Asc("€") not returning unicode value

[Tenaja]

I am trying to write a ContainsUnicode function, but asc() is giving me small (i.e. first-byte) values, and not the expected unicode values:

Code: Select all

Debug Asc(Chr(401))	; prints: 145
Debug Asc(Chr(956))	; prints: 188

(Values chosen from the Chr() help file example.)

Yes, I have "Create Unicode Executable" selected (even tested with a CompilerIf wrapper)

Am I missing something, or is there an issue with Asc() in Unicode mode? I am using PureBasic 5.31 Beta 1 x86, in W7-64.

[skywalk]

I get these in Unicode compile x86.
401
956
I get your numbers in Ascii compile mode.

[Tenaja]

I am getting 145 and 188 regardless of the string mode I compile in, both ascii and unicode, also regardless of the compiler I use, x86 or 64-bit.

What compiler & OS are you using?

[IdeasVacuum]

PB5.30 x86 WinXP
401
956

[Demivec]

Windows 8.1, x86 or x64 and using PureBasic v5.30.

401
956

[Tenaja]

I am getting 145 and 188 regardless of the string mode I compile in, both ascii and unicode, also regardless of the compiler I use, x86 or 64-bit.

I just reverted back to 5.3 (from 5.31 b1), and I am still getting 145 & 188 regardless of the unicode setting or the compiler (x86 or 64).

[luis]

What the hell is happening today ?

I got the same result as the others. Try a nice reboot, and then if still not working try to test everything:

Debug followed by compiler version, ascii/unicode mode, etc. and see if they really change when you change the compiler and / or its settings.

[skywalk]

Like luis said. Debug #PB_Compiler_Unicode for confirmation. Read a Unicode text file you confirmed in Notepad++ or a hex editor. Ascii compile will read ???? and Unicode compile will read correct contents.

[Tenaja]

What the hell is happening today ?

I got the same result as the others. Try a nice reboot, and then if still not working try to test everything:

Debug followed by compiler version, ascii/unicode mode, etc. and see if they really change when you change the compiler and / or its settings.

I rebooted this morning, just before starting the code.
As noted in my first post, I did wrap the test within a compilerif unicode...

[luis]

As noted in my first post, I did wrap the test within a compilerif unicode...

And I read it, in fact I said something different. It's yours the computer where strange things are happening.

[Little John]

I am trying to write a ContainsUnicode function, but asc() is giving me small (i.e. first-byte) values, and not the expected unicode values:

Code: Select all

Debug Asc(Chr(401))	; prints: 145
Debug Asc(Chr(956))	; prints: 188

Same here (PB 5.23 LTS) if and only if in the IDE file format is set to plain text.
Better set it to Utf8.
(should actually read UTF-8 BTW)

[IdeasVacuum]

I can confirm Little John's findings - if the source code file format is set to UTF8, all works as expected (401 and 956), set as plain text 145 and 188 are returned. I think I came across this before but now always have UTF8 set - think it's strange though, why should it matter?

[luis]

@LJ

Good catch !

Same here.

I don't see what the editor's file format has to do with a function which returns a number BTW.

[RASHAD]

Try
File ----> File Format ----> Encoding: UTF8

Edit : LJ was first
I did not read well

[Little John]

I don't see what the editor's file format has to do with a function which returns a number BTW.

I don't see it either.
However, I recalled that similar problems had been discussed here in the past.

Since ASCII compile mode is going to be dropped in a future PB version, maybe plain text file format should be dropped as well?
It's nothing but a remnant from the last millennium, isn't it?