Parent: [46a7f0] (diff)

Child: [c7a241] (diff)

Download this file

htmlparse.cpp    359 lines (305 with data), 9.8 kB

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
/* This file was copied/updated from xapian-omega-1.0.1 and modified */
/* htmlparse.cc: simple HTML parser for omega indexer
*
* Copyright 1999,2000,2001 BrightStation PLC
* Copyright 2001 Ananova Ltd
* Copyright 2002,2006 Olly Betts
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation; either version 2 of the
* License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301
* USA
*/
#include <algorithm>
using std::find;
using std::find_if;
#include "htmlparse.h"
#include <stdio.h>
#include <ctype.h>
#include <cstring>
map<string, unsigned int> HtmlParser::named_ents;
inline static bool
p_notdigit(char c)
{
return !isdigit(static_cast<unsigned char>(c));
}
inline static bool
p_notxdigit(char c)
{
return !isxdigit(static_cast<unsigned char>(c));
}
inline static bool
p_notalnum(char c)
{
return !isalnum(static_cast<unsigned char>(c));
}
inline static bool
p_notwhitespace(char c)
{
return !isspace(static_cast<unsigned char>(c));
}
inline static bool
p_nottag(char c)
{
return !isalnum(static_cast<unsigned char>(c)) &&
c != '.' && c != '-' && c != ':'; // ':' for XML namespaces.
}
inline static bool
p_whitespacegt(char c)
{
return isspace(static_cast<unsigned char>(c)) || c == '>';
}
inline static bool
p_whitespaceeqgt(char c)
{
return isspace(static_cast<unsigned char>(c)) || c == '=' || c == '>';
}
HtmlParser::HtmlParser()
{
// RECOLL: no need to initialize these entities, we use those from
// myhtmlparse
#if 0
static const struct ent { const char *n; unsigned int v; } ents[] = {
#include "namedentities.h"
{ NULL, 0 }
};
if (named_ents.empty()) {
const struct ent *i = ents;
while (i->n) {
named_ents[string(i->n)] = i->v;
++i;
}
}
#endif
}
void
HtmlParser::decode_entities(string &s)
{
// Not used for recoll. Kept here to minimize the amount of diffs
#if 0
// We need a const_iterator version of s.end() - otherwise the
// find() and find_if() templates don't work...
string::const_iterator amp = s.begin(), s_end = s.end();
while ((amp = find(amp, s_end, '&')) != s_end) {
unsigned int val = 0;
string::const_iterator end, p = amp + 1;
if (p != s_end && *p == '#') {
p++;
if (p != s_end && (*p == 'x' || *p == 'X')) {
// hex
p++;
end = find_if(p, s_end, p_notxdigit);
sscanf(s.substr(p - s.begin(), end - p).c_str(), "%x", &val);
} else {
// number
end = find_if(p, s_end, p_notdigit);
val = atoi(s.substr(p - s.begin(), end - p).c_str());
}
} else {
end = find_if(p, s_end, p_notalnum);
string code = s.substr(p - s.begin(), end - p);
map<string, unsigned int>::const_iterator i;
i = named_ents.find(code);
if (i != named_ents.end()) val = i->second;
}
if (end < s_end && *end == ';') end++;
if (val) {
string::size_type amp_pos = amp - s.begin();
if (val < 0x80) {
s.replace(amp_pos, end - amp, 1u, char(val));
} else {
// Convert unicode value val to UTF-8.
char seq[4];
unsigned len = Xapian::Unicode::nonascii_to_utf8(val, seq);
s.replace(amp_pos, end - amp, seq, len);
}
s_end = s.end();
// We've modified the string, so the iterators are no longer
// valid...
amp = s.begin() + amp_pos + 1;
} else {
amp = end;
}
}
#endif
}
void
HtmlParser::parse_html(const string &body)
{
in_script = false;
map<string,string> Param;
string::const_iterator start = body.begin();
while (true) {
// Skip through until we find an HTML tag, a comment, or the end of
// document. Ignore isolated occurences of `<' which don't start
// a tag or comment.
string::const_iterator p = start;
while (true) {
p = find(p, body.end(), '<');
if (p == body.end()) break;
unsigned char ch = *(p + 1);
// Tag, closing tag, or comment (or SGML declaration).
if ((!in_script && isalpha(ch)) || ch == '/' || ch == '!') break;
if (ch == '?') {
// PHP code or XML declaration.
// XML declaration is only valid at the start of the first line.
// FIXME: need to deal with BOMs...
if (p != body.begin() || body.size() < 20) break;
// XML declaration looks something like this:
// <?xml version="1.0" encoding="UTF-8"?>
if (p[2] != 'x' || p[3] != 'm' || p[4] != 'l') break;
if (strchr(" \t\r\n", p[5]) == NULL) break;
string::const_iterator decl_end = find(p + 6, body.end(), '?');
if (decl_end == body.end()) break;
// Default charset for XML is UTF-8.
charset = "UTF-8";
string decl(p + 6, decl_end);
size_t enc = decl.find("encoding");
if (enc == string::npos) break;
enc = decl.find_first_not_of(" \t\r\n", enc + 8);
if (enc == string::npos || enc == decl.size()) break;
if (decl[enc] != '=') break;
enc = decl.find_first_not_of(" \t\r\n", enc + 1);
if (enc == string::npos || enc == decl.size()) break;
if (decl[enc] != '"' && decl[enc] != '\'') break;
char quote = decl[enc++];
size_t enc_end = decl.find(quote, enc);
if (enc != string::npos)
charset = decl.substr(enc, enc_end - enc);
break;
}
p++;
}
// Process text up to start of tag.
if (p > start || p == body.end()) {
string text = body.substr(start - body.begin(), p - start);
decode_entities(text);
process_text(text);
}
if (p == body.end()) {
do_eof();
break;
}
start = p + 1;
if (start == body.end()) break;
if (*start == '!') {
if (++start == body.end()) break;
if (++start == body.end()) break;
// comment or SGML declaration
if (*(start - 1) == '-' && *start == '-') {
++start;
string::const_iterator close = find(start, body.end(), '>');
// An unterminated comment swallows rest of document
// (like Netscape, but unlike MSIE IIRC)
if (close == body.end()) break;
p = close;
// look for -->
while (p != body.end() && (*(p - 1) != '-' || *(p - 2) != '-'))
p = find(p + 1, body.end(), '>');
if (p != body.end()) {
// Check for htdig's "ignore this bit" comments.
if (p - start == 15 && string(start, p - 2) == "htdig_noindex") {
string::size_type i;
i = body.find("<!--/htdig_noindex-->", p + 1 - body.begin());
if (i == string::npos) break;
start = body.begin() + i + 21;
continue;
}
// If we found --> skip to there.
start = p;
} else {
// Otherwise skip to the first > we found (as Netscape does).
start = close;
}
} else {
// just an SGML declaration, perhaps giving the DTD - ignore it
start = find(start - 1, body.end(), '>');
if (start == body.end()) break;
}
++start;
} else if (*start == '?') {
if (++start == body.end()) break;
// PHP - swallow until ?> or EOF
start = find(start + 1, body.end(), '>');
// look for ?>
while (start != body.end() && *(start - 1) != '?')
start = find(start + 1, body.end(), '>');
// unterminated PHP swallows rest of document (rather arbitrarily
// but it avoids polluting the database when things go wrong)
if (start != body.end()) ++start;
} else {
// opening or closing tag
int closing = 0;
if (*start == '/') {
closing = 1;
start = find_if(start + 1, body.end(), p_notwhitespace);
}
p = start;
start = find_if(start, body.end(), p_nottag);
string tag = body.substr(p - body.begin(), start - p);
// convert tagname to lowercase
for (string::iterator i = tag.begin(); i != tag.end(); ++i)
*i = tolower(static_cast<unsigned char>(*i));
if (closing) {
closing_tag(tag);
if (in_script && tag == "script") in_script = false;
/* ignore any bogus parameters on closing tags */
p = find(start, body.end(), '>');
if (p == body.end()) break;
start = p + 1;
} else {
while (start < body.end() && *start != '>') {
string name, value;
p = find_if(start, body.end(), p_whitespaceeqgt);
name = body.substr(start - body.begin(), p - start);
p = find_if(p, body.end(), p_notwhitespace);
start = p;
if (start != body.end() && *start == '=') {
int quote;
start = find_if(start + 1, body.end(), p_notwhitespace);
p = body.end();
quote = *start;
if (quote == '"' || quote == '\'') {
start++;
p = find(start, body.end(), quote);
}
if (p == body.end()) {
// unquoted or no closing quote
p = find_if(start, body.end(), p_whitespacegt);
value = body.substr(start - body.begin(), p - start);
start = find_if(p, body.end(), p_notwhitespace);
} else {
value = body.substr(start - body.begin(), p - start);
}
if (name.size()) {
// convert parameter name to lowercase
string::iterator i;
for (i = name.begin(); i != name.end(); ++i)
*i = tolower(static_cast<unsigned char>(*i));
// in case of multiple entries, use the first
// (as Netscape does)
if (Param.find(name) == Param.end())
Param[name] = value;
}
}
}
opening_tag(tag, Param);
Param.clear();
// In <script> tags we ignore opening tags to avoid problems
// with "a<b".
if (tag == "script") in_script = true;
if (start != body.end() && *start == '>') ++start;
}
}
}
}