DSXchange

Full instructions about what you need to download are on the Passport Advantage and Fix Central sites. There are lots of fix packs and patches for version 8.1, which you will need to get from Fix Central. Let Fix Central do the searching for you.

My guess is that your friend downloaded a UNIX edition, not a Windows edition, for the server components.

The new installer does not ask for a licence file. There's a default licence and a "trust" mechanism.

No more information is required. My suggested solution will work.

You must have all your maps (a) correct and (b) compatible. For example MS1252 does not handle Chinese characters.

More work may be required if the output has Date data type.

Provide the correct format strings and use conversion functions such as DateToString() and StringToDate(). You will also need to concatenate the day number (01) to the incoming string.

You can use strings longer than one character, but you have to use Field/Record Delimiter String property rather than Field/Record Delimiter (which is limited to a single character).

There is no default for these. The default is to use Field Delimiter = comma.

You're going to encrypt and decrypt in the one flow? That's the same as taking one step forward and one step backward - a lot of effort to achieve nothing.

Anyway, there's lots of encryption algorithms to be found on the web. And you can hire a C++ programmer if you don't have the skills.

That is not the file name. That is the first three characters of the filename.

There isn't, the same way there's no way to do it within a database table. You have to do it as the rows come out.

For starters, Merge consumes rows from its reference (update) input(s). The other two stage types don't. Your Keep rule iss giving variations because you end up with unmatched masters after all the reference rows have been consumed. So what do you actually want the output to be? Both results you got...

For Version 7 the Chapter number may be different. Use the Search capability within the PDF reader. If you can't find Force Compile there, try the Parallel Job Developer's Guide.

DataStage processes fixed-width data wherever possible. It's far more efficient than having to scan for delimiters or calculated offsets.