#include "iso2022jp.h"
#include "iso646fi.h"
#include "iso88591.h"
+#include "iso885915.h"
#include "jisx0201.h"
#include "jisx0208.h"
#include "utf8.h"
using namespace std;
namespace Msp {
+namespace Codecs {
-/**
-Determines whether the given string can be successfully decoded with this
-codec. Note that this function returning true does not guarantee that the
-string was actually encoded with this codec. In particular, many 8-bit
-encodings are indistinguishable.
-*/
-bool StringCodec::detect(const string &str) const
+bool Codec::detect(const string &str) const
{
- Decoder *dec=create_decoder();
- bool result=true;
+ Decoder *dec = create_decoder();
+ bool result = true;
try
{
for(string::const_iterator i=str.begin(); i!=str.end(); )
dec->decode_char(str, i);
- dec->sync();
}
catch(const CodecError &)
{
- result=false;
+ result = false;
}
delete dec;
return result;
}
-void StringCodec::Encoder::error(const string &msg)
+void Codec::Encoder::encode(const ustring &str, string &buf)
+{
+ for(ustring::const_iterator i=str.begin(); i!=str.end(); ++i)
+ encode_char(*i, buf);
+}
+
+string Codec::Encoder::encode(const ustring &str)
+{
+ string buf;
+ encode(str, buf);
+ sync(buf);
+ return buf;
+}
+
+void Codec::Encoder::error(UnicodeChar ch, string &buf, const string &msg)
{
- switch(err_mode_)
+ switch(err_mode)
{
- case IGNORE_ERRORS: break;
- case REPLACE_ERRORS: append_replacement(); break;
- default: throw CodecError(msg);
+ case TRANSLITERATE:
+ transliterate(ch, buf);
+ case IGNORE_ERRORS:
+ break;
+ default:
+ throw CodecError(msg);
}
}
-void StringCodec::Decoder::error(const string &msg)
+
+void Codec::Decoder::decode(const string &str, ustring &buf)
{
- switch(err_mode_)
+ for(string::const_iterator i=str.begin(); i!=str.end();)
{
- case IGNORE_ERRORS: break;
- case REPLACE_ERRORS: append(0xFFFD); break;
- default: throw CodecError(msg);
+ UnicodeChar c = decode_char(str, i);
+ if(c!=-1)
+ buf += c;
}
}
-/**
-Creates a codec for the given encoding. The caller is responsible for deleting
-the codec when it's no longer needed.
-*/
-StringCodec *create_codec(const string &n)
+ustring Codec::Decoder::decode(const string &str)
+{
+ ustring buf;
+ decode(str, buf);
+ return buf;
+}
+
+UnicodeChar Codec::Decoder::error(const string &msg)
+{
+ switch(err_mode)
+ {
+ case TRANSLITERATE:
+ return 0xFFFE;
+ case IGNORE_ERRORS:
+ return -1;
+ default:
+ throw CodecError(msg);
+ }
+}
+
+Codec *create_codec(const string &n)
{
string name;
for(string::const_iterator i=n.begin(); i!=n.end(); ++i)
{
if(isupper(*i))
- name+=tolower(*i);
+ name += tolower(*i);
else if(islower(*i) || isdigit(*i))
- name+=*i;
+ name += *i;
}
if(name=="ascii") return new Ascii;
if(name=="iso2022jp") return new Iso2022Jp;
if(name=="iso646fi") return new Iso646Fi;
if(name=="iso88591" || name=="latin1") return new Iso88591;
+ if(name=="iso885915" || name=="latin9") return new Iso885915;
if(name=="jisx0201") return new JisX0201;
if(name=="jisx0208") return new JisX0208;
if(name=="utf8") return new Utf8;
- if(name=="windows1252") return new Windows1252;
+ if(name=="windows1252" || name=="cp1252") return new Windows1252;
throw InvalidParameterValue("Unknown string codec");
}
+Codec *detect_codec(const string &str)
+{
+ bool is_utf8 = true;
+ bool is_ascii = true;
+ bool is_latin1 = true;
+ unsigned utf8_mb = 0;
+
+ for(string::const_iterator i=str.begin(); i!=str.end(); ++i)
+ {
+ unsigned char c = *i;
+ if(c&0x80)
+ {
+ is_ascii = false;
+ if((c&0xC0)==0x80)
+ {
+ if((c&0xE0)==0x80)
+ is_latin1 = false;
+ if(utf8_mb)
+ --utf8_mb;
+ else
+ is_utf8 = false;
+ }
+ else if((c&0xC0)==0xC0)
+ {
+ if(utf8_mb)
+ {
+ is_utf8 = false;
+ utf8_mb = 0;
+ }
+ else
+ {
+ for(utf8_mb=1; (c>>(6-utf8_mb))&1; ++utf8_mb) ;
+ }
+ }
+ }
+ else if(utf8_mb)
+ {
+ is_utf8 = false;
+ utf8_mb = 0;
+ }
+ }
+
+ if(is_ascii)
+ return new Ascii;
+ else if(is_utf8)
+ return new Utf8;
+ else if(is_latin1)
+ return new Iso88591;
+ else
+ return new Windows1252;
+}
+
+} // namespace Codecs
} // namespace Msp