Sorry I meant to type LANG=en_US.iso886915.
This is a known limitation I learned of in IBM's UOJ class.
Note the difference below. In this sample Attribute 136 is MV with 8 @VM
marks as viewed in screen edit.
Default: LANG=en_US.utf8
[EMAIL PROTECTED] uojCounter]$ java -jar dist/uojCounter.jar
UniString: 273
UniDynArray:273
UniString: F136 1
UniDynArray: F136 1
[EMAIL PROTECTED] uojCounter]$ LANG=en_US.iso885915 java -jar
dist/uojCounter.jar
UniString: 218
UniDynArray:218
UniString: F136 1
UniDynArray: F136 9
On Mon, 2008-06-09 at 23:05 -0700, waivic wrote:
> Jeff, I try your suggestion: export LANG=iso885915. It doesn't look like
> working neither. It has the same effect with LANG variable set to C. Any
> other suggestion?
>
>
> Jeff Powell wrote:
> >
> > Try setting yourLANG variable before executing your program.
> >
> > export LANG=iso885915
> >
> > This is a Linux only problem with UOJ. It is the same for java apps and
> > servlets.
> >
> >
> > On Mon, 2008-06-02 at 16:17 -0700, waivic wrote:
> >
> >> I have a very simple file-TESTS, which only has 4 fields: @ID field, F1,
> >> F2
> >> and F3. F1 is a single value field, F2 is a multi-value field and the
> >> value
> >> is one value-mark-separated list, F3 is a single value field. It has the
> >> following sample record in the file:
> >>
> >> Field Name Value
> >> @ID t101
> >> F1 F1V1
> >> F2 F2V1:@VM:F2V2
> >> F3 F3V1
> >>
> >> I use the following UniObject for Java code segment to retrieve the
> >> record
> >> "t101" from the file:
> >>
> >> //uSession is a UniSession Oject
> >> uSession.connect();
> >>
> >> //Open a file variable to TESTS file.
> >> UniFile testFile = uSession.open("TESTS");
> >>
> >> //Read the record out
> >> String key = "t101";
> >> UniStirng uString = personFile.read(key);
> >> System.out.println(b the full record:b +uString);
> >> System.out.println(b the number of fields in the file:b
> >> +uString.dcount());
> >>
> >>
> >> When I print out the contents of the record, it looks like UOJ converts
> >> all
> >> the value marks at the field F2 into the field mark. When I use dcount()
> >> function to count the number of fields in the output, it returns 5,
> >> instead
> >> of 4. I think the correct result should be 4 since only 4 fields (@ID
> >> field, F1, F2 and F3) in the file. It seems that read() function converts
> >> all the value-marks into the field-marks. So the values at the field-F2:
> >> F2V1:@VM:F2V2 becomes F2:@FM:F2V2. That is why I got 5 instead of 4 when
> >> we
> >> use dcount() to count the number of fields in the output.
> >>
> >> We are running Unidata 7.1 in Red hat Linux server. I already set the
> >> environment variable LANG to bCb. I even tried to set LANG to
> >> ben_USb. Both
> >> times, the UOJ program returned the same result.
> >>
> >> Does anyone know how this happens? I want to keep the original
> >> value-marks
> >> and field-marks after using read() function. Please advice.
> > -------
> > u2-users mailing list
> > u2-users@listserver.u2ug.org
> > To unsubscribe please visit http://listserver.u2ug.org/
-------
u2-users mailing list
u2-users@listserver.u2ug.org
To unsubscribe please visit http://listserver.u2ug.org/
