Can someone help me to call the C++ dll (detect encoding code page for search ) Topic is solved

[viv]

The function of this library is to detect the file encoding
https://github.com/BYVoid/uchardet

I don't know C++, probably read a part of the file and check its encoding

Specifically, this section of the function
The function is to pass in a file path
return a file encoding

https://github.com/BYVoid/uchardet/blob/4e685757780cb3c652fc6c9ec759f62888969ec9/src/tools/uchardet.cpp#L52

Code: Select all

void detect(FILE * fp)
{
    uchardet_t handle = uchardet_new();

    while (!feof(fp))
    {
        size_t len = fread(buffer, 1, BUFFER_SIZE, fp);
        int retval = uchardet_handle_data(handle, buffer, len);
        if (retval != 0)
        {
            fprintf(stderr, "Handle data error.\n");
            exit(1);
        }
    }
    uchardet_data_end(handle);

    const char * charset = uchardet_get_charset(handle);
    if (*charset)
    	printf("%s\n", charset);
	else
		printf("unknown\n");
	
    uchardet_delete(handle);
}

The public functions are all in this part
https://github.com/BYVoid/uchardet/blob/4e685757780cb3c652fc6c9ec759f62888969ec9/src/uchardet.cpp#L82

The attached content is the compiled exe, dll, and my unsuccessful ahk script
Run "uchardet.exe txt.txt" directly in CMD
will return GB18030

uchardet.zip

(192.91 KiB) Downloaded 46 times

[swagfag]

Code: Select all

class UCharDet
{
	#DllLoad 'libuchardet.dll'
	
	/**
	 * Create an encoding detector.
	 * @return an instance of uchardet_t.
	 *
	 * UCHARDET_INTERFACE uchardet_t uchardet_new(void);
	 */
	Ptr := DllCall('libuchardet\uchardet_new', 'CDecl Ptr')

	/**
	 * Delete an encoding detector.
	 * @param ud [in] the uchardet_t handle to delete.
	 *
	 * UCHARDET_INTERFACE void uchardet_delete(uchardet_t ud);
	 */
	__Delete() => this.Ptr && DllCall('libuchardet\uchardet_delete', 'Ptr', this, 'CDecl')

	/**
	 * Feed data to an encoding detector.
	 * The detector is able to shortcut processing when it reaches certainty
	 * for an encoding, so you should not worry about limiting input data.
	 * As far as you should be concerned: the more the better.
	 *
	 * @param ud [in] handle of an instance of uchardet
	 * @param data [in] data
	 * @param len [in] number of byte of data
	 * @return non-zero number on failure.
	 *
	 * UCHARDET_INTERFACE int uchardet_handle_data(uchardet_t ud, const char * data, size_t len);
	 */
	HandleData(pBytes, cBytes) => DllCall('libuchardet\uchardet_handle_data', 'Ptr', this, 'Ptr', pBytes, 'Ptr', cBytes, 'CDecl Int')

	/**
	 * Notify an end of data to an encoding detector.
	 * @param ud [in] handle of an instance of uchardet
	 * 
	 * UCHARDET_INTERFACE void uchardet_data_end(uchardet_t ud);
	 */
	DataEnd() => DllCall('libuchardet\uchardet_data_end', 'Ptr', this, 'CDecl')

	/**
	 * Reset an encoding detector.
	 * @param ud [in] handle of an instance of uchardet
	 * 
	 * UCHARDET_INTERFACE void uchardet_reset(uchardet_t ud);
	 */
	Reset() => DllCall('libuchardet\uchardet_reset', 'Ptr', this, 'CDecl')

	/**
	 * Get an iconv-compatible name of the encoding that was detected.
	 * @param ud [in] handle of an instance of uchardet
	 * @return name of charset on success and "" on failure.
	 * 
	 * UCHARDET_INTERFACE const char * uchardet_get_charset(uchardet_t ud);
	 */
	GetCharset() => DllCall('libuchardet\uchardet_get_charset', 'Ptr', this, 'CDecl AStr')

	DetectBuffer(Buffer) {
		this.Reset()

		if res := this.HandleData(Buffer.Ptr, Buffer.Size)
			throw Error("Internal 'libuchardet' error, uchardet_handle_data() returned " res)

		this.DataEnd()
		charset := this.GetCharset()
		this.Reset()

		return charset
	}
}

UCD := UCharDet()
MsgBox UCD.DetectBuffer(FileRead('txt.txt', 'RAW')) ; GB18030

[viv]

@swagfag

Thank you very much, I wish you a happy life!

[Cubex]

Nice and for Autohotkey v.1 ?
Please

[swagfag]

Code: Select all

#NoEnv
#Warn ClassOverwrite
#Requires AutoHotkey v1.1.36.02

class UCharDet
{
	__DllLoad() {
		static _ := UCharDet.__DllLoad()

		if hModule := DllCall("LoadLibrary", "Str", "libuchardet.dll", "Ptr")
			DllCall("GetModuleHandleEx", "UInt", 1, "Str", "libuchardet.dll", "Ptr*", 0) ; GET_MODULE_HANDLE_EX_FLAG_PIN
		else
			throw Exception("Failed to load 'libuchardet.dll'.", -1, A_LastError)
	}
	
	/**
	 * Create an encoding detector.
	 * @return an instance of uchardet_t.
	 *
	 * UCHARDET_INTERFACE uchardet_t uchardet_new(void);
	 */
	Ptr := DllCall("libuchardet\uchardet_new", "CDecl Ptr")

	/**
	 * Delete an encoding detector.
	 * @param ud [in] the uchardet_t handle to delete.
	 *
	 * UCHARDET_INTERFACE void uchardet_delete(uchardet_t ud);
	 */
	__Delete() {
		if this.Ptr
			DllCall("libuchardet\uchardet_delete", "Ptr", this.Ptr, "CDecl")
	}

	/**
	 * Feed data to an encoding detector.
	 * The detector is able to shortcut processing when it reaches certainty
	 * for an encoding, so you should not worry about limiting input data.
	 * As far as you should be concerned: the more the better.
	 *
	 * @param ud [in] handle of an instance of uchardet
	 * @param data [in] data
	 * @param len [in] number of byte of data
	 * @return non-zero number on failure.
	 *
	 * UCHARDET_INTERFACE int uchardet_handle_data(uchardet_t ud, const char * data, size_t len);
	 */
	HandleData(pBytes, cBytes) {
		return DllCall("libuchardet\uchardet_handle_data", "Ptr", this.Ptr, "Ptr", pBytes, "Ptr", cBytes, "CDecl Int")
	}

	/**
	 * Notify an end of data to an encoding detector.
	 * @param ud [in] handle of an instance of uchardet
	 * 
	 * UCHARDET_INTERFACE void uchardet_data_end(uchardet_t ud);
	 */
	DataEnd() {
		return DllCall("libuchardet\uchardet_data_end", "Ptr", this.Ptr, "CDecl")
	}

	/**
	 * Reset an encoding detector.
	 * @param ud [in] handle of an instance of uchardet
	 * 
	 * UCHARDET_INTERFACE void uchardet_reset(uchardet_t ud);
	 */
	Reset() {
		return DllCall("libuchardet\uchardet_reset", "Ptr", this.Ptr, "CDecl")
	}

	/**
	 * Get an iconv-compatible name of the encoding that was detected.
	 * @param ud [in] handle of an instance of uchardet
	 * @return name of charset on success and "" on failure.
	 * 
	 * UCHARDET_INTERFACE const char * uchardet_get_charset(uchardet_t ud);
	 */
	GetCharset() {
		return DllCall("libuchardet\uchardet_get_charset", "Ptr", this.Ptr, "CDecl AStr")
	}

	DetectBytes(pBytes, cBytes) {
		this.Reset()

		if res := this.HandleData(pBytes, cBytes)
			throw Exception("Internal 'libuchardet' error, uchardet_handle_data() returned " res)

		this.DataEnd()
		charset := this.GetCharset()
		this.Reset()

		return charset
	}
}

UCD := new UCharDet()
f := FileOpen("txt.txt", "r")
f.Pos := 0
cBytes := f.RawRead(Bytes, f.Length)
MsgBox % UCD.DetectBytes(&Bytes, cBytes) ; GB18030

[Cubex]

@swagfag
Thank you very much. It's a 64bit version and I need a 32bit version uchardet. There is already a newer version of uchardet 0.0.8, but I can not compile it into a 32bit executable file. Damage.
Thank you so much.

[tuzi]

base on swagfag 's code, i create a lib ahk-chardet.

[vmech]

@viv, where you got this binaries ?

@tuzi, version for V2 ?