Mercurial > projects > dwt-win
changeset 177:6bfa783ac0b4
Sync with dwt-linux
author | Frank Benoit <benoit@tionex.de> |
---|---|
date | Fri, 07 Mar 2008 02:24:13 +0100 |
parents | e0ea5a76da07 |
children | e186b01aa62e |
files | dwt/dwthelper/utils.d |
diffstat | 1 files changed, 184 insertions(+), 3 deletions(-) [+] |
line wrap: on
line diff
--- a/dwt/dwthelper/utils.d Fri Mar 07 02:24:03 2008 +0100 +++ b/dwt/dwthelper/utils.d Fri Mar 07 02:24:13 2008 +0100 @@ -13,8 +13,11 @@ import tango.text.Util; import tango.text.Unicode; import tango.text.convert.Utf; +import tango.core.Exception; import tango.stdc.stdlib : exit; +import tango.util.log.Trace; + void implMissing( char[] file, uint line ){ Stderr.formatln( "implementation missing in file {} line {}", file, line ); Stderr.formatln( "exiting ..." ); @@ -48,11 +51,168 @@ alias ArrayWrapperT!(char) ArrayWrapperString; alias ArrayWrapperT!(char[]) ArrayWrapperString2; -dchar getFirstCodepoint( char[] str ){ - foreach( dchar d; str ){ - return d; +int codepointIndexToIndex( char[] str, int cpIndex ){ + int cps = cpIndex; + int res = 0; + while( cps > 0 ){ + cps--; + if( str[res] < 0x80 ){ + res+=1; + } + else if( str[res] < 0xE0 ){ + res+=2; + } + else if( str[res] & 0xF0 ){ + res+=3; + } + else{ + res+=4; + } + } + return res; +} +int indexToCodepointIndex( char[] str, int index ){ + int i = 0; + int res = 0; + while( i < index ){ + if( str[i] < 0x80 ){ + i+=1; + } + else if( str[i] < 0xE0 ){ + i+=2; + } + else if( str[i] & 0xF0 ){ + i+=3; + } + else{ + i+=4; + } + res++; + } + return res; +} + +char[] firstCodePointStr( char[] str, out int consumed ){ + dchar[1] buf; + uint ate; + dchar[] res = str.toString32( buf, &ate ); + consumed = ate; + return str[ 0 .. ate ]; +} + +dchar firstCodePoint( char[] str ){ + int dummy; + return firstCodePoint( str, dummy ); +} +dchar firstCodePoint( char[] str, out int consumed ){ + dchar[1] buf; + uint ate; + dchar[] res = str.toString32( buf, &ate ); + consumed = ate; + if( ate is 0 || res.length is 0 ){ + Trace.formatln( "dwthelper.utils {}: str.length={} str={:X2}", __LINE__, str.length, cast(ubyte[])str ); } + assert( ate > 0 ); + assert( res.length is 1 ); + return res[0]; } + +char[] dcharToString( dchar key ){ + dchar[1] buf; + buf[0] = key; + return tango.text.convert.Utf.toString( buf ); +} + +int codepointCount( char[] str ){ + scope dchar[] buf = new dchar[]( str.length ); + uint ate; + dchar[] res = tango.text.convert.Utf.toString32( str, buf, &ate ); + assert( ate is str.length ); + return res.length; +} + +alias tango.text.convert.Utf.toString16 toString16; +alias tango.text.convert.Utf.toString toString; + +int getRelativeCodePointOffset( char[] str, int startIndex, int searchRelCp ){ + int ignore; + int i = startIndex; + if( searchRelCp > 0 ){ + while( searchRelCp !is 0 ){ + + if( ( i < str.length ) + && ( str[i] & 0x80 ) is 0x00 ) + { + i+=1; + } + else if( ( i+1 < str.length ) + && (( str[i+1] & 0xC0 ) is 0x80 ) + && (( str[i ] & 0xE0 ) is 0xC0 )) + { + i+=2; + } + else if( ( i+2 < str.length ) + && (( str[i+2] & 0xC0 ) is 0x80 ) + && (( str[i+1] & 0xC0 ) is 0x80 ) + && (( str[i ] & 0xF0 ) is 0xE0 )) + { + i+=3; + } + else if(( i+3 < str.length ) + && (( str[i+3] & 0xC0 ) is 0x80 ) + && (( str[i+2] & 0xC0 ) is 0x80 ) + && (( str[i+1] & 0xC0 ) is 0x80 ) + && (( str[i ] & 0xF8 ) is 0xF0 )) + { + i+=4; + } + else{ + tango.text.convert.Utf.onUnicodeError( "invalid utf8 input", i ); + } + searchRelCp--; + } + } + else if( searchRelCp < 0 ){ + while( searchRelCp !is 0 ){ + do{ + i--; + if( i < 0 ){ + Trace.formatln( "dwthelper.utils getRelativeCodePointOffset {}: str={}, startIndex={}, searchRelCp={}", __LINE__, str, startIndex, searchRelCp ); + tango.text.convert.Utf.onUnicodeError( "invalid utf8 input", i ); + } + } while(( str[i] & 0xC0 ) is 0x80 ); + searchRelCp++; + } + } + return i - startIndex; +} +dchar getRelativeCodePoint( char[] str, int startIndex, int searchRelCp, out int relIndex ){ + relIndex = getRelativeCodePointOffset( str, startIndex, searchRelCp ); + int ignore; + return firstCodePoint( str[ startIndex+relIndex .. $ ], ignore ); +} + +int utf8AdjustOffset( char[] str, int offset ){ + if( str.length <= offset || offset <= 0 ){ + return offset; + } + while(( str[offset] & 0xC0 ) is 0x80 ){ + offset--; + } + return offset; +} + +dchar CharacterFirstToLower( char[] str ){ + int consumed; + return CharacterFirstToLower( str, consumed ); +} +dchar CharacterFirstToLower( char[] str, out int consumed ){ + dchar[1] buf; + buf[0] = firstCodePoint( str, consumed ); + dchar[] r = tango.text.Unicode.toLower( buf ); + return r[0]; +} + dchar CharacterToLower( dchar c ){ dchar[] r = tango.text.Unicode.toLower( [c] ); return r[0]; @@ -90,6 +250,15 @@ return res; } +public int lastIndexOf(char[] str, char ch){ + return lastIndexOf( str, ch, str.length ); +} +public int lastIndexOf(char[] str, char ch, int formIndex){ + int res = tango.text.Util.locatePrior( str, ch, formIndex ); + if( res is str.length ) res = -1; + return res; +} + public char[] replace( char[] str, char from, char to ){ return tango.text.Util.replace( str.dup, from, to ); } @@ -129,6 +298,14 @@ return src[ $-pattern.length .. $ ] == pattern; } +public bool equals( char[] src, char[] other ){ + return src == other; +} + +public bool equalsIgnoreCase( char[] src, char[] other ){ + return tango.text.Unicode.toFold(src) == tango.text.Unicode.toFold(other); +} + public bool startsWith( char[] src, char[] pattern ){ if( src.length < pattern.length ){ return false; @@ -139,6 +316,10 @@ return tango.text.Unicode.toLower( src ); } +public hash_t toHash( char[] src ){ + return typeid(char[]).getHash(&src); +} + static char[] toHex(uint value, bool prefix = true, int radix = 8){ return tango.text.convert.Integer.toString( value,