Au3Record: Unicode and endianness

[jbaron@…]

Hi,

When recording a script with Au3Record on a French keyboard, the saved file starts with the BOM 0xFFFE indicating a "Little Endian Unicode" sequence. Thus, most of the regular characters are actually in Little Endian order (eg "AU3" -> 41 00 55 00 33 00) but not all them; for example, the "equal" (=) character is encoded 3D 00 and when I open the file with an editor such as notepad I get the Chinese symbol 㴀 corresponding to the unicode character U+3D00. And it's the same with comma, colon, exclamation mark, accented characters and more.
Have you ever encountered such a problem?
A big Congratulations anyway for your work,
Jacques

[jchd18]

Confirmed: for instance '=' gets indeed encoded as UTF16-BE in an otherwise UTF16-LE result file.

[Jpm]

There is much more when running Under Windows 8 64-bit fr keyboard

[Jpm]

In fact the current dll does not seems to support non English keyboard ...
could be difficult to fix as the source of the .dll is not available

[jchd18]

Are you sure J-P?

I don't know which dll you're refering to as it comes to intercepting keystrokes but I know for sure from own experience that the same function can be carried over by other means and over the whole French keyboard composed character range. Then saving and Send()-ing the captured data as Unicode is straightforward.

I'm not going to discuss this further publickly. Use MP to jchd or email where you know.

[Jpm]

the dll is ASWhook.dll send by Larry
I know that can be fix

[Jpm]

In fact The recording can work only Under the same keyboard.
Definetly the present implementation just work with English keyboard.
I am on the way to fix it for any(?) keyboard at least the french one will work.

[Jpm]

Fixed by revision [7706] in version: 3.3.9.5