DBCS & _MBCS issues trying to force micron character "µ"

[DanL123]

Hi to all, esp anyone that can guide me!!

We have an application (very 'long in the tooth', VC++ 6) that has been see to show odd characters on Japanese Windows installations.

The route of the problem is a CStatic derived class that is used to display text on a tool bar. For example:

CsZ.Format("Z:%6.1fµm", Val);
...
m_ZPosVal.SetWindowText(CsZ);

where CsZ is a cstring and m_ZPosVal is the CStatic derived class, and the text string ("") is the the position in microns of a pointer.

Unfortunately, the micron symbol is converted to kanji (or similar) on Japanese Windows.

My very simple task is to embed or force the "µ" symbol to remain as-is despite language settings.

Our application is built with _MBCS defined - so that SHOULD handle single and double characters in a text string (as I understand). However, I can find no way to force this case. I have tried things like:

CsY.Format("Y:%6.0f\xB5m", Val);
&
CsY.Format("Y:%6.0f\x00B5m", Val);

as well as various _mbs functions to force the case prior to putting into cstring...

Unfortunately, unicode is far to big a change (weeks & weeks) for such a small (but important!) tweak as this.

Surely there is an easy method that I am missing somewhere!? Does anyone know an answer?

Much appreciation for any help.

[choisum]

http://www.rikai.com/library/kanjitables/kanji_codes.sjis.shtml

83 be | α β γ δ ε ζ η θ ι κ λ μ ν ξ ο 83 ce | π ρ σ τ υ φ χ ψ ω

Shift-JIS defines Greek characters around here. (0xB5 is the Unicode code point.)

Of course what you really want is Microsoft Code Page 932 reference. In case MS did something different ;)

Anyway what you want to know is this. To make it all independent of the actual Windows language (or locale rather), you could define a Unicode string with Unicode μ (0x00B5) and at runtime use wcstombs to get the correct character for the language that is in effect. (At a guess).

(Strangely enough B5 is short for Big5, Chinese code page...)

[choisum]

Actually, if the users don't actually need Japanese for your app, you can force the application to think it's running on a European/US version of Windows without affection other applications.

Under Windows XP I used Applocale (free from MS). Unfortunately this isn't officially available for Vista/Se7en.

http://en.wikipedia.org/wiki/AppLocale

[DanL123]

Hi Choisum,

Thanks for the reply & sorry for the delay. I'll try your suggestions and report back my findings. As work's gone silly at the moment - this may be a while ;-)

Seriously though, thanks for your input! Much appreciated!!!!! :-D

Dan.

[choisum]

You're welcome. I'm interested in this for my own reasons too!

Here's a test progam which converts 0x00b5 into 0x03ca.

#include <iostream>
#include<stdlib.h>
using namespace std;
int main()
{
	char sz[20];

	// Unicode for mu
	wchar_t wsz[20] = L"\x00b5";

	// Set Japanese code page
	// May not be necessary for MBCS Windows GUI program
	setlocale(LC_ALL, ".932");

	int i = wcstombs(sz, wsz, 20);

	cout << i << endl;
	cout << hex << int(sz[0]) << endl;
	cout << hex << int(sz[1]) << endl;
}