
3042: buf = in->base;
3043: len = in->length;
3044: } else {
3045: buf = in->buf->buffer->content;
3046: len = in->buf->buffer->use;
3047: }
3048: /* take into account the sequence length */
3049: if (third) len -= 2;
3050: else if (next) len --;
3051: for (;base < len;base++) {
3052: if (buf[base] == first) {
3053: if (third != 0) {
3054: if ((buf[base + 1] != next) ||
3055: (buf[base + 2] != third)) continue;
3056: } else if (next != 0) {
3057: if (buf[base + 1] != next) continue;
3058: }
3059: ctxt->checkIndex = 0;
3060: #ifdef DEBUG_PUSH
3061: if (next == 0)
3062: fprintf(stderr, "HPP: lookup '%c' found at %d\n",
3063: first, base);
3064: else if (third == 0)
3065: fprintf(stderr, "HPP: lookup '%c%c' found at %d\n",
3066: first, next, base);
3067: else
3068: fprintf(stderr, "HPP: lookup '%c%c%c' found at %d\n",
3069: first, next, third, base);
3070: #endif
3071: return(base - (in->cur - in->base));
3072: }
3073: }
3074: ctxt->checkIndex = base;
3075: #ifdef DEBUG_PUSH
3076: if (next == 0)
3077: fprintf(stderr, "HPP: lookup '%c' failed\n", first);
3078: else if (third == 0)
3079: fprintf(stderr, "HPP: lookup '%c%c' failed\n", first, next);
3080: else
3081: fprintf(stderr, "HPP: lookup '%c%c%c' failed\n", first, next, third);
3082: #endif
3083: return(-1);
3084: }
3085:
3086: /**
1.32 daniel 3087: * htmlParseTryOrFinish:
1.31 daniel 3088: * @ctxt: an HTML parser context
1.32 daniel 3089: * @terminate: last chunk indicator
1.31 daniel 3090: *
3091: * Try to progress on parsing
3092: *
3093: * Returns zero if no parsing was possible
3094: */
3095: int
1.32 daniel 3096: htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
1.31 daniel 3097: int ret = 0;
3098: htmlParserInputPtr in;
3099: int avail;
3100: xmlChar cur, next;
3101:
3102: #ifdef DEBUG_PUSH
3103: switch (ctxt->instate) {
3104: case XML_PARSER_EOF:
3105: fprintf(stderr, "HPP: try EOF\n"); break;
3106: case XML_PARSER_START:
3107: fprintf(stderr, "HPP: try START\n"); break;
3108: case XML_PARSER_MISC:
3109: fprintf(stderr, "HPP: try MISC\n");break;
3110: case XML_PARSER_COMMENT:
3111: fprintf(stderr, "HPP: try COMMENT\n");break;
3112: case XML_PARSER_PROLOG:
3113: fprintf(stderr, "HPP: try PROLOG\n");break;
3114: case XML_PARSER_START_TAG:
3115: fprintf(stderr, "HPP: try START_TAG\n");break;
3116: case XML_PARSER_CONTENT:
3117: fprintf(stderr, "HPP: try CONTENT\n");break;
3118: case XML_PARSER_CDATA_SECTION:
3119: fprintf(stderr, "HPP: try CDATA_SECTION\n");break;
3120: case XML_PARSER_END_TAG:
3121: fprintf(stderr, "HPP: try END_TAG\n");break;
3122: case XML_PARSER_ENTITY_DECL:
3123: fprintf(stderr, "HPP: try ENTITY_DECL\n");break;
3124: case XML_PARSER_ENTITY_VALUE:
3125: fprintf(stderr, "HPP: try ENTITY_VALUE\n");break;
3126: case XML_PARSER_ATTRIBUTE_VALUE:
3127: fprintf(stderr, "HPP: try ATTRIBUTE_VALUE\n");break;
3128: case XML_PARSER_DTD:
3129: fprintf(stderr, "HPP: try DTD\n");break;
3130: case XML_PARSER_EPILOG:
3131: fprintf(stderr, "HPP: try EPILOG\n");break;
3132: case XML_PARSER_PI:
3133: fprintf(stderr, "HPP: try PI\n");break;
3134: }
3135: #endif
3136:
3137: while (1) {
3138:
3139: in = ctxt->input;
3140: if (in == NULL) break;
3141: if (in->buf == NULL)
3142: avail = in->length - (in->cur - in->base);
3143: else
3144: avail = in->buf->buffer->use - (in->cur - in->base);
3145: if (avail < 1)
3146: goto done;
3147: switch (ctxt->instate) {
3148: case XML_PARSER_EOF:
3149: /*
3150: * Document parsing is done !
3151: */
3152: goto done;
3153: case XML_PARSER_START:
3154: /*
3155: * Very first chars read from the document flow.
3156: */
3157: cur = in->cur[0];
3158: if (IS_BLANK(cur)) {
3159: SKIP_BLANKS;
3160: if (in->buf == NULL)
3161: avail = in->length - (in->cur - in->base);
3162: else
3163: avail = in->buf->buffer->use - (in->cur - in->base);
3164: }
3165: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
3166: ctxt->sax->setDocumentLocator(ctxt->userData,
3167: &xmlDefaultSAXLocator);
1.46 ! daniel 3168: if ((ctxt->sax) && (ctxt->sax->startDocument) &&
! 3169: (!ctxt->disableSAX))
! 3170: ctxt->sax->startDocument(ctxt->userData);
! 3171:
1.31 daniel 3172: cur = in->cur[0];
3173: next = in->cur[1];
3174: if ((cur == '<') && (next == '!') &&
3175: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3176: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3177: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3178: (UPP(8) == 'E')) {
1.32 daniel 3179: if ((!terminate) &&
3180: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3181: goto done;
3182: #ifdef DEBUG_PUSH
3183: fprintf(stderr, "HPP: Parsing internal subset\n");
3184: #endif
3185: htmlParseDocTypeDecl(ctxt);
3186: ctxt->instate = XML_PARSER_PROLOG;
3187: #ifdef DEBUG_PUSH
3188: fprintf(stderr, "HPP: entering PROLOG\n");
3189: #endif
3190: } else {
3191: ctxt->instate = XML_PARSER_MISC;
3192: }
3193: #ifdef DEBUG_PUSH
3194: fprintf(stderr, "HPP: entering MISC\n");
3195: #endif
3196: break;
3197: case XML_PARSER_MISC:
3198: SKIP_BLANKS;
3199: if (in->buf == NULL)
3200: avail = in->length - (in->cur - in->base);
3201: else
3202: avail = in->buf->buffer->use - (in->cur - in->base);
3203: if (avail < 2)
3204: goto done;
3205: cur = in->cur[0];
3206: next = in->cur[1];
3207: if ((cur == '<') && (next == '!') &&
3208: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3209: if ((!terminate) &&
3210: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3211: goto done;
3212: #ifdef DEBUG_PUSH
3213: fprintf(stderr, "HPP: Parsing Comment\n");
3214: #endif
3215: htmlParseComment(ctxt);
3216: ctxt->instate = XML_PARSER_MISC;
3217: } else if ((cur == '<') && (next == '!') &&
3218: (UPP(2) == 'D') && (UPP(3) == 'O') &&
3219: (UPP(4) == 'C') && (UPP(5) == 'T') &&
3220: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3221: (UPP(8) == 'E')) {
1.32 daniel 3222: if ((!terminate) &&
3223: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3224: goto done;
3225: #ifdef DEBUG_PUSH
3226: fprintf(stderr, "HPP: Parsing internal subset\n");
3227: #endif
3228: htmlParseDocTypeDecl(ctxt);
3229: ctxt->instate = XML_PARSER_PROLOG;
3230: #ifdef DEBUG_PUSH
3231: fprintf(stderr, "HPP: entering PROLOG\n");
3232: #endif
3233: } else if ((cur == '<') && (next == '!') &&
3234: (avail < 9)) {
3235: goto done;
3236: } else {
3237: ctxt->instate = XML_PARSER_START_TAG;
3238: #ifdef DEBUG_PUSH
3239: fprintf(stderr, "HPP: entering START_TAG\n");
3240: #endif
3241: }
3242: break;
3243: case XML_PARSER_PROLOG:
3244: SKIP_BLANKS;
3245: if (in->buf == NULL)
3246: avail = in->length - (in->cur - in->base);
3247: else
3248: avail = in->buf->buffer->use - (in->cur - in->base);
3249: if (avail < 2)
3250: goto done;
3251: cur = in->cur[0];
3252: next = in->cur[1];
3253: if ((cur == '<') && (next == '!') &&
3254: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3255: if ((!terminate) &&
3256: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3257: goto done;
3258: #ifdef DEBUG_PUSH
3259: fprintf(stderr, "HPP: Parsing Comment\n");
3260: #endif
3261: htmlParseComment(ctxt);
3262: ctxt->instate = XML_PARSER_PROLOG;
3263: } else if ((cur == '<') && (next == '!') &&
3264: (avail < 4)) {
3265: goto done;
3266: } else {
3267: ctxt->instate = XML_PARSER_START_TAG;
3268: #ifdef DEBUG_PUSH
3269: fprintf(stderr, "HPP: entering START_TAG\n");
3270: #endif
3271: }
3272: break;
3273: case XML_PARSER_EPILOG:
3274: SKIP_BLANKS;
3275: if (in->buf == NULL)
3276: avail = in->length - (in->cur - in->base);
3277: else
3278: avail = in->buf->buffer->use - (in->cur - in->base);
3279: if (avail < 2)
3280: goto done;
3281: cur = in->cur[0];
3282: next = in->cur[1];
3283: if ((cur == '<') && (next == '!') &&
3284: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3285: if ((!terminate) &&
3286: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3287: goto done;
3288: #ifdef DEBUG_PUSH
3289: fprintf(stderr, "HPP: Parsing Comment\n");
3290: #endif
3291: htmlParseComment(ctxt);
3292: ctxt->instate = XML_PARSER_EPILOG;
3293: } else if ((cur == '<') && (next == '!') &&
3294: (avail < 4)) {
3295: goto done;
3296: } else {
3297: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3298: ctxt->sax->error(ctxt->userData,
3299: "Extra content at the end of the document\n");
3300: ctxt->wellFormed = 0;
3301: ctxt->errNo = XML_ERR_DOCUMENT_END;
3302: ctxt->instate = XML_PARSER_EOF;
3303: #ifdef DEBUG_PUSH
3304: fprintf(stderr, "HPP: entering EOF\n");
3305: #endif
3306: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3307: ctxt->sax->endDocument(ctxt->userData);
3308: goto done;
3309: }
3310: break;
3311: case XML_PARSER_START_TAG: {
3312: xmlChar *name, *oldname;
3313: int depth = ctxt->nameNr;
3314: htmlElemDescPtr info;
3315:
3316: if (avail < 2)
3317: goto done;
3318: cur = in->cur[0];
3319: if (cur != '<') {
3320: ctxt->instate = XML_PARSER_CONTENT;
3321: #ifdef DEBUG_PUSH
3322: fprintf(stderr, "HPP: entering CONTENT\n");
3323: #endif
3324: break;
3325: }
1.32 daniel 3326: if ((!terminate) &&
3327: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3328: goto done;
3329:
3330: oldname = xmlStrdup(ctxt->name);
3331: htmlParseStartTag(ctxt);
3332: name = ctxt->name;
3333: #ifdef DEBUG
3334: if (oldname == NULL)
3335: fprintf(stderr, "Start of element %s\n", name);
3336: else if (name == NULL)
3337: fprintf(stderr, "Start of element failed, was %s\n",
3338: oldname);
3339: else
3340: fprintf(stderr, "Start of element %s, was %s\n",
3341: name, oldname);
3342: #endif
3343: if (((depth == ctxt->nameNr) &&
3344: (!xmlStrcmp(oldname, ctxt->name))) ||
3345: (name == NULL)) {
3346: if (CUR == '>')
3347: NEXT;
3348: if (oldname != NULL)
3349: xmlFree(oldname);
3350: break;
3351: }
3352: if (oldname != NULL)
3353: xmlFree(oldname);
3354:
3355: /*
3356: * Lookup the info for that element.
3357: */
3358: info = htmlTagLookup(name);
3359: if (info == NULL) {
3360: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3361: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
3362: name);
3363: ctxt->wellFormed = 0;
3364: } else if (info->depr) {
3365: /***************************
3366: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
3367: ctxt->sax->warning(ctxt->userData,
3368: "Tag %s is deprecated\n",
3369: name);
3370: ***************************/
3371: }
3372:
3373: /*
3374: * Check for an Empty Element labelled the XML/SGML way
3375: */
3376: if ((CUR == '/') && (NXT(1) == '>')) {
3377: SKIP(2);
3378: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3379: ctxt->sax->endElement(ctxt->userData, name);
3380: oldname = htmlnamePop(ctxt);
3381: #ifdef DEBUG
3382: fprintf(stderr,"End of tag the XML way: popping out %s\n",
3383: oldname);
3384: #endif
3385: if (oldname != NULL)
3386: xmlFree(oldname);
3387: ctxt->instate = XML_PARSER_CONTENT;
3388: #ifdef DEBUG_PUSH
3389: fprintf(stderr, "HPP: entering CONTENT\n");
3390: #endif
3391: break;
3392: }
3393:
3394: if (CUR == '>') {
3395: NEXT;
3396: } else {
3397: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3398: ctxt->sax->error(ctxt->userData,
3399: "Couldn't find end of Start Tag %s\n",
3400: name);
3401: ctxt->wellFormed = 0;
3402:
3403: /*
3404: * end of parsing of this node.
3405: */
3406: if (!xmlStrcmp(name, ctxt->name)) {
3407: nodePop(ctxt);
3408: oldname = htmlnamePop(ctxt);
3409: #ifdef DEBUG
3410: fprintf(stderr,
3411: "End of start tag problem: popping out %s\n", oldname);
3412: #endif
3413: if (oldname != NULL)
3414: xmlFree(oldname);
3415: }
3416:
3417: ctxt->instate = XML_PARSER_CONTENT;
3418: #ifdef DEBUG_PUSH
3419: fprintf(stderr, "HPP: entering CONTENT\n");
3420: #endif
3421: break;
3422: }
3423:
3424: /*
3425: * Check for an Empty Element from DTD definition
3426: */
3427: if ((info != NULL) && (info->empty)) {
3428: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3429: ctxt->sax->endElement(ctxt->userData, name);
3430: oldname = htmlnamePop(ctxt);
3431: #ifdef DEBUG
3432: fprintf(stderr,"End of empty tag %s : popping out %s\n", name, oldname);
3433: #endif
3434: if (oldname != NULL)
3435: xmlFree(oldname);
3436: }
3437: ctxt->instate = XML_PARSER_CONTENT;
3438: #ifdef DEBUG_PUSH
3439: fprintf(stderr, "HPP: entering CONTENT\n");
3440: #endif
3441: break;
3442: }
3443: case XML_PARSER_CONTENT:
3444: /*
3445: * Handle preparsed entities and charRef
3446: */
3447: if (ctxt->token != 0) {
3448: xmlChar cur[2] = { 0 , 0 } ;
3449:
3450: cur[0] = (xmlChar) ctxt->token;
3451: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3452: ctxt->sax->characters(ctxt->userData, cur, 1);
3453: ctxt->token = 0;
3454: ctxt->checkIndex = 0;
3455: }
3456: if (avail < 2)
3457: goto done;
3458: cur = in->cur[0];
3459: next = in->cur[1];
3460: if ((cur == '<') && (next == '!') &&
3461: (in->cur[2] == '-') && (in->cur[3] == '-')) {
1.32 daniel 3462: if ((!terminate) &&
3463: (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
1.31 daniel 3464: goto done;
3465: #ifdef DEBUG_PUSH
3466: fprintf(stderr, "HPP: Parsing Comment\n");
3467: #endif
3468: htmlParseComment(ctxt);
3469: ctxt->instate = XML_PARSER_CONTENT;
3470: } else if ((cur == '<') && (next == '!') && (avail < 4)) {
3471: goto done;
3472: } else if ((cur == '<') && (next == '/')) {
3473: ctxt->instate = XML_PARSER_END_TAG;
3474: ctxt->checkIndex = 0;
3475: #ifdef DEBUG_PUSH
3476: fprintf(stderr, "HPP: entering END_TAG\n");
3477: #endif
3478: break;
3479: } else if (cur == '<') {
3480: ctxt->instate = XML_PARSER_START_TAG;
3481: ctxt->checkIndex = 0;
3482: #ifdef DEBUG_PUSH
3483: fprintf(stderr, "HPP: entering START_TAG\n");
3484: #endif
3485: break;
3486: } else if (cur == '&') {
1.32 daniel 3487: if ((!terminate) &&
3488: (htmlParseLookupSequence(ctxt, ';', 0, 0) < 0))
1.31 daniel 3489: goto done;
3490: #ifdef DEBUG_PUSH
3491: fprintf(stderr, "HPP: Parsing Reference\n");
3492: #endif
3493: /* TODO: check generation of subtrees if noent !!! */
3494: htmlParseReference(ctxt);
3495: } else {
3496: /* TODO Avoid the extra copy, handle directly !!!!!! */
3497: /*
3498: * Goal of the following test is :
3499: * - minimize calls to the SAX 'character' callback
3500: * when they are mergeable
3501: */
3502: if ((ctxt->inputNr == 1) &&
3503: (avail < HTML_PARSER_BIG_BUFFER_SIZE)) {
1.32 daniel 3504: if ((!terminate) &&
3505: (htmlParseLookupSequence(ctxt, '<', 0, 0) < 0))
1.31 daniel 3506: goto done;
3507: }
3508: ctxt->checkIndex = 0;
3509: #ifdef DEBUG_PUSH
3510: fprintf(stderr, "HPP: Parsing char data\n");
3511: #endif
3512: htmlParseCharData(ctxt, 0);
3513: }
3514: break;
3515: case XML_PARSER_END_TAG:
3516: if (avail < 2)
3517: goto done;
1.32 daniel 3518: if ((!terminate) &&
3519: (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
1.31 daniel 3520: goto done;
3521: htmlParseEndTag(ctxt);
3522: if (ctxt->nameNr == 0) {
3523: ctxt->instate = XML_PARSER_EPILOG;
3524: } else {
3525: ctxt->instate = XML_PARSER_CONTENT;
3526: }
3527: ctxt->checkIndex = 0;
3528: #ifdef DEBUG_PUSH
3529: fprintf(stderr, "HPP: entering CONTENT\n");
3530: #endif
3531: break;
3532: case XML_PARSER_CDATA_SECTION:
3533: fprintf(stderr, "HPP: internal error, state == CDATA\n");
3534: ctxt->instate = XML_PARSER_CONTENT;
3535: ctxt->checkIndex = 0;
3536: #ifdef DEBUG_PUSH
3537: fprintf(stderr, "HPP: entering CONTENT\n");
3538: #endif
3539: break;
3540: case XML_PARSER_DTD:
3541: fprintf(stderr, "HPP: internal error, state == DTD\n");
3542: ctxt->instate = XML_PARSER_CONTENT;
3543: ctxt->checkIndex = 0;
3544: #ifdef DEBUG_PUSH
3545: fprintf(stderr, "HPP: entering CONTENT\n");
3546: #endif
3547: break;
3548: case XML_PARSER_COMMENT:
3549: fprintf(stderr, "HPP: internal error, state == COMMENT\n");
3550: ctxt->instate = XML_PARSER_CONTENT;
3551: ctxt->checkIndex = 0;
3552: #ifdef DEBUG_PUSH
3553: fprintf(stderr, "HPP: entering CONTENT\n");
3554: #endif
3555: break;
3556: case XML_PARSER_PI:
3557: fprintf(stderr, "HPP: internal error, state == PI\n");
3558: ctxt->instate = XML_PARSER_CONTENT;
3559: ctxt->checkIndex = 0;
3560: #ifdef DEBUG_PUSH
3561: fprintf(stderr, "HPP: entering CONTENT\n");
3562: #endif
3563: break;
3564: case XML_PARSER_ENTITY_DECL:
3565: fprintf(stderr, "HPP: internal error, state == ENTITY_DECL\n");
3566: ctxt->instate = XML_PARSER_CONTENT;
3567: ctxt->checkIndex = 0;
3568: #ifdef DEBUG_PUSH
3569: fprintf(stderr, "HPP: entering CONTENT\n");
3570: #endif
3571: break;
3572: case XML_PARSER_ENTITY_VALUE:
3573: fprintf(stderr, "HPP: internal error, state == ENTITY_VALUE\n");
3574: ctxt->instate = XML_PARSER_CONTENT;
3575: ctxt->checkIndex = 0;
3576: #ifdef DEBUG_PUSH
3577: fprintf(stderr, "HPP: entering DTD\n");
3578: #endif
3579: break;
3580: case XML_PARSER_ATTRIBUTE_VALUE:
3581: fprintf(stderr, "HPP: internal error, state == ATTRIBUTE_VALUE\n");
3582: ctxt->instate = XML_PARSER_START_TAG;
3583: ctxt->checkIndex = 0;
3584: #ifdef DEBUG_PUSH
3585: fprintf(stderr, "HPP: entering START_TAG\n");
3586: #endif
3587: break;
3588: }
3589: }
3590: done:
3591: #ifdef DEBUG_PUSH
3592: fprintf(stderr, "HPP: done %d\n", ret);
3593: #endif
3594: return(ret);
3595: }
3596:
3597: /**
1.32 daniel 3598: * htmlParseTry:
3599: * @ctxt: an HTML parser context
3600: *
3601: * Try to progress on parsing
3602: *
3603: * Returns zero if no parsing was possible
3604: */
3605: int
3606: htmlParseTry(htmlParserCtxtPtr ctxt) {
3607: return(htmlParseTryOrFinish(ctxt, 0));
3608: }
3609:
3610: /**
1.31 daniel 3611: * htmlParseChunk:
3612: * @ctxt: an XML parser context
3613: * @chunk: an char array
3614: * @size: the size in byte of the chunk
3615: * @terminate: last chunk indicator
3616: *
3617: * Parse a Chunk of memory
3618: *
3619: * Returns zero if no error, the xmlParserErrors otherwise.
3620: */
3621: int
3622: htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
3623: int terminate) {
3624: if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
3625: (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) {
3626: int base = ctxt->input->base - ctxt->input->buf->buffer->content;
3627: int cur = ctxt->input->cur - ctxt->input->base;
3628:
3629: xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
3630: ctxt->input->base = ctxt->input->buf->buffer->content + base;
3631: ctxt->input->cur = ctxt->input->base + cur;
3632: #ifdef DEBUG_PUSH
3633: fprintf(stderr, "HPP: pushed %d\n", size);
3634: #endif
3635:
1.34 daniel 3636: if ((terminate) || (ctxt->input->buf->buffer->use > 80))
3637: htmlParseTryOrFinish(ctxt, terminate);
1.31 daniel 3638: } else if (ctxt->instate != XML_PARSER_EOF)
1.32 daniel 3639: htmlParseTryOrFinish(ctxt, terminate);
1.31 daniel 3640: if (terminate) {
3641: if ((ctxt->instate != XML_PARSER_EOF) &&
3642: (ctxt->instate != XML_PARSER_EPILOG) &&
3643: (ctxt->instate != XML_PARSER_MISC)) {
3644: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
3645: ctxt->sax->error(ctxt->userData,
3646: "Extra content at the end of the document\n");
3647: ctxt->wellFormed = 0;
3648: ctxt->errNo = XML_ERR_DOCUMENT_END;
3649: }
3650: if (ctxt->instate != XML_PARSER_EOF) {
3651: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
3652: ctxt->sax->endDocument(ctxt->userData);
3653: }
3654: ctxt->instate = XML_PARSER_EOF;
3655: }
3656: return((xmlParserErrors) ctxt->errNo);
3657: }
3658:
3659: /************************************************************************
3660: * *
3661: * User entry points *
3662: * *
3663: ************************************************************************/
3664:
3665: /**
3666: * htmlCreatePushParserCtxt :
3667: * @sax: a SAX handler
3668: * @user_data: The user data returned on SAX callbacks
3669: * @chunk: a pointer to an array of chars
3670: * @size: number of chars in the array
3671: * @filename: an optional file name or URI
3672: * @enc: an optional encoding
3673: *
3674: * Create a parser context for using the HTML parser in push mode
3675: * To allow content encoding detection, @size should be >= 4
3676: * The value of @filename is used for fetching external entities
3677: * and error/warning reports.
3678: *
3679: * Returns the new parser context or NULL
3680: */
3681: htmlParserCtxtPtr
3682: htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data,
3683: const char *chunk, int size, const char *filename,
3684: xmlCharEncoding enc) {
3685: htmlParserCtxtPtr ctxt;
3686: htmlParserInputPtr inputStream;
3687: xmlParserInputBufferPtr buf;
3688:
3689: buf = xmlAllocParserInputBuffer(enc);
3690: if (buf == NULL) return(NULL);
3691:
3692: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
3693: if (ctxt == NULL) {
3694: xmlFree(buf);
3695: return(NULL);
3696: }
3697: memset(ctxt, 0, sizeof(htmlParserCtxt));
3698: htmlInitParserCtxt(ctxt);
3699: if (sax != NULL) {
3700: if (ctxt->sax != &htmlDefaultSAXHandler)
3701: xmlFree(ctxt->sax);
3702: ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler));
3703: if (ctxt->sax == NULL) {
3704: xmlFree(buf);
3705: xmlFree(ctxt);
3706: return(NULL);
3707: }
3708: memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler));
3709: if (user_data != NULL)
3710: ctxt->userData = user_data;
3711: }
3712: if (filename == NULL) {
3713: ctxt->directory = NULL;
3714: } else {
3715: ctxt->directory = xmlParserGetDirectory(filename);
3716: }
3717:
3718: inputStream = htmlNewInputStream(ctxt);
3719: if (inputStream == NULL) {
3720: xmlFreeParserCtxt(ctxt);
3721: return(NULL);
3722: }
3723:
3724: if (filename == NULL)
3725: inputStream->filename = NULL;
3726: else
3727: inputStream->filename = xmlMemStrdup(filename);
3728: inputStream->buf = buf;
3729: inputStream->base = inputStream->buf->buffer->content;
3730: inputStream->cur = inputStream->buf->buffer->content;
3731:
3732: inputPush(ctxt, inputStream);
3733:
3734: if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
3735: (ctxt->input->buf != NULL)) {
3736: xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
3737: #ifdef DEBUG_PUSH
3738: fprintf(stderr, "HPP: pushed %d\n", size);
3739: #endif
3740: }
3741:
3742: return(ctxt);
3743: }
1.1 daniel 3744:
3745: /**
3746: * htmlSAXParseDoc :
1.14 daniel 3747: * @cur: a pointer to an array of xmlChar
1.1 daniel 3748: * @encoding: a free form C string describing the HTML document encoding, or NULL
3749: * @sax: the SAX handler block
3750: * @userData: if using SAX, this pointer will be provided on callbacks.
3751: *
3752: * parse an HTML in-memory document and build a tree.
3753: * It use the given SAX function block to handle the parsing callback.
3754: * If sax is NULL, fallback to the default DOM tree building routines.
3755: *
3756: * Returns the resulting document tree
3757: */
3758:
3759: htmlDocPtr
1.14 daniel 3760: htmlSAXParseDoc(xmlChar *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) {
1.1 daniel 3761: htmlDocPtr ret;
3762: htmlParserCtxtPtr ctxt;
3763:
3764: if (cur == NULL) return(NULL);
3765:
3766:
3767: ctxt = htmlCreateDocParserCtxt(cur, encoding);
3768: if (ctxt == NULL) return(NULL);
3769: if (sax != NULL) {
3770: ctxt->sax = sax;
3771: ctxt->userData = userData;
3772: }
3773:
3774: htmlParseDocument(ctxt);
3775: ret = ctxt->myDoc;
3776: if (sax != NULL) {
3777: ctxt->sax = NULL;
3778: ctxt->userData = NULL;
3779: }
3780: htmlFreeParserCtxt(ctxt);
3781:
3782: return(ret);
3783: }
3784:
3785: /**
3786: * htmlParseDoc :
1.14 daniel 3787: * @cur: a pointer to an array of xmlChar
1.1 daniel 3788: * @encoding: a free form C string describing the HTML document encoding, or NULL
3789: *
3790: * parse an HTML in-memory document and build a tree.
3791: *
3792: * Returns the resulting document tree
3793: */
3794:
3795: htmlDocPtr
1.14 daniel 3796: htmlParseDoc(xmlChar *cur, const char *encoding) {
1.1 daniel 3797: return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
3798: }
3799:
3800:
3801: /**
3802: * htmlCreateFileParserCtxt :
3803: * @filename: the filename
3804: * @encoding: a free form C string describing the HTML document encoding, or NULL
3805: *
3806: * Create a parser context for a file content.
3807: * Automatic support for ZLIB/Compress compressed document is provided
3808: * by default if found at compile-time.
3809: *
3810: * Returns the new parser context or NULL
3811: */
3812: htmlParserCtxtPtr
3813: htmlCreateFileParserCtxt(const char *filename, const char *encoding)
3814: {
3815: htmlParserCtxtPtr ctxt;
3816: htmlParserInputPtr inputStream;
1.5 daniel 3817: xmlParserInputBufferPtr buf;
1.1 daniel 3818: /* htmlCharEncoding enc; */
3819:
1.5 daniel 3820: buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
3821: if (buf == NULL) return(NULL);
1.1 daniel 3822:
1.11 daniel 3823: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 3824: if (ctxt == NULL) {
3825: perror("malloc");
3826: return(NULL);
3827: }
1.19 daniel 3828: memset(ctxt, 0, sizeof(htmlParserCtxt));
1.1 daniel 3829: htmlInitParserCtxt(ctxt);
1.11 daniel 3830: inputStream = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 3831: if (inputStream == NULL) {
3832: perror("malloc");
1.11 daniel 3833: xmlFree(ctxt);
1.1 daniel 3834: return(NULL);
3835: }
1.19 daniel 3836: memset(inputStream, 0, sizeof(htmlParserInput));
1.1 daniel 3837:
1.11 daniel 3838: inputStream->filename = xmlMemStrdup(filename);
1.1 daniel 3839: inputStream->line = 1;
3840: inputStream->col = 1;
1.5 daniel 3841: inputStream->buf = buf;
1.21 daniel 3842: inputStream->directory = NULL;
1.1 daniel 3843:
1.5 daniel 3844: inputStream->base = inputStream->buf->buffer->content;
3845: inputStream->cur = inputStream->buf->buffer->content;
3846: inputStream->free = NULL;
1.1 daniel 3847:
3848: inputPush(ctxt, inputStream);
3849: return(ctxt);
3850: }
3851:
3852: /**
3853: * htmlSAXParseFile :
3854: * @filename: the filename
3855: * @encoding: a free form C string describing the HTML document encoding, or NULL
3856: * @sax: the SAX handler block
3857: * @userData: if using SAX, this pointer will be provided on callbacks.
3858: *
3859: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
3860: * compressed document is provided by default if found at compile-time.
3861: * It use the given SAX function block to handle the parsing callback.
3862: * If sax is NULL, fallback to the default DOM tree building routines.
3863: *
3864: * Returns the resulting document tree
3865: */
3866:
3867: htmlDocPtr
3868: htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
3869: void *userData) {
3870: htmlDocPtr ret;
3871: htmlParserCtxtPtr ctxt;
3872:
3873: ctxt = htmlCreateFileParserCtxt(filename, encoding);
3874: if (ctxt == NULL) return(NULL);
3875: if (sax != NULL) {
3876: ctxt->sax = sax;
3877: ctxt->userData = userData;
3878: }
3879:
3880: htmlParseDocument(ctxt);
3881:
3882: ret = ctxt->myDoc;
3883: if (sax != NULL) {
3884: ctxt->sax = NULL;
3885: ctxt->userData = NULL;
3886: }
3887: htmlFreeParserCtxt(ctxt);
3888:
3889: return(ret);
3890: }
3891:
3892: /**
3893: * htmlParseFile :
3894: * @filename: the filename
3895: * @encoding: a free form C string describing the HTML document encoding, or NULL
3896: *
3897: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
3898: * compressed document is provided by default if found at compile-time.
3899: *
3900: * Returns the resulting document tree
3901: */
3902:
3903: htmlDocPtr
3904: htmlParseFile(const char *filename, const char *encoding) {
3905: return(htmlSAXParseFile(filename, encoding, NULL, NULL));
3906: }
1.39 daniel 3907:
3908: #endif /* LIBXML_HTML_ENABLED */
Webmaster