DSXchange

What "transformation" do you propose to perform upon the objects? How, precisely, do you plan to store objects in Oracle? What "dump" capabilities does Versant have? Is there an ODBC driver available? (Given that there's nothing in the ODBC API specification relating to object databases, I tend to d...

Yes. Your Informix client software requires that the sqlhosts file be accessible and contain appropriate definitions of hostname(s) supporting Informix server(s). That it's not accessible may mean simply that you have not set the INFORMIXDIR environment variable (either in dsenv script or elsewhere)...

Does your particular UNIX support positive arguments to the head command? If so, use head +2 filename. Otherwise you could use tail to retrieve all but the first two rows (use the result of (expr) `wc -l filename` - 2).

Given that ODBC is being used, I believe that the BCI (and therefore OCONV) restriction will still apply. Why OCI is not being used I have no idea.

Here's what I meant by add a line number. File A File B --------- ------------------ # UPC ORG # STORE DATE 1 01 ALB 1 001 01/24/2007 2 02 ALB 2 002 01/25/2007 3 03 ALB 3 003 01/26/2007 4 04 ALB 4 004 01/27/2007 5 05 ALB 5 005 01/28/2007 Now you've got something on which to base the join, so that th...

Who cares? Generate the maximum possible number of columns that you are likely to encounter. Say 24.

ANY join, Cartesian or otherwise, requires a key upon which to base the join.

Generate a key (line number) if you must.

Write a controlling job sequence that includes StartLoop and EndLoop activities. You can schedule this once per day. Include a mechanism (such as existence of a file) for requesting it to shut down.

Do you have an ODBC driver for Access on UNIX? If not you can't use ODBC.

"Wetware".

There are no publicly-available conversion utilities. You must apply brain power to the task, and create parallel jobs that replicate the functionality of the server jobs.

It's an alert rather than a warning. On the input to the stage YR is defined as int64; on the output from the stage YR is defined as int32 (or decimal[10,0] in the other case). You are being alerted to the fact that not every int64 value can be represented as int32. To eliminate the warning, render ...

Did you check the "search for all terms" option?

You can not give a file name ending in ".txt" as the File property in a Data Set, File Set or Lookup File Set stage. It must be one of ".ds" or ".fs". Don't worry about the ".v" - this will take care of itself once you fix the ".ds" or ".fs".

There are no hashed files in parallel jobs. A segmentation fault is an attempt to access an address in memory that either does not exist or is one that you do not own. It can, for example, be caused by trying to shoehorn 15 characters into a Char(12) column, or sometimes by trying to allocate null t...

Cartesian join (your second post) is a different requirement from union all (your first post - "append the two files in a single file"). If file1 has N lines and file2 has M lines, the approach I suggested will yield (N + M) lines. A Cartesian join will yield (N * M) lines. Which do you want? You ca...