Commit | Line | Data |
---|---|---|
0e23f538 TJ |
1 | /* |
2 | The software in this package is distributed under the GNU General | |
3 | Public License version 2 (with a special exception described below). | |
4 | ||
5 | A copy of GNU General Public License (GPL) is included in this distribution, | |
6 | in the file COPYING.GPL. | |
7 | ||
8 | As a special exception, if other files instantiate templates or use macros | |
9 | or inline functions from this file, or you compile this file and link it | |
10 | with other works to produce a work based on this file, this file | |
11 | does not by itself cause the resulting work to be covered | |
12 | by the GNU General Public License. | |
13 | ||
14 | However the source code for this file must still be made available | |
15 | in accordance with section (3) of the GNU General Public License. | |
16 | ||
17 | This exception does not invalidate any other reasons why a work based | |
18 | on this file might be covered by the GNU General Public License. | |
19 | */ | |
6a93d84a TJ |
20 | /** @file |
21 | * | |
22 | * (c) Copyright 2007-2008 by Intra2net AG | |
6a93d84a | 23 | */ |
e93545dd GE |
24 | |
25 | #include <iostream> | |
26 | #include <string> | |
27 | #include <sstream> | |
28 | #include <stdexcept> | |
5efd35b1 | 29 | #include <algorithm> |
5cd64148 | 30 | #include <cmath> // for round() |
e93545dd | 31 | |
a5f3af6e | 32 | #include <wchar.h> |
e93545dd GE |
33 | #include <stdlib.h> |
34 | #include <iconv.h> | |
35 | #include <i18n.h> | |
36 | ||
5cd64148 CH |
37 | #include <boost/numeric/conversion/cast.hpp> |
38 | ||
e93545dd GE |
39 | #include <stringfunc.hxx> |
40 | ||
41 | using namespace std; | |
42 | ||
6ab3bc95 RP |
43 | namespace I2n |
44 | { | |
6a93d84a TJ |
45 | |
46 | ||
6ab3bc95 RP |
47 | namespace |
48 | { | |
6a93d84a TJ |
49 | |
50 | const std::string hexDigitsLower("0123456789abcdef"); | |
51 | const std::string hexDigitsUpper("0123456789ABCDEF"); | |
52 | ||
53 | ||
54 | struct UpperFunc | |
55 | { | |
6ab3bc95 RP |
56 | char operator() (char c) |
57 | { | |
58 | return std::toupper(c); | |
59 | } | |
6a93d84a TJ |
60 | }; // eo struct UpperFunc |
61 | ||
62 | ||
63 | struct LowerFunc | |
64 | { | |
6ab3bc95 RP |
65 | char operator() (char c) |
66 | { | |
67 | return std::tolower(c); | |
68 | } | |
6a93d84a TJ |
69 | }; // eo struct LowerFunc |
70 | ||
71 | ||
72 | } // eo namespace <anonymous> | |
73 | ||
74 | ||
75 | ||
76 | /** | |
6ab3bc95 | 77 | * default list of Whitespaces (" \t\r\n"); |
6a93d84a | 78 | */ |
6ab3bc95 | 79 | const std::string Whitespaces = " \t\r\n"; |
6a93d84a TJ |
80 | |
81 | /** | |
82 | * default list of lineendings ("\r\n"); | |
83 | */ | |
6ab3bc95 | 84 | const std::string LineEndings= "\r\n"; |
6a93d84a TJ |
85 | |
86 | ||
87 | ||
88 | /** | |
89 | * @brief checks if a string begins with a given prefix. | |
90 | * @param[in,out] str the string which is tested | |
91 | * @param prefix the prefix which should be tested for. | |
92 | * @return @a true iff the prefix is not empty and the string begins with that prefix. | |
93 | */ | |
6ab3bc95 | 94 | bool has_prefix(const std::string& str, const std::string& prefix) |
6a93d84a | 95 | { |
6ab3bc95 RP |
96 | if (prefix.empty() || str.empty() || str.size() < prefix.size() ) |
97 | { | |
98 | return false; | |
99 | } | |
100 | return str.compare(0, prefix.size(), prefix) == 0; | |
101 | } // eo has_prefix(const std::string&,const std::string&) | |
6a93d84a TJ |
102 | |
103 | ||
104 | /** | |
105 | * @brief checks if a string ends with a given suffix. | |
106 | * @param[in,out] str the string which is tested | |
107 | * @param suffix the suffix which should be tested for. | |
108 | * @return @a true iff the suffix is not empty and the string ends with that suffix. | |
109 | */ | |
6ab3bc95 | 110 | bool has_suffix(const std::string& str, const std::string& suffix) |
6a93d84a | 111 | { |
6ab3bc95 RP |
112 | if (suffix.empty() || str.empty() || str.size() < suffix.size() ) |
113 | { | |
114 | return false; | |
115 | } | |
116 | return str.compare(str.size() - suffix.size(), suffix.size(), suffix) == 0; | |
117 | } // eo has_suffix(const std::string&,const std::string&) | |
6a93d84a TJ |
118 | |
119 | ||
120 | /** | |
121 | * cut off characters from a given list from front and end of a string. | |
122 | * @param[in,out] str the string which should be trimmed. | |
123 | * @param charlist the list of characters to remove from beginning and end of string | |
124 | * @return the result string. | |
125 | */ | |
6ab3bc95 RP |
126 | std::string trim_mod(std::string& str, const std::string& charlist) |
127 | { | |
128 | // first: trim the beginning: | |
129 | std::string::size_type pos= str.find_first_not_of (charlist); | |
130 | if (pos == std::string::npos) | |
131 | { | |
132 | // whole string consists of charlist (or is already empty) | |
133 | str.clear(); | |
134 | return str; | |
135 | } | |
136 | else if (pos>0) | |
137 | { | |
138 | // str starts with charlist | |
139 | str.erase(0,pos); | |
140 | } | |
141 | // now let's look at the tail: | |
142 | pos= str.find_last_not_of(charlist) +1; // note: we already know there is at least one other char! | |
143 | if ( pos < str.size() ) | |
144 | { | |
145 | str.erase(pos, str.size()-pos); | |
146 | } | |
147 | return str; | |
148 | } // eo trim_mod(std::string&,const std::string&) | |
6a93d84a TJ |
149 | |
150 | ||
151 | ||
152 | /** | |
153 | * removes last character from a string when it is in a list of chars to be removed. | |
154 | * @param[in,out] str the string. | |
155 | * @param what the list of chars which will be tested for. | |
156 | * @return the resulting string with last char removed (if applicable) | |
157 | */ | |
6ab3bc95 | 158 | std::string chomp_mod(std::string& str, const std::string& what) |
6a93d84a | 159 | { |
6ab3bc95 RP |
160 | if (str.empty() || what.empty() ) |
161 | { | |
162 | return str; | |
163 | } | |
164 | if (what.find(str.at (str.size()-1) ) != std::string::npos) | |
165 | { | |
166 | str.erase(str.size() - 1); | |
167 | } | |
168 | return str; | |
169 | } // eo chomp_mod(std::string&,const std::string&) | |
6a93d84a TJ |
170 | |
171 | ||
172 | /** | |
173 | * @brief converts a string to lower case. | |
174 | * @param[in,out] str the string to modify. | |
175 | * @return the string | |
176 | */ | |
6ab3bc95 | 177 | std::string to_lower_mod(std::string& str) |
6a93d84a | 178 | { |
6ab3bc95 RP |
179 | std::transform(str.begin(), str.end(), str.begin(), LowerFunc() ); |
180 | return str; | |
181 | } // eo to_lower_mod(std::string&) | |
6a93d84a TJ |
182 | |
183 | ||
184 | /** | |
185 | * @brief converts a string to upper case. | |
186 | * @param[in,out] str the string to modify. | |
187 | * @return the string | |
188 | */ | |
6ab3bc95 | 189 | std::string to_upper_mod(std::string& str) |
6a93d84a | 190 | { |
6ab3bc95 RP |
191 | std::transform( str.begin(), str.end(), str.begin(), UpperFunc() ); |
192 | return str; | |
193 | } // eo to_upper_mod(std::string&) | |
6a93d84a TJ |
194 | |
195 | ||
196 | ||
197 | /** | |
198 | * cut off characters from a given list from front and end of a string. | |
199 | * @param str the string which should be trimmed. | |
200 | * @param charlist the list of characters to remove from beginning and end of string | |
201 | * @return the result string. | |
202 | */ | |
6ab3bc95 RP |
203 | std::string trim (const std::string& str, const std::string& charlist) |
204 | { | |
205 | // first: trim the beginning: | |
206 | std::string::size_type pos0= str.find_first_not_of(charlist); | |
207 | if (pos0 == std::string::npos) | |
208 | { | |
209 | // whole string consists of charlist (or is already empty) | |
210 | return std::string(); | |
211 | } | |
212 | // now let's look at the end: | |
213 | std::string::size_type pos1= str.find_last_not_of(charlist); | |
214 | return str.substr(pos0, pos1 - pos0 + 1); | |
6a93d84a TJ |
215 | } // eo trim(const std:.string&,const std::string&) |
216 | ||
217 | ||
218 | /** | |
219 | * removes last character from a string when it is in a list of chars to be removed. | |
220 | * @param str the string. | |
221 | * @param what the list of chars which will be tested for. | |
222 | * @return the resulting string with last char removed (if applicable) | |
223 | */ | |
6ab3bc95 RP |
224 | std::string chomp (const std::string& str, const std::string& what) |
225 | { | |
226 | if (str.empty() || what.empty() ) | |
227 | { | |
228 | return str; | |
229 | } | |
230 | if (what.find(str.at (str.size()-1) ) != std::string::npos) | |
231 | { | |
232 | return str.substr(0, str.size()-1); | |
233 | } | |
234 | return str; | |
6a93d84a TJ |
235 | } // eo chomp(const std:.string&,const std::string&) |
236 | ||
237 | ||
238 | /** | |
239 | * @brief returns a lower case version of a given string. | |
240 | * @param str the string | |
241 | * @return the lower case version of the string | |
242 | */ | |
6ab3bc95 | 243 | std::string to_lower (const std::string& str) |
6a93d84a | 244 | { |
6ab3bc95 RP |
245 | std::string result(str); |
246 | return to_lower_mod(result); | |
247 | } // eo to_lower(const std::string&) | |
6a93d84a TJ |
248 | |
249 | ||
250 | /** | |
251 | * @brief returns a upper case version of a given string. | |
252 | * @param str the string | |
253 | * @return the upper case version of the string | |
254 | */ | |
6ab3bc95 | 255 | std::string to_upper(const std::string& str) |
6a93d84a | 256 | { |
6ab3bc95 RP |
257 | std::string result(str); |
258 | return to_upper_mod(result); | |
259 | } // eo to_upper(const std::string&) | |
6a93d84a TJ |
260 | |
261 | ||
262 | ||
263 | /** | |
264 | * @brief removes a given suffix from a string. | |
265 | * @param str the string. | |
266 | * @param suffix the suffix which should be removed if the string ends with it. | |
267 | * @return the string without the suffix. | |
268 | * | |
269 | * If the string ends with the suffix, it is removed. If the the string doesn't end | |
270 | * with the suffix the original string is returned. | |
271 | */ | |
6ab3bc95 | 272 | std::string remove_suffix(const std::string& str, const std::string& suffix) |
6a93d84a | 273 | { |
6ab3bc95 RP |
274 | if (has_suffix(str,suffix) ) |
275 | { | |
276 | return str.substr(0, str.size()-suffix.size() ); | |
277 | } | |
278 | return str; | |
279 | } // eo remove_suffix(const std::string&,const std::string&) | |
6a93d84a TJ |
280 | |
281 | ||
282 | ||
283 | /** | |
284 | * @brief removes a given prefix from a string. | |
285 | * @param str the string. | |
286 | * @param prefix the prefix which should be removed if the string begins with it. | |
287 | * @return the string without the prefix. | |
288 | * | |
289 | * If the string begins with the prefix, it is removed. If the the string doesn't begin | |
290 | * with the prefix the original string is returned. | |
291 | */ | |
6ab3bc95 | 292 | std::string remove_prefix(const std::string& str, const std::string& prefix) |
6a93d84a | 293 | { |
6ab3bc95 RP |
294 | if (has_prefix(str,prefix) ) |
295 | { | |
296 | return str.substr( prefix.size() ); | |
297 | } | |
298 | return str; | |
299 | } // eo remove_prefix(const std::string&,const std::string&) | |
6a93d84a TJ |
300 | |
301 | ||
302 | /** | |
303 | * split a string to key and value delimited by a given delimiter. | |
6ab3bc95 | 304 | * The resulting key and value strings are trimmed (Whitespaces removed at beginning and end). |
6a93d84a TJ |
305 | * @param str the string which should be splitted. |
306 | * @param[out] key the resulting key | |
307 | * @param[out] value the resulting value | |
308 | * @param delimiter the delimiter between key and value; default is '='. | |
309 | * @return @a true if the split was successful. | |
310 | */ | |
6ab3bc95 RP |
311 | bool pair_split( |
312 | const std::string& str, | |
313 | std::string& key, | |
314 | std::string& value, | |
315 | char delimiter) | |
316 | { | |
317 | std::string::size_type pos = str.find (delimiter); | |
318 | if (pos == std::string::npos) return false; | |
319 | key= str.substr(0,pos); | |
320 | value= str.substr(pos+1); | |
321 | trim_mod(key); | |
322 | trim_mod(value); | |
323 | return true; | |
324 | } // eo pair_split(const std::string&,std::string&,std::string&,char) | |
6a93d84a TJ |
325 | |
326 | ||
327 | /** | |
328 | * splits a string by given delimiter | |
329 | * | |
330 | * @param[in] str the string which should be splitted. | |
331 | * @param[out] result the list resulting from splitting @a str. | |
332 | * @param[in] delimiter the delimiter (word/phrase) at which @a str should be splitted. | |
333 | * @param[in] omit_empty should empty parts not be stored? | |
334 | * @param[in] trim_list list of characters the parts should be trimmed by. | |
335 | * (empty string results in no trim) | |
336 | */ | |
6ab3bc95 RP |
337 | void split_string( |
338 | const std::string& str, | |
339 | std::list<std::string>& result, | |
340 | const std::string& delimiter, | |
341 | bool omit_empty, | |
342 | const std::string& trim_list | |
6a93d84a TJ |
343 | ) |
344 | { | |
6ab3bc95 RP |
345 | std::string::size_type pos, last_pos=0; |
346 | bool delimiter_found= false; | |
347 | while ( last_pos < str.size() && last_pos != std::string::npos) | |
348 | { | |
349 | pos= str.find(delimiter, last_pos); | |
350 | std::string part; | |
351 | if (pos == std::string::npos) | |
352 | { | |
353 | part= str.substr(last_pos); | |
354 | delimiter_found= false; | |
355 | } | |
356 | else | |
357 | { | |
358 | part= str.substr(last_pos, pos-last_pos); | |
359 | delimiter_found=true; | |
360 | } | |
361 | if (pos != std::string::npos) | |
362 | { | |
363 | last_pos= pos+ delimiter.size(); | |
364 | } | |
365 | else | |
366 | { | |
367 | last_pos= std::string::npos; | |
368 | } | |
369 | if (!trim_list.empty() ) trim_mod (part, trim_list); | |
370 | if (omit_empty && part.empty() ) continue; | |
371 | result.push_back( part ); | |
372 | } | |
373 | // if the string ends with a delimiter we need to append an empty string if no omit_empty | |
374 | // was given. | |
375 | // (this way we keep the split result consistent to a join operation) | |
376 | if (delimiter_found && !omit_empty) | |
377 | { | |
378 | result.push_back(""); | |
379 | } | |
380 | } // eo split_string(const std::string&,std::list< std::string >&,const std::string&,bool,const std::string&) | |
6a93d84a TJ |
381 | |
382 | ||
338da253 CH |
383 | /** call split_string with list<string>, converts result to vector; vector is clear()-ed first |
384 | * | |
385 | * Note: Uses 3 O(n)-operations: list.size, vector.resize and std::swap_ranges; | |
386 | * not sure whether there is a better way to do this | |
387 | * */ | |
388 | void split_string( | |
389 | const std::string& str, | |
390 | std::vector<std::string>& result, | |
391 | const std::string& delimiter, | |
392 | bool omit_empty, | |
393 | const std::string& trim_list | |
394 | ) | |
395 | { | |
396 | std::list<std::string> tmp; | |
397 | split_string(str, tmp, delimiter, omit_empty, trim_list); | |
398 | std::size_t size = tmp.size(); // this is O(n) | |
399 | result.clear(); | |
400 | result.resize(size); // also O(n) | |
401 | std::swap_ranges(tmp.begin(), tmp.end(), result.begin()); // also O(n) | |
402 | } | |
403 | ||
6a93d84a TJ |
404 | /** |
405 | * splits a string by a given delimiter | |
406 | * @param str the string which should be splitted. | |
407 | * @param delimiter delimiter the delimiter (word/phrase) at which @a str should be splitted. | |
408 | * @param[in] omit_empty should empty parts not be stored? | |
409 | * @param[in] trim_list list of characters the parts should be trimmed by. | |
410 | * (empty string results in no trim) | |
411 | * @return the list resulting from splitting @a str. | |
412 | */ | |
6ab3bc95 RP |
413 | std::list<std::string> split_string( |
414 | const std::string& str, | |
415 | const std::string& delimiter, | |
416 | bool omit_empty, | |
417 | const std::string& trim_list | |
6a93d84a TJ |
418 | ) |
419 | { | |
6ab3bc95 RP |
420 | std::list<std::string> result; |
421 | split_string(str, result, delimiter, omit_empty, trim_list); | |
422 | return result; | |
423 | } // eo split_string(const std::string&,const std::string&,bool,const std::string&) | |
6a93d84a TJ |
424 | |
425 | ||
426 | /** | |
427 | * @brief joins a list of strings into a single string. | |
428 | * | |
6ab3bc95 RP |
429 | * This funtion is (basically) the reverse operation of @a split_string. |
430 | * | |
6a93d84a TJ |
431 | * @param parts the list of strings. |
432 | * @param delimiter the delimiter which is inserted between the strings. | |
433 | * @return the joined string. | |
434 | */ | |
6ab3bc95 RP |
435 | std::string join_string( |
436 | const std::list< std::string >& parts, | |
437 | const std::string& delimiter | |
6a93d84a TJ |
438 | ) |
439 | { | |
6ab3bc95 RP |
440 | std::string result; |
441 | if (! parts.empty() ) | |
442 | { | |
443 | std::list< std::string >::const_iterator it= parts.begin(); | |
444 | result = *it; | |
445 | while ( ++it != parts.end() ) | |
446 | { | |
447 | result+= delimiter; | |
448 | result+= *it; | |
449 | } | |
450 | } | |
451 | return result; | |
452 | } // eo join_string(const std::list< std::string >&,const std::string&) | |
6a93d84a TJ |
453 | |
454 | ||
376ec4fa CH |
455 | /** @brief same as join_string for list, except uses a vector */ |
456 | std::string join_string( | |
457 | const std::vector< std::string >& parts, | |
458 | const std::string& delimiter | |
459 | ) | |
460 | { | |
461 | std::string result; | |
462 | if (! parts.empty() ) | |
463 | { | |
464 | std::vector< std::string >::const_iterator it= parts.begin(); | |
465 | result = *it; | |
466 | while ( ++it != parts.end() ) | |
467 | { | |
468 | result+= delimiter; | |
469 | result+= *it; | |
470 | } | |
471 | } | |
472 | return result; | |
473 | } // eo join_string(const std::vector< std::string >&,const std::string&) | |
474 | ||
475 | ||
6a93d84a TJ |
476 | |
477 | /* | |
478 | ** conversions | |
479 | */ | |
480 | ||
481 | ||
482 | /** | |
483 | * @brief returns a hex string from a binary string. | |
484 | * @param str the (binary) string | |
485 | * @param upper_case_digits determine whether to use upper case characters for digits A-F. | |
486 | * @return the string in hex notation. | |
487 | */ | |
6ab3bc95 RP |
488 | std::string convert_binary_to_hex( |
489 | const std::string& str, | |
490 | bool upper_case_digits | |
6a93d84a TJ |
491 | ) |
492 | { | |
6ab3bc95 RP |
493 | std::string result; |
494 | std::string hexDigits(upper_case_digits ? hexDigitsUpper : hexDigitsLower); | |
495 | for ( std::string::const_iterator it= str.begin(); | |
496 | it != str.end(); | |
497 | ++it) | |
498 | { | |
499 | result.push_back( hexDigits[ ( (*it) >> 4) & 0x0f ] ); | |
500 | result.push_back( hexDigits[ (*it) & 0x0f ] ); | |
501 | } | |
502 | return result; | |
503 | } // eo convert_binary_to_hex(const std::string&,bool) | |
6a93d84a TJ |
504 | |
505 | ||
506 | /** | |
507 | * @brief converts a hex digit string to binary string. | |
508 | * @param str hex digit string | |
509 | * @return the binary string. | |
510 | * | |
511 | * The hex digit string may contains white spaces or colons which are treated | |
512 | * as delimiters between hex digit groups. | |
513 | * | |
514 | * @todo rework the handling of half nibbles (consistency)! | |
515 | */ | |
6ab3bc95 RP |
516 | std::string convert_hex_to_binary( |
517 | const std::string& str | |
6a93d84a | 518 | ) |
6ab3bc95 RP |
519 | throw (std::runtime_error) |
520 | { | |
521 | std::string result; | |
522 | char c= 0; | |
523 | bool hasNibble= false; | |
524 | bool lastWasWS= true; | |
525 | for ( std::string::const_iterator it= str.begin(); | |
526 | it != str.end(); | |
527 | ++it) | |
528 | { | |
529 | std::string::size_type p = hexDigitsLower.find( *it ); | |
530 | if (p== std::string::npos) | |
531 | { | |
532 | p= hexDigitsUpper.find( *it ); | |
533 | } | |
534 | if (p == std::string::npos) | |
535 | { | |
536 | if ( ( Whitespaces.find( *it ) != std::string::npos) // is it a whitespace? | |
6a93d84a | 537 | or ( *it == ':') // or a colon? |
6ab3bc95 RP |
538 | ) |
539 | { | |
540 | // we treat that as a valid delimiter: | |
541 | if (hasNibble) | |
6a93d84a | 542 | { |
6ab3bc95 RP |
543 | // 1 nibble before WS is treate as lower part: |
544 | result.push_back(c); | |
545 | // reset state: | |
546 | hasNibble= false; | |
6a93d84a | 547 | } |
6ab3bc95 RP |
548 | lastWasWS= true; |
549 | continue; | |
550 | } | |
551 | } | |
552 | if (p == std::string::npos ) | |
553 | { | |
554 | throw runtime_error("illegal character in hex digit string: " + str); | |
555 | } | |
556 | lastWasWS= false; | |
557 | if (hasNibble) | |
558 | { | |
559 | c<<=4; | |
560 | } | |
561 | else | |
562 | { | |
563 | c=0; | |
564 | } | |
565 | c+= (p & 0x0f); | |
566 | if (hasNibble) | |
567 | { | |
568 | //we already had a nibble, so a char is complete now: | |
569 | result.push_back( c ); | |
570 | hasNibble=false; | |
571 | } | |
572 | else | |
573 | { | |
574 | // this is the first nibble of a new char: | |
575 | hasNibble=true; | |
576 | } | |
577 | } | |
578 | if (hasNibble) | |
579 | { | |
580 | //well, there is one nibble left | |
581 | // let's do some heuristics: | |
582 | if (lastWasWS) | |
583 | { | |
584 | // if the preceeding character was a white space (or a colon) | |
585 | // we treat the nibble as lower part: | |
586 | //( this is consistent with shortened hex notations where leading zeros are not noted) | |
587 | result.push_back( c ); | |
588 | } | |
589 | else | |
590 | { | |
591 | // if it was part of a hex digit chain, we treat it as UPPER part (!!) | |
592 | result.push_back( c << 4 ); | |
593 | } | |
594 | } | |
595 | return result; | |
596 | } // eo convert_hex_to_binary(const std::string&) | |
597 | ||
598 | ||
599 | } // eo namespace I2n | |
600 | ||
601 | ||
602 | ||
6a93d84a | 603 | |
e93545dd GE |
604 | std::string iso_to_utf8(const std::string& isostring) |
605 | { | |
6ab3bc95 | 606 | string result; |
118e216e | 607 | |
6ab3bc95 | 608 | iconv_t i2utf8 = iconv_open("UTF-8", "ISO-8859-1"); |
118e216e | 609 | |
6ab3bc95 RP |
610 | if (iso_to_utf8 == (iconv_t)-1) |
611 | throw runtime_error("iconv can't convert from ISO-8859-1 to UTF-8"); | |
118e216e | 612 | |
6ab3bc95 RP |
613 | size_t in_size=isostring.size(); |
614 | size_t out_size=in_size*4; | |
118e216e | 615 | |
6ab3bc95 RP |
616 | char *buf = (char *)malloc(out_size+1); |
617 | if (buf == NULL) | |
618 | throw runtime_error("out of memory for iconv buffer"); | |
e93545dd | 619 | |
5a4ecb51 | 620 | char *in = (char *)isostring.c_str(); |
6ab3bc95 RP |
621 | char *out = buf; |
622 | iconv(i2utf8, &in, &in_size, &out, &out_size); | |
118e216e | 623 | |
6ab3bc95 | 624 | buf[isostring.size()*4-out_size]=0; |
118e216e | 625 | |
6ab3bc95 | 626 | result=buf; |
118e216e | 627 | |
6ab3bc95 RP |
628 | free(buf); |
629 | iconv_close(i2utf8); | |
118e216e | 630 | |
6ab3bc95 | 631 | return result; |
e93545dd GE |
632 | } |
633 | ||
634 | std::string utf8_to_iso(const std::string& utf8string) | |
635 | { | |
6ab3bc95 | 636 | string result; |
118e216e | 637 | |
6ab3bc95 | 638 | iconv_t utf82iso = iconv_open("ISO-8859-1","UTF-8"); |
118e216e | 639 | |
6ab3bc95 RP |
640 | if (utf82iso == (iconv_t)-1) |
641 | throw runtime_error("iconv can't convert from UTF-8 to ISO-8859-1"); | |
118e216e | 642 | |
6ab3bc95 RP |
643 | size_t in_size=utf8string.size(); |
644 | size_t out_size=in_size; | |
118e216e | 645 | |
6ab3bc95 RP |
646 | char *buf = (char *)malloc(out_size+1); |
647 | if (buf == NULL) | |
648 | throw runtime_error("out of memory for iconv buffer"); | |
e93545dd | 649 | |
5a4ecb51 | 650 | char *in = (char *)utf8string.c_str(); |
6ab3bc95 RP |
651 | char *out = buf; |
652 | iconv(utf82iso, &in, &in_size, &out, &out_size); | |
118e216e | 653 | |
6ab3bc95 | 654 | buf[utf8string.size()-out_size]=0; |
118e216e | 655 | |
6ab3bc95 | 656 | result=buf; |
118e216e | 657 | |
6ab3bc95 RP |
658 | free(buf); |
659 | iconv_close(utf82iso); | |
e93545dd | 660 | |
6ab3bc95 | 661 | return result; |
e93545dd GE |
662 | } |
663 | ||
a5f3af6e GE |
664 | wchar_t* utf8_to_wbuf(const std::string& utf8string) |
665 | { | |
6ab3bc95 | 666 | iconv_t utf82wstr = iconv_open("UCS-4LE","UTF-8"); |
a5f3af6e | 667 | |
6ab3bc95 RP |
668 | if (utf82wstr == (iconv_t)-1) |
669 | throw runtime_error("iconv can't convert from UTF-8 to UCS-4"); | |
a5f3af6e | 670 | |
6ab3bc95 RP |
671 | size_t in_size=utf8string.size(); |
672 | size_t out_size= (in_size+1)*sizeof(wchar_t); | |
a5f3af6e | 673 | |
6ab3bc95 RP |
674 | wchar_t *buf = (wchar_t *)malloc(out_size); |
675 | if (buf == NULL) | |
676 | throw runtime_error("out of memory for iconv buffer"); | |
a5f3af6e | 677 | |
5a4ecb51 | 678 | char *in = (char *)utf8string.c_str(); |
6ab3bc95 | 679 | char *out = (char*) buf; |
dbd6d77c | 680 | if (iconv(utf82wstr, &in, &in_size, &out, &out_size) == (size_t)-1) |
6ab3bc95 | 681 | throw runtime_error("error converting char encodings"); |
a5f3af6e | 682 | |
6ab3bc95 | 683 | buf[ ( (utf8string.size()+1)*sizeof(wchar_t)-out_size) /sizeof(wchar_t) ]=0; |
a5f3af6e | 684 | |
6ab3bc95 | 685 | iconv_close(utf82wstr); |
a5f3af6e | 686 | |
6ab3bc95 | 687 | return buf; |
a5f3af6e GE |
688 | } |
689 | ||
13cc4db1 | 690 | std::string utf7imap_to_utf8(const std::string& utf7imapstring) |
d116a071 | 691 | { |
6ab3bc95 | 692 | string result; |
118e216e | 693 | |
6ab3bc95 | 694 | iconv_t utf7imap2utf8 = iconv_open("UTF-8","UTF-7-IMAP"); |
118e216e | 695 | |
6ab3bc95 RP |
696 | if (utf7imap2utf8 == (iconv_t)-1) |
697 | throw runtime_error("iconv can't convert from UTF-7-IMAP to UTF-8"); | |
118e216e | 698 | |
6ab3bc95 RP |
699 | size_t in_size=utf7imapstring.size(); |
700 | size_t out_size=in_size*4; | |
118e216e | 701 | |
6ab3bc95 RP |
702 | char *buf = (char *)malloc(out_size+1); |
703 | if (buf == NULL) | |
704 | throw runtime_error("out of memory for iconv buffer"); | |
d116a071 | 705 | |
5a4ecb51 | 706 | char *in = (char *)utf7imapstring.c_str(); |
6ab3bc95 RP |
707 | char *out = buf; |
708 | iconv(utf7imap2utf8, &in, &in_size, &out, &out_size); | |
118e216e | 709 | |
6ab3bc95 | 710 | buf[utf7imapstring.size()*4-out_size]=0; |
118e216e | 711 | |
6ab3bc95 | 712 | result=buf; |
118e216e | 713 | |
6ab3bc95 RP |
714 | free(buf); |
715 | iconv_close(utf7imap2utf8); | |
118e216e | 716 | |
6ab3bc95 | 717 | return result; |
118e216e TJ |
718 | } |
719 | ||
6a2b6dd1 TJ |
720 | std::string utf8_to_utf7imap(const std::string& utf8string) |
721 | { | |
6ab3bc95 | 722 | string result; |
6a2b6dd1 | 723 | |
6ab3bc95 | 724 | iconv_t utf82utf7imap = iconv_open("UTF-7-IMAP", "UTF-8"); |
6a2b6dd1 | 725 | |
6ab3bc95 RP |
726 | if (utf82utf7imap == (iconv_t)-1) |
727 | throw runtime_error("iconv can't convert from UTF-7-IMAP to UTF-8"); | |
6a2b6dd1 | 728 | |
6ab3bc95 RP |
729 | // UTF-7 is base64 encoded, a buffer 10x as large |
730 | // as the utf-8 buffer should be enough. If not the string will be truncated. | |
731 | size_t in_size=utf8string.size(); | |
732 | size_t out_size=in_size*10; | |
6a2b6dd1 | 733 | |
6ab3bc95 RP |
734 | char *buf = (char *)malloc(out_size+1); |
735 | if (buf == NULL) | |
736 | throw runtime_error("out of memory for iconv buffer"); | |
6a2b6dd1 | 737 | |
5a4ecb51 | 738 | char *in = (char *)utf8string.c_str(); |
6ab3bc95 RP |
739 | char *out = buf; |
740 | iconv(utf82utf7imap, &in, &in_size, &out, &out_size); | |
6a2b6dd1 | 741 | |
6ab3bc95 | 742 | buf[utf8string.size()*10-out_size]= 0; |
6a2b6dd1 | 743 | |
6ab3bc95 | 744 | result=buf; |
6a2b6dd1 | 745 | |
6ab3bc95 RP |
746 | free(buf); |
747 | iconv_close(utf82utf7imap); | |
6a2b6dd1 | 748 | |
6ab3bc95 | 749 | return result; |
6a2b6dd1 TJ |
750 | } |
751 | ||
118e216e TJ |
752 | // Tokenize string by (html) tags |
753 | void tokenize_by_tag(vector<pair<string,bool> > &tokenized, const std::string &input) | |
754 | { | |
6ab3bc95 RP |
755 | string::size_type pos, len = input.size(); |
756 | bool inside_tag = false; | |
757 | string current; | |
758 | ||
759 | for (pos = 0; pos < len; pos++) | |
760 | { | |
761 | if (input[pos] == '<') | |
762 | { | |
763 | inside_tag = true; | |
764 | ||
765 | if (!current.empty() ) | |
766 | { | |
767 | tokenized.push_back( make_pair(current, false) ); | |
768 | current = ""; | |
769 | } | |
770 | ||
771 | current += input[pos]; | |
772 | } | |
773 | else if (input[pos] == '>' && inside_tag) | |
774 | { | |
775 | current += input[pos]; | |
776 | inside_tag = false; | |
777 | if (!current.empty() ) | |
778 | { | |
779 | tokenized.push_back( make_pair(current, true) ); | |
780 | current = ""; | |
781 | } | |
782 | } | |
783 | else | |
784 | current += input[pos]; | |
785 | } | |
786 | ||
787 | // String left over in buffer? | |
788 | if (!current.empty() ) | |
789 | tokenized.push_back( make_pair(current, false) ); | |
790 | } // eo tokenize_by_tag | |
118e216e | 791 | |
118e216e TJ |
792 | |
793 | std::string strip_html_tags(const std::string &input) | |
794 | { | |
6ab3bc95 RP |
795 | // Pair first: string, second: isTag |
796 | vector<pair<string,bool> > tokenized; | |
797 | tokenize_by_tag (tokenized, input); | |
118e216e | 798 | |
6ab3bc95 RP |
799 | string output; |
800 | vector<pair<string,bool> >::const_iterator token, tokens_end = tokenized.end(); | |
83d700e9 | 801 | for (token = tokenized.begin(); token != tokens_end; ++token) |
6ab3bc95 RP |
802 | if (!token->second) |
803 | output += token->first; | |
804 | ||
805 | return output; | |
806 | } // eo strip_html_tags | |
118e216e | 807 | |
118e216e TJ |
808 | |
809 | // Smart-encode HTML en | |
810 | string smart_html_entities(const std::string &input) | |
811 | { | |
6ab3bc95 RP |
812 | // Pair first: string, second: isTag |
813 | vector<pair<string,bool> > tokenized; | |
814 | tokenize_by_tag (tokenized, input); | |
815 | ||
816 | string output; | |
817 | vector<pair<string,bool> >::const_iterator token, tokens_end = tokenized.end(); | |
83d700e9 | 818 | for (token = tokenized.begin(); token != tokens_end; ++token) |
6ab3bc95 RP |
819 | { |
820 | // keep HTML tags as they are | |
821 | if (token->second) | |
822 | output += token->first; | |
823 | else | |
824 | output += html_entities(token->first); | |
825 | } | |
826 | ||
827 | return output; | |
118e216e TJ |
828 | } |
829 | ||
6ab3bc95 | 830 | |
a5f3af6e GE |
831 | string::size_type find_8bit(const std::string &str) |
832 | { | |
6ab3bc95 RP |
833 | string::size_type l=str.size(); |
834 | for (string::size_type p=0; p < l; p++) | |
835 | if (static_cast<unsigned char>(str[p]) > 127) | |
836 | return p; | |
a5f3af6e | 837 | |
6ab3bc95 | 838 | return string::npos; |
a5f3af6e GE |
839 | } |
840 | ||
118e216e TJ |
841 | // encoded UTF-8 chars into HTML entities |
842 | string html_entities(std::string str) | |
843 | { | |
6ab3bc95 RP |
844 | // Normal chars |
845 | replace_all (str, "&", "&"); | |
6ab3bc95 RP |
846 | replace_all (str, "<", "<"); |
847 | replace_all (str, ">", ">"); | |
980577e1 TJ |
848 | replace_all (str, "\"", """); |
849 | replace_all (str, "'", "'"); | |
850 | replace_all (str, "/", "/"); | |
6ab3bc95 RP |
851 | |
852 | // Umlauts | |
853 | replace_all (str, "\xC3\xA4", "ä"); | |
854 | replace_all (str, "\xC3\xB6", "ö"); | |
855 | replace_all (str, "\xC3\xBC", "ü"); | |
856 | replace_all (str, "\xC3\x84", "Ä"); | |
857 | replace_all (str, "\xC3\x96", "Ö"); | |
858 | replace_all (str, "\xC3\x9C", "Ü"); | |
859 | ||
860 | // Misc | |
861 | replace_all (str, "\xC3\x9F", "ß"); | |
862 | ||
863 | // conversion of remaining non-ASCII chars needed? | |
864 | // just do if needed because of performance | |
865 | if (find_8bit(str) != string::npos) | |
866 | { | |
867 | // convert to fixed-size encoding UTF-32 | |
868 | wchar_t* wbuf=utf8_to_wbuf(str); | |
869 | ostringstream target; | |
870 | ||
871 | // replace all non-ASCII chars with HTML representation | |
872 | for (int p=0; wbuf[p] != 0; p++) | |
873 | { | |
874 | unsigned int c=wbuf[p]; | |
875 | ||
876 | if (c <= 127) | |
877 | target << static_cast<unsigned char>(c); | |
878 | else | |
879 | target << "&#" << c << ';'; | |
880 | } | |
881 | ||
882 | free(wbuf); | |
883 | ||
884 | str=target.str(); | |
885 | } | |
886 | ||
887 | return str; | |
888 | } // eo html_entities(std::string) | |
889 | ||
554f813d GE |
890 | // convert HTML entities to something that can be viewed on a basic text console (restricted to ASCII-7) |
891 | string html_entities_to_console(std::string str) | |
892 | { | |
893 | // Normal chars | |
894 | replace_all (str, "&", "&"); | |
895 | replace_all (str, "<", "<"); | |
896 | replace_all (str, ">", ">"); | |
897 | replace_all (str, """, "\""); | |
898 | replace_all (str, "'", "'"); | |
899 | replace_all (str, "/", "/"); | |
900 | ||
901 | // Umlauts | |
902 | replace_all (str, "ä", "ae"); | |
903 | replace_all (str, "ö", "oe"); | |
904 | replace_all (str, "ü", "ue"); | |
905 | replace_all (str, "Ä", "Ae"); | |
906 | replace_all (str, "Ö", "Oe"); | |
907 | replace_all (str, "Ü", "Ue"); | |
908 | ||
909 | // Misc | |
910 | replace_all (str, "ß", "ss"); | |
911 | ||
912 | return str; | |
913 | } | |
118e216e | 914 | |
e93545dd GE |
915 | bool replace_all(string &base, const char *ist, const char *soll) |
916 | { | |
6ab3bc95 RP |
917 | string i=ist; |
918 | string s=soll; | |
919 | return replace_all(base,&i,&s); | |
e93545dd GE |
920 | } |
921 | ||
922 | bool replace_all(string &base, const string &ist, const char *soll) | |
923 | { | |
6ab3bc95 RP |
924 | string s=soll; |
925 | return replace_all(base,&ist,&s); | |
e93545dd GE |
926 | } |
927 | ||
928 | bool replace_all(string &base, const string *ist, const string *soll) | |
929 | { | |
6ab3bc95 | 930 | return replace_all(base,*ist,*soll); |
e93545dd GE |
931 | } |
932 | ||
933 | bool replace_all(string &base, const char *ist, const string *soll) | |
934 | { | |
6ab3bc95 RP |
935 | string i=ist; |
936 | return replace_all(base,&i,soll); | |
e93545dd GE |
937 | } |
938 | ||
939 | bool replace_all(string &base, const string &ist, const string &soll) | |
940 | { | |
6ab3bc95 RP |
941 | bool found_ist = false; |
942 | string::size_type a=0; | |
943 | ||
944 | if (ist.empty() ) | |
945 | throw runtime_error ("replace_all called with empty search string"); | |
e93545dd | 946 | |
6ab3bc95 RP |
947 | while ( (a=base.find(ist,a) ) != string::npos) |
948 | { | |
949 | base.replace(a,ist.size(),soll); | |
950 | a=a+soll.size(); | |
951 | found_ist = true; | |
952 | } | |
1ec2064e | 953 | |
6ab3bc95 | 954 | return found_ist; |
e93545dd GE |
955 | } |
956 | ||
b953bf36 GE |
957 | /** |
958 | * @brief replaces all characters that could be problematic or impose a security risk when being logged | |
959 | * @param str the original string | |
960 | * @param replace_with the character to replace the unsafe chars with | |
961 | * @return a string that is safe to send to syslog or other logfiles | |
962 | * | |
963 | * All chars between 0x20 (space) and 0x7E (~) (including) are considered safe for logging. | |
964 | * See e.g. RFC 5424, section 8.2 or the posix character class "printable". | |
965 | * This eliminates all possible problems with NUL, control characters, 8 bit chars, UTF8. | |
966 | * | |
967 | */ | |
968 | std::string sanitize_for_logging(const std::string &str, const char replace_with) | |
969 | { | |
970 | std::string output=str; | |
971 | ||
c0e32d64 GE |
972 | const string::size_type len = output.size(); |
973 | for (std::string::size_type p=0; p < len; p++) | |
b953bf36 GE |
974 | if (output[p] < 0x20 || output[p] > 0x7E) |
975 | output[p]=replace_with; | |
976 | ||
977 | return output; | |
978 | } | |
979 | ||
e5b21dbb | 980 | #if 0 |
e93545dd GE |
981 | string to_lower(const string &src) |
982 | { | |
6ab3bc95 | 983 | string dst = src; |
e93545dd | 984 | |
6ab3bc95 RP |
985 | string::size_type pos, end = dst.size(); |
986 | for (pos = 0; pos < end; pos++) | |
987 | dst[pos] = tolower(dst[pos]); | |
e93545dd | 988 | |
6ab3bc95 | 989 | return dst; |
e93545dd GE |
990 | } |
991 | ||
992 | string to_upper(const string &src) | |
993 | { | |
6ab3bc95 | 994 | string dst = src; |
e93545dd | 995 | |
6ab3bc95 RP |
996 | string::size_type pos, end = dst.size(); |
997 | for (pos = 0; pos < end; pos++) | |
998 | dst[pos] = toupper(dst[pos]); | |
e93545dd | 999 | |
6ab3bc95 | 1000 | return dst; |
e93545dd | 1001 | } |
e5b21dbb | 1002 | #endif |
e93545dd | 1003 | |
83809f5e | 1004 | const int MAX_UNIT_FORMAT_SYMBOLS = 6; |
d1ea9075 | 1005 | |
2cb9a9c5 | 1006 | const string shortUnitFormatSymbols[MAX_UNIT_FORMAT_SYMBOLS] = { |
d1ea9075 GMF |
1007 | " B", |
1008 | " KB", | |
1009 | " MB", | |
1010 | " GB", | |
1011 | " TB", | |
83809f5e | 1012 | " PB" |
d1ea9075 GMF |
1013 | }; |
1014 | ||
2cb9a9c5 | 1015 | const string longUnitFormatSymbols[MAX_UNIT_FORMAT_SYMBOLS] = { |
5cb766b9 GMF |
1016 | i18n_noop(" Bytes"), |
1017 | i18n_noop(" KBytes"), | |
1018 | i18n_noop(" MBytes"), | |
1019 | i18n_noop(" GBytes"), | |
1020 | i18n_noop(" TBytes"), | |
83809f5e | 1021 | i18n_noop(" PBytes") |
d1ea9075 GMF |
1022 | }; |
1023 | ||
72a94426 GMF |
1024 | |
1025 | long double rounding_upwards( | |
e91c1952 | 1026 | const long double number, |
72a94426 GMF |
1027 | const int rounding_multiplier |
1028 | ) | |
1029 | { | |
1030 | long double rounded_number; | |
1031 | rounded_number = number * rounding_multiplier; | |
1032 | rounded_number += 0.5; | |
1033 | rounded_number = (int64_t) (rounded_number); | |
1034 | rounded_number = (long double) (rounded_number) / (long double) (rounding_multiplier); | |
1035 | ||
1036 | return rounded_number; | |
1037 | } | |
1038 | ||
1039 | ||
81267544 GMF |
1040 | string nice_unit_format( |
1041 | const int64_t input, | |
70fc0674 GMF |
1042 | const UnitFormat format, |
1043 | const UnitBase base | |
81267544 | 1044 | ) |
6ab3bc95 | 1045 | { |
d1ea9075 | 1046 | // select the system of units (decimal or binary) |
81267544 | 1047 | int multiple = 0; |
a398513a | 1048 | if (base == UnitBase1000) |
81267544 GMF |
1049 | { |
1050 | multiple = 1000; | |
1051 | } | |
1052 | else | |
1053 | { | |
1054 | multiple = 1024; | |
1055 | } | |
1056 | ||
1057 | long double size = input; | |
6ab3bc95 | 1058 | |
d1ea9075 GMF |
1059 | // check the size of the input number to fit in the appropriate symbol |
1060 | int sizecount = 0; | |
81267544 | 1061 | while (size > multiple) |
6ab3bc95 | 1062 | { |
81267544 GMF |
1063 | size = size / multiple; |
1064 | sizecount++; | |
83809f5e GMF |
1065 | |
1066 | // rollback to the previous values and stop the loop when cannot | |
1067 | // represent the number length. | |
1068 | if (sizecount >= MAX_UNIT_FORMAT_SYMBOLS) | |
1069 | { | |
1070 | size = size * multiple; | |
1071 | sizecount--; | |
1072 | break; | |
1073 | } | |
6ab3bc95 RP |
1074 | } |
1075 | ||
a398513a GMF |
1076 | // round the input number "half up" to multiples of 10 |
1077 | const int rounding_multiplier = 10; | |
72a94426 | 1078 | size = rounding_upwards(size, rounding_multiplier); |
6ab3bc95 | 1079 | |
d1ea9075 | 1080 | // format the input number, placing the appropriate symbol |
6ab3bc95 | 1081 | ostringstream out; |
6ab3bc95 | 1082 | out.setf (ios::fixed); |
a398513a | 1083 | if (format == ShortUnitFormat) |
d1ea9075 GMF |
1084 | { |
1085 | out.precision(1); | |
68d37a5c | 1086 | out << size << i18n( shortUnitFormatSymbols[sizecount].c_str() ); |
d1ea9075 GMF |
1087 | } |
1088 | else | |
6ab3bc95 | 1089 | { |
d1ea9075 | 1090 | out.precision (2); |
68d37a5c | 1091 | out << size << i18n( longUnitFormatSymbols[sizecount].c_str() ); |
6ab3bc95 RP |
1092 | } |
1093 | ||
1094 | return out.str(); | |
1095 | } // eo nice_unit_format(int input) | |
1096 | ||
e93545dd | 1097 | |
5cd64148 CH |
1098 | string nice_unit_format( |
1099 | const double input, | |
1100 | const UnitFormat format, | |
1101 | const UnitBase base | |
1102 | ) | |
1103 | { | |
1104 | // round as double and cast to int64_t | |
1105 | // cast raised overflow error near max val of int64_t (~9.2e18, see unittest) | |
1106 | int64_t input_casted_and_rounded = | |
1107 | boost::numeric_cast<int64_t>( round(input) ); | |
1108 | ||
1109 | // now call other | |
1110 | return nice_unit_format( input_casted_and_rounded, format, base ); | |
1111 | } // eo nice_unit_format(double input) | |
1112 | ||
1113 | ||
47c07fba GE |
1114 | string escape(const string &s) |
1115 | { | |
6ab3bc95 RP |
1116 | string out(s); |
1117 | string::size_type p; | |
47c07fba | 1118 | |
6ab3bc95 RP |
1119 | p=0; |
1120 | while ( (p=out.find_first_of("\"\\",p) ) !=out.npos) | |
1121 | { | |
1122 | out.insert (p,"\\"); | |
1123 | p+=2; | |
1124 | } | |
47c07fba | 1125 | |
6ab3bc95 RP |
1126 | p=0; |
1127 | while ( (p=out.find_first_of("\r",p) ) !=out.npos) | |
1128 | { | |
1129 | out.replace (p,1,"\\r"); | |
1130 | p+=2; | |
1131 | } | |
47c07fba | 1132 | |
6ab3bc95 RP |
1133 | p=0; |
1134 | while ( (p=out.find_first_of("\n",p) ) !=out.npos) | |
1135 | { | |
1136 | out.replace (p,1,"\\n"); | |
1137 | p+=2; | |
1138 | } | |
47c07fba | 1139 | |
6ab3bc95 | 1140 | out='"'+out+'"'; |
47c07fba | 1141 | |
6ab3bc95 RP |
1142 | return out; |
1143 | } // eo scape(const std::string&) | |
47c07fba | 1144 | |
47c07fba | 1145 | |
6ab3bc95 RP |
1146 | string descape(const string &s, int startpos, int &endpos) |
1147 | { | |
1148 | string out; | |
1149 | ||
1150 | if (s.at(startpos) != '"') | |
1151 | throw out_of_range("value not type escaped string"); | |
1152 | ||
1153 | out=s.substr(startpos+1); | |
1154 | string::size_type p=0; | |
1155 | ||
1156 | // search for the end of the string | |
1157 | while ( (p=out.find("\"",p) ) !=out.npos) | |
1158 | { | |
1159 | int e=p-1; | |
1160 | bool escaped=false; | |
1161 | ||
1162 | // the " might be escaped with a backslash | |
1163 | while (e>=0 && out.at (e) =='\\') | |
1164 | { | |
1165 | if (escaped == false) | |
1166 | escaped=true; | |
1167 | else | |
1168 | escaped=false; | |
1169 | ||
1170 | e--; | |
1171 | } | |
1172 | ||
1173 | if (escaped==false) | |
1174 | break; | |
1175 | else | |
1176 | p++; | |
1177 | } | |
1178 | ||
1179 | // we now have the end of the string | |
1180 | out=out.substr(0,p); | |
1181 | ||
1182 | // tell calling prog about the endposition | |
1183 | endpos=startpos+p+1; | |
1184 | ||
1185 | // descape all \ stuff inside the string now | |
1186 | p=0; | |
1187 | while ( (p=out.find_first_of("\\",p) ) !=out.npos) | |
1188 | { | |
1189 | switch (out.at(p+1) ) | |
1190 | { | |
1191 | case 'r': | |
47c07fba GE |
1192 | out.replace(p,2,"\r"); |
1193 | break; | |
6ab3bc95 | 1194 | case 'n': |
47c07fba GE |
1195 | out.replace(p,2,"\n"); |
1196 | break; | |
6ab3bc95 | 1197 | default: |
47c07fba | 1198 | out.erase(p,1); |
6ab3bc95 RP |
1199 | } |
1200 | p++; | |
1201 | } | |
1202 | ||
1203 | return out; | |
1204 | } // eo descape(const std::string&,int,int&) | |
47c07fba | 1205 | |
e93545dd | 1206 | |
47c07fba GE |
1207 | string escape_shellarg(const string &input) |
1208 | { | |
6ab3bc95 RP |
1209 | string output = "'"; |
1210 | string::const_iterator it, it_end = input.end(); | |
83d700e9 | 1211 | for (it = input.begin(); it != it_end; ++it) |
6ab3bc95 RP |
1212 | { |
1213 | if ( (*it) == '\'') | |
1214 | output += "'\\'"; | |
1215 | ||
1216 | output += *it; | |
1217 | } | |
1218 | ||
1219 | output += "'"; | |
1220 | return output; | |
47c07fba | 1221 | } |