Line |
Branch |
Exec |
Source |
1 |
|
|
/***************************************************************************** |
2 |
|
|
* |
3 |
|
|
* $Id$ |
4 |
|
|
* |
5 |
|
|
* Copyright 2010 Richard Hacker (lerichi at gmx dot net) |
6 |
|
|
* |
7 |
|
|
* This file is part of the pdserv library. |
8 |
|
|
* |
9 |
|
|
* The pdserv library is free software: you can redistribute it and/or modify |
10 |
|
|
* it under the terms of the GNU Lesser General Public License as published |
11 |
|
|
* by the Free Software Foundation, either version 3 of the License, or (at |
12 |
|
|
* your option) any later version. |
13 |
|
|
* |
14 |
|
|
* The pdserv library is distributed in the hope that it will be useful, but |
15 |
|
|
* WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY |
16 |
|
|
* or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public |
17 |
|
|
* License for more details. |
18 |
|
|
* |
19 |
|
|
* You should have received a copy of the GNU Lesser General Public License |
20 |
|
|
* along with the pdserv library. If not, see <http://www.gnu.org/licenses/>. |
21 |
|
|
* |
22 |
|
|
*****************************************************************************/ |
23 |
|
|
|
24 |
|
|
#include "XmlParser.h" |
25 |
|
|
#include "Session.h" |
26 |
|
|
#include "../Debug.h" |
27 |
|
|
|
28 |
|
|
#include <algorithm> |
29 |
|
|
#include <cstring> |
30 |
|
|
#include <sstream> |
31 |
|
|
|
32 |
|
|
#undef log_debug |
33 |
|
|
#define log_debug(...) |
34 |
|
|
|
35 |
|
|
using namespace MsrProto; |
36 |
|
|
|
37 |
|
|
///////////////////////////////////////////////////////////////////////////// |
38 |
|
✗ |
XmlParser::XmlParser(size_t bufMax): bufLenMax(bufMax) |
39 |
|
|
{ |
40 |
|
✗ |
parseState = FindStart; |
41 |
|
✗ |
buf = new char[bufIncrement]; |
42 |
|
✗ |
bufLen = bufIncrement; |
43 |
|
|
|
44 |
|
✗ |
parsePos = buf; |
45 |
|
✗ |
inputEnd = buf; |
46 |
|
|
} |
47 |
|
|
|
48 |
|
|
///////////////////////////////////////////////////////////////////////////// |
49 |
|
✗ |
XmlParser::XmlParser(const char *begin, const char *end): |
50 |
|
✗ |
bufLenMax(end - begin) |
51 |
|
|
{ |
52 |
|
✗ |
parseState = FindStart; |
53 |
|
✗ |
buf = new char[bufLenMax]; |
54 |
|
✗ |
bufLen = bufLenMax; |
55 |
|
|
|
56 |
|
✗ |
parsePos = buf; |
57 |
|
✗ |
inputEnd = buf; |
58 |
|
|
|
59 |
|
✗ |
std::copy(begin, end, buf); |
60 |
|
✗ |
newData(bufLen); |
61 |
|
|
} |
62 |
|
|
|
63 |
|
|
///////////////////////////////////////////////////////////////////////////// |
64 |
|
✗ |
XmlParser::~XmlParser() |
65 |
|
|
{ |
66 |
|
✗ |
delete[] buf; |
67 |
|
|
} |
68 |
|
|
|
69 |
|
|
///////////////////////////////////////////////////////////////////////////// |
70 |
|
✗ |
size_t XmlParser::free() |
71 |
|
|
{ |
72 |
|
✗ |
return (buf + bufLen) - inputEnd; |
73 |
|
|
} |
74 |
|
|
|
75 |
|
|
///////////////////////////////////////////////////////////////////////////// |
76 |
|
✗ |
char *XmlParser::bufptr() |
77 |
|
|
{ |
78 |
|
✗ |
return inputEnd; |
79 |
|
|
} |
80 |
|
|
|
81 |
|
|
///////////////////////////////////////////////////////////////////////////// |
82 |
|
✗ |
void XmlParser::newData(size_t n) |
83 |
|
|
{ |
84 |
|
✗ |
inputEnd += n; |
85 |
|
|
} |
86 |
|
|
|
87 |
|
|
///////////////////////////////////////////////////////////////////////////// |
88 |
|
✗ |
XmlParser::Element XmlParser::nextElement() |
89 |
|
|
{ |
90 |
|
✗ |
bool finished = false; |
91 |
|
|
|
92 |
|
✗ |
for (; parsePos < inputEnd and !finished; ++parsePos) { |
93 |
|
|
// log_debug("state %i, char %c", parseState, *parsePos); |
94 |
|
|
|
95 |
|
✗ |
switch (parseState) { |
96 |
|
✗ |
case FindStart: |
97 |
|
✗ |
if (*parsePos == '<') { |
98 |
|
|
// Found beginning |
99 |
|
✗ |
attribute.clear(); |
100 |
|
✗ |
element = parsePos; |
101 |
|
✗ |
parseState = ExpectToken; |
102 |
|
|
} |
103 |
|
|
else { |
104 |
|
|
// Move forward in the buffer until '<' is found |
105 |
|
✗ |
parsePos = std::find(parsePos+1, inputEnd, '<') - 1; |
106 |
|
|
// log_debug("Found < at %zu", parsePos - buf); |
107 |
|
|
} |
108 |
|
✗ |
break; |
109 |
|
|
|
110 |
|
✗ |
case ExpectToken: |
111 |
|
|
// Expect an alpha character here |
112 |
|
✗ |
if (isalpha(*parsePos)) { |
113 |
|
✗ |
parseState = FindTokenEnd; |
114 |
|
✗ |
attribute.push_back( |
115 |
|
✗ |
std::make_pair(parsePos,(const char*)0)); |
116 |
|
|
} |
117 |
|
|
else |
118 |
|
✗ |
parseState = FindStart; |
119 |
|
✗ |
break; |
120 |
|
|
|
121 |
|
✗ |
case SkipSpace: |
122 |
|
|
// Check for end of XML Element |
123 |
|
✗ |
if (isalpha(*parsePos)) { |
124 |
|
|
// Found next token |
125 |
|
✗ |
parseState = ExpectToken; |
126 |
|
|
|
127 |
|
|
// Decrement parsePos because the pointer gets incremented |
128 |
|
|
// automatically at the loop end |
129 |
|
✗ |
--parsePos; |
130 |
|
✗ |
break; |
131 |
|
|
} |
132 |
|
|
|
133 |
|
✗ |
if (*parsePos == '/') |
134 |
|
✗ |
parseState = ExpectGT; |
135 |
|
✗ |
else if (*parsePos == '>') |
136 |
|
✗ |
finished = true; |
137 |
|
✗ |
else if (!isspace(*parsePos)) |
138 |
|
✗ |
parseState = FindStart; |
139 |
|
|
|
140 |
|
✗ |
*parsePos = '\0'; // Clobber spaces |
141 |
|
✗ |
break; |
142 |
|
|
|
143 |
|
✗ |
case FindTokenEnd: |
144 |
|
|
// Moves forward until non-alpha is found |
145 |
|
✗ |
if (isalpha(*parsePos) or *parsePos == '_') |
146 |
|
|
break; |
147 |
|
|
|
148 |
|
✗ |
if (*parsePos == '=') { |
149 |
|
✗ |
parseState = GetAttribute; |
150 |
|
✗ |
*parsePos = '\0'; |
151 |
|
|
} |
152 |
|
|
else { |
153 |
|
✗ |
parseState = SkipSpace; |
154 |
|
✗ |
--parsePos; |
155 |
|
|
} |
156 |
|
|
|
157 |
|
|
|
158 |
|
✗ |
break; |
159 |
|
|
|
160 |
|
✗ |
case GetAttribute: |
161 |
|
|
{ |
162 |
|
✗ |
const char *attrValuePos = parsePos; |
163 |
|
✗ |
if (*parsePos == '\'' or *parsePos == '"') { |
164 |
|
✗ |
quote = *parsePos; |
165 |
|
✗ |
++attrValuePos; |
166 |
|
✗ |
parseState = GetQuotedAttribute; |
167 |
|
|
} |
168 |
|
|
else { |
169 |
|
✗ |
parseState = GetUnquotedAttribute; |
170 |
|
|
} |
171 |
|
|
|
172 |
|
✗ |
attribute.rbegin()->second = attrValuePos; |
173 |
|
|
} |
174 |
|
|
|
175 |
|
✗ |
break; |
176 |
|
|
|
177 |
|
✗ |
case GetQuotedAttribute: |
178 |
|
|
{ |
179 |
|
✗ |
char *quotePos = std::find(parsePos, inputEnd, quote); |
180 |
|
✗ |
char *escape = std::find(parsePos, quotePos, '\\'); |
181 |
|
|
|
182 |
|
✗ |
parsePos = quotePos; |
183 |
|
|
|
184 |
|
|
// log_debug("quotePos=%zu escape=%zu", |
185 |
|
|
// quotePos - buf, escape - quotePos); |
186 |
|
✗ |
if (escape < quotePos) { |
187 |
|
✗ |
parsePos = escape; |
188 |
|
✗ |
escapeState = GetQuotedAttribute; |
189 |
|
✗ |
parseState = SkipEscapeChar; |
190 |
|
|
} |
191 |
|
✗ |
else if (quotePos < inputEnd) { |
192 |
|
✗ |
*parsePos = '\0'; |
193 |
|
|
|
194 |
|
✗ |
parseState = SkipSpace; |
195 |
|
|
} |
196 |
|
|
else |
197 |
|
✗ |
--parsePos; |
198 |
|
|
} |
199 |
|
✗ |
break; |
200 |
|
|
|
201 |
|
✗ |
case GetUnquotedAttribute: |
202 |
|
✗ |
if (*parsePos == '>') { |
203 |
|
✗ |
*parsePos = '\0'; |
204 |
|
✗ |
finished = true; |
205 |
|
|
} |
206 |
|
✗ |
else if (*parsePos == '/') |
207 |
|
✗ |
parseState = MaybeGT; |
208 |
|
✗ |
else if (isspace(*parsePos)) { |
209 |
|
✗ |
*parsePos = '\0'; |
210 |
|
✗ |
parseState = SkipSpace; |
211 |
|
|
} |
212 |
|
|
|
213 |
|
✗ |
break; |
214 |
|
|
|
215 |
|
✗ |
case ExpectGT: |
216 |
|
✗ |
parseState = FindStart; |
217 |
|
✗ |
finished = *parsePos == '>'; |
218 |
|
|
|
219 |
|
✗ |
break; |
220 |
|
|
|
221 |
|
✗ |
case MaybeGT: |
222 |
|
✗ |
if (*parsePos == '>') { |
223 |
|
✗ |
*(parsePos - 1) = '\0'; |
224 |
|
✗ |
finished = true; |
225 |
|
|
} |
226 |
|
|
else |
227 |
|
✗ |
parseState = GetUnquotedAttribute; |
228 |
|
✗ |
break; |
229 |
|
|
|
230 |
|
✗ |
case SkipEscapeChar: |
231 |
|
✗ |
parseState = escapeState; |
232 |
|
✗ |
break; |
233 |
|
|
} |
234 |
|
|
} |
235 |
|
|
|
236 |
|
✗ |
if (finished) { |
237 |
|
|
log_debug("found something"); |
238 |
|
✗ |
parseState = FindStart; |
239 |
|
✗ |
return Element(element+1, &attribute); |
240 |
|
|
} |
241 |
|
|
|
242 |
|
✗ |
if (parseState == FindStart) { |
243 |
|
✗ |
parsePos = inputEnd = buf; |
244 |
|
|
log_debug("resetting pointers"); |
245 |
|
|
} |
246 |
|
✗ |
else if (inputEnd == buf + bufLen) { |
247 |
|
✗ |
if (element > buf) { |
248 |
|
✗ |
size_t n = element - buf; |
249 |
|
|
|
250 |
|
✗ |
std::copy(buf + n, inputEnd, buf); |
251 |
|
|
|
252 |
|
✗ |
parsePos -= n; |
253 |
|
✗ |
inputEnd -= n; |
254 |
|
✗ |
element = buf; |
255 |
|
✗ |
for (AttributeList::iterator it = attribute.begin(); |
256 |
|
✗ |
it != attribute.end(); ++it) { |
257 |
|
✗ |
it->first -= n; |
258 |
|
✗ |
it->second -= n; |
259 |
|
|
} |
260 |
|
|
log_debug("move data to buf start"); |
261 |
|
|
} |
262 |
|
|
else { |
263 |
|
|
// Note: element points to beginning of buf here |
264 |
|
|
|
265 |
|
✗ |
bufLen += bufIncrement; |
266 |
|
|
|
267 |
|
✗ |
if (bufLen > bufLenMax) { |
268 |
|
|
// Overrun max buffer |
269 |
|
✗ |
bufLen = bufIncrement; |
270 |
|
✗ |
parseState = FindStart; |
271 |
|
|
|
272 |
|
✗ |
element = buf; |
273 |
|
✗ |
parsePos = buf; |
274 |
|
✗ |
inputEnd = buf; |
275 |
|
|
} |
276 |
|
|
|
277 |
|
✗ |
buf = new char[bufLen]; |
278 |
|
✗ |
std::copy(element, inputEnd, buf); |
279 |
|
|
|
280 |
|
✗ |
parsePos = buf + (parsePos - element); |
281 |
|
✗ |
inputEnd = buf + (inputEnd - element); |
282 |
|
✗ |
for (AttributeList::iterator it = attribute.begin(); |
283 |
|
✗ |
it != attribute.end(); ++it) { |
284 |
|
✗ |
it->first = buf + (it->first - element); |
285 |
|
✗ |
it->second = buf + (it->second - element); |
286 |
|
|
} |
287 |
|
|
|
288 |
|
✗ |
delete[] element; |
289 |
|
✗ |
element = buf; |
290 |
|
|
log_debug("increased buf to %zu", bufLen); |
291 |
|
|
} |
292 |
|
|
} |
293 |
|
|
|
294 |
|
|
log_debug("empty"); |
295 |
|
✗ |
return Element(0,0); |
296 |
|
|
} |
297 |
|
|
|
298 |
|
|
///////////////////////////////////////////////////////////////////////////// |
299 |
|
|
///////////////////////////////////////////////////////////////////////////// |
300 |
|
✗ |
XmlParser::Element::Element( |
301 |
|
✗ |
const char *command, const XmlParser::AttributeList* attr): |
302 |
|
✗ |
command(command), attribute(attr) |
303 |
|
|
{ |
304 |
|
|
} |
305 |
|
|
|
306 |
|
|
///////////////////////////////////////////////////////////////////////////// |
307 |
|
✗ |
XmlParser::Element::operator bool() const |
308 |
|
|
{ |
309 |
|
✗ |
return command != 0; |
310 |
|
|
} |
311 |
|
|
|
312 |
|
|
///////////////////////////////////////////////////////////////////////////// |
313 |
|
✗ |
const char *XmlParser::Element::getCommand() const |
314 |
|
|
{ |
315 |
|
✗ |
return command; |
316 |
|
|
} |
317 |
|
|
|
318 |
|
|
///////////////////////////////////////////////////////////////////////////// |
319 |
|
✗ |
bool XmlParser::Element::find(const char *name, const char *&value) const |
320 |
|
|
{ |
321 |
|
✗ |
for (AttributeList::const_iterator it = attribute->begin(); |
322 |
|
✗ |
it != attribute->end(); ++it) { |
323 |
|
✗ |
if (!strcasecmp(name, it->first)) { |
324 |
|
✗ |
value = it->second; |
325 |
|
✗ |
return true; |
326 |
|
|
} |
327 |
|
|
} |
328 |
|
|
|
329 |
|
✗ |
return false; |
330 |
|
|
} |
331 |
|
|
|
332 |
|
|
///////////////////////////////////////////////////////////////////////////// |
333 |
|
✗ |
bool XmlParser::Element::isEqual(const char *name, const char *s) const |
334 |
|
|
{ |
335 |
|
✗ |
const char *value; |
336 |
|
|
|
337 |
|
✗ |
if (find(name, value) and value) |
338 |
|
✗ |
return !strcasecmp(value, s); |
339 |
|
|
|
340 |
|
✗ |
return false; |
341 |
|
|
} |
342 |
|
|
|
343 |
|
|
///////////////////////////////////////////////////////////////////////////// |
344 |
|
✗ |
bool XmlParser::Element::isTrue(const char *name) const |
345 |
|
|
{ |
346 |
|
✗ |
const char *value; |
347 |
|
|
|
348 |
|
✗ |
if (!(find(name, value))) |
349 |
|
✗ |
return false; |
350 |
|
|
|
351 |
|
|
// Binary attribute, e.g <xsad sync> |
352 |
|
✗ |
if (!value) |
353 |
|
✗ |
return true; |
354 |
|
|
|
355 |
|
✗ |
size_t len = strlen(value); |
356 |
|
|
|
357 |
|
|
// Binary attribute, e.g <xsad sync=1> |
358 |
|
✗ |
if (len == 1) |
359 |
|
✗ |
return *value == '1'; |
360 |
|
|
|
361 |
|
|
// Binary attribute, e.g <xsad sync="true"> |
362 |
|
✗ |
if (len == 4) |
363 |
|
✗ |
return !strncasecmp(value, "true", 4); |
364 |
|
|
|
365 |
|
|
// Binary attribute, e.g <xsad sync='on'/> |
366 |
|
✗ |
if (len == 2) |
367 |
|
✗ |
return !strncasecmp(value, "on", 2); |
368 |
|
|
|
369 |
|
✗ |
return false; |
370 |
|
|
} |
371 |
|
|
|
372 |
|
|
///////////////////////////////////////////////////////////////////////////// |
373 |
|
✗ |
bool XmlParser::Element::getString(const char *name, std::string &s) const |
374 |
|
|
{ |
375 |
|
✗ |
const char *value; |
376 |
|
|
|
377 |
|
✗ |
s.clear(); |
378 |
|
|
|
379 |
|
✗ |
if (!(find(name, value) and value)) |
380 |
|
✗ |
return false; |
381 |
|
|
|
382 |
|
✗ |
const char *pptr, *eptr = value + strlen(value); |
383 |
|
✗ |
while ((pptr = std::find(value, eptr, '&')) != eptr) { |
384 |
|
|
// FIXME: maybe also check for escape char, e.g. \" ?? |
385 |
|
|
// this is difficult, because the quote character is not available |
386 |
|
|
// here any more :( |
387 |
|
|
|
388 |
|
✗ |
s.append(value, pptr - value); |
389 |
|
✗ |
size_t len = eptr - pptr; |
390 |
|
✗ |
if (len >= 4 and !strncmp(pptr, ">", 4)) { |
391 |
|
✗ |
s.append(1, '>'); |
392 |
|
✗ |
value = pptr + 4; |
393 |
|
|
} |
394 |
|
✗ |
else if (len >= 4 and !strncmp(pptr, "<", 4)) { |
395 |
|
✗ |
s.append(1, '<'); |
396 |
|
✗ |
value = pptr + 4; |
397 |
|
|
} |
398 |
|
✗ |
else if (len >= 5 and !strncmp(pptr, "&", 5)) { |
399 |
|
✗ |
s.append(1, '&'); |
400 |
|
✗ |
value = pptr + 5; |
401 |
|
|
} |
402 |
|
✗ |
else if (len >= 6 and !strncmp(pptr, """, 6)) { |
403 |
|
✗ |
s.append(1, '"'); |
404 |
|
✗ |
value = pptr + 6; |
405 |
|
|
} |
406 |
|
✗ |
else if (len >= 6 and !strncmp(pptr, "'", 6)) { |
407 |
|
✗ |
s.append(1, '\''); |
408 |
|
✗ |
value = pptr + 6; |
409 |
|
|
} |
410 |
|
|
else { |
411 |
|
✗ |
s.append(1, '&'); |
412 |
|
✗ |
value = pptr + 1; |
413 |
|
|
} |
414 |
|
|
} |
415 |
|
|
|
416 |
|
✗ |
s.append(value, eptr - value); |
417 |
|
✗ |
return true; |
418 |
|
|
} |
419 |
|
|
|
420 |
|
|
///////////////////////////////////////////////////////////////////////////// |
421 |
|
✗ |
bool XmlParser::Element::getUnsigned(const char *name, unsigned int &i) const |
422 |
|
|
{ |
423 |
|
✗ |
const char *value; |
424 |
|
|
|
425 |
|
✗ |
if (!(find(name, value)) or !value) |
426 |
|
✗ |
return false; |
427 |
|
|
|
428 |
|
✗ |
i = strtoul(value, 0, 0); |
429 |
|
✗ |
return true; |
430 |
|
|
} |
431 |
|
|
|
432 |
|
|
///////////////////////////////////////////////////////////////////////////// |
433 |
|
✗ |
bool XmlParser::Element::getUnsignedList(const char *name, |
434 |
|
|
std::list<unsigned int> &intList) const |
435 |
|
|
{ |
436 |
|
✗ |
const char *value; |
437 |
|
|
|
438 |
|
✗ |
if (!(find(name, value)) or !value) |
439 |
|
✗ |
return false; |
440 |
|
|
|
441 |
|
✗ |
std::istringstream is(value); |
442 |
|
✗ |
is.imbue(std::locale::classic()); |
443 |
|
|
|
444 |
|
✗ |
while (is) { |
445 |
|
✗ |
unsigned int i; |
446 |
|
✗ |
char comma; |
447 |
|
|
|
448 |
|
✗ |
is >> i; |
449 |
|
✗ |
if (is) |
450 |
|
✗ |
intList.push_back(i); |
451 |
|
✗ |
is >> comma; |
452 |
|
|
} |
453 |
|
|
|
454 |
|
✗ |
return true; |
455 |
|
3 |
} |
456 |
|
|
|