Annotation of XML/HTMLparser.c, revision 1.16
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
10: #define HAVE_FCNTL_H
11: #include <io.h>
12: #else
1.13 daniel 13: #include "config.h"
1.1 daniel 14: #endif
1.13 daniel 15:
1.1 daniel 16: #include <stdio.h>
1.13 daniel 17: #include <string.h> /* for memset() only */
18: #ifdef HAVE_CTYPE_H
1.1 daniel 19: #include <ctype.h>
1.13 daniel 20: #endif
21: #ifdef HAVE_STDLIB_H
1.1 daniel 22: #include <stdlib.h>
1.13 daniel 23: #endif
24: #ifdef HAVE_SYS_STAT_H
1.1 daniel 25: #include <sys/stat.h>
1.13 daniel 26: #endif
1.1 daniel 27: #ifdef HAVE_FCNTL_H
28: #include <fcntl.h>
29: #endif
30: #ifdef HAVE_UNISTD_H
31: #include <unistd.h>
32: #endif
33: #ifdef HAVE_ZLIB_H
34: #include <zlib.h>
35: #endif
36:
1.11 daniel 37: #include "xmlmemory.h"
1.1 daniel 38: #include "tree.h"
39: #include "HTMLparser.h"
40: #include "entities.h"
41: #include "encoding.h"
42: #include "valid.h"
43: #include "parserInternals.h"
1.5 daniel 44: #include "xmlIO.h"
45:
46: #define HTML_MAX_NAMELEN 1000
47: #define INPUT_CHUNK 50
1.1 daniel 48:
49: /* #define DEBUG */
50:
51: /************************************************************************
52: * *
53: * Parser stacks related functions and macros *
54: * *
55: ************************************************************************/
56:
57: /*
58: * Generic function for accessing stacks in the Parser Context
59: */
60:
61: #define PUSH_AND_POP(type, name) \
62: int html##name##Push(htmlParserCtxtPtr ctxt, type value) { \
63: if (ctxt->name##Nr >= ctxt->name##Max) { \
64: ctxt->name##Max *= 2; \
1.11 daniel 65: ctxt->name##Tab = (void *) xmlRealloc(ctxt->name##Tab, \
1.1 daniel 66: ctxt->name##Max * sizeof(ctxt->name##Tab[0])); \
67: if (ctxt->name##Tab == NULL) { \
68: fprintf(stderr, "realloc failed !\n"); \
69: exit(1); \
70: } \
71: } \
72: ctxt->name##Tab[ctxt->name##Nr] = value; \
73: ctxt->name = value; \
74: return(ctxt->name##Nr++); \
75: } \
76: type html##name##Pop(htmlParserCtxtPtr ctxt) { \
77: type ret; \
78: if (ctxt->name##Nr <= 0) return(0); \
79: ctxt->name##Nr--; \
80: if (ctxt->name##Nr > 0) \
81: ctxt->name = ctxt->name##Tab[ctxt->name##Nr - 1]; \
82: else \
83: ctxt->name = NULL; \
84: ret = ctxt->name##Tab[ctxt->name##Nr]; \
85: ctxt->name##Tab[ctxt->name##Nr] = 0; \
86: return(ret); \
87: } \
88:
89: PUSH_AND_POP(xmlNodePtr, node)
1.15 daniel 90: PUSH_AND_POP(xmlChar*, name)
1.1 daniel 91:
92: /*
93: * Macros for accessing the content. Those should be used only by the parser,
94: * and not exported.
95: *
96: * Dirty macros, i.e. one need to make assumption on the context to use them
97: *
1.14 daniel 98: * CUR_PTR return the current pointer to the xmlChar to be parsed.
99: * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled
1.1 daniel 100: * in ISO-Latin or UTF-8, and the current 16 bit value if compiled
101: * in UNICODE mode. This should be used internally by the parser
102: * only to compare to ASCII values otherwise it would break when
103: * running with UTF-8 encoding.
1.14 daniel 104: * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
1.1 daniel 105: * to compare on ASCII based substring.
1.14 daniel 106: * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR
1.1 daniel 107: * it should be used only to compare on ASCII based substring.
1.14 daniel 108: * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
1.1 daniel 109: * strings within the parser.
110: *
111: * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
112: *
113: * CURRENT Returns the current char value, with the full decoding of
114: * UTF-8 if we are using this mode. It returns an int.
115: * NEXT Skip to the next character, this does the proper decoding
116: * in UTF-8 mode. It also pop-up unfinished entities on the fly.
117: * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly
118: */
119:
120: #define CUR (*ctxt->input->cur)
121: #define UPPER (toupper(*ctxt->input->cur))
122: #define SKIP(val) ctxt->input->cur += (val)
123: #define NXT(val) ctxt->input->cur[(val)]
124: #define UPP(val) (toupper(ctxt->input->cur[(val)]))
125: #define CUR_PTR ctxt->input->cur
1.5 daniel 126: #define SHRINK xmlParserInputShrink(ctxt->input)
127: #define GROW xmlParserInputGrow(ctxt->input, INPUT_CHUNK)
1.1 daniel 128:
129: #define SKIP_BLANKS \
130: while (IS_BLANK(*(ctxt->input->cur))) NEXT
131:
132: #ifndef USE_UTF_8
133: #define CURRENT (*ctxt->input->cur)
1.5 daniel 134: #define NEXT { \
135: if ((*ctxt->input->cur == 0) && \
136: (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) { \
137: xmlPopInput(ctxt); \
138: } else { \
139: if (*(ctxt->input->cur) == '\n') { \
140: ctxt->input->line++; ctxt->input->col = 1; \
141: } else ctxt->input->col++; \
142: ctxt->input->cur++; \
143: if (*ctxt->input->cur == 0) \
144: xmlParserInputGrow(ctxt->input, INPUT_CHUNK); \
145: }}
146:
147: /****************************************
1.1 daniel 148: #define NEXT ((*ctxt->input->cur) ? \
149: (((*(ctxt->input->cur) == '\n') ? \
150: (ctxt->input->line++, ctxt->input->col = 1) : \
1.5 daniel 151: (ctxt->input->col++)), \
152: (ctxt->input->cur++), \
153: ((*ctxt->input->cur) ? \
154: (xmlParserInputGrow(ctxt->input, 100), \
155: ctxt->input->cur): \
156: (ctxt->input->cur))) : \
157: ((xmlParserInputGrow(ctxt->input, 100) > 0) ? \
158: ctxt->input->cur: \
159: (xmlPopInput(ctxt), ctxt->input->cur)))
160: ****************************************/
1.1 daniel 161: #else
162: #endif
163:
164:
1.5 daniel 165:
1.1 daniel 166: /************************************************************************
167: * *
168: * The list of HTML elements and their properties *
169: * *
170: ************************************************************************/
171:
172: /*
173: * Start Tag: 1 means the start tag can be ommited
174: * End Tag: 1 means the end tag can be ommited
175: * 2 means it's forbidden (empty elements)
176: * Depr: this element is deprecated
177: * DTD: 1 means that this element is valid only in the Loose DTD
178: * 2 means that this element is valid only in the Frameset DTD
179: *
180: * Name,Start Tag,End Tag, Empty, Depr., DTD, Description
181: */
182: htmlElemDesc html40ElementTable[] = {
183: { "A", 0, 0, 0, 0, 0, "anchor " },
184: { "ABBR", 0, 0, 0, 0, 0, "abbreviated form" },
185: { "ACRONYM", 0, 0, 0, 0, 0, "" },
186: { "ADDRESS", 0, 0, 0, 0, 0, "information on author " },
187: { "APPLET", 0, 0, 0, 1, 1, "Java applet " },
188: { "AREA", 0, 2, 1, 0, 0, "client-side image map area " },
189: { "B", 0, 0, 0, 0, 0, "bold text style" },
190: { "BASE", 0, 2, 1, 0, 0, "document base URI " },
191: { "BASEFONT", 0, 2, 1, 1, 1, "base font size " },
192: { "BDO", 0, 0, 0, 0, 0, "I18N BiDi over-ride " },
193: { "BIG", 0, 0, 0, 0, 0, "large text style" },
194: { "BLOCKQUOTE", 0, 0, 0, 0, 0, "long quotation " },
195: { "BODY", 1, 1, 0, 0, 0, "document body " },
196: { "BR", 0, 2, 1, 0, 0, "forced line break " },
197: { "BUTTON", 0, 0, 0, 0, 0, "push button " },
198: { "CAPTION", 0, 0, 0, 0, 0, "table caption " },
199: { "CENTER", 0, 0, 0, 1, 1, "shorthand for DIV align=center " },
200: { "CITE", 0, 0, 0, 0, 0, "citation" },
201: { "CODE", 0, 0, 0, 0, 0, "computer code fragment" },
202: { "COL", 0, 2, 1, 0, 0, "table column " },
203: { "COLGROUP", 0, 1, 0, 0, 0, "table column group " },
204: { "DD", 0, 1, 0, 0, 0, "definition description " },
205: { "DEL", 0, 0, 0, 0, 0, "deleted text " },
206: { "DFN", 0, 0, 0, 0, 0, "instance definition" },
207: { "DIR", 0, 0, 0, 1, 1, "directory list" },
208: { "DIV", 0, 0, 0, 0, 0, "generic language/style container"},
209: { "DL", 0, 0, 0, 0, 0, "definition list " },
210: { "DT", 0, 1, 0, 0, 0, "definition term " },
211: { "EM", 0, 0, 0, 0, 0, "emphasis" },
212: { "FIELDSET", 0, 0, 0, 0, 0, "form control group " },
213: { "FONT", 0, 0, 0, 1, 1, "local change to font " },
214: { "FORM", 0, 0, 0, 0, 0, "interactive form " },
215: { "FRAME", 0, 2, 1, 0, 2, "subwindow " },
216: { "FRAMESET", 0, 0, 0, 0, 2, "window subdivision" },
217: { "H1", 0, 0, 0, 0, 0, "heading " },
218: { "H2", 0, 0, 0, 0, 0, "heading " },
219: { "H3", 0, 0, 0, 0, 0, "heading " },
220: { "H4", 0, 0, 0, 0, 0, "heading " },
221: { "H5", 0, 0, 0, 0, 0, "heading " },
222: { "H6", 0, 0, 0, 0, 0, "heading " },
223: { "HEAD", 1, 1, 0, 0, 0, "document head " },
224: { "HR", 0, 2, 1, 0, 0, "horizontal rule " },
225: { "HTML", 1, 1, 0, 0, 0, "document root element " },
226: { "I", 0, 0, 0, 0, 0, "italic text style" },
227: { "IFRAME", 0, 0, 0, 0, 1, "inline subwindow " },
228: { "IMG", 0, 2, 1, 0, 0, "Embedded image " },
229: { "INPUT", 0, 2, 1, 0, 0, "form control " },
230: { "INS", 0, 0, 0, 0, 0, "inserted text" },
231: { "ISINDEX", 0, 2, 1, 1, 1, "single line prompt " },
232: { "KBD", 0, 0, 0, 0, 0, "text to be entered by the user" },
233: { "LABEL", 0, 0, 0, 0, 0, "form field label text " },
234: { "LEGEND", 0, 0, 0, 0, 0, "fieldset legend " },
235: { "LI", 0, 1, 0, 0, 0, "list item " },
236: { "LINK", 0, 2, 1, 0, 0, "a media-independent link " },
237: { "MAP", 0, 0, 0, 0, 0, "client-side image map " },
238: { "MENU", 0, 0, 0, 1, 1, "menu list " },
239: { "META", 0, 2, 1, 0, 0, "generic metainformation " },
240: { "NOFRAMES", 0, 0, 0, 0, 2, "alternate content container for non frame-based rendering " },
241: { "NOSCRIPT", 0, 0, 0, 0, 0, "alternate content container for non script-based rendering " },
242: { "OBJECT", 0, 0, 0, 0, 0, "generic embedded object " },
243: { "OL", 0, 0, 0, 0, 0, "ordered list " },
244: { "OPTGROUP", 0, 0, 0, 0, 0, "option group " },
245: { "OPTION", 0, 1, 0, 0, 0, "selectable choice " },
246: { "P", 0, 1, 0, 0, 0, "paragraph " },
247: { "PARAM", 0, 2, 1, 0, 0, "named property value " },
248: { "PRE", 0, 0, 0, 0, 0, "preformatted text " },
249: { "Q", 0, 0, 0, 0, 0, "short inline quotation " },
250: { "S", 0, 0, 0, 1, 1, "strike-through text style" },
251: { "SAMP", 0, 0, 0, 0, 0, "sample program output, scripts, etc." },
252: { "SCRIPT", 0, 0, 0, 0, 0, "script statements " },
253: { "SELECT", 0, 0, 0, 0, 0, "option selector " },
254: { "SMALL", 0, 0, 0, 0, 0, "small text style" },
255: { "SPAN", 0, 0, 0, 0, 0, "generic language/style container " },
256: { "STRIKE", 0, 0, 0, 1, 1, "strike-through text" },
257: { "STRONG", 0, 0, 0, 0, 0, "strong emphasis" },
258: { "STYLE", 0, 0, 0, 0, 0, "style info " },
259: { "SUB", 0, 0, 0, 0, 0, "subscript" },
260: { "SUP", 0, 0, 0, 0, 0, "superscript " },
261: { "TABLE", 0, 0, 0, 0, 0, " " },
262: { "TBODY", 1, 1, 0, 0, 0, "table body " },
263: { "TD", 0, 1, 0, 0, 0, "table data cell" },
264: { "TEXTAREA", 0, 0, 0, 0, 0, "multi-line text field " },
265: { "TFOOT", 0, 1, 0, 0, 0, "table footer " },
266: { "TH", 0, 1, 0, 0, 0, "table header cell" },
267: { "THEAD", 0, 1, 0, 0, 0, "table header " },
268: { "TITLE", 0, 0, 0, 0, 0, "document title " },
269: { "TR", 0, 1, 0, 0, 0, "table row " },
270: { "TT", 0, 0, 0, 0, 0, "teletype or monospaced text style" },
271: { "U", 0, 0, 0, 1, 1, "underlined text style" },
272: { "UL", 0, 0, 0, 0, 0, "unordered list " },
273: { "VAR", 0, 0, 0, 0, 0, "instance of a variable or program argument" },
274: };
275:
276: /*
277: * start tags that imply the end of a current element
278: * any tag of each line implies the end of the current element if the type of
279: * that element is in the same line
280: */
1.8 daniel 281: char *htmlEquEnd[] = {
1.1 daniel 282: "DT", "DD", "LI", "OPTION", NULL,
283: "H1", "H2", "H3", "H4", "H5", "H6", NULL,
284: "OL", "MENU", "DIR", "ADDRESS", "PRE", "LISTING", "XMP", NULL,
285: NULL
286: };
287: /*
288: * acording the HTML DTD, HR should be added to the 2nd line above, as it
289: * is not allowed within a H1, H2, H3, etc. But we should tolerate that case
290: * because many documents contain rules in headings...
291: */
292:
293: /*
294: * start tags that imply the end of current element
295: */
1.8 daniel 296: char *htmlStartClose[] = {
1.1 daniel 297: "FORM", "FORM", "P", "HR", "H1", "H2", "H3", "H4", "H5", "H6",
298: "DL", "UL", "OL", "MENU", "DIR", "ADDRESS", "PRE",
299: "LISTING", "XMP", "HEAD", NULL,
300: "HEAD", "P", NULL,
301: "TITLE", "P", NULL,
302: "BODY", "HEAD", "STYLE", "LINK", "TITLE", "P", NULL,
303: "LI", "P", "H1", "H2", "H3", "H4", "H5", "H6", "DL", "ADDRESS",
1.15 daniel 304: "PRE", "LISTING", "XMP", "HEAD", "LI", NULL,
1.1 daniel 305: "HR", "P", "HEAD", NULL,
306: "H1", "P", "HEAD", NULL,
307: "H2", "P", "HEAD", NULL,
308: "H3", "P", "HEAD", NULL,
309: "H4", "P", "HEAD", NULL,
310: "H5", "P", "HEAD", NULL,
311: "H6", "P", "HEAD", NULL,
312: "DIR", "P", "HEAD", NULL,
313: "ADDRESS", "P", "HEAD", "UL", NULL,
314: "PRE", "P", "HEAD", "UL", NULL,
315: "LISTING", "P", "HEAD", NULL,
316: "XMP", "P", "HEAD", NULL,
317: "BLOCKQUOTE", "P", "HEAD", NULL,
318: "DL", "P", "DT", "MENU", "DIR", "ADDRESS", "PRE", "LISTING",
319: "XMP", "HEAD", NULL,
320: "DT", "P", "MENU", "DIR", "ADDRESS", "PRE", "LISTING", "XMP", "HEAD", NULL,
321: "DD", "P", "MENU", "DIR", "ADDRESS", "PRE", "LISTING", "XMP", "HEAD", NULL,
322: "UL", "P", "HEAD", "OL", "MENU", "DIR", "ADDRESS", "PRE",
323: "LISTING", "XMP", NULL,
324: "OL", "P", "HEAD", "UL", NULL,
325: "MENU", "P", "HEAD", "UL", NULL,
326: "P", "P", "HEAD", "H1", "H2", "H3", "H4", "H5", "H6", NULL,
327: "DIV", "P", "HEAD", NULL,
328: "NOSCRIPT", "P", "HEAD", NULL,
329: "CENTER", "FONT", "B", "I", "P", "HEAD", NULL,
330: "A", "A", NULL,
331: "CAPTION", "P", NULL,
332: "COLGROUP", "CAPTION", "COLGROUP", "COL", "P", NULL,
333: "COL", "CAPTION", "COL", "P", NULL,
334: "TABLE", "P", "HEAD", "H1", "H2", "H3", "H4", "H5", "H6", "PRE",
335: "LISTING", "XMP", "A", NULL,
336: "TH", "TH", "TD", NULL,
1.16 ! daniel 337: "TD", "TH", "TD", "P", NULL,
! 338: "TR", "TH", "TD", "TR", "CAPTION", "COL", "COLGROUP", "P", NULL,
1.1 daniel 339: "THEAD", "CAPTION", "COL", "COLGROUP", NULL,
340: "TFOOT", "TH", "TD", "TR", "CAPTION", "COL", "COLGROUP", "THEAD",
1.16 ! daniel 341: "TBODY", "P", NULL,
1.1 daniel 342: "TBODY", "TH", "TD", "TR", "CAPTION", "COL", "COLGROUP", "THEAD",
1.16 ! daniel 343: "TFOOT", "TBODY", "P", NULL,
1.1 daniel 344: "OPTGROUP", "OPTION", NULL,
345: "FIELDSET", "LEGEND", "P", "HEAD", "H1", "H2", "H3", "H4", "H5", "H6",
346: "PRE", "LISTING", "XMP", "A", NULL,
347: NULL
348: };
349:
1.8 daniel 350: static char** htmlStartCloseIndex[100];
1.1 daniel 351: static int htmlStartCloseIndexinitialized = 0;
352:
353: /************************************************************************
354: * *
355: * functions to handle HTML specific data *
356: * *
357: ************************************************************************/
358:
359: /**
360: * htmlInitAutoClose:
361: *
362: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
363: *
364: */
365: void
366: htmlInitAutoClose(void) {
367: int index, i = 0;
368:
369: if (htmlStartCloseIndexinitialized) return;
370:
371: for (index = 0;index < 100;index ++) htmlStartCloseIndex[index] = NULL;
372: index = 0;
373: while ((htmlStartClose[i] != NULL) && (index < 100 - 1)) {
374: htmlStartCloseIndex[index++] = &htmlStartClose[i];
375: while (htmlStartClose[i] != NULL) i++;
376: i++;
377: }
378: }
379:
380: /**
381: * htmlTagLookup:
382: * @tag: The tag name
383: *
384: * Lookup the HTML tag in the ElementTable
385: *
386: * Returns the related htmlElemDescPtr or NULL if not found.
387: */
388: htmlElemDescPtr
1.14 daniel 389: htmlTagLookup(const xmlChar *tag) {
1.1 daniel 390: int i = 0;
391:
392: for (i = 0; i < (sizeof(html40ElementTable) /
393: sizeof(html40ElementTable[0]));i++) {
1.8 daniel 394: if (!xmlStrcmp(tag, BAD_CAST html40ElementTable[i].name))
1.1 daniel 395: return(&html40ElementTable[i]);
396: }
397: return(NULL);
398: }
399:
400: /**
401: * htmlCheckAutoClose:
402: * @new: The new tag name
403: * @old: The old tag name
404: *
405: * Checks wether the new tag is one of the registered valid tags for closing old.
406: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
407: *
408: * Returns 0 if no, 1 if yes.
409: */
410: int
1.14 daniel 411: htmlCheckAutoClose(const xmlChar *new, const xmlChar *old) {
1.1 daniel 412: int i, index;
1.8 daniel 413: char **close;
1.1 daniel 414:
415: if (htmlStartCloseIndexinitialized == 0) htmlInitAutoClose();
416:
417: /* inefficient, but not a big deal */
418: for (index = 0; index < 100;index++) {
419: close = htmlStartCloseIndex[index];
420: if (close == NULL) return(0);
1.8 daniel 421: if (!xmlStrcmp(BAD_CAST *close, new)) break;
1.1 daniel 422: }
423:
424: i = close - htmlStartClose;
425: i++;
426: while (htmlStartClose[i] != NULL) {
1.8 daniel 427: if (!xmlStrcmp(BAD_CAST htmlStartClose[i], old)) {
1.1 daniel 428: return(1);
429: }
430: i++;
431: }
432: return(0);
433: }
434:
435: /**
436: * htmlAutoClose:
437: * @ctxt: an HTML parser context
438: * @new: The new tag name
439: *
440: * The HTmL DtD allows a tag to implicitely close other tags.
441: * The list is kept in htmlStartClose array. This function is
442: * called when a new tag has been detected and generates the
443: * appropriates closes if possible/needed.
444: */
445: void
1.14 daniel 446: htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar *new) {
1.15 daniel 447: xmlChar *oldname;
1.1 daniel 448:
1.15 daniel 449: while ((ctxt->name != NULL) &&
450: (htmlCheckAutoClose(new, ctxt->name))) {
1.1 daniel 451: #ifdef DEBUG
1.15 daniel 452: printf("htmlAutoClose: %s closes %s\n", new, ctxt->name);
1.1 daniel 453: #endif
454: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1.15 daniel 455: ctxt->sax->endElement(ctxt->userData, ctxt->name);
456: oldname = ctxt->name;
457: htmlnamePop(ctxt);
458: xmlFree(oldname);
1.1 daniel 459: }
460: }
461:
462: /**
463: * htmlAutoCloseOnClose:
464: * @ctxt: an HTML parser context
465: * @new: The new tag name
466: *
467: * The HTmL DtD allows an ending tag to implicitely close other tags.
468: */
469: void
1.14 daniel 470: htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar *new) {
1.1 daniel 471: htmlElemDescPtr info;
1.15 daniel 472: xmlChar *oldname;
1.1 daniel 473:
1.15 daniel 474: while ((ctxt->name != NULL) &&
475: (xmlStrcmp(new, ctxt->name))) {
476: info = htmlTagLookup(ctxt->name);
1.1 daniel 477: if ((info == NULL) || (info->endTag == 1)) {
478: #ifdef DEBUG
1.15 daniel 479: printf("htmlAutoCloseOnClose: %s closes %s\n", new, ctxt->name);
1.1 daniel 480: #endif
481: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1.15 daniel 482: ctxt->sax->endElement(ctxt->userData, ctxt->name);
483: oldname = ctxt->name;
484: htmlnamePop(ctxt);
485: xmlFree(oldname);
1.1 daniel 486: } else
487: break;
488: }
489: }
490:
491: /************************************************************************
492: * *
493: * The list of HTML predefined entities *
494: * *
495: ************************************************************************/
496:
497:
498: htmlEntityDesc html40EntitiesTable[] = {
499: /*
500: * the 4 absolute ones,
501: */
502: { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
503: { 38, "amp", "ampersand, U+0026 ISOnum" },
1.4 daniel 504: { 39, "apos", "single quote" },
1.1 daniel 505: { 60, "lt", "less-than sign, U+003C ISOnum" },
506: { 62, "gt", "greater-than sign, U+003E ISOnum" },
507:
508: /*
509: * A bunch still in the 128-255 range
510: * Replacing them depend really on the charset used.
511: */
512: { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
513: { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
514: { 162, "cent", "cent sign, U+00A2 ISOnum" },
515: { 163, "pound","pound sign, U+00A3 ISOnum" },
516: { 164, "curren","currency sign, U+00A4 ISOnum" },
517: { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
518: { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
519: { 167, "sect", "section sign, U+00A7 ISOnum" },
520: { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
521: { 169, "copy", "copyright sign, U+00A9 ISOnum" },
522: { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
523: { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
524: { 172, "not", "not sign, U+00AC ISOnum" },
525: { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
526: { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
527: { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
528: { 176, "deg", "degree sign, U+00B0 ISOnum" },
529: { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
530: { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
531: { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
532: { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
533: { 181, "micro","micro sign, U+00B5 ISOnum" },
534: { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1.7 daniel 535: { 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1.1 daniel 536: { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
537: { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
538: { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1.7 daniel 539: { 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1.1 daniel 540: { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
541: { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
542: { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
543: { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
544: { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
545: { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
546: { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
547: { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
548: { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
549: { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
550: { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
551: { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
552: { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
553: { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
554: { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
555: { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
556: { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
557: { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
558: { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
559: { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
560: { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
561: { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
562: { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
563: { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
564: { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
565: { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
566: { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
567: { 215, "times","multiplication sign, U+00D7 ISOnum" },
1.7 daniel 568: { 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1.1 daniel 569: { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
570: { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
571: { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
572: { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
573: { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
574: { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
575: { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
576: { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
577: { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
578: { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
579: { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
580: { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
581: { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
582: { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
583: { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
584: { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
585: { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
586: { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
587: { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
588: { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
589: { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
590: { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
591: { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
592: { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
593: { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
594: { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
595: { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
596: { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
597: { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
598: { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
599: { 247, "divide","division sign, U+00F7 ISOnum" },
600: { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
601: { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
602: { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
603: { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
604: { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
605: { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
606: { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
607: { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
608:
609: /*
610: * Anything below should really be kept as entities references
611: */
612: { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
613:
614: { 913, "Alpha","greek capital letter alpha, U+0391" },
615: { 914, "Beta", "greek capital letter beta, U+0392" },
616: { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
617: { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
618: { 917, "Epsilon","greek capital letter epsilon, U+0395" },
619: { 918, "Zeta", "greek capital letter zeta, U+0396" },
620: { 919, "Eta", "greek capital letter eta, U+0397" },
621: { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
622: { 921, "Iota", "greek capital letter iota, U+0399" },
623: { 922, "Kappa","greek capital letter kappa, U+039A" },
624: { 923, "Lambda""greek capital letter lambda, U+039B ISOgrk3" },
625: { 924, "Mu", "greek capital letter mu, U+039C" },
626: { 925, "Nu", "greek capital letter nu, U+039D" },
627: { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
628: { 927, "Omicron","greek capital letter omicron, U+039F" },
629: { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
630: { 929, "Rho", "greek capital letter rho, U+03A1" },
631: { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
632: { 932, "Tau", "greek capital letter tau, U+03A4" },
633: { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
634: { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
635: { 935, "Chi", "greek capital letter chi, U+03A7" },
636: { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
637: { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
638:
639: { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
640: { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
641: { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
642: { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
643: { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
644: { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
645: { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
646: { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
647: { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
648: { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
649: { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
650: { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
651: { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
652: { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
653: { 959, "omicron","greek small letter omicron, U+03BF NEW" },
654: { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
655: { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
656: { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
657: { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
658: { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
659: { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
660: { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
661: { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
662: { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
663: { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
664: { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
665: { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
666: { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
667:
668: { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
669: { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
670: { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
671: { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
672: { 8254, "oline","overline = spacing overscore, U+203E NEW" },
673: { 8260, "frasl","fraction slash, U+2044 NEW" },
674:
1.7 daniel 675: { 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
1.1 daniel 676: { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
677: { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
678: { 8482, "trade","trade mark sign, U+2122 ISOnum" },
679: { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
680: { 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
681: { 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
682: { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
683: { 8595, "darr", "downwards arrow, U+2193 ISOnum" },
684: { 8596, "harr", "left right arrow, U+2194 ISOamsa" },
685: { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
686: { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
687: { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
688: { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
689: { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
690: { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
691:
692:
693: { 8704, "forall","for all, U+2200 ISOtech" },
694: { 8706, "part", "partial differential, U+2202 ISOtech" },
695: { 8707, "exist","there exists, U+2203 ISOtech" },
696: { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
697: { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
698: { 8712, "isin", "element of, U+2208 ISOtech" },
699: { 8713, "notin","not an element of, U+2209 ISOtech" },
700: { 8715, "ni", "contains as member, U+220B ISOtech" },
701: { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
702: { 8721, "sum", "n-ary sumation, U+2211 ISOamsb" },
703: { 8722, "minus","minus sign, U+2212 ISOtech" },
704: { 8727, "lowast","asterisk operator, U+2217 ISOtech" },
705: { 8730, "radic","square root = radical sign, U+221A ISOtech" },
706: { 8733, "prop", "proportional to, U+221D ISOtech" },
707: { 8734, "infin","infinity, U+221E ISOtech" },
708: { 8736, "ang", "angle, U+2220 ISOamso" },
709: { 8743, "and", "logical and = wedge, U+2227 ISOtech" },
710: { 8744, "or", "logical or = vee, U+2228 ISOtech" },
711: { 8745, "cap", "intersection = cap, U+2229 ISOtech" },
712: { 8746, "cup", "union = cup, U+222A ISOtech" },
713: { 8747, "int", "integral, U+222B ISOtech" },
714: { 8756, "there4","therefore, U+2234 ISOtech" },
715: { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
716: { 8773, "cong", "approximately equal to, U+2245 ISOtech" },
717: { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
718: { 8800, "ne", "not equal to, U+2260 ISOtech" },
719: { 8801, "equiv","identical to, U+2261 ISOtech" },
720: { 8804, "le", "less-than or equal to, U+2264 ISOtech" },
721: { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
722: { 8834, "sub", "subset of, U+2282 ISOtech" },
723: { 8835, "sup", "superset of, U+2283 ISOtech" },
724: { 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
725: { 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
726: { 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
727: { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
728: { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
729: { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
730: { 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
731: { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
732: { 8969, "rceil","right ceiling, U+2309 ISOamsc" },
733: { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
734: { 8971, "rfloor","right floor, U+230B ISOamsc" },
735: { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
736: { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
737: { 9674, "loz", "lozenge, U+25CA ISOpub" },
738:
739: { 9824, "spades","black spade suit, U+2660 ISOpub" },
740: { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
741: { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
742: { 9830, "diams","black diamond suit, U+2666 ISOpub" },
743:
744: { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
745: { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
746: { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
747: { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
748: { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
749: { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
750: { 732, "tilde","small tilde, U+02DC ISOdia" },
751:
752: { 8194, "ensp", "en space, U+2002 ISOpub" },
753: { 8195, "emsp", "em space, U+2003 ISOpub" },
754: { 8201, "thinsp","thin space, U+2009 ISOpub" },
755: { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
756: { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
757: { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
758: { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
759: { 8211, "ndash","en dash, U+2013 ISOpub" },
760: { 8212, "mdash","em dash, U+2014 ISOpub" },
761: { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
762: { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
763: { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
764: { 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
765: { 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
766: { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
767: { 8224, "dagger","dagger, U+2020 ISOpub" },
768: { 8225, "Dagger","double dagger, U+2021 ISOpub" },
769: { 8240, "permil","per mille sign, U+2030 ISOtech" },
770: { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
1.7 daniel 771: { 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
1.1 daniel 772: { 8364, "euro", "euro sign, U+20AC NEW" }
773: };
774:
775: /************************************************************************
776: * *
777: * Commodity functions to handle entities *
778: * *
779: ************************************************************************/
780:
781: /*
782: * Macro used to grow the current buffer.
783: */
784: #define growBuffer(buffer) { \
785: buffer##_size *= 2; \
1.14 daniel 786: buffer = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
1.1 daniel 787: if (buffer == NULL) { \
788: perror("realloc failed"); \
789: exit(1); \
790: } \
791: }
792:
793: /**
794: * htmlEntityLookup:
795: * @name: the entity name
796: *
797: * Lookup the given entity in EntitiesTable
798: *
799: * TODO: the linear scan is really ugly, an hash table is really needed.
800: *
801: * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
802: */
803: htmlEntityDescPtr
1.14 daniel 804: htmlEntityLookup(const xmlChar *name) {
1.1 daniel 805: int i;
806:
807: for (i = 0;i < (sizeof(html40EntitiesTable)/
808: sizeof(html40EntitiesTable[0]));i++) {
1.8 daniel 809: if (!xmlStrcmp(name, BAD_CAST html40EntitiesTable[i].name)) {
1.1 daniel 810: #ifdef DEBUG
811: printf("Found entity %s\n", name);
812: #endif
813: return(&html40EntitiesTable[i]);
814: }
815: }
816: return(NULL);
817: }
818:
819:
820: /**
821: * htmlDecodeEntities:
822: * @ctxt: the parser context
823: * @len: the len to decode (in bytes !), -1 for no size limit
1.14 daniel 824: * @end: an end marker xmlChar, 0 if none
825: * @end2: an end marker xmlChar, 0 if none
826: * @end3: an end marker xmlChar, 0 if none
1.1 daniel 827: *
828: * Subtitute the HTML entities by their value
829: *
830: * TODO: once the internal representation will be UTF-8, all entities
831: * will be substituable, in the meantime we only apply the substitution
832: * to the one with values in the 0-255 UNICODE range
833: *
834: * Returns A newly allocated string with the substitution done. The caller
835: * must deallocate it !
836: */
1.14 daniel 837: xmlChar *
1.1 daniel 838: htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
1.14 daniel 839: xmlChar end, xmlChar end2, xmlChar end3) {
840: xmlChar *buffer = NULL;
1.1 daniel 841: int buffer_size = 0;
1.14 daniel 842: xmlChar *out = NULL;
843: xmlChar *name = NULL;
1.1 daniel 844:
1.14 daniel 845: xmlChar *cur = NULL;
1.1 daniel 846: htmlEntityDescPtr ent;
1.5 daniel 847: int nbchars = 0;
1.1 daniel 848: unsigned int max = (unsigned int) len;
849:
850: /*
851: * allocate a translation buffer.
852: */
853: buffer_size = 1000;
1.14 daniel 854: buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
1.1 daniel 855: if (buffer == NULL) {
856: perror("htmlDecodeEntities: malloc failed");
857: return(NULL);
858: }
859: out = buffer;
860:
861: /*
862: * Ok loop until we reach one of the ending char or a size limit.
863: */
1.5 daniel 864: while ((nbchars < max) && (CUR != end) &&
1.1 daniel 865: (CUR != end2) && (CUR != end3)) {
866:
867: if (CUR == '&') {
868: if (NXT(1) == '#') {
869: int val = htmlParseCharRef(ctxt);
1.8 daniel 870: /* invalid for UTF-8 variable encoding !!!!! */
1.1 daniel 871: *out++ = val;
1.5 daniel 872: nbchars += 3; /* !!!! */
1.1 daniel 873: } else {
874: ent = htmlParseEntityRef(ctxt, &name);
875: if (name != NULL) {
876: if ((ent == NULL) || (ent->value <= 0) ||
877: (ent->value >= 255)) {
878: *out++ = '&';
879: cur = name;
880: while (*cur != 0) {
881: if (out - buffer > buffer_size - 100) {
882: int index = out - buffer;
883:
884: growBuffer(buffer);
885: out = &buffer[index];
886: }
887: *out++ = *cur++;
888: }
889: *out++ = ';';
890: } else {
1.8 daniel 891: /* invalid for UTF-8 variable encoding !!!!! */
1.14 daniel 892: *out++ = (xmlChar)ent->value;
1.1 daniel 893: if (out - buffer > buffer_size - 100) {
894: int index = out - buffer;
895:
896: growBuffer(buffer);
897: out = &buffer[index];
898: }
899: }
1.5 daniel 900: nbchars += 2 + xmlStrlen(name);
1.11 daniel 901: xmlFree(name);
1.1 daniel 902: }
903: }
904: } else {
1.8 daniel 905: /* invalid for UTF-8 , use COPY(out); !!!!! */
1.1 daniel 906: *out++ = CUR;
1.5 daniel 907: nbchars++;
1.1 daniel 908: if (out - buffer > buffer_size - 100) {
909: int index = out - buffer;
910:
911: growBuffer(buffer);
912: out = &buffer[index];
913: }
914: NEXT;
915: }
916: }
917: *out++ = 0;
918: return(buffer);
919: }
920:
921:
922: /************************************************************************
923: * *
924: * Commodity functions to handle encodings *
925: * *
926: ************************************************************************/
927:
928: /**
929: * htmlSwitchEncoding:
930: * @ctxt: the parser context
931: * @len: the len of @cur
932: *
933: * change the input functions when discovering the character encoding
934: * of a given entity.
935: *
936: */
937: void
938: htmlSwitchEncoding(htmlParserCtxtPtr ctxt, xmlCharEncoding enc)
939: {
940: switch (enc) {
941: case XML_CHAR_ENCODING_ERROR:
942: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
943: ctxt->sax->error(ctxt->userData, "encoding unknown\n");
944: ctxt->wellFormed = 0;
945: break;
946: case XML_CHAR_ENCODING_NONE:
947: /* let's assume it's UTF-8 without the XML decl */
948: return;
949: case XML_CHAR_ENCODING_UTF8:
950: /* default encoding, no conversion should be needed */
951: return;
952: case XML_CHAR_ENCODING_UTF16LE:
953: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
954: ctxt->sax->error(ctxt->userData,
955: "char encoding UTF16 little endian not supported\n");
956: break;
957: case XML_CHAR_ENCODING_UTF16BE:
958: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
959: ctxt->sax->error(ctxt->userData,
960: "char encoding UTF16 big endian not supported\n");
961: break;
962: case XML_CHAR_ENCODING_UCS4LE:
963: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
964: ctxt->sax->error(ctxt->userData,
965: "char encoding USC4 little endian not supported\n");
966: break;
967: case XML_CHAR_ENCODING_UCS4BE:
968: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
969: ctxt->sax->error(ctxt->userData,
970: "char encoding USC4 big endian not supported\n");
971: break;
972: case XML_CHAR_ENCODING_EBCDIC:
973: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
974: ctxt->sax->error(ctxt->userData,
975: "char encoding EBCDIC not supported\n");
976: break;
977: case XML_CHAR_ENCODING_UCS4_2143:
978: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
979: ctxt->sax->error(ctxt->userData,
980: "char encoding UCS4 2143 not supported\n");
981: break;
982: case XML_CHAR_ENCODING_UCS4_3412:
983: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
984: ctxt->sax->error(ctxt->userData,
985: "char encoding UCS4 3412 not supported\n");
986: break;
987: case XML_CHAR_ENCODING_UCS2:
988: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
989: ctxt->sax->error(ctxt->userData,
990: "char encoding UCS2 not supported\n");
991: break;
992: case XML_CHAR_ENCODING_8859_1:
993: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
994: ctxt->sax->error(ctxt->userData,
995: "char encoding ISO_8859_1 ISO Latin 1 not supported\n");
996: break;
997: case XML_CHAR_ENCODING_8859_2:
998: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
999: ctxt->sax->error(ctxt->userData,
1000: "char encoding ISO_8859_2 ISO Latin 2 not supported\n");
1001: break;
1002: case XML_CHAR_ENCODING_8859_3:
1003: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1004: ctxt->sax->error(ctxt->userData,
1005: "char encoding ISO_8859_3 not supported\n");
1006: break;
1007: case XML_CHAR_ENCODING_8859_4:
1008: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1009: ctxt->sax->error(ctxt->userData,
1010: "char encoding ISO_8859_4 not supported\n");
1011: break;
1012: case XML_CHAR_ENCODING_8859_5:
1013: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1014: ctxt->sax->error(ctxt->userData,
1015: "char encoding ISO_8859_5 not supported\n");
1016: break;
1017: case XML_CHAR_ENCODING_8859_6:
1018: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1019: ctxt->sax->error(ctxt->userData,
1020: "char encoding ISO_8859_6 not supported\n");
1021: break;
1022: case XML_CHAR_ENCODING_8859_7:
1023: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1024: ctxt->sax->error(ctxt->userData,
1025: "char encoding ISO_8859_7 not supported\n");
1026: break;
1027: case XML_CHAR_ENCODING_8859_8:
1028: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1029: ctxt->sax->error(ctxt->userData,
1030: "char encoding ISO_8859_8 not supported\n");
1031: break;
1032: case XML_CHAR_ENCODING_8859_9:
1033: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1034: ctxt->sax->error(ctxt->userData,
1035: "char encoding ISO_8859_9 not supported\n");
1036: break;
1037: case XML_CHAR_ENCODING_2022_JP:
1038: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1039: ctxt->sax->error(ctxt->userData,
1040: "char encoding ISO-2022-JPnot supported\n");
1041: break;
1042: case XML_CHAR_ENCODING_SHIFT_JIS:
1043: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1044: ctxt->sax->error(ctxt->userData,
1045: "char encoding Shift_JISnot supported\n");
1046: break;
1047: case XML_CHAR_ENCODING_EUC_JP:
1048: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1049: ctxt->sax->error(ctxt->userData,
1050: "char encoding EUC-JPnot supported\n");
1051: break;
1052: }
1053: }
1054:
1055:
1056: /************************************************************************
1057: * *
1058: * Commodity functions, cleanup needed ? *
1059: * *
1060: ************************************************************************/
1061:
1062: /**
1063: * areBlanks:
1064: * @ctxt: an HTML parser context
1.14 daniel 1065: * @str: a xmlChar *
1.1 daniel 1066: * @len: the size of @str
1067: *
1068: * Is this a sequence of blank chars that one can ignore ?
1069: *
1070: * Returns 1 if ignorable 0 otherwise.
1071: */
1072:
1.14 daniel 1073: static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
1.1 daniel 1074: int i;
1075: xmlNodePtr lastChild;
1076:
1077: for (i = 0;i < len;i++)
1078: if (!(IS_BLANK(str[i]))) return(0);
1079:
1080: if (CUR != '<') return(0);
1081: if (ctxt->node == NULL) return(0);
1082: lastChild = xmlGetLastChild(ctxt->node);
1083: if (lastChild == NULL) {
1084: if (ctxt->node->content != NULL) return(0);
1085: } else if (xmlNodeIsText(lastChild))
1086: return(0);
1087: return(1);
1088: }
1089:
1090: /**
1091: * htmlHandleEntity:
1092: * @ctxt: an HTML parser context
1093: * @entity: an XML entity pointer.
1094: *
1095: * Default handling of an HTML entity, call the parser with the
1096: * substitution string
1097: */
1098:
1099: void
1100: htmlHandleEntity(htmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1101: int len;
1102:
1103: if (entity->content == NULL) {
1104: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1105: ctxt->sax->error(ctxt->userData, "htmlHandleEntity %s: content == NULL\n",
1106: entity->name);
1107: ctxt->wellFormed = 0;
1108: return;
1109: }
1110: len = xmlStrlen(entity->content);
1111:
1112: /*
1113: * Just handle the content as a set of chars.
1114: */
1115: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1116: ctxt->sax->characters(ctxt->userData, entity->content, len);
1117:
1118: }
1119:
1120: /**
1121: * htmlNewDoc:
1122: * @URI: URI for the dtd, or NULL
1123: * @ExternalID: the external ID of the DTD, or NULL
1124: *
1125: * Returns a new document
1126: */
1127: htmlDocPtr
1.14 daniel 1128: htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
1.1 daniel 1129: xmlDocPtr cur;
1130:
1131: /*
1132: * Allocate a new document and fill the fields.
1133: */
1.11 daniel 1134: cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
1.1 daniel 1135: if (cur == NULL) {
1136: fprintf(stderr, "xmlNewDoc : malloc failed\n");
1137: return(NULL);
1138: }
1.10 daniel 1139: memset(cur, 0, sizeof(xmlDoc));
1.1 daniel 1140:
1141: cur->type = XML_DOCUMENT_NODE;
1142: cur->version = NULL;
1143: cur->intSubset = NULL;
1.8 daniel 1144: xmlCreateIntSubset(cur, BAD_CAST "HTML", ExternalID, URI);
1.1 daniel 1145: cur->name = NULL;
1146: cur->root = NULL;
1147: cur->extSubset = NULL;
1148: cur->oldNs = NULL;
1149: cur->encoding = NULL;
1150: cur->standalone = 1;
1151: cur->compression = 0;
1.12 daniel 1152: cur->ids = NULL;
1153: cur->refs = NULL;
1.1 daniel 1154: #ifndef XML_WITHOUT_CORBA
1155: cur->_private = NULL;
1156: cur->vepv = NULL;
1157: #endif
1158: return(cur);
1159: }
1160:
1161:
1162: /************************************************************************
1163: * *
1164: * The parser itself *
1165: * Relates to http://www.w3.org/TR/html40 *
1166: * *
1167: ************************************************************************/
1168:
1169: /************************************************************************
1170: * *
1171: * The parser itself *
1172: * *
1173: ************************************************************************/
1174:
1175: /**
1176: * htmlParseHTMLName:
1177: * @ctxt: an HTML parser context
1178: *
1179: * parse an HTML tag or attribute name, note that we convert it to uppercase
1180: * since HTML names are not case-sensitive.
1181: *
1182: * Returns the Tag Name parsed or NULL
1183: */
1184:
1.14 daniel 1185: xmlChar *
1.1 daniel 1186: htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1187: xmlChar *ret = NULL;
1.1 daniel 1188: int i = 0;
1.14 daniel 1189: xmlChar loc[100];
1.1 daniel 1190:
1191: if (!IS_LETTER(CUR) && (CUR != '_') &&
1192: (CUR != ':')) return(NULL);
1193:
1194: while ((i < 100) && ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)))) {
1195: if ((CUR >= 0x61) && (CUR <= 0x7a)) loc[i] = CUR - 0x20;
1196: else loc[i] = CUR;
1197: i++;
1198:
1199: NEXT;
1200: }
1201:
1202: ret = xmlStrndup(loc, i);
1203:
1204: return(ret);
1205: }
1206:
1207: /**
1208: * htmlParseName:
1209: * @ctxt: an HTML parser context
1210: *
1211: * parse an HTML name, this routine is case sensistive.
1212: *
1213: * Returns the Name parsed or NULL
1214: */
1215:
1.14 daniel 1216: xmlChar *
1.1 daniel 1217: htmlParseName(htmlParserCtxtPtr ctxt) {
1.14 daniel 1218: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1219: int len = 0;
1.1 daniel 1220:
1.5 daniel 1221: GROW;
1222: if (!IS_LETTER(CUR) && (CUR != '_')) {
1223: return(NULL);
1224: }
1.1 daniel 1225:
1226: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1227: (CUR == '.') || (CUR == '-') ||
1228: (CUR == '_') || (CUR == ':') ||
1229: (IS_COMBINING(CUR)) ||
1.5 daniel 1230: (IS_EXTENDER(CUR))) {
1231: buf[len++] = CUR;
1.1 daniel 1232: NEXT;
1.5 daniel 1233: if (len >= HTML_MAX_NAMELEN) {
1234: fprintf(stderr,
1235: "htmlParseName: reached HTML_MAX_NAMELEN limit\n");
1236: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1237: (CUR == '.') || (CUR == '-') ||
1238: (CUR == '_') || (CUR == ':') ||
1239: (IS_COMBINING(CUR)) ||
1240: (IS_EXTENDER(CUR)))
1241: NEXT;
1242: break;
1243: }
1244: }
1245: return(xmlStrndup(buf, len));
1.1 daniel 1246: }
1247:
1248: /**
1249: * htmlParseHTMLAttribute:
1250: * @ctxt: an HTML parser context
1251: *
1.5 daniel 1252: * parse an HTML attribute value (without quotes).
1.1 daniel 1253: *
1254: * Returns the Nmtoken parsed or NULL
1255: */
1256:
1.14 daniel 1257: xmlChar *
1.1 daniel 1258: htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt) {
1.14 daniel 1259: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1260: int len = 0;
1.1 daniel 1261:
1.5 daniel 1262: GROW;
1.1 daniel 1263: while ((!IS_BLANK(CUR)) && (CUR != '<') &&
1264: (CUR != '&') && (CUR != '>') &&
1.5 daniel 1265: (CUR != '\'') && (CUR != '"')) {
1266: buf[len++] = CUR;
1.1 daniel 1267: NEXT;
1.5 daniel 1268: if (len >= HTML_MAX_NAMELEN) {
1269: fprintf(stderr,
1270: "htmlParseHTMLAttribute: reached HTML_MAX_NAMELEN limit\n");
1271: while ((!IS_BLANK(CUR)) && (CUR != '<') &&
1272: (CUR != '&') && (CUR != '>') &&
1273: (CUR != '\'') && (CUR != '"'))
1274: NEXT;
1275: break;
1276: }
1277: }
1278: return(xmlStrndup(buf, len));
1.1 daniel 1279: }
1280:
1281: /**
1282: * htmlParseNmtoken:
1283: * @ctxt: an HTML parser context
1284: *
1285: * parse an HTML Nmtoken.
1286: *
1287: * Returns the Nmtoken parsed or NULL
1288: */
1289:
1.14 daniel 1290: xmlChar *
1.1 daniel 1291: htmlParseNmtoken(htmlParserCtxtPtr ctxt) {
1.14 daniel 1292: xmlChar buf[HTML_MAX_NAMELEN];
1.5 daniel 1293: int len = 0;
1.1 daniel 1294:
1.5 daniel 1295: GROW;
1.1 daniel 1296: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1297: (CUR == '.') || (CUR == '-') ||
1298: (CUR == '_') || (CUR == ':') ||
1299: (IS_COMBINING(CUR)) ||
1.5 daniel 1300: (IS_EXTENDER(CUR))) {
1301: buf[len++] = CUR;
1.1 daniel 1302: NEXT;
1.5 daniel 1303: if (len >= HTML_MAX_NAMELEN) {
1304: fprintf(stderr,
1305: "htmlParseNmtoken: reached HTML_MAX_NAMELEN limit\n");
1306: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1307: (CUR == '.') || (CUR == '-') ||
1308: (CUR == '_') || (CUR == ':') ||
1309: (IS_COMBINING(CUR)) ||
1310: (IS_EXTENDER(CUR)))
1311: NEXT;
1312: break;
1313: }
1314: }
1315: return(xmlStrndup(buf, len));
1.1 daniel 1316: }
1317:
1318: /**
1319: * htmlParseEntityRef:
1320: * @ctxt: an HTML parser context
1321: * @str: location to store the entity name
1322: *
1323: * parse an HTML ENTITY references
1324: *
1325: * [68] EntityRef ::= '&' Name ';'
1326: *
1327: * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
1328: * if non-NULL *str will have to be freed by the caller.
1329: */
1330: htmlEntityDescPtr
1.14 daniel 1331: htmlParseEntityRef(htmlParserCtxtPtr ctxt, xmlChar **str) {
1332: xmlChar *name;
1.1 daniel 1333: htmlEntityDescPtr ent = NULL;
1334: *str = NULL;
1335:
1336: if (CUR == '&') {
1337: NEXT;
1338: name = htmlParseName(ctxt);
1339: if (name == NULL) {
1340: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1341: ctxt->sax->error(ctxt->userData, "htmlParseEntityRef: no name\n");
1342: ctxt->wellFormed = 0;
1343: } else {
1.5 daniel 1344: GROW;
1.1 daniel 1345: if (CUR == ';') {
1346: NEXT;
1347: *str = name;
1348:
1349: /*
1350: * Lookup the entity in the table.
1351: */
1352: ent = htmlEntityLookup(name);
1353: } else {
1354: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1355: ctxt->sax->error(ctxt->userData,
1356: "htmlParseEntityRef: expecting ';'\n");
1357: ctxt->wellFormed = 0;
1358: if (ctxt->sax->characters != NULL) {
1.8 daniel 1359: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
1.1 daniel 1360: ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
1361: }
1.11 daniel 1362: xmlFree(name);
1.1 daniel 1363: }
1364: }
1365: }
1366: return(ent);
1367: }
1368:
1369: /**
1370: * htmlParseAttValue:
1371: * @ctxt: an HTML parser context
1372: *
1373: * parse a value for an attribute
1374: * Note: the parser won't do substitution of entities here, this
1375: * will be handled later in xmlStringGetNodeList, unless it was
1376: * asked for ctxt->replaceEntities != 0
1377: *
1378: * Returns the AttValue parsed or NULL.
1379: */
1380:
1.14 daniel 1381: xmlChar *
1.1 daniel 1382: htmlParseAttValue(htmlParserCtxtPtr ctxt) {
1.14 daniel 1383: xmlChar *ret = NULL;
1.1 daniel 1384:
1385: if (CUR == '"') {
1386: NEXT;
1387: ret = htmlDecodeEntities(ctxt, -1, '"', '<', 0);
1388: if (CUR == '<') {
1389: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1390: ctxt->sax->error(ctxt->userData,
1391: "Unescaped '<' not allowed in attributes values\n");
1392: ctxt->wellFormed = 0;
1393: }
1394: if (CUR != '"') {
1395: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1396: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
1397: ctxt->wellFormed = 0;
1398: } else
1399: NEXT;
1400: } else if (CUR == '\'') {
1401: NEXT;
1402: ret = htmlDecodeEntities(ctxt, -1, '\'', '<', 0);
1403: if (CUR == '<') {
1404: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1405: ctxt->sax->error(ctxt->userData,
1406: "Unescaped '<' not allowed in attributes values\n");
1407: ctxt->wellFormed = 0;
1408: }
1409: if (CUR != '\'') {
1410: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1411: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
1412: ctxt->wellFormed = 0;
1413: } else
1414: NEXT;
1415: } else {
1416: /*
1417: * That's an HTMLism, the attribute value may not be quoted
1418: */
1419: ret = htmlParseHTMLAttribute(ctxt);
1420: if (ret == NULL) {
1421: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1422: ctxt->sax->error(ctxt->userData, "AttValue: no value found\n");
1423: ctxt->wellFormed = 0;
1424: }
1425: }
1426:
1427: return(ret);
1428: }
1429:
1430: /**
1431: * htmlParseSystemLiteral:
1432: * @ctxt: an HTML parser context
1433: *
1434: * parse an HTML Literal
1435: *
1436: * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
1437: *
1438: * Returns the SystemLiteral parsed or NULL
1439: */
1440:
1.14 daniel 1441: xmlChar *
1.1 daniel 1442: htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 1443: const xmlChar *q;
1444: xmlChar *ret = NULL;
1.1 daniel 1445:
1446: if (CUR == '"') {
1447: NEXT;
1448: q = CUR_PTR;
1449: while ((IS_CHAR(CUR)) && (CUR != '"'))
1450: NEXT;
1451: if (!IS_CHAR(CUR)) {
1452: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1453: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
1454: ctxt->wellFormed = 0;
1455: } else {
1456: ret = xmlStrndup(q, CUR_PTR - q);
1457: NEXT;
1458: }
1459: } else if (CUR == '\'') {
1460: NEXT;
1461: q = CUR_PTR;
1462: while ((IS_CHAR(CUR)) && (CUR != '\''))
1463: NEXT;
1464: if (!IS_CHAR(CUR)) {
1465: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1466: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
1467: ctxt->wellFormed = 0;
1468: } else {
1469: ret = xmlStrndup(q, CUR_PTR - q);
1470: NEXT;
1471: }
1472: } else {
1473: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1474: ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n");
1475: ctxt->wellFormed = 0;
1476: }
1477:
1478: return(ret);
1479: }
1480:
1481: /**
1482: * htmlParsePubidLiteral:
1483: * @ctxt: an HTML parser context
1484: *
1485: * parse an HTML public literal
1486: *
1487: * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
1488: *
1489: * Returns the PubidLiteral parsed or NULL.
1490: */
1491:
1.14 daniel 1492: xmlChar *
1.1 daniel 1493: htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
1.14 daniel 1494: const xmlChar *q;
1495: xmlChar *ret = NULL;
1.1 daniel 1496: /*
1497: * Name ::= (Letter | '_') (NameChar)*
1498: */
1499: if (CUR == '"') {
1500: NEXT;
1501: q = CUR_PTR;
1502: while (IS_PUBIDCHAR(CUR)) NEXT;
1503: if (CUR != '"') {
1504: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1505: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
1506: ctxt->wellFormed = 0;
1507: } else {
1508: ret = xmlStrndup(q, CUR_PTR - q);
1509: NEXT;
1510: }
1511: } else if (CUR == '\'') {
1512: NEXT;
1513: q = CUR_PTR;
1514: while ((IS_LETTER(CUR)) && (CUR != '\''))
1515: NEXT;
1516: if (!IS_LETTER(CUR)) {
1517: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1518: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
1519: ctxt->wellFormed = 0;
1520: } else {
1521: ret = xmlStrndup(q, CUR_PTR - q);
1522: NEXT;
1523: }
1524: } else {
1525: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1526: ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n");
1527: ctxt->wellFormed = 0;
1528: }
1529:
1530: return(ret);
1531: }
1532:
1533: /**
1534: * htmlParseCharData:
1535: * @ctxt: an HTML parser context
1536: * @cdata: int indicating whether we are within a CDATA section
1537: *
1538: * parse a CharData section.
1539: * if we are within a CDATA section ']]>' marks an end of section.
1540: *
1541: * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
1542: */
1543:
1544: void
1545: htmlParseCharData(htmlParserCtxtPtr ctxt, int cdata) {
1.14 daniel 1546: const xmlChar *q;
1.1 daniel 1547:
1548: q = CUR_PTR;
1549: while ((IS_CHAR(CUR)) && (CUR != '<') &&
1550: (CUR != '&')) {
1551: if ((CUR == ']') && (NXT(1) == ']') &&
1552: (NXT(2) == '>')) {
1553: if (cdata) break;
1554: else {
1555: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1556: ctxt->sax->error(ctxt->userData,
1557: "Sequence ']]>' not allowed in content\n");
1558: ctxt->wellFormed = 0;
1559: }
1560: }
1561: NEXT;
1562: }
1563: if (q == CUR_PTR) return;
1564:
1565: /*
1566: * Ok the segment [q CUR_PTR] is to be consumed as chars.
1567: */
1568: if (ctxt->sax != NULL) {
1569: if (areBlanks(ctxt, q, CUR_PTR - q)) {
1570: if (ctxt->sax->ignorableWhitespace != NULL)
1571: ctxt->sax->ignorableWhitespace(ctxt->userData, q, CUR_PTR - q);
1572: } else {
1573: if (ctxt->sax->characters != NULL)
1574: ctxt->sax->characters(ctxt->userData, q, CUR_PTR - q);
1575: }
1576: }
1577: }
1578:
1579: /**
1580: * htmlParseExternalID:
1581: * @ctxt: an HTML parser context
1.14 daniel 1582: * @publicID: a xmlChar** receiving PubidLiteral
1.1 daniel 1583: * @strict: indicate whether we should restrict parsing to only
1584: * production [75], see NOTE below
1585: *
1586: * Parse an External ID or a Public ID
1587: *
1588: * NOTE: Productions [75] and [83] interract badly since [75] can generate
1589: * 'PUBLIC' S PubidLiteral S SystemLiteral
1590: *
1591: * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
1592: * | 'PUBLIC' S PubidLiteral S SystemLiteral
1593: *
1594: * [83] PublicID ::= 'PUBLIC' S PubidLiteral
1595: *
1596: * Returns the function returns SystemLiteral and in the second
1597: * case publicID receives PubidLiteral, is strict is off
1598: * it is possible to return NULL and have publicID set.
1599: */
1600:
1.14 daniel 1601: xmlChar *
1602: htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID, int strict) {
1603: xmlChar *URI = NULL;
1.1 daniel 1604:
1605: if ((UPPER == 'S') && (UPP(1) == 'Y') &&
1606: (UPP(2) == 'S') && (UPP(3) == 'T') &&
1607: (UPP(4) == 'E') && (UPP(5) == 'M')) {
1608: SKIP(6);
1609: if (!IS_BLANK(CUR)) {
1610: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1611: ctxt->sax->error(ctxt->userData,
1612: "Space required after 'SYSTEM'\n");
1613: ctxt->wellFormed = 0;
1614: }
1615: SKIP_BLANKS;
1616: URI = htmlParseSystemLiteral(ctxt);
1617: if (URI == NULL) {
1618: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1619: ctxt->sax->error(ctxt->userData,
1620: "htmlParseExternalID: SYSTEM, no URI\n");
1621: ctxt->wellFormed = 0;
1622: }
1623: } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
1624: (UPP(2) == 'B') && (UPP(3) == 'L') &&
1625: (UPP(4) == 'I') && (UPP(5) == 'C')) {
1626: SKIP(6);
1627: if (!IS_BLANK(CUR)) {
1628: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1629: ctxt->sax->error(ctxt->userData,
1630: "Space required after 'PUBLIC'\n");
1631: ctxt->wellFormed = 0;
1632: }
1633: SKIP_BLANKS;
1634: *publicID = htmlParsePubidLiteral(ctxt);
1635: if (*publicID == NULL) {
1636: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1637: ctxt->sax->error(ctxt->userData,
1638: "htmlParseExternalID: PUBLIC, no Public Identifier\n");
1639: ctxt->wellFormed = 0;
1640: }
1.5 daniel 1641: SKIP_BLANKS;
1642: if ((CUR == '"') || (CUR == '\'')) {
1643: URI = htmlParseSystemLiteral(ctxt);
1.1 daniel 1644: }
1645: }
1646: return(URI);
1647: }
1648:
1649: /**
1650: * htmlParseComment:
1651: * @ctxt: an HTML parser context
1652: * @create: should we create a node, or just skip the content
1653: *
1654: * Parse an XML (SGML) comment <!-- .... -->
1655: *
1656: * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
1657: */
1658: void
1659: htmlParseComment(htmlParserCtxtPtr ctxt, int create) {
1.14 daniel 1660: const xmlChar *q, *start;
1661: const xmlChar *r;
1662: xmlChar *val;
1.1 daniel 1663:
1664: /*
1665: * Check that there is a comment right here.
1666: */
1667: if ((CUR != '<') || (NXT(1) != '!') ||
1668: (NXT(2) != '-') || (NXT(3) != '-')) return;
1669:
1670: SKIP(4);
1671: start = q = CUR_PTR;
1672: NEXT;
1673: r = CUR_PTR;
1674: NEXT;
1675: while (IS_CHAR(CUR) &&
1676: ((CUR == ':') || (CUR != '>') ||
1677: (*r != '-') || (*q != '-'))) {
1678: if ((*r == '-') && (*q == '-')) {
1679: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1680: ctxt->sax->error(ctxt->userData,
1681: "Comment must not contain '--' (double-hyphen)`\n");
1682: ctxt->wellFormed = 0;
1683: }
1684: NEXT;r++;q++;
1685: }
1686: if (!IS_CHAR(CUR)) {
1687: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1688: ctxt->sax->error(ctxt->userData, "Comment not terminated \n<!--%.50s\n", start);
1689: ctxt->wellFormed = 0;
1690: } else {
1691: NEXT;
1692: if (create) {
1693: val = xmlStrndup(start, q - start);
1694: if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL))
1695: ctxt->sax->comment(ctxt->userData, val);
1.11 daniel 1696: xmlFree(val);
1.1 daniel 1697: }
1698: }
1699: }
1700:
1701: /**
1702: * htmlParseCharRef:
1703: * @ctxt: an HTML parser context
1704: *
1705: * parse Reference declarations
1706: *
1707: * [66] CharRef ::= '&#' [0-9]+ ';' |
1708: * '&#x' [0-9a-fA-F]+ ';'
1709: *
1710: * Returns the value parsed (as an int)
1711: */
1712: int
1713: htmlParseCharRef(htmlParserCtxtPtr ctxt) {
1714: int val = 0;
1715:
1716: if ((CUR == '&') && (NXT(1) == '#') &&
1717: (NXT(2) == 'x')) {
1718: SKIP(3);
1719: while (CUR != ';') {
1720: if ((CUR >= '0') && (CUR <= '9'))
1721: val = val * 16 + (CUR - '0');
1722: else if ((CUR >= 'a') && (CUR <= 'f'))
1723: val = val * 16 + (CUR - 'a') + 10;
1724: else if ((CUR >= 'A') && (CUR <= 'F'))
1725: val = val * 16 + (CUR - 'A') + 10;
1726: else {
1727: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1728: ctxt->sax->error(ctxt->userData,
1729: "htmlParseCharRef: invalid hexadecimal value\n");
1730: ctxt->wellFormed = 0;
1731: val = 0;
1732: break;
1733: }
1734: NEXT;
1735: }
1736: if (CUR == ';')
1737: NEXT;
1738: } else if ((CUR == '&') && (NXT(1) == '#')) {
1739: SKIP(2);
1740: while (CUR != ';') {
1741: if ((CUR >= '0') && (CUR <= '9'))
1742: val = val * 10 + (CUR - '0');
1743: else {
1744: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1745: ctxt->sax->error(ctxt->userData,
1746: "htmlParseCharRef: invalid decimal value\n");
1747: ctxt->wellFormed = 0;
1748: val = 0;
1749: break;
1750: }
1751: NEXT;
1752: }
1753: if (CUR == ';')
1754: NEXT;
1755: } else {
1756: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1757: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid value\n");
1758: ctxt->wellFormed = 0;
1759: }
1760: /*
1761: * Check the value IS_CHAR ...
1762: */
1763: if (IS_CHAR(val)) {
1764: return(val);
1765: } else {
1766: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1.14 daniel 1767: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid xmlChar value %d\n",
1.1 daniel 1768: val);
1769: ctxt->wellFormed = 0;
1770: }
1771: return(0);
1772: }
1773:
1774:
1775: /**
1776: * htmlParseDocTypeDecl :
1777: * @ctxt: an HTML parser context
1778: *
1779: * parse a DOCTYPE declaration
1780: *
1781: * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
1782: * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
1783: */
1784:
1785: void
1786: htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
1.14 daniel 1787: xmlChar *name;
1788: xmlChar *ExternalID = NULL;
1789: xmlChar *URI = NULL;
1.1 daniel 1790:
1791: /*
1792: * We know that '<!DOCTYPE' has been detected.
1793: */
1794: SKIP(9);
1795:
1796: SKIP_BLANKS;
1797:
1798: /*
1799: * Parse the DOCTYPE name.
1800: */
1801: name = htmlParseName(ctxt);
1802: if (name == NULL) {
1803: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1804: ctxt->sax->error(ctxt->userData, "htmlParseDocTypeDecl : no DOCTYPE name !\n");
1805: ctxt->wellFormed = 0;
1806: }
1807: /*
1808: * Check that upper(name) == "HTML" !!!!!!!!!!!!!
1809: */
1810:
1811: SKIP_BLANKS;
1812:
1813: /*
1814: * Check for SystemID and ExternalID
1815: */
1.5 daniel 1816: URI = htmlParseExternalID(ctxt, &ExternalID, 0);
1.1 daniel 1817: SKIP_BLANKS;
1818:
1819: /*
1820: * We should be at the end of the DOCTYPE declaration.
1821: */
1822: if (CUR != '>') {
1823: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1824: ctxt->sax->error(ctxt->userData, "DOCTYPE unproperly terminated\n");
1825: ctxt->wellFormed = 0;
1826: /* We shouldn't try to resynchronize ... */
1827: } else {
1828: }
1829: NEXT;
1830:
1831: /*
1832: * Create the document accordingly to the DOCTYPE
1833: */
1834: ctxt->myDoc = htmlNewDoc(URI, ExternalID);
1835:
1836: /*
1837: * Cleanup, since we don't use all those identifiers
1838: */
1.11 daniel 1839: if (URI != NULL) xmlFree(URI);
1840: if (ExternalID != NULL) xmlFree(ExternalID);
1841: if (name != NULL) xmlFree(name);
1.1 daniel 1842: }
1843:
1844: /**
1845: * htmlParseAttribute:
1846: * @ctxt: an HTML parser context
1.14 daniel 1847: * @value: a xmlChar ** used to store the value of the attribute
1.1 daniel 1848: *
1849: * parse an attribute
1850: *
1851: * [41] Attribute ::= Name Eq AttValue
1852: *
1853: * [25] Eq ::= S? '=' S?
1854: *
1855: * With namespace:
1856: *
1857: * [NS 11] Attribute ::= QName Eq AttValue
1858: *
1859: * Also the case QName == xmlns:??? is handled independently as a namespace
1860: * definition.
1861: *
1862: * Returns the attribute name, and the value in *value.
1863: */
1864:
1.14 daniel 1865: xmlChar *
1866: htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
1867: xmlChar *name, *val;
1.1 daniel 1868:
1869: *value = NULL;
1870: name = htmlParseName(ctxt);
1871: if (name == NULL) {
1872: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1873: ctxt->sax->error(ctxt->userData, "error parsing attribute name\n");
1874: ctxt->wellFormed = 0;
1875: return(NULL);
1876: }
1877:
1878: /*
1879: * read the value
1880: */
1881: SKIP_BLANKS;
1882: if (CUR == '=') {
1883: NEXT;
1884: SKIP_BLANKS;
1885: val = htmlParseAttValue(ctxt);
1886: } else {
1887: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1888: ctxt->sax->error(ctxt->userData,
1889: "Specification mandate value for attribute %s\n", name);
1890: ctxt->wellFormed = 0;
1891: return(NULL);
1892: }
1893:
1894: *value = val;
1895: return(name);
1896: }
1897:
1898: /**
1899: * htmlParseStartTag:
1900: * @ctxt: an HTML parser context
1901: *
1902: * parse a start of tag either for rule element or
1903: * EmptyElement. In both case we don't parse the tag closing chars.
1904: *
1905: * [40] STag ::= '<' Name (S Attribute)* S? '>'
1906: *
1907: * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
1908: *
1909: * With namespace:
1910: *
1911: * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
1912: *
1913: * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
1914: *
1915: * Returns the element name parsed
1916: */
1917:
1.14 daniel 1918: xmlChar *
1.1 daniel 1919: htmlParseStartTag(htmlParserCtxtPtr ctxt) {
1.14 daniel 1920: xmlChar *name;
1921: xmlChar *attname;
1922: xmlChar *attvalue;
1923: const xmlChar **atts = NULL;
1.1 daniel 1924: int nbatts = 0;
1925: int maxatts = 0;
1926: int i;
1927:
1928: if (CUR != '<') return(NULL);
1929: NEXT;
1930:
1931: name = htmlParseHTMLName(ctxt);
1932: if (name == NULL) {
1933: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1934: ctxt->sax->error(ctxt->userData,
1935: "htmlParseStartTag: invalid element name\n");
1936: ctxt->wellFormed = 0;
1937: return(NULL);
1938: }
1939:
1940: /*
1941: * Check for auto-closure of HTML elements.
1942: */
1943: htmlAutoClose(ctxt, name);
1944:
1945: /*
1946: * Now parse the attributes, it ends up with the ending
1947: *
1948: * (S Attribute)* S?
1949: */
1950: SKIP_BLANKS;
1951: while ((IS_CHAR(CUR)) &&
1952: (CUR != '>') &&
1953: ((CUR != '/') || (NXT(1) != '>'))) {
1.14 daniel 1954: const xmlChar *q = CUR_PTR;
1.1 daniel 1955:
1956: attname = htmlParseAttribute(ctxt, &attvalue);
1957: if ((attname != NULL) && (attvalue != NULL)) {
1958: /*
1959: * Well formedness requires at most one declaration of an attribute
1960: */
1961: for (i = 0; i < nbatts;i += 2) {
1962: if (!xmlStrcmp(atts[i], attname)) {
1963: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1964: ctxt->sax->error(ctxt->userData, "Attribute %s redefined\n",
1965: name);
1966: ctxt->wellFormed = 0;
1.11 daniel 1967: xmlFree(attname);
1968: xmlFree(attvalue);
1.1 daniel 1969: break;
1970: }
1971: }
1972:
1973: /*
1974: * Add the pair to atts
1975: */
1976: if (atts == NULL) {
1977: maxatts = 10;
1.14 daniel 1978: atts = (const xmlChar **) xmlMalloc(maxatts * sizeof(xmlChar *));
1.1 daniel 1979: if (atts == NULL) {
1980: fprintf(stderr, "malloc of %ld byte failed\n",
1.14 daniel 1981: maxatts * (long)sizeof(xmlChar *));
1.1 daniel 1982: return(NULL);
1983: }
1984: } else if (nbatts + 2 < maxatts) {
1985: maxatts *= 2;
1.14 daniel 1986: atts = (const xmlChar **) xmlRealloc(atts, maxatts * sizeof(xmlChar *));
1.1 daniel 1987: if (atts == NULL) {
1988: fprintf(stderr, "realloc of %ld byte failed\n",
1.14 daniel 1989: maxatts * (long)sizeof(xmlChar *));
1.1 daniel 1990: return(NULL);
1991: }
1992: }
1993: atts[nbatts++] = attname;
1994: atts[nbatts++] = attvalue;
1995: atts[nbatts] = NULL;
1996: atts[nbatts + 1] = NULL;
1997: }
1998:
1999: SKIP_BLANKS;
2000: if (q == CUR_PTR) {
2001: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2002: ctxt->sax->error(ctxt->userData,
2003: "htmlParseStartTag: problem parsing attributes\n");
2004: ctxt->wellFormed = 0;
2005: break;
2006: }
2007: }
2008:
2009: /*
2010: * SAX: Start of Element !
2011: */
1.15 daniel 2012: htmlnamePush(ctxt, xmlStrdup(name));
1.1 daniel 2013: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
2014: ctxt->sax->startElement(ctxt->userData, name, atts);
2015:
2016: if (atts != NULL) {
1.14 daniel 2017: for (i = 0;i < nbatts;i++) xmlFree((xmlChar *) atts[i]);
1.11 daniel 2018: xmlFree(atts);
1.1 daniel 2019: }
2020: return(name);
2021: }
2022:
2023: /**
2024: * htmlParseEndTag:
2025: * @ctxt: an HTML parser context
2026: * @tagname: the tag name as parsed in the opening tag.
2027: *
2028: * parse an end of tag
2029: *
2030: * [42] ETag ::= '</' Name S? '>'
2031: *
2032: * With namespace
2033: *
2034: * [NS 9] ETag ::= '</' QName S? '>'
2035: */
2036:
2037: void
1.14 daniel 2038: htmlParseEndTag(htmlParserCtxtPtr ctxt, const xmlChar *tagname) {
2039: xmlChar *name;
1.15 daniel 2040: xmlChar *oldname;
1.1 daniel 2041: int i;
2042:
2043: if ((CUR != '<') || (NXT(1) != '/')) {
2044: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2045: ctxt->sax->error(ctxt->userData, "htmlParseEndTag: '</' not found\n");
2046: ctxt->wellFormed = 0;
2047: return;
2048: }
2049: SKIP(2);
2050:
2051: name = htmlParseHTMLName(ctxt);
2052:
2053: /*
2054: * We should definitely be at the ending "S? '>'" part
2055: */
2056: SKIP_BLANKS;
2057: if ((!IS_CHAR(CUR)) || (CUR != '>')) {
2058: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2059: ctxt->sax->error(ctxt->userData, "End tag : expected '>'\n");
2060: ctxt->wellFormed = 0;
2061: } else
2062: NEXT;
2063:
2064: /*
2065: * Check that we are not closing an already closed tag,
2066: * <p><b>...</p></b> is a really common error !
2067: */
1.15 daniel 2068: for (i = ctxt->nameNr - 1;i >= 0;i--) {
2069: if ((ctxt->nameTab[i] != NULL) &&
2070: (!xmlStrcmp(tagname, ctxt->nameTab[i])))
1.1 daniel 2071: break;
2072: }
2073: if (i < 0) {
2074: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2075: ctxt->sax->error(ctxt->userData,
2076: "htmlParseEndTag: unexpected close for tag %s\n",
2077: tagname);
1.11 daniel 2078: xmlFree(name);
1.1 daniel 2079: ctxt->wellFormed = 0;
2080: return;
2081: }
2082:
2083: /*
2084: * Check for auto-closure of HTML elements.
2085: */
2086: htmlAutoCloseOnClose(ctxt, name);
2087:
2088: /*
2089: * Well formedness constraints, opening and closing must match.
2090: * With the exception that the autoclose may have popped stuff out
2091: * of the stack.
2092: */
2093: if (xmlStrcmp(name, tagname)) {
1.15 daniel 2094: if ((ctxt->name != NULL) &&
2095: (xmlStrcmp(ctxt->name, name))) {
1.1 daniel 2096: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2097: ctxt->sax->error(ctxt->userData,
2098: "Opening and ending tag mismatch: %s and %s\n",
1.15 daniel 2099: name, ctxt->name);
1.1 daniel 2100: ctxt->wellFormed = 0;
2101: }
2102: }
2103:
2104: /*
2105: * SAX: End of Tag
2106: */
2107: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
2108: ctxt->sax->endElement(ctxt->userData, name);
1.15 daniel 2109: oldname = ctxt->name;
2110: htmlnamePop(ctxt);
2111: xmlFree(oldname);
1.1 daniel 2112:
2113: if (name != NULL)
1.11 daniel 2114: xmlFree(name);
1.1 daniel 2115:
2116: return;
2117: }
2118:
2119:
2120: /**
2121: * htmlParseReference:
2122: * @ctxt: an HTML parser context
2123: *
2124: * parse and handle entity references in content,
2125: * this will end-up in a call to character() since this is either a
2126: * CharRef, or a predefined entity.
2127: */
2128: void
2129: htmlParseReference(htmlParserCtxtPtr ctxt) {
2130: htmlEntityDescPtr ent;
1.14 daniel 2131: xmlChar out[2];
2132: xmlChar *name;
1.1 daniel 2133: int val;
2134: if (CUR != '&') return;
2135:
2136: if (NXT(1) == '#') {
2137: val = htmlParseCharRef(ctxt);
1.8 daniel 2138: /* invalid for UTF-8 variable encoding !!!!! */
1.1 daniel 2139: out[0] = val;
2140: out[1] = 0;
2141: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
2142: ctxt->sax->characters(ctxt->userData, out, 1);
2143: } else {
2144: ent = htmlParseEntityRef(ctxt, &name);
2145: if (name == NULL) return; /* Shall we output & anyway ? */
2146: if ((ent == NULL) || (ent->value <= 0) || (ent->value >= 255)) {
2147: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
1.8 daniel 2148: ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
1.1 daniel 2149: ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
1.8 daniel 2150: ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1);
1.1 daniel 2151: }
2152: } else {
1.8 daniel 2153: /* invalid for UTF-8 variable encoding !!!!! */
1.1 daniel 2154: out[0] = ent->value;
2155: out[1] = 0;
2156: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
2157: ctxt->sax->characters(ctxt->userData, out, 1);
2158: }
1.11 daniel 2159: xmlFree(name);
1.1 daniel 2160: }
2161: }
2162:
2163: /**
2164: * htmlParseContent:
2165: * @ctxt: an HTML parser context
2166: * @name: the node name
2167: *
2168: * Parse a content: comment, sub-element, reference or text.
2169: *
2170: */
2171:
2172: void
1.14 daniel 2173: htmlParseContent(htmlParserCtxtPtr ctxt, const xmlChar *name) {
1.15 daniel 2174: xmlChar *currentNode;
1.1 daniel 2175:
1.15 daniel 2176: currentNode = ctxt->name;
1.1 daniel 2177: while ((CUR != '<') || (NXT(1) != '/')) {
1.14 daniel 2178: const xmlChar *test = CUR_PTR;
1.1 daniel 2179:
2180: /*
2181: * Has this node been popped out during parsing of
2182: * the next element
2183: */
1.15 daniel 2184: if (currentNode != ctxt->name) return;
1.1 daniel 2185:
2186: /*
2187: * First case : a comment
2188: */
2189: if ((CUR == '<') && (NXT(1) == '!') &&
2190: (NXT(2) == '-') && (NXT(3) == '-')) {
2191: htmlParseComment(ctxt, 1);
2192: }
2193:
2194: /*
2195: * Second case : a sub-element.
2196: */
2197: else if (CUR == '<') {
2198: htmlParseElement(ctxt);
2199: }
2200:
2201: /*
2202: * Third case : a reference. If if has not been resolved,
2203: * parsing returns it's Name, create the node
2204: */
2205: else if (CUR == '&') {
2206: htmlParseReference(ctxt);
2207: }
2208:
2209: /*
2210: * Last case, text. Note that References are handled directly.
2211: */
2212: else {
2213: htmlParseCharData(ctxt, 0);
2214: }
2215:
2216: if (test == CUR_PTR) {
2217: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2218: ctxt->sax->error(ctxt->userData,
2219: "detected an error in element content\n");
2220: ctxt->wellFormed = 0;
2221: break;
2222: }
1.5 daniel 2223: GROW;
1.1 daniel 2224: }
2225:
2226: /*
2227: * parse the end of tag: '</' should be here.
2228: */
2229: htmlParseEndTag(ctxt, name);
2230: }
2231:
2232: /**
2233: * htmlParseElement:
2234: * @ctxt: an HTML parser context
2235: *
2236: * parse an HTML element, this is highly recursive
2237: *
2238: * [39] element ::= EmptyElemTag | STag content ETag
2239: *
2240: * [41] Attribute ::= Name Eq AttValue
2241: */
2242:
2243: void
2244: htmlParseElement(htmlParserCtxtPtr ctxt) {
1.14 daniel 2245: const xmlChar *openTag = CUR_PTR;
2246: xmlChar *name;
1.15 daniel 2247: xmlChar *oldname;
1.16 ! daniel 2248: xmlChar *currentNode = NULL;
1.1 daniel 2249: htmlElemDescPtr info;
1.10 daniel 2250: htmlParserNodeInfo node_info;
1.1 daniel 2251:
2252: /* Capture start position */
1.10 daniel 2253: if (ctxt->record_info) {
2254: node_info.begin_pos = ctxt->input->consumed +
2255: (CUR_PTR - ctxt->input->base);
2256: node_info.begin_line = ctxt->input->line;
2257: }
1.1 daniel 2258:
2259: name = htmlParseStartTag(ctxt);
2260: if (name == NULL) {
2261: return;
2262: }
2263:
2264: /*
2265: * Lookup the info for that element.
2266: */
2267: info = htmlTagLookup(name);
2268: if (info == NULL) {
2269: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2270: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
2271: name);
2272: ctxt->wellFormed = 0;
2273: } else if (info->depr) {
2274: /***************************
2275: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
2276: ctxt->sax->warning(ctxt->userData, "Tag %s is deprecated\n",
2277: name);
2278: ***************************/
2279: }
2280:
2281: /*
2282: * Check for an Empty Element labelled the XML/SGML way
2283: */
2284: if ((CUR == '/') && (NXT(1) == '>')) {
2285: SKIP(2);
2286: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
2287: ctxt->sax->endElement(ctxt->userData, name);
1.15 daniel 2288: oldname = ctxt->name;
2289: htmlnamePop(ctxt);
2290: xmlFree(oldname);
1.11 daniel 2291: xmlFree(name);
1.1 daniel 2292: return;
2293: }
2294:
1.5 daniel 2295: if (CUR == '>') {
2296: NEXT;
2297: } else {
1.1 daniel 2298: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2299: ctxt->sax->error(ctxt->userData, "Couldn't find end of Start Tag\n%.30s\n",
2300: openTag);
2301: ctxt->wellFormed = 0;
2302:
2303: /*
2304: * end of parsing of this node.
2305: */
2306: nodePop(ctxt);
1.11 daniel 2307: xmlFree(name);
1.15 daniel 2308: oldname = ctxt->name;
2309: htmlnamePop(ctxt);
2310: xmlFree(oldname);
1.10 daniel 2311:
2312: /*
2313: * Capture end position and add node
2314: */
2315: if ( currentNode != NULL && ctxt->record_info ) {
2316: node_info.end_pos = ctxt->input->consumed +
2317: (CUR_PTR - ctxt->input->base);
2318: node_info.end_line = ctxt->input->line;
1.15 daniel 2319: node_info.node = ctxt->node;
1.10 daniel 2320: xmlParserAddNodeInfo(ctxt, &node_info);
2321: }
1.1 daniel 2322: return;
2323: }
2324:
2325: /*
2326: * Check for an Empty Element from DTD definition
2327: */
2328: if ((info != NULL) && (info->empty)) {
2329: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
2330: ctxt->sax->endElement(ctxt->userData, name);
1.11 daniel 2331: xmlFree(name);
1.15 daniel 2332: oldname = ctxt->name;
2333: htmlnamePop(ctxt);
2334: xmlFree(oldname);
1.1 daniel 2335: return;
2336: }
2337:
2338: /*
2339: * Parse the content of the element:
2340: */
1.15 daniel 2341: currentNode = ctxt->name;
1.1 daniel 2342: htmlParseContent(ctxt, name);
2343:
2344: /*
2345: * check whether the element get popped due to auto closure
2346: * on start tag
2347: */
1.15 daniel 2348: if (currentNode != ctxt->name) {
1.11 daniel 2349: xmlFree(name);
1.1 daniel 2350: return;
2351: }
2352:
2353: if (!IS_CHAR(CUR)) {
2354: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2355: ctxt->sax->error(ctxt->userData,
2356: "Premature end of data in tag %.30s\n", openTag);
2357: ctxt->wellFormed = 0;
2358:
2359: /*
2360: * end of parsing of this node.
2361: */
2362: nodePop(ctxt);
1.11 daniel 2363: xmlFree(name);
1.15 daniel 2364: oldname = ctxt->name;
2365: htmlnamePop(ctxt);
2366: xmlFree(oldname);
1.1 daniel 2367: return;
2368: }
2369:
1.11 daniel 2370: xmlFree(name);
1.10 daniel 2371:
2372: /*
2373: * Capture end position and add node
2374: */
2375: if ( currentNode != NULL && ctxt->record_info ) {
2376: node_info.end_pos = ctxt->input->consumed +
2377: (CUR_PTR - ctxt->input->base);
2378: node_info.end_line = ctxt->input->line;
1.15 daniel 2379: node_info.node = ctxt->node;
1.10 daniel 2380: xmlParserAddNodeInfo(ctxt, &node_info);
2381: }
1.1 daniel 2382: }
2383:
2384: /**
2385: * htmlParseDocument :
2386: * @ctxt: an HTML parser context
2387: *
2388: * parse an HTML document (and build a tree if using the standard SAX
2389: * interface).
2390: *
2391: * Returns 0, -1 in case of error. the parser context is augmented
2392: * as a result of the parsing.
2393: */
2394:
2395: int
2396: htmlParseDocument(htmlParserCtxtPtr ctxt) {
2397: htmlDefaultSAXHandlerInit();
2398: ctxt->html = 1;
2399:
1.5 daniel 2400: GROW;
1.1 daniel 2401: /*
1.9 daniel 2402: * SAX: beginning of the document processing.
1.1 daniel 2403: */
2404: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
2405: ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
2406:
2407: /*
2408: * Wipe out everything which is before the first '<'
2409: */
2410: if (IS_BLANK(CUR)) {
2411: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2412: ctxt->sax->error(ctxt->userData,
2413: "Extra spaces at the beginning of the document are not allowed\n");
2414: ctxt->wellFormed = 0;
2415: SKIP_BLANKS;
2416: }
2417:
2418: if (CUR == 0) {
2419: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2420: ctxt->sax->error(ctxt->userData, "Document is empty\n");
2421: ctxt->wellFormed = 0;
2422: }
2423:
2424:
2425: /*
2426: * Then possibly doc type declaration(s) and more Misc
2427: * (doctypedecl Misc*)?
2428: */
2429: if ((CUR == '<') && (NXT(1) == '!') &&
2430: (UPP(2) == 'D') && (UPP(3) == 'O') &&
2431: (UPP(4) == 'C') && (UPP(5) == 'T') &&
2432: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
2433: (UPP(8) == 'E')) {
2434: htmlParseDocTypeDecl(ctxt);
2435: }
2436: SKIP_BLANKS;
2437:
2438: /*
2439: * Create the document if not done already.
2440: */
2441: if (ctxt->myDoc == NULL) {
2442: ctxt->myDoc = htmlNewDoc(NULL, NULL);
2443: }
2444:
2445: /*
2446: * Time to start parsing the tree itself
2447: */
2448: htmlParseElement(ctxt);
2449:
2450: /*
2451: * SAX: end of the document processing.
2452: */
2453: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
2454: ctxt->sax->endDocument(ctxt->userData);
2455: if (! ctxt->wellFormed) return(-1);
2456: return(0);
2457: }
2458:
2459:
2460: /********************************************************************************
2461: * *
2462: * Parser contexts handling *
2463: * *
2464: ********************************************************************************/
2465:
2466: /**
2467: * xmlInitParserCtxt:
2468: * @ctxt: an HTML parser context
2469: *
2470: * Initialize a parser context
2471: */
2472:
2473: void
2474: htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
2475: {
2476: htmlSAXHandler *sax;
2477:
1.11 daniel 2478: sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
1.1 daniel 2479: if (sax == NULL) {
2480: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
2481: }
2482:
2483: /* Allocate the Input stack */
1.11 daniel 2484: ctxt->inputTab = (htmlParserInputPtr *) xmlMalloc(5 * sizeof(htmlParserInputPtr));
1.1 daniel 2485: ctxt->inputNr = 0;
2486: ctxt->inputMax = 5;
2487: ctxt->input = NULL;
2488: ctxt->version = NULL;
2489: ctxt->encoding = NULL;
2490: ctxt->standalone = -1;
2491:
2492: /* Allocate the Node stack */
1.11 daniel 2493: ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
1.1 daniel 2494: ctxt->nodeNr = 0;
2495: ctxt->nodeMax = 10;
2496: ctxt->node = NULL;
2497:
1.15 daniel 2498: /* Allocate the Name stack */
2499: ctxt->nameTab = (xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
2500: ctxt->nameNr = 0;
2501: ctxt->nameMax = 10;
2502: ctxt->name = NULL;
2503:
1.1 daniel 2504: if (sax == NULL) ctxt->sax = &htmlDefaultSAXHandler;
2505: else {
2506: ctxt->sax = sax;
2507: memcpy(sax, &htmlDefaultSAXHandler, sizeof(htmlSAXHandler));
2508: }
2509: ctxt->userData = ctxt;
2510: ctxt->myDoc = NULL;
2511: ctxt->wellFormed = 1;
2512: ctxt->replaceEntities = 0;
2513: ctxt->html = 1;
2514: ctxt->record_info = 0;
2515: xmlInitNodeInfoSeq(&ctxt->node_seq);
2516: }
2517:
2518: /**
2519: * htmlFreeParserCtxt:
2520: * @ctxt: an HTML parser context
2521: *
2522: * Free all the memory used by a parser context. However the parsed
2523: * document in ctxt->myDoc is not freed.
2524: */
2525:
2526: void
2527: htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
2528: {
2529: htmlParserInputPtr input;
1.15 daniel 2530: xmlChar *oldname;
1.1 daniel 2531:
2532: if (ctxt == NULL) return;
2533:
2534: while ((input = inputPop(ctxt)) != NULL) {
2535: xmlFreeInputStream(input);
2536: }
2537:
1.11 daniel 2538: if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab);
1.15 daniel 2539: while ((oldname = ctxt->name) != NULL) {
2540: htmlnamePop(ctxt);
2541: xmlFree(oldname);
2542: }
2543: if (ctxt->nameTab != NULL) xmlFree(ctxt->nameTab);
1.11 daniel 2544: if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab);
2545: if (ctxt->version != NULL) xmlFree((char *) ctxt->version);
1.1 daniel 2546: if ((ctxt->sax != NULL) && (ctxt->sax != &htmlDefaultSAXHandler))
1.11 daniel 2547: xmlFree(ctxt->sax);
2548: xmlFree(ctxt);
1.1 daniel 2549: }
2550:
2551: /**
2552: * htmlCreateDocParserCtxt :
1.14 daniel 2553: * @cur: a pointer to an array of xmlChar
1.1 daniel 2554: * @encoding: a free form C string describing the HTML document encoding, or NULL
2555: *
2556: * Create a parser context for an HTML document.
2557: *
2558: * Returns the new parser context or NULL
2559: */
2560: htmlParserCtxtPtr
1.14 daniel 2561: htmlCreateDocParserCtxt(xmlChar *cur, const char *encoding) {
1.1 daniel 2562: htmlParserCtxtPtr ctxt;
2563: htmlParserInputPtr input;
2564: /* htmlCharEncoding enc; */
2565:
1.11 daniel 2566: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 2567: if (ctxt == NULL) {
2568: perror("malloc");
2569: return(NULL);
2570: }
2571: htmlInitParserCtxt(ctxt);
1.11 daniel 2572: input = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 2573: if (input == NULL) {
2574: perror("malloc");
1.11 daniel 2575: xmlFree(ctxt);
1.1 daniel 2576: return(NULL);
2577: }
2578:
2579: /*
2580: * plug some encoding conversion routines here. !!!
2581: if (encoding != NULL) {
2582: enc = htmlDetectCharEncoding(cur);
2583: htmlSwitchEncoding(ctxt, enc);
2584: }
2585: */
2586:
2587: input->filename = NULL;
2588: input->line = 1;
2589: input->col = 1;
2590: input->base = cur;
2591: input->cur = cur;
2592: input->free = NULL;
1.5 daniel 2593: input->buf = NULL;
1.1 daniel 2594:
2595: inputPush(ctxt, input);
2596: return(ctxt);
2597: }
2598:
2599: /********************************************************************************
2600: * *
2601: * User entry points *
2602: * *
2603: ********************************************************************************/
2604:
2605: /**
2606: * htmlSAXParseDoc :
1.14 daniel 2607: * @cur: a pointer to an array of xmlChar
1.1 daniel 2608: * @encoding: a free form C string describing the HTML document encoding, or NULL
2609: * @sax: the SAX handler block
2610: * @userData: if using SAX, this pointer will be provided on callbacks.
2611: *
2612: * parse an HTML in-memory document and build a tree.
2613: * It use the given SAX function block to handle the parsing callback.
2614: * If sax is NULL, fallback to the default DOM tree building routines.
2615: *
2616: * Returns the resulting document tree
2617: */
2618:
2619: htmlDocPtr
1.14 daniel 2620: htmlSAXParseDoc(xmlChar *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) {
1.1 daniel 2621: htmlDocPtr ret;
2622: htmlParserCtxtPtr ctxt;
2623:
2624: if (cur == NULL) return(NULL);
2625:
2626:
2627: ctxt = htmlCreateDocParserCtxt(cur, encoding);
2628: if (ctxt == NULL) return(NULL);
2629: if (sax != NULL) {
2630: ctxt->sax = sax;
2631: ctxt->userData = userData;
2632: }
2633:
2634: htmlParseDocument(ctxt);
2635: ret = ctxt->myDoc;
2636: if (sax != NULL) {
2637: ctxt->sax = NULL;
2638: ctxt->userData = NULL;
2639: }
2640: htmlFreeParserCtxt(ctxt);
2641:
2642: return(ret);
2643: }
2644:
2645: /**
2646: * htmlParseDoc :
1.14 daniel 2647: * @cur: a pointer to an array of xmlChar
1.1 daniel 2648: * @encoding: a free form C string describing the HTML document encoding, or NULL
2649: *
2650: * parse an HTML in-memory document and build a tree.
2651: *
2652: * Returns the resulting document tree
2653: */
2654:
2655: htmlDocPtr
1.14 daniel 2656: htmlParseDoc(xmlChar *cur, const char *encoding) {
1.1 daniel 2657: return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
2658: }
2659:
2660:
2661: /**
2662: * htmlCreateFileParserCtxt :
2663: * @filename: the filename
2664: * @encoding: a free form C string describing the HTML document encoding, or NULL
2665: *
2666: * Create a parser context for a file content.
2667: * Automatic support for ZLIB/Compress compressed document is provided
2668: * by default if found at compile-time.
2669: *
2670: * Returns the new parser context or NULL
2671: */
2672: htmlParserCtxtPtr
2673: htmlCreateFileParserCtxt(const char *filename, const char *encoding)
2674: {
2675: htmlParserCtxtPtr ctxt;
2676: htmlParserInputPtr inputStream;
1.5 daniel 2677: xmlParserInputBufferPtr buf;
1.1 daniel 2678: /* htmlCharEncoding enc; */
2679:
1.5 daniel 2680: buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
2681: if (buf == NULL) return(NULL);
1.1 daniel 2682:
1.11 daniel 2683: ctxt = (htmlParserCtxtPtr) xmlMalloc(sizeof(htmlParserCtxt));
1.1 daniel 2684: if (ctxt == NULL) {
2685: perror("malloc");
2686: return(NULL);
2687: }
2688: htmlInitParserCtxt(ctxt);
1.11 daniel 2689: inputStream = (htmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1.1 daniel 2690: if (inputStream == NULL) {
2691: perror("malloc");
1.11 daniel 2692: xmlFree(ctxt);
1.1 daniel 2693: return(NULL);
2694: }
2695:
1.11 daniel 2696: inputStream->filename = xmlMemStrdup(filename);
1.1 daniel 2697: inputStream->line = 1;
2698: inputStream->col = 1;
1.5 daniel 2699: inputStream->buf = buf;
1.1 daniel 2700:
1.5 daniel 2701: inputStream->base = inputStream->buf->buffer->content;
2702: inputStream->cur = inputStream->buf->buffer->content;
2703: inputStream->free = NULL;
1.1 daniel 2704:
2705: inputPush(ctxt, inputStream);
2706: return(ctxt);
2707: }
2708:
2709: /**
2710: * htmlSAXParseFile :
2711: * @filename: the filename
2712: * @encoding: a free form C string describing the HTML document encoding, or NULL
2713: * @sax: the SAX handler block
2714: * @userData: if using SAX, this pointer will be provided on callbacks.
2715: *
2716: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
2717: * compressed document is provided by default if found at compile-time.
2718: * It use the given SAX function block to handle the parsing callback.
2719: * If sax is NULL, fallback to the default DOM tree building routines.
2720: *
2721: * Returns the resulting document tree
2722: */
2723:
2724: htmlDocPtr
2725: htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
2726: void *userData) {
2727: htmlDocPtr ret;
2728: htmlParserCtxtPtr ctxt;
2729:
2730: ctxt = htmlCreateFileParserCtxt(filename, encoding);
2731: if (ctxt == NULL) return(NULL);
2732: if (sax != NULL) {
2733: ctxt->sax = sax;
2734: ctxt->userData = userData;
2735: }
2736:
2737: htmlParseDocument(ctxt);
2738:
2739: ret = ctxt->myDoc;
2740: if (sax != NULL) {
2741: ctxt->sax = NULL;
2742: ctxt->userData = NULL;
2743: }
2744: htmlFreeParserCtxt(ctxt);
2745:
2746: return(ret);
2747: }
2748:
2749: /**
2750: * htmlParseFile :
2751: * @filename: the filename
2752: * @encoding: a free form C string describing the HTML document encoding, or NULL
2753: *
2754: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
2755: * compressed document is provided by default if found at compile-time.
2756: *
2757: * Returns the resulting document tree
2758: */
2759:
2760: htmlDocPtr
2761: htmlParseFile(const char *filename, const char *encoding) {
2762: return(htmlSAXParseFile(filename, encoding, NULL, NULL));
2763: }
Webmaster