detect_handledata - Man Page

Detecting character set and measuring accuracy of charset

Snopsys

#include <chardet.h>

short chardet_handledata (Detect ** handle, const char * inbuf, DetectObj ** outbuf);

short chardet_handledata_r (Detect ** handle, const char * inbuf, size_t inlen, DetectObj ** outbuf);

Description

Storing charset and accuracy of inbuf to outbuf

The detect_handledata API is deprecated becase this api is not binary safe. Use or replace to detect_handledata_r api.

Arguments

handle

Detect handle resource that allocated by detect_init api.

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 Values

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) {
	Detect    * d;
	DetectObj * obj;
	int i, arrayNum;
	char *str[] = {
		"this is ascii",
		"이건 euc-kr 입니다."
	};

	arrayNum = sizeof (str) / sizeof (str[0]);

	if ( (d = detect_init ()) == NULL ) {
		fprintf (stderr, "chardet handle initialize failed\n");
		return CHARDET_MEM_ALLOCATED_FAIL;
	}

	for ( i=0; i<arrayNum; i++ ) {
		detect_reset (&d);

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

		//switch (detect_handledata (&d, str[i], &obj))
		switch (detect_handledata (&d, str[i], strlen(str[i]), &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);
	}
	detect_destroy (&d);

    return 0;
}