Files
itgmania212121/stepmania/src/RageUtil.cpp
T

1025 lines
22 KiB
C++
Raw Normal View History

2003-02-16 04:01:45 +00:00
#include "global.h"
/*
-----------------------------------------------------------------------------
File: RageUtil
Desc: See header.
2002-05-19 01:59:48 +00:00
Copyright (c) 2001-2002 by the person(s) listed below. All rights reserved.
Chris Danford
-----------------------------------------------------------------------------
*/
2003-07-04 18:03:20 +00:00
#include "RageMath.h"
2001-11-03 10:52:42 +00:00
#include "RageUtil.h"
2003-07-03 02:41:27 +00:00
#include "RageUtil_FileDB.h"
#include "RageLog.h"
2003-07-22 07:47:27 +00:00
#include "arch/arch.h"
2002-07-23 01:41:40 +00:00
#include <numeric>
2003-01-03 05:26:15 +00:00
#include <time.h>
#include <math.h>
2003-07-22 07:47:27 +00:00
#include "RageFile.h"
2003-01-06 02:48:14 +00:00
#include <map>
2003-07-03 09:22:03 +00:00
#include <sys/types.h>
#include <sys/stat.h>
2003-01-03 05:26:15 +00:00
unsigned long randseed = time(NULL);
RandomGen::RandomGen( unsigned long seed_ )
{
seed = seed_;
if(seed == 0)
seed = time(NULL);
}
int RandomGen::operator() ( int maximum )
{
seed = 1664525L * seed + 1013904223L;
2003-08-03 04:47:53 +00:00
return (seed >> 2) % maximum;
}
void fapproach( float& val, float other_val, float to_move )
{
ASSERT( to_move >= 0 );
if( val == other_val )
return;
float fDelta = other_val - val;
float fSign = fDelta / fabsf( fDelta );
float fToMove = fSign*to_move;
if( fabsf(fToMove) > fabsf(fDelta) )
fToMove = fDelta; // snap
val += fToMove;
}
2003-02-05 20:18:02 +00:00
/* Return a positive x mod y. */
float fmodfp(float x, float y)
{
x = fmodf(x, y); /* x is [-y,y] */
x += y; /* x is [0,y*2] */
x = fmodf(x, y); /* x is [0,y] */
return x;
}
2002-12-20 06:28:22 +00:00
int power_of_two(int input)
{
int value = 1;
while ( value < input ) value <<= 1;
return value;
}
2003-01-03 05:26:15 +00:00
bool IsAnInt( const CString &s )
2001-12-28 10:15:59 +00:00
{
2003-05-22 21:13:02 +00:00
if( !s.size() )
2001-12-28 10:15:59 +00:00
return false;
2002-12-17 06:47:19 +00:00
for( int i=0; s[i]; i++ )
2001-12-28 10:15:59 +00:00
if( s[i] < '0' || s[i] > '9' )
return false;
return true;
}
2003-01-03 22:51:01 +00:00
bool IsHexVal( const CString &s )
{
2003-05-22 21:13:29 +00:00
if( !s.size() )
2003-01-03 22:51:01 +00:00
return false;
for( int i=0; s[i]; i++ )
if( !(s[i] >= '0' && s[i] <= '9') &&
!(toupper(s[i]) >= 'A' && toupper(s[i]) <= 'F'))
return false;
return true;
}
2002-04-28 20:42:32 +00:00
float TimeToSeconds( CString sHMS )
{
CStringArray arrayBits;
split( sHMS, ":", arrayBits, false );
2002-10-31 02:11:52 +00:00
while( arrayBits.size() < 3 )
2002-10-24 19:55:09 +00:00
arrayBits.insert(arrayBits.begin(), "0" ); // pad missing bits
2002-04-28 20:42:32 +00:00
float fSeconds = 0;
fSeconds += atoi( arrayBits[0] ) * 60 * 60;
fSeconds += atoi( arrayBits[1] ) * 60;
fSeconds += (float)atof( arrayBits[2] );
return fSeconds;
}
2001-12-28 10:15:59 +00:00
2002-08-01 13:42:56 +00:00
CString SecondsToTime( float fSecs )
{
const int iMinsDisplay = (int)fSecs/60;
const int iSecsDisplay = (int)fSecs - iMinsDisplay*60;
const float fLeftoverDisplay = (fSecs - iMinsDisplay*60 - iSecsDisplay) * 100;
CString sReturn = ssprintf( "%02d:%02d:%02.0f", iMinsDisplay, iSecsDisplay, min(99.0f,fLeftoverDisplay) );
if( iMinsDisplay >= 60 )
{
2002-10-16 18:25:14 +00:00
/* Oops. Probably a really long endless course. Do "hh:mm.ss"; use a period
* to differentiate between "mm:ss:ms". I'd much prefer reversing those, since
* it makes much more sense to do "mm:ss.ms", but people would probably complain
* about "arcade accuracy" ...
*/
sReturn = ssprintf( "%02d:%02d.%02d", iMinsDisplay/60, iMinsDisplay%60, iSecsDisplay );
2002-10-16 18:25:14 +00:00
}
ASSERT( sReturn.GetLength() <= 8 );
return sReturn;
2002-08-01 13:42:56 +00:00
}
2001-12-28 10:15:59 +00:00
2002-08-29 21:02:04 +00:00
CString ssprintf( const char *fmt, ...)
2001-11-03 10:52:42 +00:00
{
va_list va;
va_start(va, fmt);
return vssprintf(fmt, va);
}
2002-08-29 21:02:04 +00:00
CString vssprintf( const char *fmt, va_list argList)
2001-11-03 10:52:42 +00:00
{
CString str;
str.FormatV(fmt, argList);
return str;
}
#ifdef WIN32
2003-07-14 22:23:04 +00:00
#ifdef _XBOX
#include "D3DX8Core.h"
2003-07-14 22:23:04 +00:00
#else
#include "Dxerr8.h"
#pragma comment(lib, "Dxerr8.lib")
#endif
2001-11-03 10:52:42 +00:00
CString hr_ssprintf( int hr, const char *fmt, ...)
{
va_list va;
va_start(va, fmt);
CString s = vssprintf( fmt, va );
va_end(va);
2003-07-14 22:23:04 +00:00
#ifdef _XBOX
/* XXX: This doesn't work for DirectInput or DirectSound; why is it needed?
* XBox uses DI and DS, too, right? -glenn */
char szError[1024] = "";
D3DXGetErrorString( hr, szError, sizeof(szError) );
2003-07-14 22:23:04 +00:00
#else
const char *szError = DXGetErrorString8( hr );
#endif
return s + ssprintf( " (%s)", szError );
}
2002-12-21 08:40:40 +00:00
CString werr_ssprintf( int err, const char *fmt, ...)
{
2003-05-22 05:28:37 +00:00
char buf[1024] = "";
#ifndef _XBOX
2002-12-21 08:40:40 +00:00
FormatMessage(FORMAT_MESSAGE_FROM_SYSTEM,
0, err, 0, buf, sizeof(buf), NULL);
2003-05-22 05:28:37 +00:00
#endif
2002-12-21 08:40:40 +00:00
va_list va;
va_start(va, fmt);
CString s = vssprintf( fmt, va );
va_end(va);
2002-12-21 08:44:26 +00:00
return s += ssprintf( " (%s)", buf );
2002-12-21 08:40:40 +00:00
}
#endif
2001-11-03 10:52:42 +00:00
2002-06-24 22:04:31 +00:00
CString join( const CString &Deliminator, const CStringArray& Source)
2001-11-03 10:52:42 +00:00
{
2002-10-31 02:11:52 +00:00
if( Source.empty() )
2002-06-27 17:49:10 +00:00
return "";
2001-11-03 10:52:42 +00:00
CString csTmp;
// Loop through the Array and Append the Deliminator
2002-10-31 02:11:52 +00:00
for( unsigned iNum = 0; iNum < Source.size()-1; iNum++ ) {
2002-10-24 19:55:09 +00:00
csTmp += Source[iNum];
2001-11-03 10:52:42 +00:00
csTmp += Deliminator;
}
2002-10-24 19:55:09 +00:00
csTmp += Source.back();
2002-06-24 22:04:31 +00:00
return csTmp;
2001-11-03 10:52:42 +00:00
}
2003-01-05 03:59:07 +00:00
template <class S>
void do_split( const S &Source, const S &Deliminator, vector<S> &AddIt, const bool bIgnoreEmpty )
2001-11-03 10:52:42 +00:00
{
2003-01-05 03:59:07 +00:00
unsigned startpos = 0;
2001-11-03 10:52:42 +00:00
do {
2003-01-15 20:37:15 +00:00
unsigned pos = Source.find(Deliminator, startpos);
2003-01-05 03:59:07 +00:00
if ( pos == Source.npos ) pos=Source.size();
2002-08-30 00:28:20 +00:00
2003-01-05 03:59:07 +00:00
S AddCString = Source.substr(startpos, pos-startpos);
2003-01-03 05:26:15 +00:00
if( AddCString.empty() && bIgnoreEmpty )
2002-08-30 00:28:20 +00:00
; // do nothing
else
2002-10-31 04:23:39 +00:00
AddIt.push_back(AddCString);
2001-11-03 10:52:42 +00:00
2003-01-05 03:59:07 +00:00
startpos=pos+Deliminator.size();
} while ( startpos <= Source.size() );
}
void split( const CString &Source, const CString &Deliminator, CStringArray &AddIt, const bool bIgnoreEmpty )
{
do_split(Source, Deliminator, AddIt, bIgnoreEmpty );
}
2003-01-28 00:45:01 +00:00
void split( const wstring &Source, const wstring &Deliminator, vector<wstring> &AddIt, const bool bIgnoreEmpty )
2003-01-05 03:59:07 +00:00
{
do_split(Source, Deliminator, AddIt, bIgnoreEmpty );
2001-11-03 10:52:42 +00:00
}
2003-01-16 22:49:22 +00:00
/*
* foo\fum\ -> "foo\fum\", "", ""
* c:\foo\bar.txt -> "c:\foo\", "bar", ".txt"
* \\foo\fum -> "\\foo\", "fum", ""
*/
void splitpath( CString Path, CString& Dir, CString& Filename, CString& Ext )
2001-11-03 10:52:42 +00:00
{
2003-01-16 22:49:22 +00:00
Dir = Filename = Ext = "";
2001-11-03 10:52:42 +00:00
2003-01-16 22:49:22 +00:00
CStringArray mat;
2001-11-03 10:52:42 +00:00
/* One level of escapes for the regex, one for C. Ew.
* This is really:
* ^(.*[\\/])?(.*)$ */
static Regex sep("^(.*[\\\\/])?(.*)$");
bool check = sep.Compare(Path, mat);
ASSERT(check);
2003-01-16 22:49:22 +00:00
Dir = mat[0];
Path = mat[1];
/* ^(.*)(\.[^\.]+)$ */
static Regex SplitExt("^(.*)(\\.[^\\.]+)$");
if(SplitExt.Compare(Path, mat))
2003-01-16 22:49:22 +00:00
{
Filename = mat[0];
Ext = mat[1];
} else
Filename = Path;
2001-11-03 10:52:42 +00:00
}
void splitrelpath( const CString &Path, CString& Dir, CString& FName, CString& Ext )
2001-12-28 10:15:59 +00:00
{
2002-09-01 00:37:10 +00:00
/* Find the last slash or backslash. */
int Last = max(Path.ReverseFind('/'), Path.ReverseFind('\\'));
2001-12-28 10:15:59 +00:00
2002-09-01 00:37:10 +00:00
/* Set 'Last' to the first character of the filename. If we have
* no directory separators, this is the entire string, so set it
* to 0. */
if(Last == -1) Last = 0;
else Last++;
2001-12-28 10:15:59 +00:00
2002-09-01 00:37:10 +00:00
CString sFNameAndExt = Path.Right(Path.GetLength()-Last);
2001-12-28 10:15:59 +00:00
// subtract the FNameAndExt from Path
Dir = Path.Left( Path.GetLength()-sFNameAndExt.GetLength() ); // don't subtract out the trailing slash
CStringArray sFNameAndExtBits;
split( sFNameAndExt, ".", sFNameAndExtBits, false );
2002-09-01 00:37:10 +00:00
if( sFNameAndExt.GetLength() == 0 ) // no file at the end of this path
2001-12-28 10:15:59 +00:00
{
FName = "";
Ext = "";
}
2002-10-31 02:11:52 +00:00
else if( sFNameAndExtBits.size() == 1 ) // file doesn't have extension
2001-12-28 10:15:59 +00:00
{
FName = sFNameAndExtBits[0];
Ext = "";
}
2002-10-31 02:11:52 +00:00
else if( sFNameAndExtBits.size() > 1 ) // file has extension and possibly multiple periods
2001-12-28 10:15:59 +00:00
{
2002-10-31 02:11:52 +00:00
Ext = sFNameAndExtBits[ sFNameAndExtBits.size()-1 ];
2001-12-28 10:15:59 +00:00
// subtract the Ext and last period from FNameAndExt
FName = sFNameAndExt.Left( sFNameAndExt.GetLength()-Ext.GetLength()-1 );
}
}
2003-09-04 19:28:39 +00:00
CString SetExtension( const CString &path, const CString &ext )
{
CString Dir, FName, OldExt;
splitpath( path, Dir, FName, OldExt );
return Dir + FName + "." + ext;
}
CString GetExtension( CString sPath )
{
CString Dir, FName, Ext;
splitrelpath( sPath, Dir, FName, Ext );
Ext.MakeLower();
return Ext;
}
2003-09-05 06:20:55 +00:00
CString GetCwd()
{
char buf[PATH_MAX];
bool ret = getcwd(buf, PATH_MAX) != NULL;
ASSERT(ret);
return buf;
}
2002-08-30 00:28:20 +00:00
/* mkdir -p. Doesn't fail if Path already exists and is a directory. */
bool CreateDirectories( CString Path )
{
CStringArray parts;
CString curpath;
2003-07-22 07:47:27 +00:00
split(Path, SLASH, parts);
2002-08-30 00:28:20 +00:00
2002-10-31 02:11:52 +00:00
for(unsigned i = 0; i < parts.size(); ++i)
2002-08-30 00:28:20 +00:00
{
2003-07-22 07:47:27 +00:00
curpath += parts[i] + SLASH;
2003-07-28 23:26:40 +00:00
if( mkdir(curpath, 0755) == 0 )
2002-08-30 00:28:20 +00:00
continue;
2002-12-17 06:47:19 +00:00
if(errno != EEXIST)
2003-09-06 22:32:38 +00:00
{
if( LOG )
LOG->Warn("Couldn't create %s: %s", curpath.c_str(), strerror(errno) );
2003-10-10 04:16:44 +00:00
/* When creating a directory that already exists over Samba, Windows is
* returning ENOENT instead of EEXIST. Continue if we get that error, but
* log it just in case it was a real ENOENT. (Otherwise, we'll fail to
* create a directory when we could have succeeded.) */
if( errno != ENOENT )
return false;
else
continue;
2003-09-06 22:32:38 +00:00
}
2002-08-30 00:28:20 +00:00
/* Make sure it's a directory. */
if( !IsADirectory(curpath) )
2003-09-06 22:32:38 +00:00
{
if( LOG )
LOG->Warn("Couldn't create %s: path exists and is not a directory", curpath.c_str() );
2002-08-30 00:28:20 +00:00
return false;
2003-09-06 22:32:38 +00:00
}
2002-08-30 00:28:20 +00:00
}
return true;
}
2001-12-28 10:15:59 +00:00
2003-09-08 07:21:41 +00:00
bool Rename( const char *oldname, const char *newname )
{
return 0 == rename( oldname, newname );
}
2003-01-23 04:30:08 +00:00
#if 0
2002-05-19 06:13:16 +00:00
void GetDirListing( CString sPath, CStringArray &AddTo, bool bOnlyDirs, bool bReturnPathToo )
2001-11-03 10:52:42 +00:00
{
2002-05-19 06:13:16 +00:00
CString sDir, sThrowAway;
splitrelpath( sPath, sDir, sThrowAway, sThrowAway );
2002-10-18 19:46:00 +00:00
/* XXX: We should use Find* to get a file list only, and handle
* wildcard matching ourself. Windows matching is braindead. For
* example, *.dwi matches "foo.dwi~". */
2001-11-03 10:52:42 +00:00
WIN32_FIND_DATA fd;
HANDLE hFind = ::FindFirstFile( sPath, &fd );
2002-03-06 08:25:09 +00:00
if( INVALID_HANDLE_VALUE == hFind ) // no files found
return;
do
2001-11-03 10:52:42 +00:00
{
2002-03-06 08:25:09 +00:00
if( bOnlyDirs && !(fd.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) )
continue; // skip
CString sDirName( fd.cFileName );
if( sDirName == "." || sDirName == ".." )
continue;
2002-05-19 06:13:16 +00:00
if( bReturnPathToo )
2002-10-31 04:23:39 +00:00
AddTo.push_back( sDir + sDirName );
2002-05-19 06:13:16 +00:00
else
2002-10-31 04:23:39 +00:00
AddTo.push_back( sDirName );
2002-05-19 06:13:16 +00:00
2002-03-06 08:25:09 +00:00
} while( ::FindNextFile( hFind, &fd ) );
::FindClose( hFind );
2001-11-03 10:52:42 +00:00
}
2003-01-23 04:30:08 +00:00
#endif
2001-11-03 10:52:42 +00:00
/* Reference: http://www.theorem.com/java/CRC32.java, rewritten by Glenn Maynard.
* Public domain. */
unsigned int GetHashForString ( CString s )
2002-05-19 01:59:48 +00:00
{
static unsigned tab[256];
static bool initted = false;
if(!initted)
{
initted = true;
const unsigned POLY = 0xEDB88320;
2002-07-23 01:41:40 +00:00
for(int i = 0; i < 256; ++i)
{
tab[i] = i;
for(int j = 0; j < 8; ++j)
{
if(tab[i] & 1) tab[i] = (tab[i] >> 1) ^ POLY;
else tab[i] >>= 1;
}
}
}
2002-07-23 01:41:40 +00:00
unsigned crc = 0;
for(unsigned i = 0; i < s.size(); ++i)
crc = (crc >> 8) ^ tab[(crc ^ s[i]) & 0xFF];
return crc;
2002-05-19 01:59:48 +00:00
}
2002-08-20 21:00:56 +00:00
unsigned int GetHashForFile( CString sPath )
2002-05-19 01:59:48 +00:00
{
2002-08-20 21:00:56 +00:00
unsigned int hash = 0;
2002-05-19 01:59:48 +00:00
hash += GetHashForString( sPath );
hash += GetFileSizeInBytes( sPath );
2003-09-23 00:09:10 +00:00
hash += GetFileModTime( sPath );
2002-05-19 01:59:48 +00:00
2002-08-20 21:00:56 +00:00
return hash;
2002-05-19 01:59:48 +00:00
}
2002-08-20 21:00:56 +00:00
unsigned int GetHashForDirectory( CString sDir )
2002-05-19 01:59:48 +00:00
{
2002-08-20 21:00:56 +00:00
unsigned int hash = 0;
2002-05-19 01:59:48 +00:00
hash += GetHashForFile( sDir );
CStringArray arrayFiles;
2003-07-22 07:47:27 +00:00
GetDirListing( sDir+"*", arrayFiles, false );
2002-10-31 02:11:52 +00:00
for( unsigned i=0; i<arrayFiles.size(); i++ )
2002-05-19 01:59:48 +00:00
{
const CString sFilePath = sDir + arrayFiles[i];
hash += GetHashForFile( sFilePath );
}
2002-08-20 21:00:56 +00:00
return hash;
2002-05-19 01:59:48 +00:00
}
2002-10-24 08:17:09 +00:00
bool CompareCStringsAsc(const CString &str1, const CString &str2)
2001-11-03 10:52:42 +00:00
{
2002-11-08 20:12:19 +00:00
return str1.CompareNoCase( str2 ) < 0;
2001-11-03 10:52:42 +00:00
}
2002-10-24 08:17:09 +00:00
bool CompareCStringsDesc(const CString &str1, const CString &str2)
2001-11-03 10:52:42 +00:00
{
2002-11-08 20:12:19 +00:00
return str1.CompareNoCase( str2 ) > 0;
2001-11-03 10:52:42 +00:00
}
2002-10-24 08:17:09 +00:00
void SortCStringArray( CStringArray &arrayCStrings, const bool bSortAscending )
2001-11-03 10:52:42 +00:00
{
2002-10-24 08:40:27 +00:00
sort( arrayCStrings.begin(), arrayCStrings.end(),
2002-10-24 08:17:09 +00:00
bSortAscending?CompareCStringsAsc:CompareCStringsDesc);
2001-11-03 10:52:42 +00:00
}
float calc_mean(const float *start, const float *end)
{
return accumulate(start, end, 0.f) / distance(start, end);
}
float calc_stddev(const float *start, const float *end)
{
/* Calculate the mean. */
float mean = calc_mean(start, end);
/* Calculate stddev. */
float dev = 0.0f;
for( const float *i=start; i != end; ++i )
dev += (*i - mean) * (*i - mean);
dev /= distance(start, end) - 1;
dev = sqrtf(dev);
return dev;
}
2002-05-27 08:23:27 +00:00
void TrimLeft(CString &str, const char *s)
{
int n = 0;
while(n < int(str.size()) && strchr(s, str[n]))
n++;
str.erase(str.begin(), str.begin()+n);
}
2002-05-27 08:23:27 +00:00
void TrimRight(CString &str, const char *s)
{
2002-11-01 20:48:36 +00:00
int n = str.size();
while(n > 0 && strchr(s, str[n-1]))
n--;
2002-11-01 20:48:36 +00:00
/* Delete from n to the end. If n == str.size(), nothing is deleted;
* if n == 0, the whole string is erased. */
str.erase(str.begin()+n, str.end());
}
2003-01-01 22:02:41 +00:00
2003-02-14 22:28:29 +00:00
void StripCrnl(CString &s)
{
2003-07-10 01:55:37 +00:00
while( s.size() && (s[s.size()-1] == '\r' || s[s.size()-1] == '\n') )
2003-02-14 22:28:29 +00:00
s.erase(s.size()-1);
}
2003-01-01 22:02:41 +00:00
/* path is a .redir pathname. Read it and return the real one. */
CString DerefRedir(const CString &path)
{
CString sDir, sFName, sExt;
splitrelpath( path, sDir, sFName, sExt );
if(sExt != "redir")
{
CString path2 = path;
ResolvePath( path2 );
return path2;
}
2003-01-01 22:02:41 +00:00
CString sNewFileName;
{
ifstream file(path);
2003-01-01 22:02:41 +00:00
getline(file, sNewFileName);
}
2003-02-16 05:12:27 +00:00
StripCrnl(sNewFileName);
2003-01-01 22:02:41 +00:00
/* Empty is invalid. */
if(sNewFileName == "")
return "";
FixSlashesInPlace( sNewFileName );
CString path2 = sDir+sNewFileName;
CollapsePath( path2 );
ResolvePath( path2 );
2003-09-23 05:39:47 +00:00
if( !DoesFileExist(path2) )
RageException::Throw( "The redirect '%s' references a file '%s' which doesn't exist.", path.c_str(), path2.c_str() );
return path2;
2003-01-01 22:02:41 +00:00
}
2003-01-05 02:48:15 +00:00
CString GetRedirContents(const CString &path)
{
CString sDir, sFName, sExt;
splitrelpath( path, sDir, sFName, sExt );
ASSERT( sExt == "redir"); // don't ever call this on a non-redirect
CString sNewFileName;
{
ifstream file(path);
getline(file, sNewFileName);
}
StripCrnl(sNewFileName);
return sNewFileName;
}
#if 1
/* PCRE */
#include "pcre/pcre.h"
void Regex::Compile()
{
const char *error;
int offset;
reg = pcre_compile(pattern.c_str(), PCRE_CASELESS, &error, &offset, NULL);
if(reg == NULL)
RageException::Throw("Invalid regex: '%s' (%s)", pattern.c_str(), error);
int ret = pcre_fullinfo( (pcre *) reg, NULL, PCRE_INFO_CAPTURECOUNT, &backrefs);
ASSERT(ret >= 0);
backrefs++;
ASSERT(backrefs < 128);
}
void Regex::Set(const CString &str)
{
Release();
pattern=str;
Compile();
}
void Regex::Release()
{
pcre_free(reg);
reg = NULL;
pattern = "";
}
Regex::Regex(const CString &str)
{
reg = NULL;
Set(str);
}
Regex::Regex(const Regex &rhs)
{
reg = NULL;
Set(rhs.pattern);
}
Regex &Regex::operator=(const Regex &rhs)
{
if(this != &rhs) Set(rhs.pattern);
return *this;
}
Regex::~Regex()
{
Release();
}
bool Regex::Compare(const CString &str)
{
int mat[128*3];
int ret = pcre_exec( (pcre *) reg, NULL,
str.data(), str.size(), 0, 0, mat, 128*3);
if( ret < -1 )
RageException::Throw("Unexpected return from pcre_exec('%s'): %i",
pattern.c_str(), ret);
return ret >= 0;
}
bool Regex::Compare(const CString &str, vector<CString> &matches)
{
matches.clear();
int mat[128*3];
int ret = pcre_exec( (pcre *) reg, NULL,
str.data(), str.size(), 0, 0, mat, 128*3);
if( ret < -1 )
RageException::Throw("Unexpected return from pcre_exec('%s'): %i",
pattern.c_str(), ret);
if(ret == -1)
return false;
for(unsigned i = 1; i < backrefs; ++i)
{
const int start = mat[i*2], end = mat[i*2+1];
if(start == -1)
matches.push_back(""); /* no match */
else
matches.push_back(str.substr(start, end - start));
}
return true;
}
#else
/* GNU regex */
#include "regex.h"
2003-01-23 21:03:57 +00:00
void Regex::Compile()
2003-01-05 02:48:15 +00:00
{
2003-01-23 21:03:57 +00:00
reg = new regex_t;
int ret = regcomp((regex_t *) reg, pattern.c_str(), REG_EXTENDED|REG_ICASE);
if(ret != 0)
RageException::Throw("Invalid regex: '%s'", pattern.c_str());
/* Count the number of backreferences. */
backrefs = 0;
for(int i = 0; i < int(pattern.size()); ++i)
if(pattern[i] == '(') backrefs++;
ASSERT(backrefs+1 < 128);
2003-01-05 02:48:15 +00:00
}
2003-03-27 20:16:17 +00:00
void Regex::Set(const CString &str)
{
Release();
pattern=str;
Compile();
}
2003-01-23 21:03:57 +00:00
void Regex::Release()
{
delete (regex_t *)reg;
reg = NULL;
pattern = "";
2003-01-23 21:03:57 +00:00
}
2003-03-27 20:16:17 +00:00
Regex::Regex(const CString &str)
2003-01-23 21:03:57 +00:00
{
reg = NULL;
2003-03-27 20:16:17 +00:00
Set(str);
2003-01-23 21:03:57 +00:00
}
Regex::Regex(const Regex &rhs)
{
reg = NULL;
Set(rhs.pattern);
}
Regex &Regex::operator=(const Regex &rhs)
{
if(this != &rhs) Set(rhs.pattern);
return *this;
2003-01-23 21:03:57 +00:00
}
2003-01-05 02:48:15 +00:00
Regex::~Regex()
2003-01-05 02:48:15 +00:00
{
2003-01-23 21:03:57 +00:00
Release();
2003-01-05 02:48:15 +00:00
}
bool Regex::Compare(const CString &str)
2003-01-05 02:48:15 +00:00
{
return regexec((regex_t *) reg, str.c_str(), 0, NULL, 0) != REG_NOMATCH;
}
2003-01-05 02:48:15 +00:00
bool Regex::Compare(const CString &str, vector<CString> &matches)
{
matches.clear();
2003-01-05 02:48:15 +00:00
regmatch_t mat[128];
int ret = regexec((regex_t *) reg, str.c_str(), 128, mat, 0);
2003-01-05 02:48:15 +00:00
if(ret == REG_NOMATCH)
2003-01-05 02:48:15 +00:00
return false;
for(unsigned i = 1; i < backrefs+1; ++i)
2003-01-05 02:48:15 +00:00
{
if(mat[i].rm_so == -1)
matches.push_back(""); /* no match */
else
matches.push_back(str.substr(mat[i].rm_so, mat[i].rm_eo - mat[i].rm_so));
}
return true;
}
#endif
2003-01-05 02:48:15 +00:00
2003-01-05 05:31:59 +00:00
/* UTF-8 decoding code from glib. */
char *utf8_find_next_char (const char *p, const char *end)
{
2003-01-06 01:19:22 +00:00
if (end)
for (++p; p < end && (*p & 0xc0) == 0x80; ++p)
;
else if (*p) {
for (++p; (*p & 0xc0) == 0x80; ++p)
;
}
return (p == end) ? NULL : (char *)p;
2003-01-05 05:31:59 +00:00
}
2003-01-06 01:19:22 +00:00
int masks[7] = { 0, 0x7f, 0x1f, 0x0f, 0x07, 0x03, 0x01 };
int utf8_get_char_len (const char *pp)
{
2003-05-18 04:36:03 +00:00
const unsigned char *p = (const unsigned char *) pp;
if (*p < 128) return 1;
else if ((*p & 0xe0) == 0xc0) return 2;
else if ((*p & 0xf0) == 0xe0) return 3;
else if ((*p & 0xf8) == 0xf0) return 4;
else if ((*p & 0xfc) == 0xf8) return 5;
else if ((*p & 0xfe) == 0xfc) return 6;
return -1;
}
bool utf8_is_valid(const CString &str)
{
unsigned pos = 0;
while(pos < str.size())
{
int len = utf8_get_char_len(str.c_str() + pos);
if(len == -1)
return false;
2003-06-30 08:38:24 +00:00
if( utf8_get_char( str.c_str() + pos ) == INVALID_CHAR )
return false;
2003-05-18 04:36:03 +00:00
pos += len;
}
return true;
2003-01-06 01:19:22 +00:00
}
2003-01-05 05:31:59 +00:00
2003-01-28 00:45:01 +00:00
wchar_t utf8_get_char (const char *p)
2003-01-05 05:31:59 +00:00
{
2003-01-06 01:19:22 +00:00
int len = utf8_get_char_len(p);
if(len == -1)
2003-01-28 00:45:01 +00:00
return INVALID_CHAR;
2003-01-06 01:19:22 +00:00
int mask = masks[len];
2003-01-28 00:45:01 +00:00
wchar_t result = wchar_t(p[0] & mask);
2003-01-06 01:19:22 +00:00
for (int i = 1; i < len; ++i) {
2003-01-05 05:31:59 +00:00
if ((p[i] & 0xc0) != 0x80)
2003-01-28 00:45:01 +00:00
return INVALID_CHAR;
2003-01-05 05:31:59 +00:00
result <<= 6;
result |= p[i] & 0x3f;
}
return result;
}
2003-01-28 00:45:01 +00:00
const wchar_t INVALID_CHAR = 0xFFFF;
2003-01-06 04:23:41 +00:00
2003-01-28 00:45:01 +00:00
wstring CStringToWstring(const CString &str)
2003-01-05 05:31:59 +00:00
{
const char *ptr = str.c_str(), *end = str.c_str()+str.size();
2003-01-28 00:45:01 +00:00
wstring ret;
2003-01-05 05:31:59 +00:00
2003-01-06 01:19:22 +00:00
while(ptr && ptr != end)
2003-01-05 05:31:59 +00:00
{
2003-01-28 00:45:01 +00:00
wchar_t c = utf8_get_char (ptr);
2003-01-06 04:23:41 +00:00
if(c == -1)
2003-01-07 08:16:55 +00:00
ret += INVALID_CHAR;
2003-01-06 04:23:41 +00:00
else
2003-01-07 08:16:55 +00:00
ret += c;
2003-01-05 05:31:59 +00:00
ptr = utf8_find_next_char (ptr, end);
}
return ret;
}
2003-01-06 01:35:05 +00:00
2003-01-28 00:45:01 +00:00
CString WStringToCString(const wstring &str)
{
CString ret;
for(unsigned i = 0; i < str.size(); ++i)
2003-01-28 00:45:01 +00:00
ret.append(WcharToUTF8(str[i]));
return ret;
}
2003-01-28 00:45:01 +00:00
static int unichar_to_utf8 (wchar_t c, char *outbuf)
2003-01-06 01:35:05 +00:00
{
unsigned int len = 0;
int first;
if (c < 0x80) {
first = 0;
len = 1;
} else if (c < 0x800) {
first = 0xc0;
len = 2;
} else if (c < 0x10000) {
first = 0xe0;
len = 3;
} else if (c < 0x200000) {
first = 0xf0;
len = 4;
} else if (c < 0x4000000) {
first = 0xf8;
len = 5;
} else {
first = 0xfc;
len = 6;
}
if (outbuf)
{
for (int i = len - 1; i > 0; --i)
{
outbuf[i] = char((c & 0x3f) | 0x80);
c >>= 6;
}
outbuf[0] = char(c | first);
}
return len;
}
2003-01-28 00:45:01 +00:00
CString WcharToUTF8( wchar_t c )
2003-01-06 01:35:05 +00:00
{
char buf[6];
int cnt = unichar_to_utf8(c, buf);
return CString(buf, cnt);
}
2003-01-06 02:48:14 +00:00
2003-02-15 00:21:40 +00:00
2003-01-06 02:48:14 +00:00
/* Replace &#nnnn; (decimal) &xnnnn; (hex) with corresponding UTF-8 characters. */
void Replace_Unicode_Markers( CString &Text )
{
unsigned start = 0;
while(start < Text.size())
{
/* Look for &#digits; */
bool hex = false;
unsigned pos = Text.find("&#", start);
if(pos == Text.npos) {
hex = true;
pos = Text.find("&x", start);
}
if(pos == Text.npos) break;
start = pos+1;
unsigned p = pos;
p += 2;
/* Found &# or &x. Is it followed by digits and a semicolon? */
if(p >= Text.size()) continue;
int numdigits = 0;
while(p < Text.size() &&
(hex && isxdigit(Text[p])) || (!hex && isdigit(Text[p])))
{
p++;
numdigits++;
}
if(!numdigits) continue; /* must have at least one digit */
if(p >= Text.size() || Text[p] != ';') continue;
p++;
int num;
if(hex) sscanf(Text.c_str()+pos, "&x%x;", &num);
else sscanf(Text.c_str()+pos, "&#%i;", &num);
2003-01-28 00:45:01 +00:00
if(num > 0xFFFF)
num = INVALID_CHAR;
Text.replace(pos, p-pos, WcharToUTF8(wchar_t(num)));
2003-01-06 02:48:14 +00:00
}
}
void ReplaceText( CString &Text, const map<CString,CString> &m )
{
2003-02-14 23:06:01 +00:00
basic_string<char,char_traits_char_nocase> txt(Text);
2003-01-18 07:26:31 +00:00
2003-01-06 02:48:14 +00:00
for(map<CString,CString>::const_iterator it = m.begin(); it != m.end(); ++it)
{
unsigned start = 0;
while(1)
{
unsigned pos = txt.find(it->first, start);
if(pos == txt.npos)
break;
2003-01-18 07:26:31 +00:00
txt.replace(pos, it->first.size(), it->second);
2003-01-06 02:48:14 +00:00
start = pos+it->second.size();
}
}
2003-01-18 07:26:31 +00:00
Text = txt.c_str();
2003-01-06 02:48:14 +00:00
}
2003-01-09 23:53:14 +00:00
/* Form a string to identify a wchar_t with ASCII. */
CString WcharDisplayText(wchar_t c)
{
CString chr;
chr = ssprintf("U+%4.4x", c);
if(c < 128) chr += ssprintf(" ('%c')", char(c));
return chr;
}
/* Return the last named component of dir:
* a/b/c -> c
* a/b/c/ -> c
*/
CString Basename(CString dir)
{
TrimRight(dir, "/\\");
unsigned pos = dir.find_last_of("/\\");
if(pos != dir.npos)
return dir.substr(pos+1);
return dir;
}
CString Capitalize( CString s )
{
if( s.GetLength()==0 )
return "";
CString s1 = s.Left(1);
s1.MakeUpper();
CString s2 = s.Right( s.GetLength()-1 );
return s1+s2;
}
2003-10-20 03:57:02 +00:00
char char_traits_char_nocase::uptab[256] =
{
'\x00','\x01','\x02','\x03','\x04','\x05','\x06','\x07','\x08','\x09','\x0A','\x0B','\x0C','\x0D','\x0E','\x0F',
'\x10','\x11','\x12','\x13','\x14','\x15','\x16','\x17','\x18','\x19','\x1A','\x1B','\x1C','\x1D','\x1E','\x1F',
'\x20','\x21','\x22','\x23','\x24','\x25','\x26','\x27','\x28','\x29','\x2A','\x2B','\x2C','\x2D','\x2E','\x2F',
'\x30','\x31','\x32','\x33','\x34','\x35','\x36','\x37','\x38','\x39','\x3A','\x3B','\x3C','\x3D','\x3E','\x3F',
'\x40','\x41','\x42','\x43','\x44','\x45','\x46','\x47','\x48','\x49','\x4A','\x4B','\x4C','\x4D','\x4E','\x4F',
'\x50','\x51','\x52','\x53','\x54','\x55','\x56','\x57','\x58','\x59','\x5A','\x5B','\x5C','\x5D','\x5E','\x5F',
'\x60','\x41','\x42','\x43','\x44','\x45','\x46','\x47','\x48','\x49','\x4A','\x4B','\x4C','\x4D','\x4E','\x4F',
'\x50','\x51','\x52','\x53','\x54','\x55','\x56','\x57','\x58','\x59','\x5A','\x7B','\x7C','\x7D','\x7E','\x7F',
'\x80','\x81','\x82','\x83','\x84','\x85','\x86','\x87','\x88','\x89','\x8A','\x8B','\x8C','\x8D','\x8E','\x8F',
'\x90','\x91','\x92','\x93','\x94','\x95','\x96','\x97','\x98','\x99','\x9A','\x9B','\x9C','\x9D','\x9E','\x9F',
'\xA0','\xA1','\xA2','\xA3','\xA4','\xA5','\xA6','\xA7','\xA8','\xA9','\xAA','\xAB','\xAC','\xAD','\xAE','\xAF',
'\xB0','\xB1','\xB2','\xB3','\xB4','\xB5','\xB6','\xB7','\xB8','\xB9','\xBA','\xBB','\xBC','\xBD','\xBE','\xBF',
'\xC0','\xC1','\xC2','\xC3','\xC4','\xC5','\xC6','\xC7','\xC8','\xC9','\xCA','\xCB','\xCC','\xCD','\xCE','\xCF',
'\xD0','\xD1','\xD2','\xD3','\xD4','\xD5','\xD6','\xD7','\xD8','\xD9','\xDA','\xDB','\xDC','\xDD','\xDE','\xDF',
'\xE0','\xE1','\xE2','\xE3','\xE4','\xE5','\xE6','\xE7','\xE8','\xE9','\xEA','\xEB','\xEC','\xED','\xEE','\xEF',
'\xF0','\xF1','\xF2','\xF3','\xF4','\xF5','\xF6','\xF7','\xF8','\xF9','\xFA','\xFB','\xFC','\xFD','\xFE','\xFF',
};