ICU 72.1 72.1
Namespaces | Typedefs | Functions
ucsdet.h File Reference

C API: Charset Detection API. More...

#include "unicode/utypes.h"
#include "unicode/uenum.h"
#include "unicode/localpointer.h"

Go to the source code of this file.

Namespaces

namespace  icu
 File coll.h.
 

Typedefs

typedef struct UCharsetDetector UCharsetDetector
 Structure representing a charset detector. More...
 
typedef struct UCharsetMatch UCharsetMatch
 Opaque structure representing a match that was identified from a charset detection operation. More...
 

Functions

U_CAPI UCharsetDetectorucsdet_open (UErrorCode *status)
 Open a charset detector. More...
 
U_CAPI void ucsdet_close (UCharsetDetector *ucsd)
 Close a charset detector. More...
 
U_CAPI void ucsdet_setText (UCharsetDetector *ucsd, const char *textIn, int32_t len, UErrorCode *status)
 Set the input byte data whose charset is to detected. More...
 
U_CAPI void ucsdet_setDeclaredEncoding (UCharsetDetector *ucsd, const char *encoding, int32_t length, UErrorCode *status)
 Set the declared encoding for charset detection. More...
 
U_CAPI const UCharsetMatchucsdet_detect (UCharsetDetector *ucsd, UErrorCode *status)
 Return the charset that best matches the supplied input data. More...
 
U_CAPI const UCharsetMatch ** ucsdet_detectAll (UCharsetDetector *ucsd, int32_t *matchesFound, UErrorCode *status)
 Find all charset matches that appear to be consistent with the input, returning an array of results. More...
 
U_CAPI const char * ucsdet_getName (const UCharsetMatch *ucsm, UErrorCode *status)
 Get the name of the charset represented by a UCharsetMatch. More...
 
U_CAPI int32_t ucsdet_getConfidence (const UCharsetMatch *ucsm, UErrorCode *status)
 Get a confidence number for the quality of the match of the byte data with the charset. More...
 
U_CAPI const char * ucsdet_getLanguage (const UCharsetMatch *ucsm, UErrorCode *status)
 Get the RFC 3066 code for the language of the input data. More...
 
U_CAPI int32_t ucsdet_getUChars (const UCharsetMatch *ucsm, UChar *buf, int32_t cap, UErrorCode *status)
 Get the entire input text as a UChar string, placing it into a caller-supplied buffer. More...
 
U_CAPI UEnumerationucsdet_getAllDetectableCharsets (const UCharsetDetector *ucsd, UErrorCode *status)
 Get an iterator over the set of all detectable charsets - over the charsets that are known to the charset detection service. More...
 
U_CAPI UBool ucsdet_isInputFilterEnabled (const UCharsetDetector *ucsd)
 Test whether input filtering is enabled for this charset detector. More...
 
U_CAPI UBool ucsdet_enableInputFilter (UCharsetDetector *ucsd, UBool filter)
 Enable filtering of input text. More...
 
U_CAPI UEnumerationucsdet_getDetectableCharsets (const UCharsetDetector *ucsd, UErrorCode *status)
 Get an iterator over the set of detectable charsets - over the charsets that are enabled by the specified charset detector. More...
 
U_CAPI void ucsdet_setDetectableCharset (UCharsetDetector *ucsd, const char *encoding, UBool enabled, UErrorCode *status)
 Enable or disable individual charset encoding. More...
 

Detailed Description

C API: Charset Detection API.

This API provides a facility for detecting the charset or encoding of character data in an unknown text format. The input data can be from an array of bytes.

Character set detection is at best an imprecise operation. The detection process will attempt to identify the charset that best matches the characteristics of the byte data, but the process is partly statistical in nature, and the results can not be guaranteed to always be correct.

For best accuracy in charset detection, the input data should be primarily in a single language, and a minimum of a few hundred bytes worth of plain text in the language are needed. The detection process will attempt to ignore html or xml style markup that could otherwise obscure the content.

An alternative to the ICU Charset Detector is the Compact Encoding Detector, https://github.com/google/compact_enc_det. It often gives more accurate results, especially with short input samples.

Definition in file ucsdet.h.

Typedef Documentation

◆ UCharsetDetector

Structure representing a charset detector.

Stable:
ICU 3.6

Definition at line 63 of file ucsdet.h.

◆ UCharsetMatch

typedef struct UCharsetMatch UCharsetMatch

Opaque structure representing a match that was identified from a charset detection operation.

Stable:
ICU 3.6

Definition at line 71 of file ucsdet.h.

Function Documentation

◆ ucsdet_close()

U_CAPI void ucsdet_close ( UCharsetDetector ucsd)

Close a charset detector.

All storage and any other resources owned by this charset detector will be released. Failure to close a charset detector when finished with it can result in memory leaks in the application.

Parameters
ucsdThe charset detector to be closed.
Stable:
ICU 3.6

◆ ucsdet_detect()

U_CAPI const UCharsetMatch * ucsdet_detect ( UCharsetDetector ucsd,
UErrorCode status 
)

Return the charset that best matches the supplied input data.

Note though, that because the detection only looks at the start of the input data, there is a possibility that the returned charset will fail to handle the full set of input data.

The returned UCharsetMatch object is owned by the UCharsetDetector. It will remain valid until the detector input is reset, or until the detector is closed.

The function will fail if

  • no charset appears to match the data.
  • no input text has been provided
Parameters
ucsdthe charset detector to be used.
statusany error conditions are reported back in this variable.
Returns
a UCharsetMatch representing the best matching charset, or NULL if no charset matches the byte data.
Stable:
ICU 3.6

◆ ucsdet_detectAll()

U_CAPI const UCharsetMatch ** ucsdet_detectAll ( UCharsetDetector ucsd,
int32_t *  matchesFound,
UErrorCode status 
)

Find all charset matches that appear to be consistent with the input, returning an array of results.

The results are ordered with the best quality match first.

Because the detection only looks at a limited amount of the input byte data, some of the returned charsets may fail to handle the all of input data.

The returned UCharsetMatch objects are owned by the UCharsetDetector. They will remain valid until the detector is closed or modified

Return an error if

  • no charsets appear to match the input data.
  • no input text has been provided
Parameters
ucsdthe charset detector to be used.
matchesFoundpointer to a variable that will be set to the number of charsets identified that are consistent with the input data. Output only.
statusany error conditions are reported back in this variable.
Returns
A pointer to an array of pointers to UCharSetMatch objects. This array, and the UCharSetMatch instances to which it refers, are owned by the UCharsetDetector, and will remain valid until the detector is closed or modified.
Stable:
ICU 3.6

◆ ucsdet_enableInputFilter()

U_CAPI UBool ucsdet_enableInputFilter ( UCharsetDetector ucsd,
UBool  filter 
)

Enable filtering of input text.

If filtering is enabled, text within angle brackets ("<" and ">") will be removed before detection, which will remove most HTML or xml markup.

Parameters
ucsdthe charset detector to be modified.
filtertrue to enable input text filtering.
Returns
The previous setting.
Stable:
ICU 3.6

◆ ucsdet_getAllDetectableCharsets()

U_CAPI UEnumeration * ucsdet_getAllDetectableCharsets ( const UCharsetDetector ucsd,
UErrorCode status 
)

Get an iterator over the set of all detectable charsets - over the charsets that are known to the charset detection service.

The returned UEnumeration provides access to the names of the charsets.

The state of the Charset detector that is passed in does not affect the result of this function, but requiring a valid, open charset detector as a parameter insures that the charset detection service has been safely initialized and that the required detection data is available.

Note: Multiple different charset encodings in a same family may use a single shared name in this implementation. For example, this method returns an array including "ISO-8859-1" (ISO Latin 1), but not including "windows-1252" (Windows Latin 1). However, actual detection result could be "windows-1252" when the input data matches Latin 1 code points with any points only available in "windows-1252".

Parameters
ucsda Charset detector.
statusAny error conditions are reported back in this variable.
Returns
an iterator providing access to the detectable charset names.
Stable:
ICU 3.6

◆ ucsdet_getConfidence()

U_CAPI int32_t ucsdet_getConfidence ( const UCharsetMatch ucsm,
UErrorCode status 
)

Get a confidence number for the quality of the match of the byte data with the charset.

Confidence numbers range from zero to 100, with 100 representing complete confidence and zero representing no confidence.

The confidence values are somewhat arbitrary. They define an an ordering within the results for any single detection operation but are not generally comparable between the results for different input.

A confidence value of ten does have a general meaning - it is used for charsets that can represent the input data, but for which there is no other indication that suggests that the charset is the correct one. Pure 7 bit ASCII data, for example, is compatible with a great many charsets, most of which will appear as possible matches with a confidence of 10.

Parameters
ucsmThe charset match object.
statusAny error conditions are reported back in this variable.
Returns
A confidence number for the charset match.
Stable:
ICU 3.6

◆ ucsdet_getDetectableCharsets()

U_CAPI UEnumeration * ucsdet_getDetectableCharsets ( const UCharsetDetector ucsd,
UErrorCode status 
)

Get an iterator over the set of detectable charsets - over the charsets that are enabled by the specified charset detector.

The returned UEnumeration provides access to the names of the charsets.

Parameters
ucsda Charset detector.
statusAny error conditions are reported back in this variable.
Returns
an iterator providing access to the detectable charset names by the specified charset detector.
Internal:
Do not use. This API is for internal use only.

◆ ucsdet_getLanguage()

U_CAPI const char * ucsdet_getLanguage ( const UCharsetMatch ucsm,
UErrorCode status 
)

Get the RFC 3066 code for the language of the input data.

The Charset Detection service is intended primarily for detecting charsets, not language. For some, but not all, charsets, a language is identified as a byproduct of the detection process, and that is what is returned by this function.

CAUTION:

  1. Language information is not available for input data encoded in all charsets. In particular, no language is identified for UTF-8 input data.
  2. Closely related languages may sometimes be confused.

If more accurate language detection is required, a linguistic analysis package should be used.

The storage for the returned name string is owned by the UCharsetMatch, and will remain valid while the UCharsetMatch is valid.

Parameters
ucsmThe charset match object.
statusAny error conditions are reported back in this variable.
Returns
The RFC 3066 code for the language of the input data, or an empty string if the language could not be determined.
Stable:
ICU 3.6

◆ ucsdet_getName()

U_CAPI const char * ucsdet_getName ( const UCharsetMatch ucsm,
UErrorCode status 
)

Get the name of the charset represented by a UCharsetMatch.

The storage for the returned name string is owned by the UCharsetMatch, and will remain valid while the UCharsetMatch is valid.

The name returned is suitable for use with the ICU conversion APIs.

Parameters
ucsmThe charset match object.
statusAny error conditions are reported back in this variable.
Returns
The name of the matching charset.
Stable:
ICU 3.6

◆ ucsdet_getUChars()

U_CAPI int32_t ucsdet_getUChars ( const UCharsetMatch ucsm,
UChar buf,
int32_t  cap,
UErrorCode status 
)

Get the entire input text as a UChar string, placing it into a caller-supplied buffer.

A terminating NUL character will be appended to the buffer if space is available.

The number of UChars in the output string, not including the terminating NUL, is returned.

If the supplied buffer is smaller than required to hold the output, the contents of the buffer are undefined. The full output string length (in UChars) is returned as always, and can be used to allocate a buffer of the correct size.

Parameters
ucsmThe charset match object.
bufA UChar buffer to be filled with the converted text data.
capThe capacity of the buffer in UChars.
statusAny error conditions are reported back in this variable.
Returns
The number of UChars in the output string.
Stable:
ICU 3.6

◆ ucsdet_isInputFilterEnabled()

U_CAPI UBool ucsdet_isInputFilterEnabled ( const UCharsetDetector ucsd)

Test whether input filtering is enabled for this charset detector.

Input filtering removes text that appears to be HTML or xml markup from the input before applying the code page detection heuristics.

Parameters
ucsdThe charset detector to check.
Returns
true if filtering is enabled.
Stable:
ICU 3.6

◆ ucsdet_open()

U_CAPI UCharsetDetector * ucsdet_open ( UErrorCode status)

Open a charset detector.

Parameters
statusAny error conditions occurring during the open operation are reported back in this variable.
Returns
the newly opened charset detector.
Stable:
ICU 3.6

◆ ucsdet_setDeclaredEncoding()

U_CAPI void ucsdet_setDeclaredEncoding ( UCharsetDetector ucsd,
const char *  encoding,
int32_t  length,
UErrorCode status 
)

Set the declared encoding for charset detection.

The declared encoding of an input text is an encoding obtained by the user from an http header or xml declaration or similar source that can be provided as an additional hint to the charset detector.

How and whether the declared encoding will be used during the detection process is TBD.

Parameters
ucsdthe charset detector to be used.
encodingan encoding for the current data obtained from a header or declaration or other source outside of the byte data itself.
lengththe length of the encoding name, or -1 if the name string is NUL terminated.
statusany error conditions are reported back in this variable.
Stable:
ICU 3.6

◆ ucsdet_setDetectableCharset()

U_CAPI void ucsdet_setDetectableCharset ( UCharsetDetector ucsd,
const char *  encoding,
UBool  enabled,
UErrorCode status 
)

Enable or disable individual charset encoding.

A name of charset encoding must be included in the names returned by ucsdet_getAllDetectableCharsets().

Parameters
ucsda Charset detector.
encodingencoding the name of charset encoding.
enabledtrue to enable, or false to disable the charset encoding.
statusreceives the return status. When the name of charset encoding is not supported, U_ILLEGAL_ARGUMENT_ERROR is set.
Internal:
Do not use. This API is for internal use only.

◆ ucsdet_setText()

U_CAPI void ucsdet_setText ( UCharsetDetector ucsd,
const char *  textIn,
int32_t  len,
UErrorCode status 
)

Set the input byte data whose charset is to detected.

Ownership of the input text byte array remains with the caller. The input string must not be altered or deleted until the charset detector is either closed or reset to refer to different input text.

Parameters
ucsdthe charset detector to be used.
textInthe input text of unknown encoding. .
lenthe length of the input text, or -1 if the text is NUL terminated.
statusany error conditions are reported back in this variable.
Stable:
ICU 3.6