2 * Copyright (C) 2005-2013 Team XBMC
5 * This Program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2, or (at your option)
10 * This Program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
15 * You should have received a copy of the GNU General Public License
16 * along with XBMC; see the file COPYING. If not, see
17 * <http://www.gnu.org/licenses/>.
20 //-----------------------------------------------------------------------
22 // File: StringUtils.cpp
24 // Purpose: ATL split string utility
25 // Author: Paul J. Weiss
27 // Modified to use J O'Leary's CStdString class by kraqh3d
29 //------------------------------------------------------------------------
32 #include "StringUtils.h"
33 #include "utils/RegExp.h"
34 #include "utils/fstrcmp.h"
41 #define FORMAT_BLOCK_SIZE 2048 // # of bytes to increment per try
45 const char* ADDON_GUID_RE = "^(\\{){0,1}[0-9a-fA-F]{8}\\-[0-9a-fA-F]{4}\\-[0-9a-fA-F]{4}\\-[0-9a-fA-F]{4}\\-[0-9a-fA-F]{12}(\\}){0,1}$";
47 /* empty string for use in returns by ref */
48 const CStdString StringUtils::EmptyString = "";
49 const std::string StringUtils::Empty = "";
50 CStdString StringUtils::m_lastUUID = "";
52 string StringUtils::Format(const char *fmt, ...)
56 string str = FormatV(fmt, args);
62 string StringUtils::FormatV(const char *fmt, va_list args)
67 int size = FORMAT_BLOCK_SIZE;
70 char *cstr = reinterpret_cast<char*>(malloc(sizeof(char) * size));
76 va_copy(argCopy, args);
78 int nActual = vsnprintf(cstr, size, fmt, argCopy);
81 if (nActual > -1 && nActual < size) // We got a valid result
83 string str(cstr, nActual);
87 if (nActual > -1) // Exactly what we will need (glibc 2.1)
89 else // Let's try to double the size (glibc 2.0)
92 char *new_cstr = reinterpret_cast<char*>(realloc(cstr, sizeof(char) * size));
106 wstring StringUtils::Format(const wchar_t *fmt, ...)
110 wstring str = FormatV(fmt, args);
116 wstring StringUtils::FormatV(const wchar_t *fmt, va_list args)
121 int size = FORMAT_BLOCK_SIZE;
124 wchar_t *cstr = reinterpret_cast<wchar_t*>(malloc(sizeof(wchar_t) * size));
130 va_copy(argCopy, args);
132 int nActual = vswprintf(cstr, size, fmt, argCopy);
135 if (nActual > -1 && nActual < size) // We got a valid result
137 wstring str(cstr, nActual);
141 if (nActual > -1) // Exactly what we will need (glibc 2.1)
143 else // Let's try to double the size (glibc 2.0)
146 wchar_t *new_cstr = reinterpret_cast<wchar_t*>(realloc(cstr, sizeof(wchar_t) * size));
147 if (new_cstr == NULL)
159 void StringUtils::ToUpper(string &str)
161 transform(str.begin(), str.end(), str.begin(), ::toupper);
164 void StringUtils::ToLower(string &str)
166 transform(str.begin(), str.end(), str.begin(), ::tolower);
169 bool StringUtils::EqualsNoCase(const std::string &str1, const std::string &str2)
171 return EqualsNoCase(str1.c_str(), str2.c_str());
174 bool StringUtils::EqualsNoCase(const std::string &str1, const char *s2)
176 return EqualsNoCase(str1.c_str(), s2);
179 bool StringUtils::EqualsNoCase(const char *s1, const char *s2)
181 char c2; // we need only one char outside the loop
184 const char c1 = *s1++; // const local variable should help compiler to optimize
186 if (c1 != c2 && ::tolower(c1) != ::tolower(c2)) // This includes the possibility that one of the characters is the null-terminator, which implies a string mismatch.
188 } while (c2 != '\0'); // At this point, we know c1 == c2, so there's no need to test them both.
192 string StringUtils::Left(const string &str, size_t count)
194 count = max((size_t)0, min(count, str.size()));
195 return str.substr(0, count);
198 string StringUtils::Mid(const string &str, size_t first, size_t count /* = string::npos */)
200 if (first + count > str.size())
201 count = str.size() - first;
203 if (first > str.size())
206 ASSERT(first + count <= str.size());
208 return str.substr(first, count);
211 string StringUtils::Right(const string &str, size_t count)
213 count = max((size_t)0, min(count, str.size()));
214 return str.substr(str.size() - count);
217 std::string& StringUtils::Trim(std::string &str)
220 return TrimRight(str);
223 // hack to ensure that std::string::iterator will be dereferenced as _unsigned_ char
224 // without this hack "TrimX" functions failed on Win32 with UTF-8 strings
225 static int isspace_c(char c)
227 return ::isspace((unsigned char)c);
230 std::string& StringUtils::TrimLeft(std::string &str)
232 str.erase(str.begin(), ::find_if(str.begin(), str.end(), ::not1(::ptr_fun(isspace_c))));
236 std::string& StringUtils::TrimLeft(std::string &str, const std::string& chars)
238 size_t nidx = str.find_first_not_of(chars);
239 str.substr(nidx == str.npos ? 0 : nidx).swap(str);
243 std::string& StringUtils::TrimRight(std::string &str)
245 str.erase(::find_if(str.rbegin(), str.rend(), ::not1(::ptr_fun(isspace_c))).base(), str.end());
249 std::string& StringUtils::TrimRight(std::string &str, const std::string& chars)
251 size_t nidx = str.find_last_not_of(chars);
252 str.erase(str.npos == nidx ? 0 : ++nidx);
256 std::string& StringUtils::RemoveDuplicatedSpacesAndTabs(std::string& str)
258 std::string::iterator it = str.begin();
259 bool onSpace = false;
260 while(it != str.end())
283 int StringUtils::Replace(string &str, char oldChar, char newChar)
285 int replacedChars = 0;
286 for (string::iterator it = str.begin(); it != str.end(); it++)
295 return replacedChars;
298 int StringUtils::Replace(std::string &str, const std::string &oldStr, const std::string &newStr)
303 int replacedChars = 0;
306 while (index < str.size() && (index = str.find(oldStr, index)) != string::npos)
308 str.replace(index, oldStr.size(), newStr);
309 index += newStr.size();
313 return replacedChars;
316 bool StringUtils::StartsWith(const std::string &str1, const std::string &str2)
318 return str1.compare(0, str2.size(), str2) == 0;
321 bool StringUtils::StartsWith(const std::string &str1, const char *s2)
323 return StartsWith(str1.c_str(), s2);
326 bool StringUtils::StartsWith(const char *s1, const char *s2)
338 bool StringUtils::StartsWithNoCase(const std::string &str1, const std::string &str2)
340 return StartsWithNoCase(str1.c_str(), str2.c_str());
343 bool StringUtils::StartsWithNoCase(const std::string &str1, const char *s2)
345 return StartsWithNoCase(str1.c_str(), s2);
348 bool StringUtils::StartsWithNoCase(const char *s1, const char *s2)
352 if (::tolower(*s1) != ::tolower(*s2))
360 bool StringUtils::EndsWith(const std::string &str1, const std::string &str2)
362 if (str1.size() < str2.size())
364 return str1.compare(str1.size() - str2.size(), str2.size(), str2) == 0;
367 bool StringUtils::EndsWith(const std::string &str1, const char *s2)
369 size_t len2 = strlen(s2);
370 if (str1.size() < len2)
372 return str1.compare(str1.size() - len2, len2, s2) == 0;
375 bool StringUtils::EndsWithNoCase(const std::string &str1, const std::string &str2)
377 if (str1.size() < str2.size())
379 const char *s1 = str1.c_str() + str1.size() - str2.size();
380 const char *s2 = str2.c_str();
383 if (::tolower(*s1) != ::tolower(*s2))
391 bool StringUtils::EndsWithNoCase(const std::string &str1, const char *s2)
393 size_t len2 = strlen(s2);
394 if (str1.size() < len2)
396 const char *s1 = str1.c_str() + str1.size() - len2;
399 if (::tolower(*s1) != ::tolower(*s2))
407 void StringUtils::JoinString(const CStdStringArray &strings, const CStdString& delimiter, CStdString& result)
410 for(CStdStringArray::const_iterator it = strings.begin(); it != strings.end(); it++ )
411 result += (*it) + delimiter;
414 result.Delete(result.size()-delimiter.size(), delimiter.size());
417 CStdString StringUtils::JoinString(const CStdStringArray &strings, const CStdString& delimiter)
420 JoinString(strings, delimiter, result);
424 CStdString StringUtils::Join(const vector<string> &strings, const CStdString& delimiter)
426 CStdStringArray strArray;
427 for (unsigned int index = 0; index < strings.size(); index++)
428 strArray.push_back(strings.at(index));
430 return JoinString(strArray, delimiter);
433 // Splits the string input into pieces delimited by delimiter.
434 // if 2 delimiters are in a row, it will include the empty string between them.
435 // added MaxStrings parameter to restrict the number of returned substrings (like perl and python)
436 int StringUtils::SplitString(const CStdString& input, const CStdString& delimiter, CStdStringArray &results, unsigned int iMaxStrings /* = 0 */)
440 int sizeS2 = delimiter.GetLength();
441 int isize = input.GetLength();
445 vector<unsigned int> positions;
447 newPos = input.Find (delimiter, 0);
451 results.push_back(input);
455 while ( newPos > iPos )
457 positions.push_back(newPos);
459 newPos = input.Find (delimiter, iPos + sizeS2);
462 // numFound is the number of delimiters which is one less
463 // than the number of substrings
464 unsigned int numFound = positions.size();
465 if (iMaxStrings > 0 && numFound >= iMaxStrings)
466 numFound = iMaxStrings - 1;
468 for ( unsigned int i = 0; i <= numFound; i++ )
476 s = input.Mid( i, positions[i] );
480 int offset = positions[i - 1] + sizeS2;
481 if ( offset < isize )
484 s = input.Mid(offset);
486 s = input.Mid( positions[i - 1] + sizeS2,
487 positions[i] - positions[i - 1] - sizeS2 );
490 results.push_back(s);
492 // return the number of substrings
493 return results.size();
496 CStdStringArray StringUtils::SplitString(const CStdString& input, const CStdString& delimiter, unsigned int iMaxStrings /* = 0 */)
498 CStdStringArray result;
499 SplitString(input, delimiter, result, iMaxStrings);
503 vector<string> StringUtils::Split(const std::string& input, const std::string& delimiter, unsigned int iMaxStrings /* = 0 */)
505 CStdStringArray result;
506 SplitString(input, delimiter, result, iMaxStrings);
508 vector<string> strArray;
509 for (unsigned int index = 0; index < result.size(); index++)
510 strArray.push_back(result.at(index));
515 // returns the number of occurrences of strFind in strInput.
516 int StringUtils::FindNumber(const CStdString& strInput, const CStdString &strFind)
518 int pos = strInput.Find(strFind, 0);
523 pos = strInput.Find(strFind, pos + 1);
528 // Compares separately the numeric and alphabetic parts of a string.
529 // returns negative if left < right, positive if left > right
530 // and 0 if they are identical (essentially calculates left - right)
531 int64_t StringUtils::AlphaNumericCompare(const wchar_t *left, const wchar_t *right)
533 wchar_t *l = (wchar_t *)left;
534 wchar_t *r = (wchar_t *)right;
538 const collate<wchar_t>& coll = use_facet< collate<wchar_t> >( locale() );
540 while (*l != 0 && *r != 0)
542 // check if we have a numerical value
543 if (*l >= L'0' && *l <= L'9' && *r >= L'0' && *r <= L'9')
547 while (*ld >= L'0' && *ld <= L'9' && ld < l + 15)
548 { // compare only up to 15 digits
554 while (*rd >= L'0' && *rd <= L'9' && rd < r + 15)
555 { // compare only up to 15 digits
557 rnum += *rd++ - L'0';
559 // do we have numbers?
561 { // yes - and they're different!
568 // do case less comparison
570 if (lc >= L'A' && lc <= L'Z')
573 if (rc >= L'A' && rc <= L'Z')
576 // ok, do a normal comparison, taking current locale into account. Add special case stuff (eg '(' characters)) in here later
577 if ((cmp_res = coll.compare(&lc, &lc + 1, &rc, &rc + 1)) != 0)
591 return 0; // files are the same
594 int StringUtils::DateStringToYYYYMMDD(const CStdString &dateString)
596 CStdStringArray days;
597 int splitCount = StringUtils::SplitString(dateString, "-", days);
599 return atoi(days[0].c_str());
600 else if (splitCount == 2)
601 return atoi(days[0].c_str())*100+atoi(days[1].c_str());
602 else if (splitCount == 3)
603 return atoi(days[0].c_str())*10000+atoi(days[1].c_str())*100+atoi(days[2].c_str());
608 long StringUtils::TimeStringToSeconds(const CStdString &timeString)
610 CStdString strCopy(timeString);
611 StringUtils::Trim(strCopy);
612 if(StringUtils::EndsWithNoCase(strCopy, " min"))
614 // this is imdb format of "XXX min"
615 return 60 * atoi(strCopy.c_str());
619 CStdStringArray secs;
620 StringUtils::SplitString(strCopy, ":", secs);
622 for (unsigned int i = 0; i < 3 && i < secs.size(); i++)
625 timeInSecs += atoi(secs[i]);
631 CStdString StringUtils::SecondsToTimeString(long lSeconds, TIME_FORMAT format)
633 int hh = lSeconds / 3600;
634 lSeconds = lSeconds % 3600;
635 int mm = lSeconds / 60;
636 int ss = lSeconds % 60;
638 if (format == TIME_FORMAT_GUESS)
639 format = (hh >= 1) ? TIME_FORMAT_HH_MM_SS : TIME_FORMAT_MM_SS;
641 if (format & TIME_FORMAT_HH)
642 strHMS.AppendFormat("%02.2i", hh);
643 else if (format & TIME_FORMAT_H)
644 strHMS.AppendFormat("%i", hh);
645 if (format & TIME_FORMAT_MM)
646 strHMS.AppendFormat(strHMS.empty() ? "%02.2i" : ":%02.2i", mm);
647 if (format & TIME_FORMAT_SS)
648 strHMS.AppendFormat(strHMS.empty() ? "%02.2i" : ":%02.2i", ss);
652 bool StringUtils::IsNaturalNumber(const CStdString& str)
655 // allow whitespace,digits,whitespace
656 while (i < str.size() && isspace((unsigned char) str[i]))
658 while (i < str.size() && isdigit((unsigned char) str[i]))
662 while (i < str.size() && isspace((unsigned char) str[i]))
664 return i == str.size() && n > 0;
667 bool StringUtils::IsInteger(const CStdString& str)
670 // allow whitespace,-,digits,whitespace
671 while (i < str.size() && isspace((unsigned char) str[i]))
673 if (i < str.size() && str[i] == '-')
675 while (i < str.size() && isdigit((unsigned char) str[i]))
679 while (i < str.size() && isspace((unsigned char) str[i]))
681 return i == str.size() && n > 0;
684 void StringUtils::RemoveCRLF(CStdString& strLine)
686 while ( strLine.size() && (strLine.Right(1) == "\n" || strLine.Right(1) == "\r") )
688 strLine = strLine.Left(std::max(0, (int)strLine.size() - 1));
692 CStdString StringUtils::SizeToString(int64_t size)
695 const char prefixes[] = {' ','k', 'M', 'G', 'T', 'P', 'E', 'Z', 'Y'};
697 double s = (double)size;
698 while (i < sizeof(prefixes)/sizeof(prefixes[0]) && s >= 1000.0)
705 strLabel = StringUtils::Format("%.0lf %cB ", s, prefixes[i]);
707 strLabel = StringUtils::Format("%.1lf %cB", s, prefixes[i]);
709 strLabel = StringUtils::Format("%.2lf %cB", s, prefixes[i]);
714 // return -1 if not, else return the utf8 char length.
715 int IsUTF8Letter(const unsigned char *str)
718 // unicode -> utf8 table: http://www.utf8-chartable.de/
719 // latin characters in unicode: http://en.wikipedia.org/wiki/Latin_characters_in_Unicode
720 unsigned char ch = str[0];
723 if ((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z'))
727 unsigned char ch2 = str[1];
730 // check latin 1 letter table: http://en.wikipedia.org/wiki/C1_Controls_and_Latin-1_Supplement
731 if (ch == 0xC3 && ch2 >= 0x80 && ch2 <= 0xBF && ch2 != 0x97 && ch2 != 0xB7)
733 // check latin extended A table: http://en.wikipedia.org/wiki/Latin_Extended-A
734 if (ch >= 0xC4 && ch <= 0xC7 && ch2 >= 0x80 && ch2 <= 0xBF)
736 // check latin extended B table: http://en.wikipedia.org/wiki/Latin_Extended-B
737 // and International Phonetic Alphabet: http://en.wikipedia.org/wiki/IPA_Extensions_(Unicode_block)
738 if (((ch == 0xC8 || ch == 0xC9) && ch2 >= 0x80 && ch2 <= 0xBF)
739 || (ch == 0xCA && ch2 >= 0x80 && ch2 <= 0xAF))
744 size_t StringUtils::FindWords(const char *str, const char *wordLowerCase)
746 // NOTE: This assumes word is lowercase!
747 unsigned char *s = (unsigned char *)str;
750 // start with a compare
751 unsigned char *c = s;
752 unsigned char *w = (unsigned char *)wordLowerCase;
754 while (same && *c && *w)
756 unsigned char lc = *c++;
757 if (lc >= 'A' && lc <= 'Z')
760 if (lc != *w++) // different
763 if (same && *w == 0) // only the same if word has been exhausted
764 return (const char *)s - str;
766 // otherwise, skip current word (composed by latin letters) or number
768 if (*s >= '0' && *s <= '9')
771 while (*s >= '0' && *s <= '9') ++s;
773 else if ((l = IsUTF8Letter(s)) > 0)
776 while ((l = IsUTF8Letter(s)) > 0) s += l;
780 while (*s && *s == ' ') s++;
782 // and repeat until we're done
785 return CStdString::npos;
788 // assumes it is called from after the first open bracket is found
789 int StringUtils::FindEndBracket(const CStdString &str, char opener, char closer, int startPos)
792 for (unsigned int i = startPos; i < str.size(); i++)
794 if (str[i] == opener)
796 else if (str[i] == closer)
804 return (int)CStdString::npos;
807 void StringUtils::WordToDigits(CStdString &word)
809 static const char word_to_letter[] = "22233344455566677778889999";
811 for (unsigned int i = 0; i < word.size(); ++i)
812 { // NB: This assumes ascii, which probably needs extending at some point.
813 char letter = word[i];
814 if ((letter >= 'a' && letter <= 'z')) // assume contiguous letter range
816 word[i] = word_to_letter[letter-'a'];
818 else if (letter < '0' || letter > '9') // We want to keep 0-9!
820 word[i] = ' '; // replace everything else with a space
825 CStdString StringUtils::CreateUUID()
827 /* This function generate a DCE 1.1, ISO/IEC 11578:1996 and IETF RFC-4122
828 * Version 4 conform local unique UUID based upon random number generation.
831 char *pUuidStr = UuidStrTmp;
834 static bool m_uuidInitialized = false;
835 if (!m_uuidInitialized)
837 /* use current time as the seed for rand()*/
839 m_uuidInitialized = true;
842 /*Data1 - 8 characters.*/
843 for(i = 0; i < 8; i++, pUuidStr++)
844 ((*pUuidStr = (rand() % 16)) < 10) ? *pUuidStr += 48 : *pUuidStr += 55;
846 /*Data2 - 4 characters.*/
848 for(i = 0; i < 4; i++, pUuidStr++)
849 ((*pUuidStr = (rand() % 16)) < 10) ? *pUuidStr += 48 : *pUuidStr += 55;
851 /*Data3 - 4 characters.*/
853 for(i = 0; i < 4; i++, pUuidStr++)
854 ((*pUuidStr = (rand() % 16)) < 10) ? *pUuidStr += 48 : *pUuidStr += 55;
856 /*Data4 - 4 characters.*/
858 for(i = 0; i < 4; i++, pUuidStr++)
859 ((*pUuidStr = (rand() % 16)) < 10) ? *pUuidStr += 48 : *pUuidStr += 55;
861 /*Data5 - 12 characters.*/
863 for(i = 0; i < 12; i++, pUuidStr++)
864 ((*pUuidStr = (rand() % 16)) < 10) ? *pUuidStr += 48 : *pUuidStr += 55;
868 m_lastUUID = UuidStrTmp;
872 bool StringUtils::ValidateUUID(const CStdString &uuid)
875 guidRE.RegComp(ADDON_GUID_RE);
876 return (guidRE.RegFind(uuid.c_str()) == 0);
879 double StringUtils::CompareFuzzy(const CStdString &left, const CStdString &right)
881 return (0.5 + fstrcmp(left.c_str(), right.c_str(), 0.0) * (left.length() + right.length())) / 2.0;
884 int StringUtils::FindBestMatch(const CStdString &str, const CStdStringArray &strings, double &matchscore)
890 for (CStdStringArray::const_iterator it = strings.begin(); it != strings.end(); it++, i++)
892 int maxlength = max(str.length(), it->length());
893 double score = StringUtils::CompareFuzzy(str, *it) / maxlength;
894 if (score > matchscore)
903 bool StringUtils::ContainsKeyword(const CStdString &str, const CStdStringArray &keywords)
905 for (CStdStringArray::const_iterator it = keywords.begin(); it != keywords.end(); it++)
907 if (str.find(*it) != str.npos)
913 size_t StringUtils::utf8_strlen(const char *s)
918 if ((*s++ & 0xC0) != 0x80)
924 std::string StringUtils::Paramify(const std::string ¶m)
926 std::string result = param;
928 StringUtils::Replace(result, "\\", "\\\\");
929 // escape double quotes
930 StringUtils::Replace(result, "\"", "\\\"");
932 // add double quotes around the whole string
933 return "\"" + result + "\"";
936 void StringUtils::Tokenize(const std::string& input, std::vector<std::string>& tokens, const std::string& delimiters)
938 // Tokenize ripped from http://www.linuxselfhelp.com/HOWTO/C++Programming-HOWTO-7.html
939 // Skip delimiters at beginning.
940 string::size_type lastPos = input.find_first_not_of(delimiters, 0);
941 // Find first "non-delimiter".
942 string::size_type pos = input.find_first_of(delimiters, lastPos);
944 while (string::npos != pos || string::npos != lastPos)
946 // Found a token, add it to the vector.
947 tokens.push_back(input.substr(lastPos, pos - lastPos));
948 // Skip delimiters. Note the "not_of"
949 lastPos = input.find_first_not_of(delimiters, pos);
950 // Find next "non-delimiter"
951 pos = input.find_first_of(delimiters, lastPos);