detect - Man Page

Detecting character set and measuring accuracy of charset

Synopsis

#include <chardet.h>

short detect (char * inbuf, DetectObj ** outbuf);

short detect_r (char * inbuf, size_t inlen, DetectObj ** outbuf);

Description

Storing charset and accuracy of inbuf to outbuf

The detect API is deprecated becase this api is not binary safe. Use or replace to detect_r api.

Arguments

inbuf

input string for detecting

inlen

length of input string for detecting

outbuf

Stroing inforamtion of inbuf. The structure of outbuf is follows.

	typedef struct DetectObject {
		char * encoding;
		float confidence;
	} DetectObj;

The outbuf variable is must initialized by detect_obj_init API before calling this detect api.

Return Value

Returns following condition as case by case.

CHARDET_SUCCESS: Detecting success
CHARDET_NO_RESULT: Detection failure
CHARDET_NULL_OBJECT: Don't initializing outbuf with chardet_obj_init
CHARDET_OUT_OF_MEMORY: Occuring out of memory at internal API

Example

#include <chardet.h>

int main (void) {
	DetectObj *obj;
	char * checkstr = "안녕하세요";

	if ( (obj = detect_obj_init ()) == NULL ) {
		fprintf (stderr, "Memory Allocation failed\n");
		return CHARDET_MEM_ALLOCATED_FAIL;
	}

	//switch (detect (checkstr, &obj))
	switch (detect_r (checkstr, strlen (checkstr), &obj))
	{
		case CHARDET_OUT_OF_MEMORY :
			fprintf (stderr, "On handle processing, occured out of memory\n");
			detect_obj_free (&obj);
			return CHARDET_OUT_OF_MEMORY;
		case CHARDET_NULL_OBJECT :
			fprintf (stderr,
					"2st argument of chardet() is must memory allocation "
					"with detect_obj_init API\n");
			return CHARDET_NULL_OBJECT;
	}

	printf ("encoding: %s, confidence: %f\n", obj->encoding, obj->confidence);
	detect_obj_free (&obj);

    return 0;
}