Converting ASCII to EBCDIC 1

[meckeard]

Hi guys,

I am in need of some help in regards to converting ASCII to EBCDIC in a VB 6 app.

The following article returns a results but it doesn't seem to match any EBCDIC or ASCII table that I find:

http://support.microsoft.com/default.aspx?scid=kb;en-us;216399

Has anyone done something like this before?

As far as a third party tool, I can't use one. This must be done in-house

Thanks!

 

[Hypetia]

The reverse conversion (EBCDIC to ASCII) was recently discussed in detail in thread222-1288126.

 

[strongm]

My first post in that thread illustrates doing it both ways.

 

[meckeard]

OK, dumb question...how to I verify that it's working correctly? I'm looking at conversaion tables and my results don't match. It's quite possible I'm not truly understanding what my results should be though.

Thanks,
Mark

 

[strongm]

One of the problems is that there are various different EBCDIC - so you have to use the right code page for the particular EBCDIC you are targetting. The one I use in the example is IBM's

 

[Swi]

Here is a link:

http://msdn.microsoft.com/library/default.asp?url=/library/en-us/intl/unicode_81rn.asp

Swi

 

[strongm]

As you can see, quite a lot of EBCDICs (but quite why I was using IBM's Greek EBCDIC for my example now escapes me ...)

 

[meckeard]

Strongm,

I think I don't know what to supply.

Let's use this chart for example:

http://www.natural-innovations.com/computing/asciiebcdic.html

Looking at this, what ASCII value can I supply and what should be my result be?

Thanks,
Mark

 

[Swi]

What are you getting the data from? A mainframe? If so look into what single byte character set translation table you are using and it should tell you the code page number.

Swi

 

[meckeard]

Maybe more detail is in order. Sorry, I try hard to keep my posts to a minimum so I don't innundate people with too much info.

I have hundreds of records in a text file that needs to be converted to EBCDIC so that a vendor can import it in to their system. Unfortunately, details are sketchy at this point, so I'm just trying to get a head start.

Here is a sample record from my text file:

"5555-5555",37082," 18193"," ",0.00,10108290," 15"," 18193","20061117"

My goal is to write a VB app that opens this file, reads each line, converts it to the EBCDIC equivalent and writes it to a new file. Another process picks up this new file an hour later and sends it to the vendor for importing.

Hopefully that sheds some light on what I'm processing and what I'm trying to achieve.

Thanks,
Mark

 

[Swi]

You will need to find out from the vendor what code page that they would like the EBCDIC data in.

Swi

 

[meckeard]

They are telling me IBM.

With that said, can anyone tell me the best way to test my code?

Thanks,
Mark

 

[Swi]

I would send them the following list and put the onus on them to tell you which type:

http://msdn.microsoft.com/library/default.asp?url=/library/en-us/intl/unicode_81rn.asp

Swi

 

[strongm]

I'd suspect code page 37.

 

[Swi]

strongm is correct (as usual).

Swi

 

[meckeard]

IBM it is. The vendor has confirmed this.

Using my string above and the code provided by strongm, let's assume I pass in my first value of "5555-5555". As it stands now, my result is:

????<square - no key to represent this>????

This doesn't seem right since I can randomly change a number within this string and still get the same results.

Here is the code that I'm using:

Private Function etoa(strSource As String) As String
Dim buffer() As Byte
Dim buffer2() As Byte
Dim lLenb As Long

ReDim buffer(Len(strSource) - 1)
ReDim buffer2(LenB(strSource) - 1)
lLenb = LenB(strSource) - 1
buffer = StrConv(strSource, vbFromUnicode)
MultiByteToWideChar CP_EBCDIC, 0&, buffer(0), -1, buffer2(0), lLenb
etoa = buffer2
End Function

Thanks,
Mark

 

[strongm]

>my result is:


You have to bear in mind that VB is ASCII based - so when it tries to display an EBCDIC character it actually tries to display an ASCII character.

This can be confusing because the ASCII character is NOT the ASCII character originally represented before the conversion to EBCDIC.

Blame IBM for all this rubbish ...

 

[meckeard]

Let me start by saying thanks to you all for your help and insight.

So I guess I'm still left not knowing if my code works. Can I tell once it's written to a text file? That's what I'll be writing to and what will get sent to the vendor.

Thanks,
Mark

 

[strongm]

You need to examine the actual value of each character

 

[BobRodes]

And a good way to do that is to download some kind of hex editor. I use XVI32, which is freeware and can be downloaded from

http://www.chmaas.handshake.de/delphi/freeware/xvi32/xvi32.htm.