/* Copyright (C) 2015-2020, Dirk Krause SPDX-License-Identifier: BSD-3-Clause */ /* WARNING: This file was generated by the dkct program (see http://dktools.sourceforge.net/ for details). Changes you make here will be lost if dkct is run again! You should modify the original source and run dkct on it. Original source: dk4utf16.ctr */ #ifndef DK4UTF16_H_INCLUDED /** Avoid multiple inclusions. */ #define DK4UTF16_H_INCLUDED 1 #line 10 "dk4utf16.ctr" /** @file Encode and decode 32 bit characters in UTF-16 encoding. CRT on Windows: Not used. The character to encode to UTF-16 must be in range U+00000000 to U+0010FFFF. (the hexadecimal numbers range defined by UNICODE). The character must not be in range U+0000D800 to U+0000DFFF which is reserved for surrogate pairs. Characters up to U+0000FFFF are converted to 16 bits and passed "as is". For characters above, we substract U+00010000 and split the result in two 10-bit blocks. We prepend bits 110110 to the higher 10 bits, this results in the "high surrogate". We prepend bits 110111 to the lower 10 bits, this results in the "low surrogate". When decoding we have to check for a byte order marker (BOM) at start of stream. The character U+0000FEFF as first character in the stream can be used to mark encoding and/or byte order. If we read an UTF-16 encoded stream and have 0xFEFF as first character we can leave byte order as is. If the first character is 0xFFFE we have to change byte order in each word. */ #ifndef DK4CONF_H_INCLUDED #if DK4_BUILDING_DKTOOLS4 #include "dk4conf.h" #else #include #endif #endif #ifndef DK4CONF_H_INCLUDED #if DK4_BUILDING_DKTOOLS4 #include "dk4types.h" #else #include #endif #endif #ifndef DK4ERROR_H_INCLUDED #if DK4_BUILDING_DKTOOLS4 #include "dk4error.h" #else #include #endif #endif #ifndef DK4DEC_H_INCLUDED #if DK4_BUILDING_DKTOOLS4 #include "dk4edstm.h" #else #include #endif #endif /** Decode UTF-16 encoded data. */ typedef struct { dk4_c32_t uc; /**< Unicode character to produce. */ int state; /**< Decoder state. */ } dk4_utf16_decoder_t; /** Decode UTF-16 encoded bytes. */ typedef struct { dk4_utf16_decoder_t dec; /**< Decoder UTF-16 to 32-bit char. */ int msb; /**< Flag: Most significant byte comes first. */ int state; /**< Decoder state. */ dk4_c16_t saved; /**< Partial input character saved. */ } dk4_utf16_byte_decoder_t; #ifdef __cplusplus extern "C" { #endif /** Initialize the UTF-16 decoder. @param ptr Decoder to initialize. */ void dk4utf16_init(dk4_utf16_decoder_t *ptr); /** Add one word to the decoder. @param ptr Decoder to use. @param word Word to add. @return Operation result, one from: DK4_EDSTM_ACCEPT if the byte was added to the next output without completing output, DK4_EDSTM_FINISHED, if the byte was added to the next output completing the output, or DK4_EDSTM_ERROR on error. */ int dk4utf16_add(dk4_utf16_decoder_t *ptr, dk4_c16_t word); /** Retrieve a completed Unicode character from the decoder. @param ptr Decoder to obtain character from. */ dk4_c32_t dk4utf16_get(dk4_utf16_decoder_t const *ptr); /** Check whether the decoder is empty. @param ptr Decoder to check. @return 1 on success, 0 on error. */ int dk4utf16_is_empty(dk4_utf16_decoder_t const *ptr); /** UTF-16 encode UNICODE character. @param rb Result buffer, should be at least 3 dk4_c16_t. @param pszrb Size of the result buffer, in: available, out: used. @param c Character to encode. @param erp Error report, may be NULL. @return 1 on success, 0 on error. Error codes: - DK4_E_INVALID_ARGUMENTS
if rb or pszrb is NULL, - DK4_E_BUFFER_TOO_SMALL
if the buffer size is less than 3, - DK4_E_SYNTAX
if c is out of the encodable range. */ int dk4utf16_encode(dk4_c16_t *rb, size_t *pszrb, dk4_c32_t c, dk4_er_t *erp); /** Initialize the UTF-16 decoder. @param ptr Decoder to initialize. @param msb Flag: Most significant byte comes first. */ void dk4utf16_byte_init(dk4_utf16_byte_decoder_t *ptr, int msb); /** Add one byte to the decoder. @param ptr Decoder to use. @param byte Byte to add. @return Operation result, one from: DK4_EDSTM_ACCEPT if the byte was added to the next output without completing output, DK4_EDSTM_FINISHED, if the byte was added to the next output completing the output, or DK4_EDSTM_ERROR on error. */ int dk4utf16_byte_add(dk4_utf16_byte_decoder_t *ptr, unsigned char byte); /** Retrieve a completed Unicode character from the decoder. @param ptr Decoder to obtain character from. */ dk4_c32_t dk4utf16_byte_get(dk4_utf16_byte_decoder_t *ptr); /** Check whether a decoder is empty. @param ptr Decoder to check. @return 1 on success, 0 on error. */ int dk4utf16_byte_is_empty(dk4_utf16_byte_decoder_t *ptr); /** UTF-16 encode UNICODE character, write to bytes. @param rb Result buffer, should be at least 5 bytes. @param pszrb Size of the result buffer (number of bytes), in: available, out: used. @param c Character to encode. @param msb Flag: Write most significant byte first. @param erp Error report, may be NULL. @return 1 on success, 0 on error. Error codes: - DK4_E_INVALID_ARGUMENTS
if rb or pszrb is NULL, - DK4_E_BUFFER_TOO_SMALL
if the buffer size is less than 3, - DK4_E_SYNTAX
if c is out of the encodable range. */ int dk4utf16_byte_encode( unsigned char *rb, size_t *pszrb, dk4_c32_t c, int msb, dk4_er_t *erp ); #ifdef __cplusplus } #endif #endif