/src/icu/icu4c/source/common/ucnv_err.cpp

Source
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
 *****************************************************************************
 *
 *   Copyright (C) 1998-2016, International Business Machines
 *   Corporation and others.  All Rights Reserved.
 *
 *****************************************************************************
 *
 *  ucnv_err.c
 *  Implements error behaviour functions called by T_UConverter_{from,to}Unicode
 *
 *
*   Change history:
*
*   06/29/2000  helena      Major rewrite of the callback APIs.
*/

#include "unicode/utypes.h"

#if !UCONFIG_NO_CONVERSION

#include "unicode/ucnv_err.h"
#include "unicode/ucnv_cb.h"
#include "ucnv_cnv.h"
#include "cmemory.h"
#include "unicode/ucnv.h"
#include "ustrfmt.h"

#define VALUE_STRING_LENGTH 48
/*Magic # 32 = 4(number of char in value string) * 8(max number of bytes per char for any converter) */
#define UNICODE_PERCENT_SIGN_CODEPOINT  0x0025
#define UNICODE_U_CODEPOINT             0x0055
#define UNICODE_X_CODEPOINT             0x0058
#define UNICODE_RS_CODEPOINT            0x005C
#define UNICODE_U_LOW_CODEPOINT         0x0075
#define UNICODE_X_LOW_CODEPOINT         0x0078
#define UNICODE_AMP_CODEPOINT           0x0026
#define UNICODE_HASH_CODEPOINT          0x0023
#define UNICODE_SEMICOLON_CODEPOINT     0x003B
#define UNICODE_PLUS_CODEPOINT          0x002B
#define UNICODE_LEFT_CURLY_CODEPOINT    0x007B
#define UNICODE_RIGHT_CURLY_CODEPOINT   0x007D
#define UNICODE_SPACE_CODEPOINT         0x0020
#define UCNV_PRV_ESCAPE_ICU         0
#define UCNV_PRV_ESCAPE_C           'C'
#define UCNV_PRV_ESCAPE_XML_DEC     'D'
#define UCNV_PRV_ESCAPE_XML_HEX     'X'
#define UCNV_PRV_ESCAPE_JAVA        'J'
#define UCNV_PRV_ESCAPE_UNICODE     'U'
#define UCNV_PRV_ESCAPE_CSS2        'S'
#define UCNV_PRV_STOP_ON_ILLEGAL    'i'

/*
 * IS_DEFAULT_IGNORABLE_CODE_POINT
 * This is to check if a code point has the default ignorable unicode property.
 * As such, this list needs to be updated if the ignorable code point list ever
 * changes.
 * To avoid dependency on other code, this list is hard coded here.
 * When an ignorable code point is found and is unmappable, the default callbacks
 * will ignore them.
 * For a list of the default ignorable code points, use this link:
 * https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5B%3ADI%3A%5D&abb=on&g=&i=
 *
 * This list should be sync with the one in CharsetCallback.java
 */
#define IS_DEFAULT_IGNORABLE_CODE_POINT(c) ( \
    (c == 0x00AD) || \
    (c == 0x034F) || \
    (c == 0x061C) || \
    (c == 0x115F) || \
    (c == 0x1160) || \
    (0x17B4 <= c && c <= 0x17B5) || \
    (0x180B <= c && c <= 0x180F) || \
    (0x200B <= c && c <= 0x200F) || \
    (0x202A <= c && c <= 0x202E) || \
    (0x2060 <= c && c <= 0x206F) || \
    (c == 0x3164) || \
    (0xFE00 <= c && c <= 0xFE0F) || \
    (c == 0xFEFF) || \
    (c == 0xFFA0) || \
    (0xFFF0 <= c && c <= 0xFFF8) || \
    (0x1BCA0 <= c && c <= 0x1BCA3) || \
    (0x1D173 <= c && c <= 0x1D17A) || \
    (0xE0000 <= c && c <= 0xE0FFF))


/*Function Pointer STOPS at the ILLEGAL_SEQUENCE */
U_CAPI void    U_EXPORT2
UCNV_FROM_U_CALLBACK_STOP (
                  const void *context,
                  UConverterFromUnicodeArgs *fromUArgs,
                  const char16_t* codeUnits,
                  int32_t length,
                  UChar32 codePoint,
                  UConverterCallbackReason reason,
                  UErrorCode * err)
{
    (void)context;
    (void)fromUArgs;
    (void)codeUnits;
    (void)length;
    if (reason == UCNV_UNASSIGNED && IS_DEFAULT_IGNORABLE_CODE_POINT(codePoint))
    {
        /*
         * Skip if the codepoint has unicode property of default ignorable.
         */
        *err = U_ZERO_ERROR;
    }
    /* the caller must have set the error code accordingly */
}


/*Function Pointer STOPS at the ILLEGAL_SEQUENCE */
U_CAPI void    U_EXPORT2
UCNV_TO_U_CALLBACK_STOP (
                   const void *context,
                   UConverterToUnicodeArgs *toUArgs,
                   const char* codePoints,
                   int32_t length,
                   UConverterCallbackReason reason,
                   UErrorCode * err)
{
    /* the caller must have set the error code accordingly */
    (void)context; (void)toUArgs; (void)codePoints; (void)length; (void)reason; (void)err;
}

U_CAPI void    U_EXPORT2
UCNV_FROM_U_CALLBACK_SKIP (                  
                  const void *context,
                  UConverterFromUnicodeArgs *fromUArgs,
                  const char16_t* codeUnits,
                  int32_t length,
                  UChar32 codePoint,
                  UConverterCallbackReason reason,
                  UErrorCode * err)
{
    (void)fromUArgs;
    (void)codeUnits;
    (void)length;
    if (reason <= UCNV_IRREGULAR)
    {
        if (reason == UCNV_UNASSIGNED && IS_DEFAULT_IGNORABLE_CODE_POINT(codePoint))
        {
            /*
             * Skip if the codepoint has unicode property of default ignorable.
             */
            *err = U_ZERO_ERROR;
        }
        else if (context == nullptr || (*((char*)context) == UCNV_PRV_STOP_ON_ILLEGAL && reason == UCNV_UNASSIGNED))
        {
            *err = U_ZERO_ERROR;
        }
        /* else the caller must have set the error code accordingly. */
    }
    /* else ignore the reset, close and clone calls. */
}

U_CAPI void    U_EXPORT2
UCNV_FROM_U_CALLBACK_SUBSTITUTE (
                  const void *context,
                  UConverterFromUnicodeArgs *fromArgs,
                  const char16_t* codeUnits,
                  int32_t length,
                  UChar32 codePoint,
                  UConverterCallbackReason reason,
                  UErrorCode * err)
{
    (void)codeUnits;
    (void)length;
    if (reason <= UCNV_IRREGULAR)
    {
        if (reason == UCNV_UNASSIGNED && IS_DEFAULT_IGNORABLE_CODE_POINT(codePoint))
        {
            /*
             * Skip if the codepoint has unicode property of default ignorable.
             */
            *err = U_ZERO_ERROR;
        }
        else if (context == nullptr || (*((char*)context) == UCNV_PRV_STOP_ON_ILLEGAL && reason == UCNV_UNASSIGNED))
        {
            *err = U_ZERO_ERROR;
            ucnv_cbFromUWriteSub(fromArgs, 0, err);
        }
        /* else the caller must have set the error code accordingly. */
    }
    /* else ignore the reset, close and clone calls. */
}

/*uses uprv_itou to get a unicode escape sequence of the offensive sequence,
 *uses a clean copy (resetted) of the converter, to convert that unicode
 *escape sequence to the target codepage (if conversion failure happens then
 *we revert to substituting with subchar)
 */
U_CAPI void    U_EXPORT2
UCNV_FROM_U_CALLBACK_ESCAPE (
                         const void *context,
                         UConverterFromUnicodeArgs *fromArgs,
                         const char16_t *codeUnits,
                         int32_t length,
                         UChar32 codePoint,
                         UConverterCallbackReason reason,
                         UErrorCode * err)
{

  char16_t valueString[VALUE_STRING_LENGTH];
  int32_t valueStringLength = 0;
  int32_t i = 0;

  const char16_t *myValueSource = nullptr;
  UErrorCode err2 = U_ZERO_ERROR;
  UConverterFromUCallback original = nullptr;
  const void *originalContext;

  UConverterFromUCallback ignoredCallback = nullptr;
  const void *ignoredContext;
  
  if (reason > UCNV_IRREGULAR)
  {
      return;
  }
  else if (reason == UCNV_UNASSIGNED && IS_DEFAULT_IGNORABLE_CODE_POINT(codePoint))
  {
      /*
       * Skip if the codepoint has unicode property of default ignorable.
       */
      *err = U_ZERO_ERROR;
      return;
  }

  ucnv_setFromUCallBack (fromArgs->converter,
                     (UConverterFromUCallback) UCNV_FROM_U_CALLBACK_SUBSTITUTE,
                     nullptr,
                     &original,
                     &originalContext,
                     &err2);
  
  if (U_FAILURE (err2))
  {
    *err = err2;
    return;
  } 
  if(context==nullptr)
  { 
      while (i < length)
      {
        valueString[valueStringLength++] = (char16_t) UNICODE_PERCENT_SIGN_CODEPOINT;  /* adding % */
        valueString[valueStringLength++] = (char16_t) UNICODE_U_CODEPOINT; /* adding U */
        valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[i++], 16, 4);
      }
  }
  else
  {
      switch(*((char*)context))
      {
      case UCNV_PRV_ESCAPE_JAVA:
          while (i < length)
          {
              valueString[valueStringLength++] = (char16_t) UNICODE_RS_CODEPOINT;    /* adding \ */
              valueString[valueStringLength++] = (char16_t) UNICODE_U_LOW_CODEPOINT; /* adding u */
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[i++], 16, 4);
          }
          break;

      case UCNV_PRV_ESCAPE_C:
          valueString[valueStringLength++] = (char16_t) UNICODE_RS_CODEPOINT;    /* adding \ */

          if(length==2){
              valueString[valueStringLength++] = (char16_t) UNICODE_U_CODEPOINT; /* adding U */
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 16, 8);

          }
          else{
              valueString[valueStringLength++] = (char16_t) UNICODE_U_LOW_CODEPOINT; /* adding u */
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[0], 16, 4);
          }
          break;

      case UCNV_PRV_ESCAPE_XML_DEC:

          valueString[valueStringLength++] = (char16_t) UNICODE_AMP_CODEPOINT;   /* adding & */
          valueString[valueStringLength++] = (char16_t) UNICODE_HASH_CODEPOINT;  /* adding # */
          if(length==2){
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 10, 0);
          }
          else{
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[0], 10, 0);
          }
          valueString[valueStringLength++] = (char16_t) UNICODE_SEMICOLON_CODEPOINT; /* adding ; */
          break;

      case UCNV_PRV_ESCAPE_XML_HEX:

          valueString[valueStringLength++] = (char16_t) UNICODE_AMP_CODEPOINT;   /* adding & */
          valueString[valueStringLength++] = (char16_t) UNICODE_HASH_CODEPOINT;  /* adding # */
          valueString[valueStringLength++] = (char16_t) UNICODE_X_LOW_CODEPOINT; /* adding x */
          if(length==2){
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 16, 0);
          }
          else{
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[0], 16, 0);
          }
          valueString[valueStringLength++] = (char16_t) UNICODE_SEMICOLON_CODEPOINT; /* adding ; */
          break;

      case UCNV_PRV_ESCAPE_UNICODE:
          valueString[valueStringLength++] = (char16_t) UNICODE_LEFT_CURLY_CODEPOINT;    /* adding { */
          valueString[valueStringLength++] = (char16_t) UNICODE_U_CODEPOINT;    /* adding U */
          valueString[valueStringLength++] = (char16_t) UNICODE_PLUS_CODEPOINT; /* adding + */
          if (length == 2) {
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 16, 4);
          } else {
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[0], 16, 4);
          }
          valueString[valueStringLength++] = (char16_t) UNICODE_RIGHT_CURLY_CODEPOINT;    /* adding } */
          break;

      case UCNV_PRV_ESCAPE_CSS2:
          valueString[valueStringLength++] = (char16_t) UNICODE_RS_CODEPOINT;    /* adding \ */
          valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 16, 0);
          /* Always add space character, because the next character might be whitespace,
             which would erroneously be considered the termination of the escape sequence. */
          valueString[valueStringLength++] = (char16_t) UNICODE_SPACE_CODEPOINT;
          break;

      default:
          while (i < length)
          {
              valueString[valueStringLength++] = (char16_t) UNICODE_PERCENT_SIGN_CODEPOINT;  /* adding % */
              valueString[valueStringLength++] = (char16_t) UNICODE_U_CODEPOINT;             /* adding U */
              valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[i++], 16, 4);
          }
      }
  }  
  myValueSource = valueString;

  /* reset the error */
  *err = U_ZERO_ERROR;

  ucnv_cbFromUWriteUChars(fromArgs, &myValueSource, myValueSource+valueStringLength, 0, err);

  ucnv_setFromUCallBack (fromArgs->converter,
                         original,
                         originalContext,
                         &ignoredCallback,
                         &ignoredContext,
                         &err2);
  if (U_FAILURE (err2))
  {
      *err = err2;
      return;
  }
}



U_CAPI void  U_EXPORT2
UCNV_TO_U_CALLBACK_SKIP (
                 const void *context,
                 UConverterToUnicodeArgs *toArgs,
                 const char* codeUnits,
                 int32_t length,
                 UConverterCallbackReason reason,
                 UErrorCode * err)
{
    (void)toArgs;
    (void)codeUnits;
    (void)length;
    if (reason <= UCNV_IRREGULAR)
    {
        if (context == nullptr || (*((char*)context) == UCNV_PRV_STOP_ON_ILLEGAL && reason == UCNV_UNASSIGNED))
        {
            *err = U_ZERO_ERROR;
        }
        /* else the caller must have set the error code accordingly. */
    }
    /* else ignore the reset, close and clone calls. */
}

U_CAPI void    U_EXPORT2
UCNV_TO_U_CALLBACK_SUBSTITUTE (
                 const void *context,
                 UConverterToUnicodeArgs *toArgs,
                 const char* codeUnits,
                 int32_t length,
                 UConverterCallbackReason reason,
                 UErrorCode * err)
{
    (void)codeUnits;
    (void)length;
    if (reason <= UCNV_IRREGULAR)
    {
        if (context == nullptr || (*((char*)context) == UCNV_PRV_STOP_ON_ILLEGAL && reason == UCNV_UNASSIGNED))
        {
            *err = U_ZERO_ERROR;
            ucnv_cbToUWriteSub(toArgs,0,err);
        }
        /* else the caller must have set the error code accordingly. */
    }
    /* else ignore the reset, close and clone calls. */
}

/*uses uprv_itou to get a unicode escape sequence of the offensive sequence,
 *and uses that as the substitution sequence
 */
U_CAPI void   U_EXPORT2
UCNV_TO_U_CALLBACK_ESCAPE (
                 const void *context,
                 UConverterToUnicodeArgs *toArgs,
                 const char* codeUnits,
                 int32_t length,
                 UConverterCallbackReason reason,
                 UErrorCode * err)
{
    char16_t uniValueString[VALUE_STRING_LENGTH];
    int32_t valueStringLength = 0;
    int32_t i = 0;

    if (reason > UCNV_IRREGULAR)
    {
        return;
    }

    if(context==nullptr)
    {    
        while (i < length)
        {
            uniValueString[valueStringLength++] = (char16_t) UNICODE_PERCENT_SIGN_CODEPOINT; /* adding % */
            uniValueString[valueStringLength++] = (char16_t) UNICODE_X_CODEPOINT;    /* adding X */
            valueStringLength += uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t) codeUnits[i++], 16, 2);
        }
    }
    else
    {
        switch(*((char*)context))
        {
        case UCNV_PRV_ESCAPE_XML_DEC:
            while (i < length)
            {
                uniValueString[valueStringLength++] = (char16_t) UNICODE_AMP_CODEPOINT;   /* adding & */
                uniValueString[valueStringLength++] = (char16_t) UNICODE_HASH_CODEPOINT;  /* adding # */
                valueStringLength += uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t)codeUnits[i++], 10, 0);
                uniValueString[valueStringLength++] = (char16_t) UNICODE_SEMICOLON_CODEPOINT; /* adding ; */
            }
            break;

        case UCNV_PRV_ESCAPE_XML_HEX:
            while (i < length)
            {
                uniValueString[valueStringLength++] = (char16_t) UNICODE_AMP_CODEPOINT;   /* adding & */
                uniValueString[valueStringLength++] = (char16_t) UNICODE_HASH_CODEPOINT;  /* adding # */
                uniValueString[valueStringLength++] = (char16_t) UNICODE_X_LOW_CODEPOINT; /* adding x */
                valueStringLength += uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t)codeUnits[i++], 16, 0);
                uniValueString[valueStringLength++] = (char16_t) UNICODE_SEMICOLON_CODEPOINT; /* adding ; */
            }
            break;
        case UCNV_PRV_ESCAPE_C:
            while (i < length)
            {
                uniValueString[valueStringLength++] = (char16_t) UNICODE_RS_CODEPOINT;    /* adding \ */
                uniValueString[valueStringLength++] = (char16_t) UNICODE_X_LOW_CODEPOINT; /* adding x */
                valueStringLength += uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t)codeUnits[i++], 16, 2);
            }
            break;
        default:
            while (i < length)
            {
                uniValueString[valueStringLength++] = (char16_t) UNICODE_PERCENT_SIGN_CODEPOINT; /* adding % */
                uniValueString[valueStringLength++] = (char16_t) UNICODE_X_CODEPOINT;    /* adding X */
                uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t) codeUnits[i++], 16, 2);
                valueStringLength += 2;
            }
        }
    }
    /* reset the error */
    *err = U_ZERO_ERROR;

    ucnv_cbToUWriteUChars(toArgs, uniValueString, valueStringLength, 0, err);
}

#endif

Coverage Report

Created: 2025-12-07 06:36

Line	Count	Source
1		// © 2016 and later: Unicode, Inc. and others.
2		// License & terms of use: http://www.unicode.org/copyright.html
3		/*
4		*****************************************************************************
5		*
6		* Copyright (C) 1998-2016, International Business Machines
7		* Corporation and others. All Rights Reserved.
8		*
9		*****************************************************************************
10		*
11		* ucnv_err.c
12		* Implements error behaviour functions called by T_UConverter_{from,to}Unicode
13		*
14		*
15		* Change history:
16		*
17		* 06/29/2000 helena Major rewrite of the callback APIs.
18		*/
19
20		#include "unicode/utypes.h"
21
22		#if !UCONFIG_NO_CONVERSION
23
24		#include "unicode/ucnv_err.h"
25		#include "unicode/ucnv_cb.h"
26		#include "ucnv_cnv.h"
27		#include "cmemory.h"
28		#include "unicode/ucnv.h"
29		#include "ustrfmt.h"
30
31	0	#define VALUE_STRING_LENGTH 48
32		/Magic # 32 = 4(number of char in value string) 8(max number of bytes per char for any converter) */
33	0	#define UNICODE_PERCENT_SIGN_CODEPOINT 0x0025
34	0	#define UNICODE_U_CODEPOINT 0x0055
35	0	#define UNICODE_X_CODEPOINT 0x0058
36	0	#define UNICODE_RS_CODEPOINT 0x005C
37	0	#define UNICODE_U_LOW_CODEPOINT 0x0075
38	0	#define UNICODE_X_LOW_CODEPOINT 0x0078
39	0	#define UNICODE_AMP_CODEPOINT 0x0026
40	0	#define UNICODE_HASH_CODEPOINT 0x0023
41	0	#define UNICODE_SEMICOLON_CODEPOINT 0x003B
42	0	#define UNICODE_PLUS_CODEPOINT 0x002B
43	0	#define UNICODE_LEFT_CURLY_CODEPOINT 0x007B
44	0	#define UNICODE_RIGHT_CURLY_CODEPOINT 0x007D
45	0	#define UNICODE_SPACE_CODEPOINT 0x0020
46		#define UCNV_PRV_ESCAPE_ICU 0
47	0	#define UCNV_PRV_ESCAPE_C 'C'
48	0	#define UCNV_PRV_ESCAPE_XML_DEC 'D'
49	0	#define UCNV_PRV_ESCAPE_XML_HEX 'X'
50	0	#define UCNV_PRV_ESCAPE_JAVA 'J'
51	0	#define UCNV_PRV_ESCAPE_UNICODE 'U'
52	0	#define UCNV_PRV_ESCAPE_CSS2 'S'
53	0	#define UCNV_PRV_STOP_ON_ILLEGAL 'i'
54
55		/*
56		* IS_DEFAULT_IGNORABLE_CODE_POINT
57		* This is to check if a code point has the default ignorable unicode property.
58		* As such, this list needs to be updated if the ignorable code point list ever
59		* changes.
60		* To avoid dependency on other code, this list is hard coded here.
61		* When an ignorable code point is found and is unmappable, the default callbacks
62		* will ignore them.
63		* For a list of the default ignorable code points, use this link:
64		* https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5B%3ADI%3A%5D&abb=on&g=&i=
65		*
66		* This list should be sync with the one in CharsetCallback.java
67		*/
68	38.6M	#define IS_DEFAULT_IGNORABLE_CODE_POINT(c) ( \
69	38.6M	(c == 0x00AD) \|\| \
70	38.6M	(c == 0x034F) \|\| \
71	38.6M	(c == 0x061C) \|\| \
72	38.6M	(c == 0x115F) \|\| \
73	38.6M	(c == 0x1160) \|\| \
74	38.6M	(0x17B4 <= c && c <= 0x17B5) \|\| \
75	38.6M	(0x180B <= c && c <= 0x180F) \|\| \
76	38.6M	(0x200B <= c && c <= 0x200F) \|\| \
77	38.6M	(0x202A <= c && c <= 0x202E) \|\| \
78	38.6M	(0x2060 <= c && c <= 0x206F) \|\| \
79	38.6M	(c == 0x3164) \|\| \
80	38.6M	(0xFE00 <= c && c <= 0xFE0F) \|\| \
81	38.6M	(c == 0xFEFF) \|\| \
82	38.6M	(c == 0xFFA0) \|\| \
83	38.6M	(0xFFF0 <= c && c <= 0xFFF8) \|\| \
84	38.6M	(0x1BCA0 <= c && c <= 0x1BCA3) \|\| \
85	38.6M	(0x1D173 <= c && c <= 0x1D17A) \|\| \
86	38.6M	(0xE0000 <= c && c <= 0xE0FFF))
87
88
89		/Function Pointer STOPS at the ILLEGAL_SEQUENCE /
90		U_CAPI void U_EXPORT2
91		UCNV_FROM_U_CALLBACK_STOP (
92		const void *context,
93		UConverterFromUnicodeArgs *fromUArgs,
94		const char16_t* codeUnits,
95		int32_t length,
96		UChar32 codePoint,
97		UConverterCallbackReason reason,
98		UErrorCode * err)
99	0	{
100	0	(void)context;
101	0	(void)fromUArgs;
102	0	(void)codeUnits;
103	0	(void)length;
104	0	if (reason == UCNV_UNASSIGNED && IS_DEFAULT_IGNORABLE_CODE_POINT(codePoint))
105	0	{
106		/*
107		* Skip if the codepoint has unicode property of default ignorable.
108		*/
109	0	*err = U_ZERO_ERROR;
110	0	}
111		/* the caller must have set the error code accordingly */
112	0	}
113
114
115		/Function Pointer STOPS at the ILLEGAL_SEQUENCE /
116		U_CAPI void U_EXPORT2
117		UCNV_TO_U_CALLBACK_STOP (
118		const void *context,
119		UConverterToUnicodeArgs *toUArgs,
120		const char* codePoints,
121		int32_t length,
122		UConverterCallbackReason reason,
123		UErrorCode * err)
124	0	{
125		/* the caller must have set the error code accordingly */
126	0	(void)context; (void)toUArgs; (void)codePoints; (void)length; (void)reason; (void)err;
127	0	}
128
129		U_CAPI void U_EXPORT2
130		UCNV_FROM_U_CALLBACK_SKIP (
131		const void *context,
132		UConverterFromUnicodeArgs *fromUArgs,
133		const char16_t* codeUnits,
134		int32_t length,
135		UChar32 codePoint,
136		UConverterCallbackReason reason,
137		UErrorCode * err)
138	0	{
139	0	(void)fromUArgs;
140	0	(void)codeUnits;
141	0	(void)length;
142	0	if (reason <= UCNV_IRREGULAR)
143	0	{
144	0	if (reason == UCNV_UNASSIGNED && IS_DEFAULT_IGNORABLE_CODE_POINT(codePoint))
145	0	{
146		/*
147		* Skip if the codepoint has unicode property of default ignorable.
148		*/
149	0	*err = U_ZERO_ERROR;
150	0	}
151	0	else if (context == nullptr \|\| (((char)context) == UCNV_PRV_STOP_ON_ILLEGAL && reason == UCNV_UNASSIGNED))
152	0	{
153	0	*err = U_ZERO_ERROR;
154	0	}
155		/* else the caller must have set the error code accordingly. */
156	0	}
157		/* else ignore the reset, close and clone calls. */
158	0	}
159
160		U_CAPI void U_EXPORT2
161		UCNV_FROM_U_CALLBACK_SUBSTITUTE (
162		const void *context,
163		UConverterFromUnicodeArgs *fromArgs,
164		const char16_t* codeUnits,
165		int32_t length,
166		UChar32 codePoint,
167		UConverterCallbackReason reason,
168		UErrorCode * err)
169	64.2M	{
170	64.2M	(void)codeUnits;
171	64.2M	(void)length;
172	64.2M	if (reason <= UCNV_IRREGULAR)
173	64.2M	{
174	64.2M	if (reason == UCNV_UNASSIGNED && IS_DEFAULT_IGNORABLE_CODE_POINT(codePoint))
175	66.8k	{
176		/*
177		* Skip if the codepoint has unicode property of default ignorable.
178		*/
179	66.8k	*err = U_ZERO_ERROR;
180	66.8k	}
181	64.1M	else if (context == nullptr \|\| (((char)context) == UCNV_PRV_STOP_ON_ILLEGAL && reason == UCNV_UNASSIGNED))
182	64.1M	{
183	64.1M	*err = U_ZERO_ERROR;
184	64.1M	ucnv_cbFromUWriteSub(fromArgs, 0, err);
185	64.1M	}
186		/* else the caller must have set the error code accordingly. */
187	64.2M	}
188		/* else ignore the reset, close and clone calls. */
189	64.2M	}
190
191		/*uses uprv_itou to get a unicode escape sequence of the offensive sequence,
192		*uses a clean copy (resetted) of the converter, to convert that unicode
193		*escape sequence to the target codepage (if conversion failure happens then
194		*we revert to substituting with subchar)
195		*/
196		U_CAPI void U_EXPORT2
197		UCNV_FROM_U_CALLBACK_ESCAPE (
198		const void *context,
199		UConverterFromUnicodeArgs *fromArgs,
200		const char16_t *codeUnits,
201		int32_t length,
202		UChar32 codePoint,
203		UConverterCallbackReason reason,
204		UErrorCode * err)
205	0	{
206
207	0	char16_t valueString[VALUE_STRING_LENGTH];
208	0	int32_t valueStringLength = 0;
209	0	int32_t i = 0;
210
211	0	const char16_t *myValueSource = nullptr;
212	0	UErrorCode err2 = U_ZERO_ERROR;
213	0	UConverterFromUCallback original = nullptr;
214	0	const void *originalContext;
215
216	0	UConverterFromUCallback ignoredCallback = nullptr;
217	0	const void *ignoredContext;
218
219	0	if (reason > UCNV_IRREGULAR)
220	0	{
221	0	return;
222	0	}
223	0	else if (reason == UCNV_UNASSIGNED && IS_DEFAULT_IGNORABLE_CODE_POINT(codePoint))
224	0	{
225		/*
226		* Skip if the codepoint has unicode property of default ignorable.
227		*/
228	0	*err = U_ZERO_ERROR;
229	0	return;
230	0	}
231
232	0	ucnv_setFromUCallBack (fromArgs->converter,
233	0	(UConverterFromUCallback) UCNV_FROM_U_CALLBACK_SUBSTITUTE,
234	0	nullptr,
235	0	&original,
236	0	&originalContext,
237	0	&err2);
238
239	0	if (U_FAILURE (err2))
240	0	{
241	0	*err = err2;
242	0	return;
243	0	}
244	0	if(context==nullptr)
245	0	{
246	0	while (i < length)
247	0	{
248	0	valueString[valueStringLength++] = (char16_t) UNICODE_PERCENT_SIGN_CODEPOINT; /* adding % */
249	0	valueString[valueStringLength++] = (char16_t) UNICODE_U_CODEPOINT; /* adding U */
250	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[i++], 16, 4);
251	0	}
252	0	}
253	0	else
254	0	{
255	0	switch(((char)context))
256	0	{
257	0	case UCNV_PRV_ESCAPE_JAVA:
258	0	while (i < length)
259	0	{
260	0	valueString[valueStringLength++] = (char16_t) UNICODE_RS_CODEPOINT; /* adding \ */
261	0	valueString[valueStringLength++] = (char16_t) UNICODE_U_LOW_CODEPOINT; /* adding u */
262	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[i++], 16, 4);
263	0	}
264	0	break;
265
266	0	case UCNV_PRV_ESCAPE_C:
267	0	valueString[valueStringLength++] = (char16_t) UNICODE_RS_CODEPOINT; /* adding \ */
268
269	0	if(length==2){
270	0	valueString[valueStringLength++] = (char16_t) UNICODE_U_CODEPOINT; /* adding U */
271	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 16, 8);
272
273	0	}
274	0	else{
275	0	valueString[valueStringLength++] = (char16_t) UNICODE_U_LOW_CODEPOINT; /* adding u */
276	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[0], 16, 4);
277	0	}
278	0	break;
279
280	0	case UCNV_PRV_ESCAPE_XML_DEC:
281
282	0	valueString[valueStringLength++] = (char16_t) UNICODE_AMP_CODEPOINT; /* adding & */
283	0	valueString[valueStringLength++] = (char16_t) UNICODE_HASH_CODEPOINT; /* adding # */
284	0	if(length==2){
285	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 10, 0);
286	0	}
287	0	else{
288	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[0], 10, 0);
289	0	}
290	0	valueString[valueStringLength++] = (char16_t) UNICODE_SEMICOLON_CODEPOINT; /* adding ; */
291	0	break;
292
293	0	case UCNV_PRV_ESCAPE_XML_HEX:
294
295	0	valueString[valueStringLength++] = (char16_t) UNICODE_AMP_CODEPOINT; /* adding & */
296	0	valueString[valueStringLength++] = (char16_t) UNICODE_HASH_CODEPOINT; /* adding # */
297	0	valueString[valueStringLength++] = (char16_t) UNICODE_X_LOW_CODEPOINT; /* adding x */
298	0	if(length==2){
299	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 16, 0);
300	0	}
301	0	else{
302	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[0], 16, 0);
303	0	}
304	0	valueString[valueStringLength++] = (char16_t) UNICODE_SEMICOLON_CODEPOINT; /* adding ; */
305	0	break;
306
307	0	case UCNV_PRV_ESCAPE_UNICODE:
308	0	valueString[valueStringLength++] = (char16_t) UNICODE_LEFT_CURLY_CODEPOINT; /* adding { */
309	0	valueString[valueStringLength++] = (char16_t) UNICODE_U_CODEPOINT; /* adding U */
310	0	valueString[valueStringLength++] = (char16_t) UNICODE_PLUS_CODEPOINT; /* adding + */
311	0	if (length == 2) {
312	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 16, 4);
313	0	} else {
314	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[0], 16, 4);
315	0	}
316	0	valueString[valueStringLength++] = (char16_t) UNICODE_RIGHT_CURLY_CODEPOINT; /* adding } */
317	0	break;
318
319	0	case UCNV_PRV_ESCAPE_CSS2:
320	0	valueString[valueStringLength++] = (char16_t) UNICODE_RS_CODEPOINT; /* adding \ */
321	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, codePoint, 16, 0);
322		/* Always add space character, because the next character might be whitespace,
323		which would erroneously be considered the termination of the escape sequence. */
324	0	valueString[valueStringLength++] = (char16_t) UNICODE_SPACE_CODEPOINT;
325	0	break;
326
327	0	default:
328	0	while (i < length)
329	0	{
330	0	valueString[valueStringLength++] = (char16_t) UNICODE_PERCENT_SIGN_CODEPOINT; /* adding % */
331	0	valueString[valueStringLength++] = (char16_t) UNICODE_U_CODEPOINT; /* adding U */
332	0	valueStringLength += uprv_itou (valueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint16_t)codeUnits[i++], 16, 4);
333	0	}
334	0	}
335	0	}
336	0	myValueSource = valueString;
337
338		/* reset the error */
339	0	*err = U_ZERO_ERROR;
340
341	0	ucnv_cbFromUWriteUChars(fromArgs, &myValueSource, myValueSource+valueStringLength, 0, err);
342
343	0	ucnv_setFromUCallBack (fromArgs->converter,
344	0	original,
345	0	originalContext,
346	0	&ignoredCallback,
347	0	&ignoredContext,
348	0	&err2);
349	0	if (U_FAILURE (err2))
350	0	{
351	0	*err = err2;
352	0	return;
353	0	}
354	0	}
355
356
357
358		U_CAPI void U_EXPORT2
359		UCNV_TO_U_CALLBACK_SKIP (
360		const void *context,
361		UConverterToUnicodeArgs *toArgs,
362		const char* codeUnits,
363		int32_t length,
364		UConverterCallbackReason reason,
365		UErrorCode * err)
366	0	{
367	0	(void)toArgs;
368	0	(void)codeUnits;
369	0	(void)length;
370	0	if (reason <= UCNV_IRREGULAR)
371	0	{
372	0	if (context == nullptr \|\| (((char)context) == UCNV_PRV_STOP_ON_ILLEGAL && reason == UCNV_UNASSIGNED))
373	0	{
374	0	*err = U_ZERO_ERROR;
375	0	}
376		/* else the caller must have set the error code accordingly. */
377	0	}
378		/* else ignore the reset, close and clone calls. */
379	0	}
380
381		U_CAPI void U_EXPORT2
382		UCNV_TO_U_CALLBACK_SUBSTITUTE (
383		const void *context,
384		UConverterToUnicodeArgs *toArgs,
385		const char* codeUnits,
386		int32_t length,
387		UConverterCallbackReason reason,
388		UErrorCode * err)
389	4.01M	{
390	4.01M	(void)codeUnits;
391	4.01M	(void)length;
392	4.01M	if (reason <= UCNV_IRREGULAR)
393	4.01M	{
394	4.01M	if (context == nullptr \|\| (((char)context) == UCNV_PRV_STOP_ON_ILLEGAL && reason == UCNV_UNASSIGNED))
395	4.01M	{
396	4.01M	*err = U_ZERO_ERROR;
397	4.01M	ucnv_cbToUWriteSub(toArgs,0,err);
398	4.01M	}
399		/* else the caller must have set the error code accordingly. */
400	4.01M	}
401		/* else ignore the reset, close and clone calls. */
402	4.01M	}
403
404		/*uses uprv_itou to get a unicode escape sequence of the offensive sequence,
405		*and uses that as the substitution sequence
406		*/
407		U_CAPI void U_EXPORT2
408		UCNV_TO_U_CALLBACK_ESCAPE (
409		const void *context,
410		UConverterToUnicodeArgs *toArgs,
411		const char* codeUnits,
412		int32_t length,
413		UConverterCallbackReason reason,
414		UErrorCode * err)
415	0	{
416	0	char16_t uniValueString[VALUE_STRING_LENGTH];
417	0	int32_t valueStringLength = 0;
418	0	int32_t i = 0;
419
420	0	if (reason > UCNV_IRREGULAR)
421	0	{
422	0	return;
423	0	}
424
425	0	if(context==nullptr)
426	0	{
427	0	while (i < length)
428	0	{
429	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_PERCENT_SIGN_CODEPOINT; /* adding % */
430	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_X_CODEPOINT; /* adding X */
431	0	valueStringLength += uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t) codeUnits[i++], 16, 2);
432	0	}
433	0	}
434	0	else
435	0	{
436	0	switch(((char)context))
437	0	{
438	0	case UCNV_PRV_ESCAPE_XML_DEC:
439	0	while (i < length)
440	0	{
441	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_AMP_CODEPOINT; /* adding & */
442	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_HASH_CODEPOINT; /* adding # */
443	0	valueStringLength += uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t)codeUnits[i++], 10, 0);
444	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_SEMICOLON_CODEPOINT; /* adding ; */
445	0	}
446	0	break;
447
448	0	case UCNV_PRV_ESCAPE_XML_HEX:
449	0	while (i < length)
450	0	{
451	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_AMP_CODEPOINT; /* adding & */
452	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_HASH_CODEPOINT; /* adding # */
453	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_X_LOW_CODEPOINT; /* adding x */
454	0	valueStringLength += uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t)codeUnits[i++], 16, 0);
455	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_SEMICOLON_CODEPOINT; /* adding ; */
456	0	}
457	0	break;
458	0	case UCNV_PRV_ESCAPE_C:
459	0	while (i < length)
460	0	{
461	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_RS_CODEPOINT; /* adding \ */
462	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_X_LOW_CODEPOINT; /* adding x */
463	0	valueStringLength += uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t)codeUnits[i++], 16, 2);
464	0	}
465	0	break;
466	0	default:
467	0	while (i < length)
468	0	{
469	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_PERCENT_SIGN_CODEPOINT; /* adding % */
470	0	uniValueString[valueStringLength++] = (char16_t) UNICODE_X_CODEPOINT; /* adding X */
471	0	uprv_itou (uniValueString + valueStringLength, VALUE_STRING_LENGTH - valueStringLength, (uint8_t) codeUnits[i++], 16, 2);
472	0	valueStringLength += 2;
473	0	}
474	0	}
475	0	}
476		/* reset the error */
477	0	*err = U_ZERO_ERROR;
478
479	0	ucnv_cbToUWriteUChars(toArgs, uniValueString, valueStringLength, 0, err);
480	0	}
481
482		#endif