ucnv_err.h File Reference

C API: UConverter predefined error callbacks. More...

#include "unicode/utypes.h"

Go to the source code of this file.

Data Structures
struct	UConverterFromUnicodeArgs
	The structure for the fromUnicode callback function parameter. More...
struct	UConverterToUnicodeArgs
	The structure for the toUnicode callback function parameter. More...

Macros
#define	UCNV_SUB_STOP_ON_ILLEGAL   "i"
	FROM_U, TO_U context options for sub callback.
#define	UCNV_SKIP_STOP_ON_ILLEGAL   "i"
	FROM_U, TO_U context options for skip callback.
#define	UCNV_ESCAPE_ICU   NULL
	FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to ICU (UXXXX)
#define	UCNV_ESCAPE_JAVA   "J"
	FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to JAVA (\uXXXX)
#define	UCNV_ESCAPE_C   "C"
	FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to C (\uXXXX \UXXXXXXXX) TO_U_CALLBACK_ESCAPE option to escape the character value according to C (\xXXXX)
#define	UCNV_ESCAPE_XML_DEC   "D"
	FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to XML Decimal escape (&#DDDD;) TO_U_CALLBACK_ESCAPE context option to escape the character value according to XML Decimal escape (&#DDDD;).
#define	UCNV_ESCAPE_XML_HEX   "X"
	FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to XML Hex escape (&#xXXXX;) TO_U_CALLBACK_ESCAPE context option to escape the character value according to XML Hex escape (&#xXXXX;).
#define	UCNV_ESCAPE_UNICODE   "U"
	FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to Unicode (U+XXXXX)
#define	UCNV_ESCAPE_CSS2   "S"
	FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to CSS2 conventions (\HH..H<space>, that is, a backslash, 1..6 hex digits, and a space)

Typedefs
typedef struct UConverter	UConverter

Enumerations
enum	UConverterCallbackReason {   UCNV_UNASSIGNED = 0 , UCNV_ILLEGAL = 1 , UCNV_IRREGULAR = 2 , UCNV_RESET = 3 ,   UCNV_CLOSE = 4 , UCNV_CLONE = 5 }
	The process condition code to be used with the callbacks. More...

Functions
U_CAPI void	UCNV_FROM_U_CALLBACK_STOP (const void *context, UConverterFromUnicodeArgs *fromUArgs, const UChar *codeUnits, int32_t length, UChar32 codePoint, UConverterCallbackReason reason, UErrorCode *err)
	DO NOT CALL THIS FUNCTION DIRECTLY! This From Unicode callback STOPS at the ILLEGAL_SEQUENCE, returning the error code back to the caller immediately.
U_CAPI void	UCNV_TO_U_CALLBACK_STOP (const void *context, UConverterToUnicodeArgs *toUArgs, const char *codeUnits, int32_t length, UConverterCallbackReason reason, UErrorCode *err)
	DO NOT CALL THIS FUNCTION DIRECTLY! This To Unicode callback STOPS at the ILLEGAL_SEQUENCE, returning the error code back to the caller immediately.
U_CAPI void	UCNV_FROM_U_CALLBACK_SKIP (const void *context, UConverterFromUnicodeArgs *fromUArgs, const UChar *codeUnits, int32_t length, UChar32 codePoint, UConverterCallbackReason reason, UErrorCode *err)
	DO NOT CALL THIS FUNCTION DIRECTLY! This From Unicode callback skips any ILLEGAL_SEQUENCE, or skips only UNASSIGNED_SEQUENCE depending on the context parameter simply ignoring those characters.
U_CAPI void	UCNV_FROM_U_CALLBACK_SUBSTITUTE (const void *context, UConverterFromUnicodeArgs *fromUArgs, const UChar *codeUnits, int32_t length, UChar32 codePoint, UConverterCallbackReason reason, UErrorCode *err)
	DO NOT CALL THIS FUNCTION DIRECTLY! This From Unicode callback will Substitute the ILLEGAL SEQUENCE, or UNASSIGNED_SEQUENCE depending on context parameter, with the current substitution string for the converter.
U_CAPI void	UCNV_FROM_U_CALLBACK_ESCAPE (const void *context, UConverterFromUnicodeArgs *fromUArgs, const UChar *codeUnits, int32_t length, UChar32 codePoint, UConverterCallbackReason reason, UErrorCode *err)
	DO NOT CALL THIS FUNCTION DIRECTLY! This From Unicode callback will Substitute the ILLEGAL SEQUENCE with the hexadecimal representation of the illegal codepoints.
U_CAPI void	UCNV_TO_U_CALLBACK_SKIP (const void *context, UConverterToUnicodeArgs *toUArgs, const char *codeUnits, int32_t length, UConverterCallbackReason reason, UErrorCode *err)
	DO NOT CALL THIS FUNCTION DIRECTLY! This To Unicode callback skips any ILLEGAL_SEQUENCE, or skips only UNASSIGNED_SEQUENCE depending on the context parameter simply ignoring those characters.
U_CAPI void	UCNV_TO_U_CALLBACK_SUBSTITUTE (const void *context, UConverterToUnicodeArgs *toUArgs, const char *codeUnits, int32_t length, UConverterCallbackReason reason, UErrorCode *err)
	DO NOT CALL THIS FUNCTION DIRECTLY! This To Unicode callback will Substitute the ILLEGAL SEQUENCE,or UNASSIGNED_SEQUENCE depending on context parameter, with the Unicode substitution character, U+FFFD.
U_CAPI void	UCNV_TO_U_CALLBACK_ESCAPE (const void *context, UConverterToUnicodeArgs *toUArgs, const char *codeUnits, int32_t length, UConverterCallbackReason reason, UErrorCode *err)
	DO NOT CALL THIS FUNCTION DIRECTLY! This To Unicode callback will Substitute the ILLEGAL SEQUENCE with the hexadecimal representation of the illegal bytes (in the format XNN, e.g.

Detailed Description

C API: UConverter predefined error callbacks.

Error Behaviour Functions

Defines some error behaviour functions called by ucnv_{from,to}Unicode These are provided as part of ICU and many are stable, but they can also be considered only as an example of what can be done with callbacks. You may of course write your own.

If you want to write your own, you may also find the functions from ucnv_cb.h useful when writing your own callbacks.

These functions, although public, should NEVER be called directly. They should be used as parameters to the ucnv_setFromUCallback and ucnv_setToUCallback functions, to set the behaviour of a converter when it encounters ILLEGAL/UNMAPPED/INVALID sequences.

usage example: 'STOP' doesn't need any context, but newContext could be set to something other than 'NULL' if needed. The available contexts in this header can modify the default behavior of the callback.

UErrorCode err = U_ZERO_ERROR;
UConverter *myConverter = ucnv_open("ibm-949", &err);
const void *oldContext;
UConverterFromUCallback oldAction;
 
 
if (U_SUCCESS(err))
{
    ucnv_setFromUCallBack(myConverter,
                     UCNV_FROM_U_CALLBACK_STOP,
                     NULL,
                     &oldAction,
                     &oldContext,
                     &status);
}

The code above tells "myConverter" to stop when it encounters an ILLEGAL/TRUNCATED/INVALID sequences when it is used to convert from Unicode -> Codepage. The behavior from Codepage to Unicode is not changed, and ucnv_setToUCallBack would need to be called in order to change that behavior too.

Here is an example with a context:

UErrorCode err = U_ZERO_ERROR;
UConverter *myConverter = ucnv_open("ibm-949", &err);
const void *oldContext;
UConverterFromUCallback oldAction;
 
 
if (U_SUCCESS(err))
{
    ucnv_setToUCallBack(myConverter,
                     UCNV_TO_U_CALLBACK_SUBSTITUTE,
                     UCNV_SUB_STOP_ON_ILLEGAL,
                     &oldAction,
                     &oldContext,
                     &status);
}

The code above tells "myConverter" to stop when it encounters an ILLEGAL/TRUNCATED/INVALID sequences when it is used to convert from Codepage -> Unicode. Any unmapped and legal characters will be substituted to be the default substitution character.

Definition in file ucnv_err.h.

Macro Definition Documentation

UCNV_ESCAPE_C

#define UCNV_ESCAPE_C   "C"

FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to C (\uXXXX \UXXXXXXXX) TO_U_CALLBACK_ESCAPE option to escape the character value according to C (\xXXXX)

Stable:

ICU 2.0

Definition at line 125 of file ucnv_err.h.

UCNV_ESCAPE_CSS2

#define UCNV_ESCAPE_CSS2   "S"

FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to CSS2 conventions (\HH..H<space>, that is, a backslash, 1..6 hex digits, and a space)

Stable:

ICU 4.0

Definition at line 149 of file ucnv_err.h.

UCNV_ESCAPE_ICU

#define UCNV_ESCAPE_ICU   NULL

FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to ICU (UXXXX)

Stable:

ICU 2.0

Definition at line 114 of file ucnv_err.h.

UCNV_ESCAPE_JAVA

#define UCNV_ESCAPE_JAVA   "J"

FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to JAVA (\uXXXX)

Stable:

ICU 2.0

Definition at line 119 of file ucnv_err.h.

UCNV_ESCAPE_UNICODE

#define UCNV_ESCAPE_UNICODE   "U"

FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to Unicode (U+XXXXX)

Stable:

ICU 2.0

Definition at line 142 of file ucnv_err.h.

UCNV_ESCAPE_XML_DEC

#define UCNV_ESCAPE_XML_DEC   "D"

FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to XML Decimal escape (&#DDDD;) TO_U_CALLBACK_ESCAPE context option to escape the character value according to XML Decimal escape (&#DDDD;).

Stable:

ICU 2.0

Definition at line 131 of file ucnv_err.h.

UCNV_ESCAPE_XML_HEX

#define UCNV_ESCAPE_XML_HEX   "X"

FROM_U_CALLBACK_ESCAPE context option to escape the code unit according to XML Hex escape (&#xXXXX;) TO_U_CALLBACK_ESCAPE context option to escape the character value according to XML Hex escape (&#xXXXX;).

Stable:

ICU 2.0

Definition at line 137 of file ucnv_err.h.

UCNV_SKIP_STOP_ON_ILLEGAL

#define UCNV_SKIP_STOP_ON_ILLEGAL   "i"

FROM_U, TO_U context options for skip callback.

Stable:

ICU 2.0

Definition at line 108 of file ucnv_err.h.

UCNV_SUB_STOP_ON_ILLEGAL

#define UCNV_SUB_STOP_ON_ILLEGAL   "i"

FROM_U, TO_U context options for sub callback.

Stable:

ICU 2.0

Definition at line 102 of file ucnv_err.h.

Typedef Documentation

UConverter

typedef struct UConverter UConverter

Stable:

ICU 2.0

Definition at line 96 of file ucnv_err.h.

Enumeration Type Documentation

UConverterCallbackReason

enum UConverterCallbackReason

The process condition code to be used with the callbacks.

Codes which are greater than UCNV_IRREGULAR should be passed on to any chained callbacks.

Stable:

ICU 2.0

Enumerator
UCNV_UNASSIGNED	The code point is unassigned. The error code U_INVALID_CHAR_FOUND will be set.
UCNV_ILLEGAL	The code point is illegal. For example, \x81\x2E is illegal in SJIS because \x2E is not a valid trail byte for the \x81 lead byte. Also, starting with Unicode 3.0.1, non-shortest byte sequences in UTF-8 (like \xC1\xA1 instead of \x61 for U+0061) are also illegal, not just irregular. The error code U_ILLEGAL_CHAR_FOUND will be set.
UCNV_IRREGULAR	The codepoint is not a regular sequence in the encoding. For example, \xED\xA0\x80..\xED\xBF\xBF are irregular UTF-8 byte sequences for single surrogate code points. The error code U_INVALID_CHAR_FOUND will be set.
UCNV_RESET	The callback is called with this reason when a 'reset' has occurred. Callback should reset all state.
UCNV_CLOSE	Called when the converter is closed. The callback should release any allocated memory.
UCNV_CLONE	Called when ucnv_safeClone() is called on the converter. the pointer available as the 'context' is an alias to the original converters' context pointer. If the context must be owned by the new converter, the callback must clone the data and call ucnv_setFromUCallback (or setToUCallback) with the correct pointer. Stable: ICU 2.2

Definition at line 157 of file ucnv_err.h.

Function Documentation

UCNV_FROM_U_CALLBACK_ESCAPE()

U_CAPI void UCNV_FROM_U_CALLBACK_ESCAPE	(	const void *	context,
		UConverterFromUnicodeArgs *	fromUArgs,
		const UChar *	codeUnits,
		int32_t	length,
		UChar32	codePoint,
		UConverterCallbackReason	reason,
		UErrorCode *	err
	)

DO NOT CALL THIS FUNCTION DIRECTLY! This From Unicode callback will Substitute the ILLEGAL SEQUENCE with the hexadecimal representation of the illegal codepoints.

Parameters

context	The function currently recognizes the callback options: UCNV_ESCAPE_ICU: Substitutes the ILLEGAL SEQUENCE with the hexadecimal representation in the format UXXXX, e.g. "%uFFFE%u00AC%uC8FE"). In the Event the converter doesn't support the characters {%,U}[A-F][0-9], it will substitute the illegal sequence with the substitution characters. Note that codeUnit(32bit int eg: unit of a surrogate pair) is represented as UD84DUDC56 UCNV_ESCAPE_JAVA: Substitutes the ILLEGAL SEQUENCE with the hexadecimal representation in the format \uXXXX, e.g. "\\uFFFE\\u00AC\\uC8FE"). In the Event the converter doesn't support the characters {\,u}[A-F][0-9], it will substitute the illegal sequence with the substitution characters. Note that codeUnit(32bit int eg: unit of a surrogate pair) is represented as \uD84D\uDC56 UCNV_ESCAPE_C: Substitutes the ILLEGAL SEQUENCE with the hexadecimal representation in the format \uXXXX, e.g. "\\uFFFE\\u00AC\\uC8FE"). In the Event the converter doesn't support the characters {\,u,U}[A-F][0-9], it will substitute the illegal sequence with the substitution characters. Note that codeUnit(32bit int eg: unit of a surrogate pair) is represented as \U00023456 UCNV_ESCAPE_XML_DEC: Substitutes the ILLEGAL SEQUENCE with the decimal representation in the format &#DDDDDDDD;, e.g. "¬죾"). In the Event the converter doesn't support the characters {&,#}[0-9], it will substitute the illegal sequence with the substitution characters. Note that codeUnit(32bit int eg: unit of a surrogate pair) is represented as 𣑖 and Zero padding is ignored. UCNV_ESCAPE_XML_HEX:Substitutes the ILLEGAL SEQUENCE with the decimal representation in the format &#xXXXX; e.g. "¬죾"). In the Event the converter doesn't support the characters {&,#,x}[0-9], it will substitute the illegal sequence with the substitution characters. Note that codeUnit(32bit int eg: unit of a surrogate pair) is represented as 𣑖
fromUArgs	Information about the conversion in progress
codeUnits	Points to 'length' UChars of the concerned Unicode sequence
length	Size (in bytes) of the concerned codepage sequence
codePoint	Single UChar32 (UTF-32) containing the concerend Unicode codepoint.
reason	Defines the reason the callback was invoked
err	Return value will be set to success if the callback was handled, otherwise this value will be set to a failure status.

Stable:

ICU 2.0

UCNV_FROM_U_CALLBACK_SKIP()

U_CAPI void UCNV_FROM_U_CALLBACK_SKIP	(	const void *	context,
		UConverterFromUnicodeArgs *	fromUArgs,
		const UChar *	codeUnits,
		int32_t	length,
		UChar32	codePoint,
		UConverterCallbackReason	reason,
		UErrorCode *	err
	)

DO NOT CALL THIS FUNCTION DIRECTLY! This From Unicode callback skips any ILLEGAL_SEQUENCE, or skips only UNASSIGNED_SEQUENCE depending on the context parameter simply ignoring those characters.

Parameters

context	The function currently recognizes the callback options: UCNV_SKIP_STOP_ON_ILLEGAL: STOPS at the ILLEGAL_SEQUENCE, returning the error code back to the caller immediately. NULL: Skips any ILLEGAL_SEQUENCE
fromUArgs	Information about the conversion in progress
codeUnits	Points to 'length' UChars of the concerned Unicode sequence
length	Size (in bytes) of the concerned codepage sequence
codePoint	Single UChar32 (UTF-32) containing the concerend Unicode codepoint.
reason	Defines the reason the callback was invoked
err	Return value will be set to success if the callback was handled, otherwise this value will be set to a failure status.

Stable:

ICU 2.0

UCNV_FROM_U_CALLBACK_STOP()

U_CAPI void UCNV_FROM_U_CALLBACK_STOP	(	const void *	context,
		UConverterFromUnicodeArgs *	fromUArgs,
		const UChar *	codeUnits,
		int32_t	length,
		UChar32	codePoint,
		UConverterCallbackReason	reason,
		UErrorCode *	err
	)

DO NOT CALL THIS FUNCTION DIRECTLY! This From Unicode callback STOPS at the ILLEGAL_SEQUENCE, returning the error code back to the caller immediately.

Parameters

context	Pointer to the callback's private data
fromUArgs	Information about the conversion in progress
codeUnits	Points to 'length' UChars of the concerned Unicode sequence
length	Size (in bytes) of the concerned codepage sequence
codePoint	Single UChar32 (UTF-32) containing the concerend Unicode codepoint.
reason	Defines the reason the callback was invoked
err	This should always be set to a failure status prior to calling.

Stable:

ICU 2.0

UCNV_FROM_U_CALLBACK_SUBSTITUTE()

U_CAPI void UCNV_FROM_U_CALLBACK_SUBSTITUTE	(	const void *	context,
		UConverterFromUnicodeArgs *	fromUArgs,
		const UChar *	codeUnits,
		int32_t	length,
		UChar32	codePoint,
		UConverterCallbackReason	reason,
		UErrorCode *	err
	)

DO NOT CALL THIS FUNCTION DIRECTLY! This From Unicode callback will Substitute the ILLEGAL SEQUENCE, or UNASSIGNED_SEQUENCE depending on context parameter, with the current substitution string for the converter.

This is the default callback.

Parameters

context	The function currently recognizes the callback options: UCNV_SUB_STOP_ON_ILLEGAL: STOPS at the ILLEGAL_SEQUENCE, returning the error code back to the caller immediately. NULL: Substitutes any ILLEGAL_SEQUENCE
fromUArgs	Information about the conversion in progress
codeUnits	Points to 'length' UChars of the concerned Unicode sequence
length	Size (in bytes) of the concerned codepage sequence
codePoint	Single UChar32 (UTF-32) containing the concerend Unicode codepoint.
reason	Defines the reason the callback was invoked
err	Return value will be set to success if the callback was handled, otherwise this value will be set to a failure status.

See also

ucnv_setSubstChars

Stable:

ICU 2.0

UCNV_TO_U_CALLBACK_ESCAPE()

U_CAPI void UCNV_TO_U_CALLBACK_ESCAPE	(	const void *	context,
		UConverterToUnicodeArgs *	toUArgs,
		const char *	codeUnits,
		int32_t	length,
		UConverterCallbackReason	reason,
		UErrorCode *	err
	)

DO NOT CALL THIS FUNCTION DIRECTLY! This To Unicode callback will Substitute the ILLEGAL SEQUENCE with the hexadecimal representation of the illegal bytes (in the format XNN, e.g.

"%XFF%X0A%XC8%X03").

Parameters

context	This function currently recognizes the callback options: UCNV_ESCAPE_ICU, UCNV_ESCAPE_JAVA, UCNV_ESCAPE_C, UCNV_ESCAPE_XML_DEC, UCNV_ESCAPE_XML_HEX and UCNV_ESCAPE_UNICODE.
toUArgs	Information about the conversion in progress
codeUnits	Points to 'length' bytes of the concerned codepage sequence
length	Size (in bytes) of the concerned codepage sequence
reason	Defines the reason the callback was invoked
err	Return value will be set to success if the callback was handled, otherwise this value will be set to a failure status.

Stable:

ICU 2.0

UCNV_TO_U_CALLBACK_SKIP()

U_CAPI void UCNV_TO_U_CALLBACK_SKIP	(	const void *	context,
		UConverterToUnicodeArgs *	toUArgs,
		const char *	codeUnits,
		int32_t	length,
		UConverterCallbackReason	reason,
		UErrorCode *	err
	)

DO NOT CALL THIS FUNCTION DIRECTLY! This To Unicode callback skips any ILLEGAL_SEQUENCE, or skips only UNASSIGNED_SEQUENCE depending on the context parameter simply ignoring those characters.

Parameters

context	The function currently recognizes the callback options: UCNV_SKIP_STOP_ON_ILLEGAL: STOPS at the ILLEGAL_SEQUENCE, returning the error code back to the caller immediately. NULL: Skips any ILLEGAL_SEQUENCE
toUArgs	Information about the conversion in progress
codeUnits	Points to 'length' bytes of the concerned codepage sequence
length	Size (in bytes) of the concerned codepage sequence
reason	Defines the reason the callback was invoked
err	Return value will be set to success if the callback was handled, otherwise this value will be set to a failure status.

Stable:

ICU 2.0

UCNV_TO_U_CALLBACK_STOP()

U_CAPI void UCNV_TO_U_CALLBACK_STOP	(	const void *	context,
		UConverterToUnicodeArgs *	toUArgs,
		const char *	codeUnits,
		int32_t	length,
		UConverterCallbackReason	reason,
		UErrorCode *	err
	)

DO NOT CALL THIS FUNCTION DIRECTLY! This To Unicode callback STOPS at the ILLEGAL_SEQUENCE, returning the error code back to the caller immediately.

Parameters

context	Pointer to the callback's private data
toUArgs	Information about the conversion in progress
codeUnits	Points to 'length' bytes of the concerned codepage sequence
length	Size (in bytes) of the concerned codepage sequence
reason	Defines the reason the callback was invoked
err	This should always be set to a failure status prior to calling.

Stable:

ICU 2.0

UCNV_TO_U_CALLBACK_SUBSTITUTE()

U_CAPI void UCNV_TO_U_CALLBACK_SUBSTITUTE	(	const void *	context,
		UConverterToUnicodeArgs *	toUArgs,
		const char *	codeUnits,
		int32_t	length,
		UConverterCallbackReason	reason,
		UErrorCode *	err
	)

DO NOT CALL THIS FUNCTION DIRECTLY! This To Unicode callback will Substitute the ILLEGAL SEQUENCE,or UNASSIGNED_SEQUENCE depending on context parameter, with the Unicode substitution character, U+FFFD.

Parameters

context	The function currently recognizes the callback options: UCNV_SUB_STOP_ON_ILLEGAL: STOPS at the ILLEGAL_SEQUENCE, returning the error code back to the caller immediately. NULL: Substitutes any ILLEGAL_SEQUENCE
toUArgs	Information about the conversion in progress
codeUnits	Points to 'length' bytes of the concerned codepage sequence
length	Size (in bytes) of the concerned codepage sequence
reason	Defines the reason the callback was invoked
err	Return value will be set to success if the callback was handled, otherwise this value will be set to a failure status.

Stable:

ICU 2.0