---------------------------------------------------- ---------------------------- -- XML/Ada - An XML suite for Ada95 -- -- -- -- Copyright (C) 2001-2012, AdaCore -- -- -- -- This library is free software; you can redistribute it and/or modify it -- -- under terms of the GNU General Public License as published by the Free -- -- Software Foundation; either version 3, or (at your option) any later -- -- version. This library is distributed in the hope that it will be useful, -- -- but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHAN- -- -- TABILITY or FITNESS FOR A PARTICULAR PURPOSE. -- -- -- -- -- -- -- -- -- -- -- -- You should have received a copy of the GNU General Public License and -- -- a copy of the GCC Runtime Library Exception along with this program; -- -- see the files COPYING3 and COPYING.RUNTIME respectively. If not, see -- -- . -- -- -- ------------------------------------------------------------------------------ -- This package implements a basic 8bit encoding. -- Only code points from 16#00# to 16#FF# can be encoded in such strings. -- These are the standard Ada Strings. -- -- However, then can be used to read files that contain accented characters, -- in combination with Unicode.CCS.Iso_8859_1 for instance with Unicode.CES.Utf32; with Unicode.CCS; with Unchecked_Deallocation; package Unicode.CES.Basic_8bit is ----------- -- Types -- ----------- subtype basic_8bit_string is String; type basic_8bit_string_access is access basic_8bit_string; -- A heigh bit string, undefined byte-order ------------------------------------------- -- Conversion to and from byte sequences -- ------------------------------------------- procedure Encode (Char : unicode_char; Output : in out byte_sequence; Index : in out Natural); -- Return the byte sequence representing Char in the 8bit character -- encoding form -- Invalid_Encoding is raised if Char can not be converted. procedure Read (Str : basic_8bit_string; Index : in out Positive; Char : out unicode_char); -- Return the character starting at location Index in Str function Width (Char : unicode_char) return Natural; -- Return the number of bytes occupied by the 8bit representation of Char function Length (Str : basic_8bit_string) return Natural; -- Return the number of characters in Str ------------------------------------------ -- Conversion to and from 8bit-encoding -- ------------------------------------------ function From_Utf32 (Str : Unicode.CES.Utf32.utf32_le_string) return basic_8bit_string; -- Return a new string, from a utf32-encoded string. function To_Utf32 (Str : basic_8bit_string) return Unicode.CES.Utf32.utf32_le_string; -- Return a new utf32-encoded string, from a standard Ada string. --------------------------------------------- -- Byte order and character set conversion -- --------------------------------------------- function To_Unicode_LE (Str : basic_8bit_string; Cs : Unicode.CCS.character_set := Unicode.CCS.Unicode_Character_Set; Order : byte_order := Default_Byte_Order) return basic_8bit_string; -- Convert Str to a Unicode string, assuming it contains code points from -- the character set CS. -- Byte-order is irrelevant for 8bit strings, but is kept for interface -- compatibility with other similar functions function To_CS (Str : basic_8bit_string; Cs : Unicode.CCS.character_set := Unicode.CCS.Unicode_Character_Set; Order : byte_order := Default_Byte_Order) return basic_8bit_string; -- Convert Str to the character set Cs, assuming it contains Unicode -- characters. --------------------- -- Encoding Scheme -- --------------------- Basic_8bit_Encoding : constant encoding_scheme := (BOM => unknown, Read => Read'access, Width => Width'access, Encode => encode_function'(Encode'access), Length => Length'access); ------------------ -- Deallocation -- ------------------ procedure Free is new Unchecked_Deallocation (basic_8bit_string, basic_8bit_string_access); private pragma inline (Width); end Unicode.CES.Basic_8bit;