utf8rewind
1.3.0
System library for processing UTF-8 encoded text
|
Public interface for UTF-8 functions. More...
Go to the source code of this file.
Macros | |
#define | UTF8_VERSION_MAKE(_major, _minor, _bugfix) ((_major) * 10000) + ((_minor) * 100) + (_bugfix) |
Macro for creating a version number from a major, minor and bugfix number. More... | |
#define | UTF8_VERSION_MAJOR 1 |
The major version number of this release. More... | |
#define | UTF8_VERSION_MINOR 3 |
The minor version number of this release. More... | |
#define | UTF8_VERSION_BUGFIX 0 |
The bugfix version number of this release. More... | |
#define | UTF8_VERSION UTF8_VERSION_MAKE(UTF8_VERSION_MAJOR, UTF8_VERSION_MINOR, UTF8_VERSION_BUGFIX) |
The version number as an integer. More... | |
#define | UTF8_VERSION_STRING "1.3.0" |
The verion number as a string. More... | |
#define | UTF8_VERSION_GUARD(_major, _minor, _bugfix) (UTF8_VERSION >= UTF8_VERSION_MAKE(_major, _minor, _bugfix)) |
Check if feature is supported by the current release. More... | |
#define | UTF8_ERR_NONE (0) |
No errors. More... | |
#define | UTF8_ERR_INVALID_DATA (-1) |
Input data is invalid. More... | |
#define | UTF8_ERR_INVALID_FLAG (-2) |
Input flag is invalid. More... | |
#define | UTF8_ERR_NOT_ENOUGH_SPACE (-3) |
Not enough space in buffer to store result. More... | |
#define | UTF8_ERR_OVERLAPPING_PARAMETERS (-4) |
Input and output buffers overlap in memory. More... | |
#define | UTF8_NORMALIZE_COMPOSE 0x00000001 |
Normalize input to Normalization Form C (NFC). More... | |
#define | UTF8_NORMALIZE_DECOMPOSE 0x00000002 |
Normalize input to Normalization Form D (NFD). More... | |
#define | UTF8_NORMALIZE_COMPATIBILITY 0x00000004 |
Change Normalization Form from NFC to NFKC or from NFD to NFKD. More... | |
#define | UTF8_NORMALIZATION_RESULT_YES (0) |
Text is stable and does not have to be normalized. More... | |
#define | UTF8_NORMALIZATION_RESULT_MAYBE (1) |
Text is unstable, but normalization may be skipped. More... | |
#define | UTF8_NORMALIZATION_RESULT_NO (2) |
Text is unstable and must be normalized. More... | |
#define | UTF8_WCHAR_SIZE (2) |
Specifies the size of the wchar_t type. On Windows this is 2, on POSIX systems it is 4. If not specified on the command line, the compiler tries to automatically determine the size of the wchar_t type based on the environment. More... | |
#define | UTF8_WCHAR_UTF16 (1) |
The wchar_t type is treated as UTF-16 (2 bytes). More... | |
#define | UTF8_API |
Calling convention for public functions. More... | |
Typedefs | |
typedef uint16_t | utf16_t |
UTF-16 encoded code point. More... | |
typedef uint32_t | unicode_t |
UTF-32 encoded code point. More... | |
Functions | |
UTF8_API size_t | utf8len (const char *text) |
Get the length in code points of a UTF-8 encoded string. More... | |
UTF8_API size_t | utf16toutf8 (const utf16_t *input, size_t inputSize, char *target, size_t targetSize, int32_t *errors) |
Convert a UTF-16 encoded string to a UTF-8 encoded string. More... | |
UTF8_API size_t | utf32toutf8 (const unicode_t *input, size_t inputSize, char *target, size_t targetSize, int32_t *errors) |
Convert a UTF-32 encoded string to a UTF-8 encoded string. More... | |
UTF8_API size_t | widetoutf8 (const wchar_t *input, size_t inputSize, char *target, size_t targetSize, int32_t *errors) |
Convert a wide string to a UTF-8 encoded string. More... | |
UTF8_API size_t | utf8toutf16 (const char *input, size_t inputSize, utf16_t *target, size_t targetSize, int32_t *errors) |
Convert a UTF-8 encoded string to a UTF-16 encoded string. More... | |
UTF8_API size_t | utf8toutf32 (const char *input, size_t inputSize, unicode_t *target, size_t targetSize, int32_t *errors) |
Convert a UTF-8 encoded string to a UTF-32 encoded string. More... | |
UTF8_API size_t | utf8towide (const char *input, size_t inputSize, wchar_t *target, size_t targetSize, int32_t *errors) |
Convert a UTF-8 encoded string to a wide string. More... | |
UTF8_API const char * | utf8seek (const char *text, size_t textSize, const char *textStart, off_t offset, int direction) |
Seek into a UTF-8 encoded string. More... | |
UTF8_API size_t | utf8toupper (const char *input, size_t inputSize, char *target, size_t targetSize, int32_t *errors) |
Convert UTF-8 encoded text to uppercase. More... | |
UTF8_API size_t | utf8tolower (const char *input, size_t inputSize, char *target, size_t targetSize, int32_t *errors) |
Convert UTF-8 encoded text to lowercase. More... | |
UTF8_API size_t | utf8totitle (const char *input, size_t inputSize, char *target, size_t targetSize, int32_t *errors) |
Convert UTF-8 encoded text to titlecase. More... | |
UTF8_API uint8_t | utf8isnormalized (const char *input, size_t inputSize, size_t flags, size_t *offset) |
Check if a string is stable in the specified Unicode Normalization Form. More... | |
UTF8_API size_t | utf8normalize (const char *input, size_t inputSize, char *target, size_t targetSize, size_t flags, int32_t *errors) |
Normalize a string to the specified Unicode Normalization Form. More... | |
Public interface for UTF-8 functions.
#define UTF8_NORMALIZE_COMPOSE 0x00000001 |
Normalize input to Normalization Form C (NFC).
#define UTF8_NORMALIZE_DECOMPOSE 0x00000002 |
Normalize input to Normalization Form D (NFD).
#define UTF8_NORMALIZE_COMPATIBILITY 0x00000004 |
Change Normalization Form from NFC to NFKC or from NFD to NFKD.
#define UTF8_NORMALIZATION_RESULT_YES (0) |
Text is stable and does not have to be normalized.
#define UTF8_NORMALIZATION_RESULT_MAYBE (1) |
Text is unstable, but normalization may be skipped.
#define UTF8_NORMALIZATION_RESULT_NO (2) |
Text is unstable and must be normalized.
UTF-16 encoded code point.
UTF-32 encoded code point.
UTF8_API size_t utf8len | ( | const char * | text | ) |
UTF8_API size_t utf16toutf8 | ( | const utf16_t * | input, |
size_t | inputSize, | ||
char * | target, | ||
size_t | targetSize, | ||
int32_t * | errors | ||
) |
Convert a UTF-16 encoded string to a UTF-8 encoded string.
Example:
[in] | input | UTF-16 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |
UTF8_API size_t utf32toutf8 | ( | const unicode_t * | input, |
size_t | inputSize, | ||
char * | target, | ||
size_t | targetSize, | ||
int32_t * | errors | ||
) |
Convert a UTF-32 encoded string to a UTF-8 encoded string.
Example:
[in] | input | UTF-32 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |
UTF8_API size_t widetoutf8 | ( | const wchar_t * | input, |
size_t | inputSize, | ||
char * | target, | ||
size_t | targetSize, | ||
int32_t * | errors | ||
) |
Convert a wide string to a UTF-8 encoded string.
Depending on the platform, wide strings are either UTF-16 or UTF-32 encoded. This function takes a wide string as input and automatically calls the correct conversion function.
This allows for a cross-platform treatment of wide text and is preferable to using the UTF-16 or UTF-32 versions directly.
Example:
[in] | input | Wide-encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |
UTF8_API size_t utf8toutf16 | ( | const char * | input, |
size_t | inputSize, | ||
utf16_t * | target, | ||
size_t | targetSize, | ||
int32_t * | errors | ||
) |
Convert a UTF-8 encoded string to a UTF-16 encoded string.
Erroneous byte sequences such as missing or illegal bytes or overlong encoding of code points (e.g. using five bytes to encode a sequence that can be represented by two bytes) are converted to the replacement character U+FFFD.
Example:
[in] | input | UTF-8 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |
UTF8_API size_t utf8toutf32 | ( | const char * | input, |
size_t | inputSize, | ||
unicode_t * | target, | ||
size_t | targetSize, | ||
int32_t * | errors | ||
) |
Convert a UTF-8 encoded string to a UTF-32 encoded string.
Erroneous byte sequences such as missing or illegal bytes or overlong encoding of code points (e.g. using five bytes to encode a sequence that can be represented by two bytes) are converted to the replacement character U+FFFD.
Example:
[in] | input | UTF-8 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |
UTF8_API size_t utf8towide | ( | const char * | input, |
size_t | inputSize, | ||
wchar_t * | target, | ||
size_t | targetSize, | ||
int32_t * | errors | ||
) |
Convert a UTF-8 encoded string to a wide string.
Depending on the platform, wide strings are either UTF-16 or UTF-32 encoded. This function takes a UTF-8 encoded string as input and automatically calls the correct conversion function.
This allows for a cross-platform treatment of wide text and is preferable to using the UTF-16 or UTF-32 versions directly.
Erroneous byte sequences such as missing or illegal bytes or overlong encoding of code points (e.g. using five bytes to encode a sequence that can be represented by two bytes) are converted to the replacement character U+FFFD.
Example:
[in] | input | UTF-8 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |
UTF8_API const char* utf8seek | ( | const char * | text, |
size_t | textSize, | ||
const char * | textStart, | ||
off_t | offset, | ||
int | direction | ||
) |
Seek into a UTF-8 encoded string.
Working with UTF-8 encoded strings can be tricky due to the nature of the variable-length encoding. Because one character no longer equals one byte, it can be difficult to skip around in a UTF-8 encoded string without decoding the code points.
This function provides an interface similar to fseek
in order to enable skipping to another part of the string.
textStart
must come before text
in memory when seeking from the current or end position.Example:
[in] | text | Input string. |
[in] | textSize | Size of input string in bytes. |
[in] | textStart | Start of input string. |
[in] | offset | Requested offset in code points. |
[in] | direction | Direction to seek in.
|
UTF8_API size_t utf8toupper | ( | const char * | input, |
size_t | inputSize, | ||
char * | target, | ||
size_t | targetSize, | ||
int32_t * | errors | ||
) |
Convert UTF-8 encoded text to uppercase.
This function allows conversion of UTF-8 encoded strings to uppercase without first changing the encoding to UTF-32. Conversion is fully compliant with the Unicode 7.0 standard.
Although most code points can be converted in-place, there are notable exceptions. For example, U+00DF (LATIN SMALL LETTER SHARP S) maps to "U+0053 U+0053" (LATIN CAPITAL LETTER S and LATIN CAPITAL LETTER S) when converted to uppercase. Therefor, it is advised to first determine the size in bytes of the output by calling the function with a NULL output buffer.
Only a handful of scripts make a distinction between upper and lowercase. In addition to modern scripts, such as Latin, Greek, Armenian and Cyrillic, a few historic or archaic scripts have case. The vast majority of scripts do not have case distinctions.
toUpper(toLower(x)) != toLower(toUpper(x))
.Example:
[in] | input | UTF-8 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |
UTF8_API size_t utf8tolower | ( | const char * | input, |
size_t | inputSize, | ||
char * | target, | ||
size_t | targetSize, | ||
int32_t * | errors | ||
) |
Convert UTF-8 encoded text to lowercase.
This function allows conversion of UTF-8 encoded strings to lowercase without first changing the encoding to UTF-32. Conversion is fully compliant with the Unicode 7.0 standard.
Although most code points can be converted to lowercase in-place, there are notable exceptions. For example, U+0130 (LATIN CAPITAL LETTER I WITH DOT ABOVE) maps to "U+0069 U+0307" (LATIN SMALL LETTER I and COMBINING DOT ABOVE) when converted to lowercase. Therefor, it is advised to first determine the size in bytes of the output by calling the function with a NULL output buffer.
Only a handful of scripts make a distinction between upper- and lowercase. In addition to modern scripts, such as Latin, Greek, Armenian and Cyrillic, a few historic or archaic scripts have case. The vast majority of scripts do not have case distinctions.
toUpper(toLower(x)) != toLower(toUpper(x))
.Example:
[in] | input | UTF-8 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |
UTF8_API size_t utf8totitle | ( | const char * | input, |
size_t | inputSize, | ||
char * | target, | ||
size_t | targetSize, | ||
int32_t * | errors | ||
) |
Convert UTF-8 encoded text to titlecase.
This function allows conversion of UTF-8 encoded strings to titlecase without first changing the encoding to UTF-32. Conversion is fully compliant with the Unicode 7.0 standard.
Titlecase requires a bit more explanation than uppercase and lowercase, because it is not a common text transformation. Titlecase uses uppercase for the first letter of each word and lowercase for the rest. Words are defined as "collections of code points with general category Lu, Ll, Lt, Lm or Lo according to the Unicode database".
Effectively, any type of punctuation can break up a word, even if this is not grammatically valid. This happens because the titlecasing algorithm does not and cannot take grammar rules into account.
Text | Titlecase |
---|---|
The running man | The Running Man |
NATO Alliance | Nato Alliance |
You're amazing at building libraries | You'Re Amazing At Building Libraries |
Although most code points can be converted to titlecase in-place, there are notable exceptions. For example, U+00DF (LATIN SMALL LETTER SHARP S) maps to "U+0053 U+0073" (LATIN CAPITAL LETTER S and LATIN SMALL LETTER S) when converted to titlecase. Therefor, it is advised to first determine the size in bytes of the output by calling the function with a NULL output buffer.
Only a handful of scripts make a distinction between upper- and lowercase. In addition to modern scripts, such as Latin, Greek, Armenian and Cyrillic, a few historic or archaic scripts have case. The vast majority of scripts do not have case distinctions.
toUpper(toLower(x)) != toLower(toUpper(x))
.Example:
[in] | input | UTF-8 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |
UTF8_API uint8_t utf8isnormalized | ( | const char * | input, |
size_t | inputSize, | ||
size_t | flags, | ||
size_t * | offset | ||
) |
Check if a string is stable in the specified Unicode Normalization Form.
This function can be used as a preprocessing step, before attempting to normalize a string. Normalization is a very expensive process, it is often cheaper to first determine if the string is unstable in the requested normalization form.
The result of the check will be YES if the string is stable and MAYBE or NO if it is unstable. If the result is MAYBE, the string does not necessarily have to be normalized.
If the result is unstable, the offset parameter is set to the offset for the first unstable code point. If the string is stable, the offset is equivalent to the length of the string in bytes.
You must specify the desired Unicode Normalization Form by using a combination of flags:
Unicode | Flags |
---|---|
Normalization Form C (NFC) | UTF8_NORMALIZE_COMPOSE |
Normalization Form KC (NFKC) | UTF8_NORMALIZE_COMPOSE + UTF8_NORMALIZE_COMPATIBILITY |
Normalization Form D (NFD) | UTF8_NORMALIZE_DECOMPOSE |
Normalization Form KD (NFKD) | UTF8_NORMALIZE_DECOMPOSE + UTF8_NORMALIZE_COMPATIBILITY |
For more information, please review Unicode Standard Annex #15 - Unicode Normalization Forms.
Example:
[in] | input | UTF-8 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[in] | flags | Desired normalization form. Must be a combination of UTF8_NORMALIZE_COMPOSE, UTF8_NORMALIZE_DECOMPOSE and UTF8_NORMALIZE_COMPATIBILITY. |
[out] | offset | Offset to first unstable code point or length of input in bytes if stable. |
UTF8_NORMALIZATION_RESULT_YES | Input is stable and does not have to be normalized. |
UTF8_NORMALIZATION_RESULT_MAYBE | Input is unstable, but normalization may be skipped. |
UTF8_NORMALIZATION_RESULT_NO | Input is unstable and must be normalized. |
UTF8_API size_t utf8normalize | ( | const char * | input, |
size_t | inputSize, | ||
char * | target, | ||
size_t | targetSize, | ||
size_t | flags, | ||
int32_t * | errors | ||
) |
Normalize a string to the specified Unicode Normalization Form.
The Unicode standard defines two standards for equivalence between characters: canonical and compatibility equivalence. Canonically equivalent characters and sequence represent the same abstract character and must be rendered with the same appearance and behavior. Compatibility equivalent characters have a weaker equivalence and may be rendered differently.
Unicode Normalization Forms are formally defined standards that can be used to test whether any two strings of characters are equivalent to each other. This equivalence may be canonical or compatibility.
The algorithm puts all combining marks into a specified order and uses the rules for decomposition and composition to transform the string into one of four Unicode Normalization Forms. A binary comparison can then be used to determine equivalence.
These are the Unicode Normalization Forms:
Form | Description |
---|---|
Normalization Form D (NFD) | Canonical decomposition |
Normalization Form C (NFC) | Canonical decomposition, followed by canonical composition |
Normalization Form KD (NFKD) | Compatibility decomposition |
Normalization Form KC (NFKC) | Compatibility decomposition, followed by canonical composition |
utf8normalize
can be used to transform text into one of these forms. You must specify the desired Unicode Normalization Form by using a combination of flags:
Form | Flags |
---|---|
Normalization Form D (NFD) | UTF8_NORMALIZE_DECOMPOSE |
Normalization Form C (NFC) | UTF8_NORMALIZE_COMPOSE |
Normalization Form KD (NFKD) | UTF8_NORMALIZE_DECOMPOSE + UTF8_NORMALIZE_COMPATIBILITY |
Normalization Form KC (NFKC) | UTF8_NORMALIZE_COMPOSE + UTF8_NORMALIZE_COMPATIBILITY |
For more information, please review Unicode Standard Annex #15 - Unicode Normalization Forms.
Example:
[in] | input | UTF-8 encoded string. |
[in] | inputSize | Size of the input in bytes. |
[out] | target | Output buffer for the result, can be NULL. |
[in] | targetSize | Size of the output buffer in bytes. |
[in] | flags | Desired normalization form. Must be a combination of UTF8_NORMALIZE_COMPOSE, UTF8_NORMALIZE_DECOMPOSE and UTF8_NORMALIZE_COMPATIBILITY. |
[out] | errors | Output for errors. |
UTF8_ERR_NONE | No errors. |
UTF8_ERR_INVALID_FLAG | Invalid combination of flags was specified. |
UTF8_ERR_INVALID_DATA | Failed to decode data. |
UTF8_ERR_OVERLAPPING_PARAMETERS | Input and output buffers overlap in memory. |
UTF8_ERR_NOT_ENOUGH_SPACE | Target buffer size is insufficient for result. |