Problem with nsExec and Characterset

I have a problem with nsExec and the logging of german umlaute into the logwindow. I guess this applies to all non-english versions of windows. Have a look at the screenshot of my XP-System.

Can I do anything in the script to get a correct output in the logwindow or will this have to be fixed in nsExec.dll by a characterset conversion before the string will be logged ?

There was a similar issue on


Can anyone test this with Win9x ?

Script

Here is the script

Works fine for me with Hebrew. Make sure your language for non-Unicode applications is set right in the Regional and Language Options control panel.

like this ?

I am very sure that these are the default settings for Germany and I have never touched them. And CMD.exe is able to produce a correct output - as you can see in my last screenshot...

See the attached screenshot with my 'extended' (?) language settings.

Here is another example, which makes the whole thing even more confusing...

I have made a small application ExecCMD, which passes the commandline to nsExec. The four tests are echoing a string with an umlaut.

The echo-command is given at the command prompt and in a batchfile with and without nsExec.

After changing the Codepage in my batch-editor from ANSI to OEM (DOS), I get another result: it seems that nsExec cannot handle OEM-Chars but logs ANSI-chars instead ?

The Codepage for CMD.exe is 850 (OEM Multilingual-Latin 1)

nsExec treats the output as ANSI as it does no conversion and doesn't use Unicode enabled functions.

I tried creating a console application to output text with the OEM codepage, but my console refused to show any type of Hebrew. I also tried creating an application to print out some Hebrew with Unicode but unlike the OEM codepage where the characters came out wrong, it wouldn't even print them this time.

I don't want to invest anymore time in this right now. You should submit a bug report so it sticks and doesn't fade away in the forum. Maybe someday I or someone else will get to it.

Thanks, kichik.

It's rather a cosmetical problem - at least for the german language, because one can guess the meaning of the crippled characters. For other languages it could be worse...

How about that:



Anyway, I have posted the problem at the bugtracker: