godot/thirdparty/icu4c/common/uinvchar.cpp

// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*
*   Copyright (C) 1999-2010, International Business Machines
*   Corporation and others.  All Rights Reserved.
*
*******************************************************************************
*   file name:  uinvchar.c
*   encoding:   UTF-8
*   tab size:   8 (not used)
*   indentation:2
*
*   created on: 2004sep14
*   created by: Markus W. Scherer
*
*   Functions for handling invariant characters, moved here from putil.c
*   for better modularization.
*/

#include "unicode/utypes.h"
#include "unicode/ustring.h"
#include "udataswp.h"
#include "cstring.h"
#include "cmemory.h"
#include "uassert.h"
#include "uinvchar.h"

/* invariant-character handling --------------------------------------------- */

/*
 * These maps for ASCII to/from EBCDIC map invariant characters (see utypes.h)
 * appropriately for most EBCDIC codepages.
 *
 * They currently also map most other ASCII graphic characters,
 * appropriately for codepages 37 and 1047.
 * Exceptions: The characters for []^ have different codes in 37 & 1047.
 * Both versions are mapped to ASCII.
 *
 *    ASCII 37 1047
 * [     5B BA   AD
 * ]     5D BB   BD
 * ^     5E B0   5F
 *
 * There are no mappings for variant characters from Unicode to EBCDIC.
 *
 * Currently, C0 control codes are also included in these maps.
 * Exceptions: S/390 Open Edition swaps LF and NEL codes compared with other
 * EBCDIC platforms; both codes (15 and 25) are mapped to ASCII LF (0A),
 * but there is no mapping for ASCII LF back to EBCDIC.
 *
 *    ASCII EBCDIC S/390-OE
 * LF    0A     25       15
 * NEL   85     15       25
 *
 * The maps below explicitly exclude the variant
 * control and graphical characters that are in ASCII-based
 * codepages at 0x80 and above.
 * "No mapping" is expressed by mapping to a 00 byte.
 *
 * These tables do not establish a converter or a codepage.
 */

static const uint8_t asciiFromEbcdic[256]= …;

static const uint8_t ebcdicFromAscii[256]= …;

/* Same as asciiFromEbcdic[] except maps all letters to lowercase. */
static const uint8_t lowercaseAsciiFromEbcdic[256]= …;

/*
 * Bit sets indicating which characters of the ASCII repertoire
 * (by ASCII/Unicode code) are "invariant".
 * See utypes.h for more details.
 *
 * As invariant are considered the characters of the ASCII repertoire except
 * for the following:
 * 21  '!' <exclamation mark>
 * 23  '#' <number sign>
 * 24  '$' <dollar sign>
 *
 * 40  '@' <commercial at>
 *
 * 5b  '[' <left bracket>
 * 5c  '\' <backslash>
 * 5d  ']' <right bracket>
 * 5e  '^' <circumflex>
 *
 * 60  '`' <grave accent>
 *
 * 7b  '{' <left brace>
 * 7c  '|' <vertical line>
 * 7d  '}' <right brace>
 * 7e  '~' <tilde>
 */
static const uint32_t invariantChars[4]= …;

/*
 * test unsigned types (or values known to be non-negative) for invariant characters,
 * tests ASCII-family character values
 */
#define UCHAR_IS_INVARIANT(c) …

/* test signed types for invariant characters, adds test for positive values */
#define SCHAR_IS_INVARIANT(c) …

#if U_CHARSET_FAMILY==U_ASCII_FAMILY
#define CHAR_TO_UCHAR(c) …
#define UCHAR_TO_CHAR(c) …
#elif U_CHARSET_FAMILY==U_EBCDIC_FAMILY
#define CHAR_TO_UCHAR …
#define UCHAR_TO_CHAR …
#else
#   error U_CHARSET_FAMILY is not valid
#endif


U_CAPI void U_EXPORT2
u_charsToUChars(const char *cs, char16_t *us, int32_t length) { … }

U_CAPI void U_EXPORT2
u_UCharsToChars(const char16_t *us, char *cs, int32_t length) { … }

U_CAPI UBool U_EXPORT2
uprv_isInvariantString(const char *s, int32_t length) { … }

U_CAPI UBool U_EXPORT2
uprv_isInvariantUString(const char16_t *s, int32_t length) { … }

/* UDataSwapFn implementations used in udataswp.c ------- */

/* convert ASCII to EBCDIC and verify that all characters are invariant */
U_CAPI int32_t U_EXPORT2
uprv_ebcdicFromAscii(const UDataSwapper *ds,
                     const void *inData, int32_t length, void *outData,
                     UErrorCode *pErrorCode) { … }

/* this function only checks and copies ASCII strings without conversion */
U_CFUNC int32_t
uprv_copyAscii(const UDataSwapper *ds,
               const void *inData, int32_t length, void *outData,
               UErrorCode *pErrorCode) { … }

/* convert EBCDIC to ASCII and verify that all characters are invariant */
U_CFUNC int32_t
uprv_asciiFromEbcdic(const UDataSwapper *ds,
                     const void *inData, int32_t length, void *outData,
                     UErrorCode *pErrorCode) { … }

/* this function only checks and copies EBCDIC strings without conversion */
U_CFUNC int32_t
uprv_copyEbcdic(const UDataSwapper *ds,
                const void *inData, int32_t length, void *outData,
                UErrorCode *pErrorCode) { … }

U_CFUNC UBool
uprv_isEbcdicAtSign(char c) { … }

/* compare invariant strings; variant characters compare less than others and unlike each other */
U_CFUNC int32_t
uprv_compareInvAscii(const UDataSwapper *ds,
                     const char *outString, int32_t outLength,
                     const char16_t *localString, int32_t localLength) { … }

U_CFUNC int32_t
uprv_compareInvEbcdic(const UDataSwapper *ds,
                      const char *outString, int32_t outLength,
                      const char16_t *localString, int32_t localLength) { … }

U_CAPI int32_t U_EXPORT2
uprv_compareInvEbcdicAsAscii(const char *s1, const char *s2) { … }

U_CAPI char U_EXPORT2
uprv_ebcdicToAscii(char c) { … }

U_CAPI char U_EXPORT2
uprv_ebcdicToLowercaseAscii(char c) { … }

U_CAPI uint8_t* U_EXPORT2
uprv_aestrncpy(uint8_t *dst, const uint8_t *src, int32_t n)
{ … }

U_CAPI uint8_t* U_EXPORT2
uprv_eastrncpy(uint8_t *dst, const uint8_t *src, int32_t n)
{ … }