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 | 37 | #include <boost/numeric/conversion/cast.hpp> |
3f5c5ccd | 38 | #include <boost/foreach.hpp> |
5cd64148 | 39 | |
e93545dd GE |
40 | #include <stringfunc.hxx> |
41 | ||
42 | using namespace std; | |
43 | ||
6ab3bc95 RP |
44 | namespace I2n |
45 | { | |
6a93d84a TJ |
46 | |
47 | ||
6ab3bc95 RP |
48 | namespace |
49 | { | |
6a93d84a TJ |
50 | |
51 | const std::string hexDigitsLower("0123456789abcdef"); | |
52 | const std::string hexDigitsUpper("0123456789ABCDEF"); | |
53 | ||
54 | ||
55 | struct UpperFunc | |
56 | { | |
6ab3bc95 RP |
57 | char operator() (char c) |
58 | { | |
59 | return std::toupper(c); | |
60 | } | |
6a93d84a TJ |
61 | }; // eo struct UpperFunc |
62 | ||
63 | ||
64 | struct LowerFunc | |
65 | { | |
6ab3bc95 RP |
66 | char operator() (char c) |
67 | { | |
68 | return std::tolower(c); | |
69 | } | |
6a93d84a TJ |
70 | }; // eo struct LowerFunc |
71 | ||
72 | ||
73 | } // eo namespace <anonymous> | |
74 | ||
75 | ||
76 | ||
77 | /** | |
6ab3bc95 | 78 | * default list of Whitespaces (" \t\r\n"); |
6a93d84a | 79 | */ |
6ab3bc95 | 80 | const std::string Whitespaces = " \t\r\n"; |
6a93d84a TJ |
81 | |
82 | /** | |
83 | * default list of lineendings ("\r\n"); | |
84 | */ | |
6ab3bc95 | 85 | const std::string LineEndings= "\r\n"; |
6a93d84a TJ |
86 | |
87 | ||
88 | ||
89 | /** | |
90 | * @brief checks if a string begins with a given prefix. | |
91 | * @param[in,out] str the string which is tested | |
92 | * @param prefix the prefix which should be tested for. | |
93 | * @return @a true iff the prefix is not empty and the string begins with that prefix. | |
94 | */ | |
6ab3bc95 | 95 | bool has_prefix(const std::string& str, const std::string& prefix) |
6a93d84a | 96 | { |
6ab3bc95 RP |
97 | if (prefix.empty() || str.empty() || str.size() < prefix.size() ) |
98 | { | |
99 | return false; | |
100 | } | |
101 | return str.compare(0, prefix.size(), prefix) == 0; | |
102 | } // eo has_prefix(const std::string&,const std::string&) | |
6a93d84a TJ |
103 | |
104 | ||
105 | /** | |
106 | * @brief checks if a string ends with a given suffix. | |
107 | * @param[in,out] str the string which is tested | |
108 | * @param suffix the suffix which should be tested for. | |
109 | * @return @a true iff the suffix is not empty and the string ends with that suffix. | |
110 | */ | |
6ab3bc95 | 111 | bool has_suffix(const std::string& str, const std::string& suffix) |
6a93d84a | 112 | { |
6ab3bc95 RP |
113 | if (suffix.empty() || str.empty() || str.size() < suffix.size() ) |
114 | { | |
115 | return false; | |
116 | } | |
117 | return str.compare(str.size() - suffix.size(), suffix.size(), suffix) == 0; | |
118 | } // eo has_suffix(const std::string&,const std::string&) | |
6a93d84a TJ |
119 | |
120 | ||
121 | /** | |
122 | * cut off characters from a given list from front and end of a string. | |
123 | * @param[in,out] str the string which should be trimmed. | |
124 | * @param charlist the list of characters to remove from beginning and end of string | |
125 | * @return the result string. | |
126 | */ | |
6ab3bc95 RP |
127 | std::string trim_mod(std::string& str, const std::string& charlist) |
128 | { | |
129 | // first: trim the beginning: | |
130 | std::string::size_type pos= str.find_first_not_of (charlist); | |
131 | if (pos == std::string::npos) | |
132 | { | |
133 | // whole string consists of charlist (or is already empty) | |
134 | str.clear(); | |
135 | return str; | |
136 | } | |
137 | else if (pos>0) | |
138 | { | |
139 | // str starts with charlist | |
140 | str.erase(0,pos); | |
141 | } | |
142 | // now let's look at the tail: | |
143 | pos= str.find_last_not_of(charlist) +1; // note: we already know there is at least one other char! | |
144 | if ( pos < str.size() ) | |
145 | { | |
146 | str.erase(pos, str.size()-pos); | |
147 | } | |
148 | return str; | |
149 | } // eo trim_mod(std::string&,const std::string&) | |
6a93d84a TJ |
150 | |
151 | ||
152 | ||
153 | /** | |
154 | * removes last character from a string when it is in a list of chars to be removed. | |
155 | * @param[in,out] str the string. | |
156 | * @param what the list of chars which will be tested for. | |
157 | * @return the resulting string with last char removed (if applicable) | |
158 | */ | |
6ab3bc95 | 159 | std::string chomp_mod(std::string& str, const std::string& what) |
6a93d84a | 160 | { |
6ab3bc95 RP |
161 | if (str.empty() || what.empty() ) |
162 | { | |
163 | return str; | |
164 | } | |
165 | if (what.find(str.at (str.size()-1) ) != std::string::npos) | |
166 | { | |
167 | str.erase(str.size() - 1); | |
168 | } | |
169 | return str; | |
170 | } // eo chomp_mod(std::string&,const std::string&) | |
6a93d84a TJ |
171 | |
172 | ||
173 | /** | |
174 | * @brief converts a string to lower case. | |
175 | * @param[in,out] str the string to modify. | |
176 | * @return the string | |
177 | */ | |
6ab3bc95 | 178 | std::string to_lower_mod(std::string& str) |
6a93d84a | 179 | { |
6ab3bc95 RP |
180 | std::transform(str.begin(), str.end(), str.begin(), LowerFunc() ); |
181 | return str; | |
182 | } // eo to_lower_mod(std::string&) | |
6a93d84a TJ |
183 | |
184 | ||
185 | /** | |
186 | * @brief converts a string to upper case. | |
187 | * @param[in,out] str the string to modify. | |
188 | * @return the string | |
189 | */ | |
6ab3bc95 | 190 | std::string to_upper_mod(std::string& str) |
6a93d84a | 191 | { |
6ab3bc95 RP |
192 | std::transform( str.begin(), str.end(), str.begin(), UpperFunc() ); |
193 | return str; | |
194 | } // eo to_upper_mod(std::string&) | |
6a93d84a TJ |
195 | |
196 | ||
197 | ||
198 | /** | |
199 | * cut off characters from a given list from front and end of a string. | |
200 | * @param str the string which should be trimmed. | |
201 | * @param charlist the list of characters to remove from beginning and end of string | |
202 | * @return the result string. | |
203 | */ | |
6ab3bc95 RP |
204 | std::string trim (const std::string& str, const std::string& charlist) |
205 | { | |
206 | // first: trim the beginning: | |
207 | std::string::size_type pos0= str.find_first_not_of(charlist); | |
208 | if (pos0 == std::string::npos) | |
209 | { | |
210 | // whole string consists of charlist (or is already empty) | |
211 | return std::string(); | |
212 | } | |
213 | // now let's look at the end: | |
214 | std::string::size_type pos1= str.find_last_not_of(charlist); | |
215 | return str.substr(pos0, pos1 - pos0 + 1); | |
6a93d84a TJ |
216 | } // eo trim(const std:.string&,const std::string&) |
217 | ||
218 | ||
219 | /** | |
220 | * removes last character from a string when it is in a list of chars to be removed. | |
221 | * @param str the string. | |
222 | * @param what the list of chars which will be tested for. | |
223 | * @return the resulting string with last char removed (if applicable) | |
224 | */ | |
6ab3bc95 RP |
225 | std::string chomp (const std::string& str, const std::string& what) |
226 | { | |
227 | if (str.empty() || what.empty() ) | |
228 | { | |
229 | return str; | |
230 | } | |
231 | if (what.find(str.at (str.size()-1) ) != std::string::npos) | |
232 | { | |
233 | return str.substr(0, str.size()-1); | |
234 | } | |
235 | return str; | |
6a93d84a TJ |
236 | } // eo chomp(const std:.string&,const std::string&) |
237 | ||
238 | ||
239 | /** | |
240 | * @brief returns a lower case version of a given string. | |
241 | * @param str the string | |
242 | * @return the lower case version of the string | |
243 | */ | |
6ab3bc95 | 244 | std::string to_lower (const std::string& str) |
6a93d84a | 245 | { |
6ab3bc95 RP |
246 | std::string result(str); |
247 | return to_lower_mod(result); | |
248 | } // eo to_lower(const std::string&) | |
6a93d84a TJ |
249 | |
250 | ||
251 | /** | |
252 | * @brief returns a upper case version of a given string. | |
253 | * @param str the string | |
254 | * @return the upper case version of the string | |
255 | */ | |
6ab3bc95 | 256 | std::string to_upper(const std::string& str) |
6a93d84a | 257 | { |
6ab3bc95 RP |
258 | std::string result(str); |
259 | return to_upper_mod(result); | |
260 | } // eo to_upper(const std::string&) | |
6a93d84a TJ |
261 | |
262 | ||
263 | ||
264 | /** | |
265 | * @brief removes a given suffix from a string. | |
266 | * @param str the string. | |
267 | * @param suffix the suffix which should be removed if the string ends with it. | |
268 | * @return the string without the suffix. | |
269 | * | |
270 | * If the string ends with the suffix, it is removed. If the the string doesn't end | |
271 | * with the suffix the original string is returned. | |
272 | */ | |
6ab3bc95 | 273 | std::string remove_suffix(const std::string& str, const std::string& suffix) |
6a93d84a | 274 | { |
6ab3bc95 RP |
275 | if (has_suffix(str,suffix) ) |
276 | { | |
277 | return str.substr(0, str.size()-suffix.size() ); | |
278 | } | |
279 | return str; | |
280 | } // eo remove_suffix(const std::string&,const std::string&) | |
6a93d84a TJ |
281 | |
282 | ||
283 | ||
284 | /** | |
285 | * @brief removes a given prefix from a string. | |
286 | * @param str the string. | |
287 | * @param prefix the prefix which should be removed if the string begins with it. | |
288 | * @return the string without the prefix. | |
289 | * | |
290 | * If the string begins with the prefix, it is removed. If the the string doesn't begin | |
291 | * with the prefix the original string is returned. | |
292 | */ | |
6ab3bc95 | 293 | std::string remove_prefix(const std::string& str, const std::string& prefix) |
6a93d84a | 294 | { |
6ab3bc95 RP |
295 | if (has_prefix(str,prefix) ) |
296 | { | |
297 | return str.substr( prefix.size() ); | |
298 | } | |
299 | return str; | |
300 | } // eo remove_prefix(const std::string&,const std::string&) | |
6a93d84a TJ |
301 | |
302 | ||
303 | /** | |
304 | * split a string to key and value delimited by a given delimiter. | |
6ab3bc95 | 305 | * The resulting key and value strings are trimmed (Whitespaces removed at beginning and end). |
6a93d84a TJ |
306 | * @param str the string which should be splitted. |
307 | * @param[out] key the resulting key | |
308 | * @param[out] value the resulting value | |
309 | * @param delimiter the delimiter between key and value; default is '='. | |
310 | * @return @a true if the split was successful. | |
311 | */ | |
6ab3bc95 RP |
312 | bool pair_split( |
313 | const std::string& str, | |
314 | std::string& key, | |
315 | std::string& value, | |
316 | char delimiter) | |
317 | { | |
318 | std::string::size_type pos = str.find (delimiter); | |
319 | if (pos == std::string::npos) return false; | |
320 | key= str.substr(0,pos); | |
321 | value= str.substr(pos+1); | |
322 | trim_mod(key); | |
323 | trim_mod(value); | |
324 | return true; | |
325 | } // eo pair_split(const std::string&,std::string&,std::string&,char) | |
6a93d84a TJ |
326 | |
327 | ||
328 | /** | |
329 | * splits a string by given delimiter | |
330 | * | |
331 | * @param[in] str the string which should be splitted. | |
332 | * @param[out] result the list resulting from splitting @a str. | |
333 | * @param[in] delimiter the delimiter (word/phrase) at which @a str should be splitted. | |
334 | * @param[in] omit_empty should empty parts not be stored? | |
335 | * @param[in] trim_list list of characters the parts should be trimmed by. | |
336 | * (empty string results in no trim) | |
337 | */ | |
6ab3bc95 RP |
338 | void split_string( |
339 | const std::string& str, | |
340 | std::list<std::string>& result, | |
341 | const std::string& delimiter, | |
342 | bool omit_empty, | |
343 | const std::string& trim_list | |
6a93d84a TJ |
344 | ) |
345 | { | |
6ab3bc95 RP |
346 | std::string::size_type pos, last_pos=0; |
347 | bool delimiter_found= false; | |
348 | while ( last_pos < str.size() && last_pos != std::string::npos) | |
349 | { | |
350 | pos= str.find(delimiter, last_pos); | |
351 | std::string part; | |
352 | if (pos == std::string::npos) | |
353 | { | |
354 | part= str.substr(last_pos); | |
355 | delimiter_found= false; | |
356 | } | |
357 | else | |
358 | { | |
359 | part= str.substr(last_pos, pos-last_pos); | |
360 | delimiter_found=true; | |
361 | } | |
362 | if (pos != std::string::npos) | |
363 | { | |
364 | last_pos= pos+ delimiter.size(); | |
365 | } | |
366 | else | |
367 | { | |
368 | last_pos= std::string::npos; | |
369 | } | |
370 | if (!trim_list.empty() ) trim_mod (part, trim_list); | |
371 | if (omit_empty && part.empty() ) continue; | |
372 | result.push_back( part ); | |
373 | } | |
374 | // if the string ends with a delimiter we need to append an empty string if no omit_empty | |
375 | // was given. | |
376 | // (this way we keep the split result consistent to a join operation) | |
377 | if (delimiter_found && !omit_empty) | |
378 | { | |
379 | result.push_back(""); | |
380 | } | |
381 | } // eo split_string(const std::string&,std::list< std::string >&,const std::string&,bool,const std::string&) | |
6a93d84a TJ |
382 | |
383 | ||
338da253 CH |
384 | /** call split_string with list<string>, converts result to vector; vector is clear()-ed first |
385 | * | |
386 | * Note: Uses 3 O(n)-operations: list.size, vector.resize and std::swap_ranges; | |
387 | * not sure whether there is a better way to do this | |
388 | * */ | |
389 | void split_string( | |
390 | const std::string& str, | |
391 | std::vector<std::string>& result, | |
392 | const std::string& delimiter, | |
393 | bool omit_empty, | |
394 | const std::string& trim_list | |
395 | ) | |
396 | { | |
397 | std::list<std::string> tmp; | |
398 | split_string(str, tmp, delimiter, omit_empty, trim_list); | |
399 | std::size_t size = tmp.size(); // this is O(n) | |
400 | result.clear(); | |
401 | result.resize(size); // also O(n) | |
402 | std::swap_ranges(tmp.begin(), tmp.end(), result.begin()); // also O(n) | |
403 | } | |
404 | ||
6a93d84a TJ |
405 | /** |
406 | * splits a string by a given delimiter | |
407 | * @param str the string which should be splitted. | |
408 | * @param delimiter delimiter the delimiter (word/phrase) at which @a str should be splitted. | |
409 | * @param[in] omit_empty should empty parts not be stored? | |
410 | * @param[in] trim_list list of characters the parts should be trimmed by. | |
411 | * (empty string results in no trim) | |
412 | * @return the list resulting from splitting @a str. | |
413 | */ | |
6ab3bc95 RP |
414 | std::list<std::string> split_string( |
415 | const std::string& str, | |
416 | const std::string& delimiter, | |
417 | bool omit_empty, | |
418 | const std::string& trim_list | |
6a93d84a TJ |
419 | ) |
420 | { | |
6ab3bc95 RP |
421 | std::list<std::string> result; |
422 | split_string(str, result, delimiter, omit_empty, trim_list); | |
423 | return result; | |
424 | } // eo split_string(const std::string&,const std::string&,bool,const std::string&) | |
6a93d84a TJ |
425 | |
426 | ||
427 | /** | |
428 | * @brief joins a list of strings into a single string. | |
429 | * | |
6ab3bc95 RP |
430 | * This funtion is (basically) the reverse operation of @a split_string. |
431 | * | |
6a93d84a TJ |
432 | * @param parts the list of strings. |
433 | * @param delimiter the delimiter which is inserted between the strings. | |
434 | * @return the joined string. | |
435 | */ | |
6ab3bc95 RP |
436 | std::string join_string( |
437 | const std::list< std::string >& parts, | |
438 | const std::string& delimiter | |
6a93d84a TJ |
439 | ) |
440 | { | |
6ab3bc95 RP |
441 | std::string result; |
442 | if (! parts.empty() ) | |
443 | { | |
444 | std::list< std::string >::const_iterator it= parts.begin(); | |
445 | result = *it; | |
446 | while ( ++it != parts.end() ) | |
447 | { | |
448 | result+= delimiter; | |
449 | result+= *it; | |
450 | } | |
451 | } | |
452 | return result; | |
453 | } // eo join_string(const std::list< std::string >&,const std::string&) | |
6a93d84a TJ |
454 | |
455 | ||
376ec4fa CH |
456 | /** @brief same as join_string for list, except uses a vector */ |
457 | std::string join_string( | |
458 | const std::vector< std::string >& parts, | |
459 | const std::string& delimiter | |
460 | ) | |
461 | { | |
462 | std::string result; | |
463 | if (! parts.empty() ) | |
464 | { | |
465 | std::vector< std::string >::const_iterator it= parts.begin(); | |
466 | result = *it; | |
467 | while ( ++it != parts.end() ) | |
468 | { | |
469 | result+= delimiter; | |
470 | result+= *it; | |
471 | } | |
472 | } | |
473 | return result; | |
474 | } // eo join_string(const std::vector< std::string >&,const std::string&) | |
475 | ||
476 | ||
6a93d84a TJ |
477 | |
478 | /* | |
479 | ** conversions | |
480 | */ | |
481 | ||
482 | ||
483 | /** | |
484 | * @brief returns a hex string from a binary string. | |
485 | * @param str the (binary) string | |
486 | * @param upper_case_digits determine whether to use upper case characters for digits A-F. | |
487 | * @return the string in hex notation. | |
488 | */ | |
6ab3bc95 RP |
489 | std::string convert_binary_to_hex( |
490 | const std::string& str, | |
491 | bool upper_case_digits | |
6a93d84a TJ |
492 | ) |
493 | { | |
6ab3bc95 RP |
494 | std::string result; |
495 | std::string hexDigits(upper_case_digits ? hexDigitsUpper : hexDigitsLower); | |
496 | for ( std::string::const_iterator it= str.begin(); | |
497 | it != str.end(); | |
498 | ++it) | |
499 | { | |
500 | result.push_back( hexDigits[ ( (*it) >> 4) & 0x0f ] ); | |
501 | result.push_back( hexDigits[ (*it) & 0x0f ] ); | |
502 | } | |
503 | return result; | |
504 | } // eo convert_binary_to_hex(const std::string&,bool) | |
6a93d84a TJ |
505 | |
506 | ||
507 | /** | |
508 | * @brief converts a hex digit string to binary string. | |
509 | * @param str hex digit string | |
510 | * @return the binary string. | |
511 | * | |
512 | * The hex digit string may contains white spaces or colons which are treated | |
513 | * as delimiters between hex digit groups. | |
514 | * | |
515 | * @todo rework the handling of half nibbles (consistency)! | |
516 | */ | |
6ab3bc95 RP |
517 | std::string convert_hex_to_binary( |
518 | const std::string& str | |
6a93d84a | 519 | ) |
6ab3bc95 RP |
520 | throw (std::runtime_error) |
521 | { | |
522 | std::string result; | |
523 | char c= 0; | |
524 | bool hasNibble= false; | |
525 | bool lastWasWS= true; | |
526 | for ( std::string::const_iterator it= str.begin(); | |
527 | it != str.end(); | |
528 | ++it) | |
529 | { | |
530 | std::string::size_type p = hexDigitsLower.find( *it ); | |
531 | if (p== std::string::npos) | |
532 | { | |
533 | p= hexDigitsUpper.find( *it ); | |
534 | } | |
535 | if (p == std::string::npos) | |
536 | { | |
537 | if ( ( Whitespaces.find( *it ) != std::string::npos) // is it a whitespace? | |
6a93d84a | 538 | or ( *it == ':') // or a colon? |
6ab3bc95 RP |
539 | ) |
540 | { | |
541 | // we treat that as a valid delimiter: | |
542 | if (hasNibble) | |
6a93d84a | 543 | { |
6ab3bc95 RP |
544 | // 1 nibble before WS is treate as lower part: |
545 | result.push_back(c); | |
546 | // reset state: | |
547 | hasNibble= false; | |
6a93d84a | 548 | } |
6ab3bc95 RP |
549 | lastWasWS= true; |
550 | continue; | |
551 | } | |
552 | } | |
553 | if (p == std::string::npos ) | |
554 | { | |
555 | throw runtime_error("illegal character in hex digit string: " + str); | |
556 | } | |
557 | lastWasWS= false; | |
558 | if (hasNibble) | |
559 | { | |
560 | c<<=4; | |
561 | } | |
562 | else | |
563 | { | |
564 | c=0; | |
565 | } | |
566 | c+= (p & 0x0f); | |
567 | if (hasNibble) | |
568 | { | |
569 | //we already had a nibble, so a char is complete now: | |
570 | result.push_back( c ); | |
571 | hasNibble=false; | |
572 | } | |
573 | else | |
574 | { | |
575 | // this is the first nibble of a new char: | |
576 | hasNibble=true; | |
577 | } | |
578 | } | |
579 | if (hasNibble) | |
580 | { | |
581 | //well, there is one nibble left | |
582 | // let's do some heuristics: | |
583 | if (lastWasWS) | |
584 | { | |
585 | // if the preceeding character was a white space (or a colon) | |
586 | // we treat the nibble as lower part: | |
587 | //( this is consistent with shortened hex notations where leading zeros are not noted) | |
588 | result.push_back( c ); | |
589 | } | |
590 | else | |
591 | { | |
592 | // if it was part of a hex digit chain, we treat it as UPPER part (!!) | |
593 | result.push_back( c << 4 ); | |
594 | } | |
595 | } | |
596 | return result; | |
597 | } // eo convert_hex_to_binary(const std::string&) | |
598 | ||
599 | ||
1a0267e5 CH |
600 | static list<string>& alloc_template_starts() |
601 | { | |
602 | static list<string> result; | |
603 | if (result.empty()) | |
604 | { | |
605 | result.push_back("std::list"); | |
606 | result.push_back("std::vector"); | |
607 | } | |
608 | return result; | |
609 | } | |
610 | ||
611 | string shorten_stl_types(const string &input) | |
612 | { | |
613 | string output = input; | |
614 | ||
615 | // first: replace fixed string for std::string | |
616 | replace_all(output, "std::basic_string<char, std::char_traits<char>, std::allocator<char> >", | |
617 | "std::string"); | |
618 | ||
619 | // loop over list/vector/... that have an allocator, e.g. | |
620 | // std::list< some_type_here, std::allocator<some_type_here> > | |
621 | string::size_type start, comma, end, len, start_text_len; | |
622 | int n_open_brackets; | |
623 | string allocator_text; | |
624 | BOOST_FOREACH(const string &start_text, alloc_template_starts()) | |
625 | { | |
626 | start = 0; | |
627 | comma = 0; | |
628 | end = 0; | |
629 | start_text_len = start_text.length(); | |
630 | while( (start=output.find(start_text+"<", start)) != string::npos ) | |
631 | { | |
632 | len = output.length(); | |
633 | start += start_text_len+1; // start next iter and tests here after opening bracket | |
634 | ||
635 | // now comes the tricky part: find matching ',' and the closing '>' even if "subtype" is template again | |
636 | comma = start; | |
637 | n_open_brackets = 1; // the bracket right after start_text counts as first | |
638 | while (comma < len && n_open_brackets > 0) | |
639 | { | |
640 | if (output[comma] == ',' && n_open_brackets == 1) | |
641 | break; | |
642 | else if (output[comma] == '<') | |
643 | ++n_open_brackets; | |
644 | else if (output[comma] == '>') | |
645 | --n_open_brackets; | |
646 | ++comma; | |
647 | } | |
648 | end = comma+1; | |
649 | while (end < len && n_open_brackets > 0) | |
650 | { | |
651 | if (output[end] == '<') | |
652 | ++n_open_brackets; | |
653 | else if (output[end] == '>') | |
654 | { | |
655 | --n_open_brackets; | |
656 | if (n_open_brackets == 0) | |
657 | break; // do not increment end | |
658 | } | |
659 | ++end; | |
660 | } | |
661 | ||
662 | // check that start < comma < end < len && n_open_brackets == 0 | |
663 | if (start >= comma || comma >= end || end >= len || n_open_brackets != 0) | |
664 | continue; // input seems to be of unexpected form | |
665 | ||
666 | // check that type in allocator is same as until comma | |
667 | string type = output.substr(start, comma-start); | |
668 | if (type[type.length()-1] == '>') | |
669 | allocator_text = string("std::allocator<") + type + " > "; | |
670 | else | |
671 | allocator_text = string("std::allocator<") + type + "> "; | |
672 | if (output.substr(comma+2, end-comma-2) == allocator_text) | |
673 | output.replace(comma+2, end-comma-2, "_alloc_"); | |
674 | } | |
675 | } | |
676 | ||
677 | return output; | |
678 | } | |
679 | ||
6ab3bc95 RP |
680 | } // eo namespace I2n |
681 | ||
682 | ||
683 | ||
6a93d84a | 684 | |
e93545dd GE |
685 | std::string iso_to_utf8(const std::string& isostring) |
686 | { | |
6ab3bc95 | 687 | string result; |
118e216e | 688 | |
6ab3bc95 | 689 | iconv_t i2utf8 = iconv_open("UTF-8", "ISO-8859-1"); |
118e216e | 690 | |
6ab3bc95 RP |
691 | if (iso_to_utf8 == (iconv_t)-1) |
692 | throw runtime_error("iconv can't convert from ISO-8859-1 to UTF-8"); | |
118e216e | 693 | |
6ab3bc95 RP |
694 | size_t in_size=isostring.size(); |
695 | size_t out_size=in_size*4; | |
118e216e | 696 | |
6ab3bc95 RP |
697 | char *buf = (char *)malloc(out_size+1); |
698 | if (buf == NULL) | |
699 | throw runtime_error("out of memory for iconv buffer"); | |
e93545dd | 700 | |
5a4ecb51 | 701 | char *in = (char *)isostring.c_str(); |
6ab3bc95 RP |
702 | char *out = buf; |
703 | iconv(i2utf8, &in, &in_size, &out, &out_size); | |
118e216e | 704 | |
6ab3bc95 | 705 | buf[isostring.size()*4-out_size]=0; |
118e216e | 706 | |
6ab3bc95 | 707 | result=buf; |
118e216e | 708 | |
6ab3bc95 RP |
709 | free(buf); |
710 | iconv_close(i2utf8); | |
118e216e | 711 | |
6ab3bc95 | 712 | return result; |
e93545dd GE |
713 | } |
714 | ||
715 | std::string utf8_to_iso(const std::string& utf8string) | |
716 | { | |
6ab3bc95 | 717 | string result; |
118e216e | 718 | |
6ab3bc95 | 719 | iconv_t utf82iso = iconv_open("ISO-8859-1","UTF-8"); |
118e216e | 720 | |
6ab3bc95 RP |
721 | if (utf82iso == (iconv_t)-1) |
722 | throw runtime_error("iconv can't convert from UTF-8 to ISO-8859-1"); | |
118e216e | 723 | |
6ab3bc95 RP |
724 | size_t in_size=utf8string.size(); |
725 | size_t out_size=in_size; | |
118e216e | 726 | |
6ab3bc95 RP |
727 | char *buf = (char *)malloc(out_size+1); |
728 | if (buf == NULL) | |
729 | throw runtime_error("out of memory for iconv buffer"); | |
e93545dd | 730 | |
5a4ecb51 | 731 | char *in = (char *)utf8string.c_str(); |
6ab3bc95 RP |
732 | char *out = buf; |
733 | iconv(utf82iso, &in, &in_size, &out, &out_size); | |
118e216e | 734 | |
6ab3bc95 | 735 | buf[utf8string.size()-out_size]=0; |
118e216e | 736 | |
6ab3bc95 | 737 | result=buf; |
118e216e | 738 | |
6ab3bc95 RP |
739 | free(buf); |
740 | iconv_close(utf82iso); | |
e93545dd | 741 | |
6ab3bc95 | 742 | return result; |
e93545dd GE |
743 | } |
744 | ||
a5f3af6e GE |
745 | wchar_t* utf8_to_wbuf(const std::string& utf8string) |
746 | { | |
6ab3bc95 | 747 | iconv_t utf82wstr = iconv_open("UCS-4LE","UTF-8"); |
a5f3af6e | 748 | |
6ab3bc95 RP |
749 | if (utf82wstr == (iconv_t)-1) |
750 | throw runtime_error("iconv can't convert from UTF-8 to UCS-4"); | |
a5f3af6e | 751 | |
6ab3bc95 RP |
752 | size_t in_size=utf8string.size(); |
753 | size_t out_size= (in_size+1)*sizeof(wchar_t); | |
a5f3af6e | 754 | |
6ab3bc95 RP |
755 | wchar_t *buf = (wchar_t *)malloc(out_size); |
756 | if (buf == NULL) | |
757 | throw runtime_error("out of memory for iconv buffer"); | |
a5f3af6e | 758 | |
5a4ecb51 | 759 | char *in = (char *)utf8string.c_str(); |
6ab3bc95 | 760 | char *out = (char*) buf; |
dbd6d77c | 761 | if (iconv(utf82wstr, &in, &in_size, &out, &out_size) == (size_t)-1) |
6ab3bc95 | 762 | throw runtime_error("error converting char encodings"); |
a5f3af6e | 763 | |
6ab3bc95 | 764 | buf[ ( (utf8string.size()+1)*sizeof(wchar_t)-out_size) /sizeof(wchar_t) ]=0; |
a5f3af6e | 765 | |
6ab3bc95 | 766 | iconv_close(utf82wstr); |
a5f3af6e | 767 | |
6ab3bc95 | 768 | return buf; |
a5f3af6e GE |
769 | } |
770 | ||
13cc4db1 | 771 | std::string utf7imap_to_utf8(const std::string& utf7imapstring) |
d116a071 | 772 | { |
6ab3bc95 | 773 | string result; |
118e216e | 774 | |
6ab3bc95 | 775 | iconv_t utf7imap2utf8 = iconv_open("UTF-8","UTF-7-IMAP"); |
118e216e | 776 | |
6ab3bc95 RP |
777 | if (utf7imap2utf8 == (iconv_t)-1) |
778 | throw runtime_error("iconv can't convert from UTF-7-IMAP to UTF-8"); | |
118e216e | 779 | |
6ab3bc95 RP |
780 | size_t in_size=utf7imapstring.size(); |
781 | size_t out_size=in_size*4; | |
118e216e | 782 | |
6ab3bc95 RP |
783 | char *buf = (char *)malloc(out_size+1); |
784 | if (buf == NULL) | |
785 | throw runtime_error("out of memory for iconv buffer"); | |
d116a071 | 786 | |
5a4ecb51 | 787 | char *in = (char *)utf7imapstring.c_str(); |
6ab3bc95 RP |
788 | char *out = buf; |
789 | iconv(utf7imap2utf8, &in, &in_size, &out, &out_size); | |
118e216e | 790 | |
6ab3bc95 | 791 | buf[utf7imapstring.size()*4-out_size]=0; |
118e216e | 792 | |
6ab3bc95 | 793 | result=buf; |
118e216e | 794 | |
6ab3bc95 RP |
795 | free(buf); |
796 | iconv_close(utf7imap2utf8); | |
118e216e | 797 | |
6ab3bc95 | 798 | return result; |
118e216e TJ |
799 | } |
800 | ||
6a2b6dd1 TJ |
801 | std::string utf8_to_utf7imap(const std::string& utf8string) |
802 | { | |
6ab3bc95 | 803 | string result; |
6a2b6dd1 | 804 | |
6ab3bc95 | 805 | iconv_t utf82utf7imap = iconv_open("UTF-7-IMAP", "UTF-8"); |
6a2b6dd1 | 806 | |
6ab3bc95 RP |
807 | if (utf82utf7imap == (iconv_t)-1) |
808 | throw runtime_error("iconv can't convert from UTF-7-IMAP to UTF-8"); | |
6a2b6dd1 | 809 | |
6ab3bc95 RP |
810 | // UTF-7 is base64 encoded, a buffer 10x as large |
811 | // as the utf-8 buffer should be enough. If not the string will be truncated. | |
812 | size_t in_size=utf8string.size(); | |
813 | size_t out_size=in_size*10; | |
6a2b6dd1 | 814 | |
6ab3bc95 RP |
815 | char *buf = (char *)malloc(out_size+1); |
816 | if (buf == NULL) | |
817 | throw runtime_error("out of memory for iconv buffer"); | |
6a2b6dd1 | 818 | |
5a4ecb51 | 819 | char *in = (char *)utf8string.c_str(); |
6ab3bc95 RP |
820 | char *out = buf; |
821 | iconv(utf82utf7imap, &in, &in_size, &out, &out_size); | |
6a2b6dd1 | 822 | |
6ab3bc95 | 823 | buf[utf8string.size()*10-out_size]= 0; |
6a2b6dd1 | 824 | |
6ab3bc95 | 825 | result=buf; |
6a2b6dd1 | 826 | |
6ab3bc95 RP |
827 | free(buf); |
828 | iconv_close(utf82utf7imap); | |
6a2b6dd1 | 829 | |
6ab3bc95 | 830 | return result; |
6a2b6dd1 TJ |
831 | } |
832 | ||
118e216e TJ |
833 | // Tokenize string by (html) tags |
834 | void tokenize_by_tag(vector<pair<string,bool> > &tokenized, const std::string &input) | |
835 | { | |
6ab3bc95 RP |
836 | string::size_type pos, len = input.size(); |
837 | bool inside_tag = false; | |
838 | string current; | |
839 | ||
840 | for (pos = 0; pos < len; pos++) | |
841 | { | |
842 | if (input[pos] == '<') | |
843 | { | |
844 | inside_tag = true; | |
845 | ||
846 | if (!current.empty() ) | |
847 | { | |
848 | tokenized.push_back( make_pair(current, false) ); | |
849 | current = ""; | |
850 | } | |
851 | ||
852 | current += input[pos]; | |
853 | } | |
854 | else if (input[pos] == '>' && inside_tag) | |
855 | { | |
856 | current += input[pos]; | |
857 | inside_tag = false; | |
858 | if (!current.empty() ) | |
859 | { | |
860 | tokenized.push_back( make_pair(current, true) ); | |
861 | current = ""; | |
862 | } | |
863 | } | |
864 | else | |
865 | current += input[pos]; | |
866 | } | |
867 | ||
868 | // String left over in buffer? | |
869 | if (!current.empty() ) | |
870 | tokenized.push_back( make_pair(current, false) ); | |
871 | } // eo tokenize_by_tag | |
118e216e | 872 | |
118e216e TJ |
873 | |
874 | std::string strip_html_tags(const std::string &input) | |
875 | { | |
6ab3bc95 RP |
876 | // Pair first: string, second: isTag |
877 | vector<pair<string,bool> > tokenized; | |
878 | tokenize_by_tag (tokenized, input); | |
118e216e | 879 | |
6ab3bc95 RP |
880 | string output; |
881 | vector<pair<string,bool> >::const_iterator token, tokens_end = tokenized.end(); | |
83d700e9 | 882 | for (token = tokenized.begin(); token != tokens_end; ++token) |
6ab3bc95 RP |
883 | if (!token->second) |
884 | output += token->first; | |
885 | ||
886 | return output; | |
887 | } // eo strip_html_tags | |
118e216e | 888 | |
118e216e TJ |
889 | |
890 | // Smart-encode HTML en | |
891 | string smart_html_entities(const std::string &input) | |
892 | { | |
6ab3bc95 RP |
893 | // Pair first: string, second: isTag |
894 | vector<pair<string,bool> > tokenized; | |
895 | tokenize_by_tag (tokenized, input); | |
896 | ||
897 | string output; | |
898 | vector<pair<string,bool> >::const_iterator token, tokens_end = tokenized.end(); | |
83d700e9 | 899 | for (token = tokenized.begin(); token != tokens_end; ++token) |
6ab3bc95 RP |
900 | { |
901 | // keep HTML tags as they are | |
902 | if (token->second) | |
903 | output += token->first; | |
904 | else | |
905 | output += html_entities(token->first); | |
906 | } | |
907 | ||
908 | return output; | |
118e216e TJ |
909 | } |
910 | ||
6ab3bc95 | 911 | |
a5f3af6e GE |
912 | string::size_type find_8bit(const std::string &str) |
913 | { | |
6ab3bc95 RP |
914 | string::size_type l=str.size(); |
915 | for (string::size_type p=0; p < l; p++) | |
916 | if (static_cast<unsigned char>(str[p]) > 127) | |
917 | return p; | |
a5f3af6e | 918 | |
6ab3bc95 | 919 | return string::npos; |
a5f3af6e GE |
920 | } |
921 | ||
118e216e TJ |
922 | // encoded UTF-8 chars into HTML entities |
923 | string html_entities(std::string str) | |
924 | { | |
6ab3bc95 RP |
925 | // Normal chars |
926 | replace_all (str, "&", "&"); | |
6ab3bc95 RP |
927 | replace_all (str, "<", "<"); |
928 | replace_all (str, ">", ">"); | |
980577e1 TJ |
929 | replace_all (str, "\"", """); |
930 | replace_all (str, "'", "'"); | |
931 | replace_all (str, "/", "/"); | |
6ab3bc95 RP |
932 | |
933 | // Umlauts | |
934 | replace_all (str, "\xC3\xA4", "ä"); | |
935 | replace_all (str, "\xC3\xB6", "ö"); | |
936 | replace_all (str, "\xC3\xBC", "ü"); | |
937 | replace_all (str, "\xC3\x84", "Ä"); | |
938 | replace_all (str, "\xC3\x96", "Ö"); | |
939 | replace_all (str, "\xC3\x9C", "Ü"); | |
940 | ||
941 | // Misc | |
942 | replace_all (str, "\xC3\x9F", "ß"); | |
943 | ||
944 | // conversion of remaining non-ASCII chars needed? | |
945 | // just do if needed because of performance | |
946 | if (find_8bit(str) != string::npos) | |
947 | { | |
948 | // convert to fixed-size encoding UTF-32 | |
949 | wchar_t* wbuf=utf8_to_wbuf(str); | |
950 | ostringstream target; | |
951 | ||
952 | // replace all non-ASCII chars with HTML representation | |
953 | for (int p=0; wbuf[p] != 0; p++) | |
954 | { | |
955 | unsigned int c=wbuf[p]; | |
956 | ||
957 | if (c <= 127) | |
958 | target << static_cast<unsigned char>(c); | |
959 | else | |
960 | target << "&#" << c << ';'; | |
961 | } | |
962 | ||
963 | free(wbuf); | |
964 | ||
965 | str=target.str(); | |
966 | } | |
967 | ||
968 | return str; | |
969 | } // eo html_entities(std::string) | |
970 | ||
554f813d GE |
971 | // convert HTML entities to something that can be viewed on a basic text console (restricted to ASCII-7) |
972 | string html_entities_to_console(std::string str) | |
973 | { | |
974 | // Normal chars | |
975 | replace_all (str, "&", "&"); | |
976 | replace_all (str, "<", "<"); | |
977 | replace_all (str, ">", ">"); | |
978 | replace_all (str, """, "\""); | |
979 | replace_all (str, "'", "'"); | |
980 | replace_all (str, "/", "/"); | |
981 | ||
982 | // Umlauts | |
983 | replace_all (str, "ä", "ae"); | |
984 | replace_all (str, "ö", "oe"); | |
985 | replace_all (str, "ü", "ue"); | |
986 | replace_all (str, "Ä", "Ae"); | |
987 | replace_all (str, "Ö", "Oe"); | |
988 | replace_all (str, "Ü", "Ue"); | |
989 | ||
990 | // Misc | |
991 | replace_all (str, "ß", "ss"); | |
992 | ||
993 | return str; | |
994 | } | |
118e216e | 995 | |
3f5c5ccd CH |
996 | // find_html_comments + remove_html_comments(str, comments) |
997 | void remove_html_comments(string &str) | |
998 | { | |
46dd1321 | 999 | vector<CommentZone> comments = find_html_comments(str); |
3f5c5ccd CH |
1000 | remove_html_comments(str, comments); |
1001 | } | |
1002 | ||
1003 | // find all html comments, behaving correctly if they are nested; ignores comment tags ("<!--FOO .... BAR-->") | |
1004 | // If there are invalid comments ("-->" before "<!--" or different number of closing and opening tags), | |
1005 | // then the unknown index of corresponding start/end tag will be represented by a string::npos | |
1006 | // Indices are from start of start tag until first index after closing tag | |
46dd1321 | 1007 | vector<CommentZone> find_html_comments(const std::string &str) |
3f5c5ccd CH |
1008 | { |
1009 | static const string START = "<!--"; | |
1010 | static const string CLOSE = "-->"; | |
1011 | static const string::size_type START_LEN = START.length(); | |
1012 | static const string::size_type CLOSE_LEN = CLOSE.length(); | |
1013 | ||
46dd1321 TJ |
1014 | vector<CommentZone> comments; |
1015 | ||
3f5c5ccd CH |
1016 | // in order to find nested comments, need either recursion or a stack |
1017 | vector<string::size_type> starts; // stack of start tags | |
1018 | ||
1019 | string::size_type pos = 0; | |
1020 | string::size_type len = str.length(); | |
1021 | string::size_type next_start, next_close; | |
1022 | ||
1023 | while (pos < len) // not really needed but just in case | |
1024 | { | |
1025 | next_start = str.find(START, pos); | |
1026 | next_close = str.find(CLOSE, pos); | |
1027 | ||
1028 | if ( (next_start == string::npos) && (next_close == string::npos) ) | |
1029 | break; // we are done | |
1030 | ||
1031 | else if ( (next_start == string::npos) || (next_close < next_start) ) // close one comment (pop) | |
1032 | { | |
1033 | if (starts.empty()) // closing tag without a start | |
1034 | comments.push_back(CommentZone(string::npos, next_close+CLOSE_LEN)); | |
1035 | else | |
1036 | { | |
1037 | comments.push_back(CommentZone(starts.back(), next_close+CLOSE_LEN)); | |
1038 | starts.pop_back(); | |
1039 | } | |
1040 | pos = next_close + CLOSE_LEN; | |
1041 | } | |
1042 | ||
1043 | else if ( (next_close == string::npos) || (next_start < next_close) ) // start a new comment (push) | |
1044 | { | |
1045 | starts.push_back(next_start); | |
1046 | pos = next_start + START_LEN; | |
1047 | } | |
1048 | } | |
1049 | ||
1050 | // add comments that have no closing tag from back to front (important for remove_html_comments!) | |
1051 | while (!starts.empty()) | |
1052 | { | |
1053 | comments.push_back(CommentZone(starts.back(), string::npos)); | |
1054 | starts.pop_back(); | |
1055 | } | |
46dd1321 TJ |
1056 | |
1057 | return comments; | |
3f5c5ccd CH |
1058 | } |
1059 | ||
1060 | // remove all html comments foundby find_html_comments | |
1061 | void remove_html_comments(std::string &str, const vector<CommentZone> &comments) | |
1062 | { | |
1063 | // remember position where last removal started | |
1064 | string::size_type last_removal_start = str.length(); | |
1065 | ||
1066 | // Go from back to front to not mess up indices. | |
1067 | // This requires that bigger comments, that contain smaller comments, come AFTER | |
1068 | // the small contained comments in the comments vector (i.e. comments are ordered by | |
1069 | // their closing tag, not their opening tag). This is true for results from find_html_comments | |
1070 | BOOST_REVERSE_FOREACH(const CommentZone &comment, comments) | |
1071 | { | |
1072 | if (comment.first == string::npos) | |
1073 | { | |
1074 | str = str.replace(0, comment.second, ""); // comment starts "before" str --> delete from start | |
1075 | break; // there can be no more | |
1076 | } | |
1077 | else if (comment.first >= last_removal_start) | |
1078 | { | |
1079 | continue; // this comment is inside another comment that we have removed already | |
1080 | } | |
1081 | else if (comment.second == string::npos) // comment ends "after" str --> delete until end | |
1082 | { | |
1083 | str = str.replace(comment.first, string::npos, ""); | |
1084 | last_removal_start = comment.first; | |
1085 | } | |
1086 | else | |
1087 | { | |
1088 | str = str.replace(comment.first, comment.second-comment.first, ""); | |
1089 | last_removal_start = comment.first; | |
1090 | } | |
1091 | } | |
1092 | } | |
1093 | ||
e93545dd GE |
1094 | bool replace_all(string &base, const char *ist, const char *soll) |
1095 | { | |
6ab3bc95 RP |
1096 | string i=ist; |
1097 | string s=soll; | |
1098 | return replace_all(base,&i,&s); | |
e93545dd GE |
1099 | } |
1100 | ||
1101 | bool replace_all(string &base, const string &ist, const char *soll) | |
1102 | { | |
6ab3bc95 RP |
1103 | string s=soll; |
1104 | return replace_all(base,&ist,&s); | |
e93545dd GE |
1105 | } |
1106 | ||
1107 | bool replace_all(string &base, const string *ist, const string *soll) | |
1108 | { | |
6ab3bc95 | 1109 | return replace_all(base,*ist,*soll); |
e93545dd GE |
1110 | } |
1111 | ||
1112 | bool replace_all(string &base, const char *ist, const string *soll) | |
1113 | { | |
6ab3bc95 RP |
1114 | string i=ist; |
1115 | return replace_all(base,&i,soll); | |
e93545dd GE |
1116 | } |
1117 | ||
1118 | bool replace_all(string &base, const string &ist, const string &soll) | |
1119 | { | |
6ab3bc95 RP |
1120 | bool found_ist = false; |
1121 | string::size_type a=0; | |
1122 | ||
1123 | if (ist.empty() ) | |
1124 | throw runtime_error ("replace_all called with empty search string"); | |
e93545dd | 1125 | |
6ab3bc95 RP |
1126 | while ( (a=base.find(ist,a) ) != string::npos) |
1127 | { | |
1128 | base.replace(a,ist.size(),soll); | |
1129 | a=a+soll.size(); | |
1130 | found_ist = true; | |
1131 | } | |
1ec2064e | 1132 | |
6ab3bc95 | 1133 | return found_ist; |
e93545dd GE |
1134 | } |
1135 | ||
b953bf36 GE |
1136 | /** |
1137 | * @brief replaces all characters that could be problematic or impose a security risk when being logged | |
1138 | * @param str the original string | |
1139 | * @param replace_with the character to replace the unsafe chars with | |
1140 | * @return a string that is safe to send to syslog or other logfiles | |
1141 | * | |
1142 | * All chars between 0x20 (space) and 0x7E (~) (including) are considered safe for logging. | |
1143 | * See e.g. RFC 5424, section 8.2 or the posix character class "printable". | |
1144 | * This eliminates all possible problems with NUL, control characters, 8 bit chars, UTF8. | |
1145 | * | |
1146 | */ | |
1147 | std::string sanitize_for_logging(const std::string &str, const char replace_with) | |
1148 | { | |
1149 | std::string output=str; | |
1150 | ||
c0e32d64 GE |
1151 | const string::size_type len = output.size(); |
1152 | for (std::string::size_type p=0; p < len; p++) | |
b953bf36 GE |
1153 | if (output[p] < 0x20 || output[p] > 0x7E) |
1154 | output[p]=replace_with; | |
1155 | ||
1156 | return output; | |
1157 | } | |
1158 | ||
e5b21dbb | 1159 | #if 0 |
e93545dd GE |
1160 | string to_lower(const string &src) |
1161 | { | |
6ab3bc95 | 1162 | string dst = src; |
e93545dd | 1163 | |
6ab3bc95 RP |
1164 | string::size_type pos, end = dst.size(); |
1165 | for (pos = 0; pos < end; pos++) | |
1166 | dst[pos] = tolower(dst[pos]); | |
e93545dd | 1167 | |
6ab3bc95 | 1168 | return dst; |
e93545dd GE |
1169 | } |
1170 | ||
1171 | string to_upper(const string &src) | |
1172 | { | |
6ab3bc95 | 1173 | string dst = src; |
e93545dd | 1174 | |
6ab3bc95 RP |
1175 | string::size_type pos, end = dst.size(); |
1176 | for (pos = 0; pos < end; pos++) | |
1177 | dst[pos] = toupper(dst[pos]); | |
e93545dd | 1178 | |
6ab3bc95 | 1179 | return dst; |
e93545dd | 1180 | } |
e5b21dbb | 1181 | #endif |
e93545dd | 1182 | |
83809f5e | 1183 | const int MAX_UNIT_FORMAT_SYMBOLS = 6; |
d1ea9075 | 1184 | |
2cb9a9c5 | 1185 | const string shortUnitFormatSymbols[MAX_UNIT_FORMAT_SYMBOLS] = { |
d1ea9075 GMF |
1186 | " B", |
1187 | " KB", | |
1188 | " MB", | |
1189 | " GB", | |
1190 | " TB", | |
83809f5e | 1191 | " PB" |
d1ea9075 GMF |
1192 | }; |
1193 | ||
2cb9a9c5 | 1194 | const string longUnitFormatSymbols[MAX_UNIT_FORMAT_SYMBOLS] = { |
5cb766b9 GMF |
1195 | i18n_noop(" Bytes"), |
1196 | i18n_noop(" KBytes"), | |
1197 | i18n_noop(" MBytes"), | |
1198 | i18n_noop(" GBytes"), | |
1199 | i18n_noop(" TBytes"), | |
83809f5e | 1200 | i18n_noop(" PBytes") |
d1ea9075 GMF |
1201 | }; |
1202 | ||
72a94426 GMF |
1203 | |
1204 | long double rounding_upwards( | |
e91c1952 | 1205 | const long double number, |
72a94426 GMF |
1206 | const int rounding_multiplier |
1207 | ) | |
1208 | { | |
1209 | long double rounded_number; | |
1210 | rounded_number = number * rounding_multiplier; | |
1211 | rounded_number += 0.5; | |
1212 | rounded_number = (int64_t) (rounded_number); | |
1213 | rounded_number = (long double) (rounded_number) / (long double) (rounding_multiplier); | |
1214 | ||
1215 | return rounded_number; | |
1216 | } | |
1217 | ||
1218 | ||
81267544 GMF |
1219 | string nice_unit_format( |
1220 | const int64_t input, | |
70fc0674 GMF |
1221 | const UnitFormat format, |
1222 | const UnitBase base | |
81267544 | 1223 | ) |
6ab3bc95 | 1224 | { |
d1ea9075 | 1225 | // select the system of units (decimal or binary) |
81267544 | 1226 | int multiple = 0; |
a398513a | 1227 | if (base == UnitBase1000) |
81267544 GMF |
1228 | { |
1229 | multiple = 1000; | |
1230 | } | |
1231 | else | |
1232 | { | |
1233 | multiple = 1024; | |
1234 | } | |
1235 | ||
1236 | long double size = input; | |
6ab3bc95 | 1237 | |
d1ea9075 GMF |
1238 | // check the size of the input number to fit in the appropriate symbol |
1239 | int sizecount = 0; | |
81267544 | 1240 | while (size > multiple) |
6ab3bc95 | 1241 | { |
81267544 GMF |
1242 | size = size / multiple; |
1243 | sizecount++; | |
83809f5e GMF |
1244 | |
1245 | // rollback to the previous values and stop the loop when cannot | |
1246 | // represent the number length. | |
1247 | if (sizecount >= MAX_UNIT_FORMAT_SYMBOLS) | |
1248 | { | |
1249 | size = size * multiple; | |
1250 | sizecount--; | |
1251 | break; | |
1252 | } | |
6ab3bc95 RP |
1253 | } |
1254 | ||
a398513a GMF |
1255 | // round the input number "half up" to multiples of 10 |
1256 | const int rounding_multiplier = 10; | |
72a94426 | 1257 | size = rounding_upwards(size, rounding_multiplier); |
6ab3bc95 | 1258 | |
d1ea9075 | 1259 | // format the input number, placing the appropriate symbol |
6ab3bc95 | 1260 | ostringstream out; |
6ab3bc95 | 1261 | out.setf (ios::fixed); |
a398513a | 1262 | if (format == ShortUnitFormat) |
d1ea9075 GMF |
1263 | { |
1264 | out.precision(1); | |
68d37a5c | 1265 | out << size << i18n( shortUnitFormatSymbols[sizecount].c_str() ); |
d1ea9075 GMF |
1266 | } |
1267 | else | |
6ab3bc95 | 1268 | { |
d1ea9075 | 1269 | out.precision (2); |
68d37a5c | 1270 | out << size << i18n( longUnitFormatSymbols[sizecount].c_str() ); |
6ab3bc95 RP |
1271 | } |
1272 | ||
1273 | return out.str(); | |
1274 | } // eo nice_unit_format(int input) | |
1275 | ||
e93545dd | 1276 | |
5cd64148 CH |
1277 | string nice_unit_format( |
1278 | const double input, | |
1279 | const UnitFormat format, | |
1280 | const UnitBase base | |
1281 | ) | |
1282 | { | |
1283 | // round as double and cast to int64_t | |
1284 | // cast raised overflow error near max val of int64_t (~9.2e18, see unittest) | |
1285 | int64_t input_casted_and_rounded = | |
1286 | boost::numeric_cast<int64_t>( round(input) ); | |
1287 | ||
1288 | // now call other | |
1289 | return nice_unit_format( input_casted_and_rounded, format, base ); | |
1290 | } // eo nice_unit_format(double input) | |
1291 | ||
1292 | ||
47c07fba GE |
1293 | string escape(const string &s) |
1294 | { | |
6ab3bc95 RP |
1295 | string out(s); |
1296 | string::size_type p; | |
47c07fba | 1297 | |
6ab3bc95 RP |
1298 | p=0; |
1299 | while ( (p=out.find_first_of("\"\\",p) ) !=out.npos) | |
1300 | { | |
1301 | out.insert (p,"\\"); | |
1302 | p+=2; | |
1303 | } | |
47c07fba | 1304 | |
6ab3bc95 RP |
1305 | p=0; |
1306 | while ( (p=out.find_first_of("\r",p) ) !=out.npos) | |
1307 | { | |
1308 | out.replace (p,1,"\\r"); | |
1309 | p+=2; | |
1310 | } | |
47c07fba | 1311 | |
6ab3bc95 RP |
1312 | p=0; |
1313 | while ( (p=out.find_first_of("\n",p) ) !=out.npos) | |
1314 | { | |
1315 | out.replace (p,1,"\\n"); | |
1316 | p+=2; | |
1317 | } | |
47c07fba | 1318 | |
6ab3bc95 | 1319 | out='"'+out+'"'; |
47c07fba | 1320 | |
6ab3bc95 RP |
1321 | return out; |
1322 | } // eo scape(const std::string&) | |
47c07fba | 1323 | |
47c07fba | 1324 | |
6ab3bc95 RP |
1325 | string descape(const string &s, int startpos, int &endpos) |
1326 | { | |
1327 | string out; | |
1328 | ||
1329 | if (s.at(startpos) != '"') | |
1330 | throw out_of_range("value not type escaped string"); | |
1331 | ||
1332 | out=s.substr(startpos+1); | |
1333 | string::size_type p=0; | |
1334 | ||
1335 | // search for the end of the string | |
1336 | while ( (p=out.find("\"",p) ) !=out.npos) | |
1337 | { | |
1338 | int e=p-1; | |
1339 | bool escaped=false; | |
1340 | ||
1341 | // the " might be escaped with a backslash | |
1342 | while (e>=0 && out.at (e) =='\\') | |
1343 | { | |
1344 | if (escaped == false) | |
1345 | escaped=true; | |
1346 | else | |
1347 | escaped=false; | |
1348 | ||
1349 | e--; | |
1350 | } | |
1351 | ||
1352 | if (escaped==false) | |
1353 | break; | |
1354 | else | |
1355 | p++; | |
1356 | } | |
1357 | ||
1358 | // we now have the end of the string | |
1359 | out=out.substr(0,p); | |
1360 | ||
1361 | // tell calling prog about the endposition | |
1362 | endpos=startpos+p+1; | |
1363 | ||
1364 | // descape all \ stuff inside the string now | |
1365 | p=0; | |
1366 | while ( (p=out.find_first_of("\\",p) ) !=out.npos) | |
1367 | { | |
1368 | switch (out.at(p+1) ) | |
1369 | { | |
1370 | case 'r': | |
47c07fba GE |
1371 | out.replace(p,2,"\r"); |
1372 | break; | |
6ab3bc95 | 1373 | case 'n': |
47c07fba GE |
1374 | out.replace(p,2,"\n"); |
1375 | break; | |
6ab3bc95 | 1376 | default: |
47c07fba | 1377 | out.erase(p,1); |
6ab3bc95 RP |
1378 | } |
1379 | p++; | |
1380 | } | |
1381 | ||
1382 | return out; | |
1383 | } // eo descape(const std::string&,int,int&) | |
47c07fba | 1384 | |
e93545dd | 1385 | |
47c07fba GE |
1386 | string escape_shellarg(const string &input) |
1387 | { | |
6ab3bc95 RP |
1388 | string output = "'"; |
1389 | string::const_iterator it, it_end = input.end(); | |
83d700e9 | 1390 | for (it = input.begin(); it != it_end; ++it) |
6ab3bc95 RP |
1391 | { |
1392 | if ( (*it) == '\'') | |
1393 | output += "'\\'"; | |
1394 | ||
1395 | output += *it; | |
1396 | } | |
1397 | ||
1398 | output += "'"; | |
1399 | return output; | |
47c07fba | 1400 | } |