indigo. i18n. textcodec

public this( char[]  msg )

Creates a new exception with the message msg.

protected this()

Constructs a new TextCodec and registers it.  This constructor needs fully working mibEnum(), name() and aliases() functions before it may be called.  It is not reentrant.

abstract protected void[] convertFromUtf16( inout  wchar[]  input, void[]  buffer )

Encodes the input, and returns the result. input is replaced by an in-place-slice of trailing bytes that could not be converted because they were incomplete. buffer should be used to avoid memory allocation.  Subclasses of TextCodec must implement this function.  If there are characters in the input that cannot be encoded, a CodecException must be thrown.

abstract protected wchar[] convertToUtf16( inout  void[]  input, wchar[]  buffer )

Decodes the input, and returns the UTF-16 result. input is replaced by an in-place-slice of trailing bytes that could not be converted because they were incomplete. buffer should be used to avoid memory allocation.  Subclasses of TextCodec must implement this function.  If there are invalid bytes in the input, a CodecException must be thrown.  The returned string must be null-terminated (but the null not included in the returned array).

void[] convertFromUtf8( inout  char[]  input, void[]  buffer, inout  wchar[]  utf16Buffer )

Works like convertFromUtf16(), but uses UTF-8 as the input encoding.  This function has a default implementation that converts the input into UTF-16 and calls convertFromUtf16(); derived classes can override it if they know a faster way.

char[] convertToUtf8( inout  void[]  input, char[]  buffer, inout  wchar[]  utf16Buffer )

Works like convertToUtf16(), but uses UTF-8 as the output encoding.  This function has a default implementation that calls convertToUtf16() and converts the output to UTF-8; derived classes can override it if they know a faster way.

static TextCodec codecForMib( int  mib )

Returns the TextCodec that matches the mibEnum value mib.  If there is no such codec, null is returned.

See also codecForName().

TODO Change implementation as soon as DMD problem is fixed.

static TextCodec codecForName( char[]  name )

Returns the TextCodec that matches name.  If there is no such codec, null is returned. name is compared to the codecs’ name and aliases.

See also codecForMib().

FIXME This should be case-insensitive at least, better a fuzzy matching (for example “UTF-8” should also be recognized as “utf8”).

TODO Change implementation as soon as DMD problem is fixed.

final int canEncode( dchar  ch )

Returns true if this codec can fully encode the character ch, false otherwise.

final int canEncode( wchar[]  str )

Returns true if this codec can fully encode the string str, false otherwise.

final int canEncode( char[]  str )

Returns true if this codec can fully encode the string str, false otherwise.

final void[] fromUtf16( wchar[]  str,    void[]  buffer  =  null )

Encodes the string str, which is in UTF-16 format.  If you specify a buffer, it will be used to store the result into to avoid reallocation.

final void[] fromUtf8( char[]  str,    void[]  buffer  =  null )

Encodes the string str, which is in UTF-8 format.  If you specify a buffer, it will be used to store the result into to avoid reallocation.

final wchar[] toUtf16( void[]  str,    wchar[]  buffer  =  null )

Decodes the string str and returns the UTF-16 encoded result.  This function throws an exception if str contains an incomplete encoding at the end.

final char[] toUtf8( void[]  str,    char[]  buffer  =  null )

Decodes the string str and returns the UTF-8 encoded result.  This function throws an exception if str contains an incomplete encoding at the end.

int moreNeeded( void[]  input )

Returns the number of bytes that must be added to input for it to be a complete character according to this encoding.  If an implementation knows the exact number of bytes required then it will return this number; if it only knows the minimum then it will return the negative of this value.  If input is null or zero-length, then the return value corresponds to the minimum length of a character under the encoding.

This is useful for multibyte encodings, to determine if more bytes need to be read from a stream before a character is extracted.  The function should be called only on the beginning or whole of a single character, or on a null or empty string.

size_t maxCharSize()

Returns the maximum length in bytes of a character under this encoding.

The default implementation simply returns 1, and can be used as is for encodings where every character is a single byte.