DSXchange

Is there any preceding message about not being able to open the file named in the Sequential File stage? The ds_seqgetnext function is the one that asks an output link from a Sequential File stage to deliver the next row from its text file. Is this a recurring error (that is, occurs on every run)? I...

Given that the error message is from the ds_seqput function, it can be seen that the error has occurred when a Sequential File stage is attempting to write (not read). If the Sequential File stage is writing to a named pipe, this can mean that the reader process on the other end of the pipe has stop...

The "two tools" solution is excellent. DataStage (6.0 or later) can reformat the data into a fixed-width format required by INTEGRITY (within these fields data can be free format, but rules must have been created for making sense of these). From DataStage you can invoke INTEGRITY through a stage, wh...

How was the file produced? Can you import column definitions for the text file either from a COBOL file definition or a DCLGen description? This should tell you what *should* be in the text file, from which you should be able to Load the metadata directly into the CFF stage to process that data file...

This one's going to be difficult to track down - I'd be involving your support providers ASAP and prohibiting each from blaming the other. Numeric precision for DataStage is set by the EXACTNUMERIC configuration parameter (in the uvconfig file). Minimum and default value for this parameter is 15. Th...

When you buy the tool, you will discover that DataStage, too, has a rich set of documentation. Some manuals install with the client software. Manuals for the optional ("plug-in") components come with those components; you can install these with the regular set if you wish. In addition, DataStage has...

Ah, yes, but Jim's solution does not require you to create a Routine. (There is no capacity for loop constructs in column derivation expressions - or any other kind of DataStage expression for that matter.)

Credit for this one belongs to Jim Tsimis, one of the original technical people involved with the original development of DataStage at VMARK. It is hugely faster than any other technique for this result.

Replacing the pipe characters with @VM (a system variable) can be achieved with the Convert function. Process the whole line as a single column, apply Convert("|",@VM,InLink.ColName) as a transform function and write the result into a hashed file using sessionID as the key value. Use "normalize on" ...

Are you trying to import into an older version of DataStage from a file exported from a more recent version of DataStage? Take a look at the export file itself. Do all the ODBC stages have a property called STAGETYPE? Were any of the jobs imported earlier and containing an ODBC stage successful? Is ...

Assuming your output is also a sequential file, try the following column derivation expression. You will also need a stage variable (called EOL in my example) that has been initialized to the correct end-of-line sequence for your platform (that is Char(13) for UNIX, or Char(10):Char(13) for Windows)...

You can even improve on that. INITIALIZE the stage variable with a call to GetMaxKey. The stage variable expression in the Transformer stage simply assigns its current value to itself. The expression that generates the value for the surrogate key column on the output link has the form StageVariableN...

In DataStage 6.0 server jobs you can also accomplish this by splitting the data stream with a Link Partitioner stage. Active stages (for example Transformer stages) on the separate links will execute as separate processes. It is also possible (recommended) to take advantage of link row buffering to ...

A couple of points about Ken's code. First, if you're using NLS and your TIME locale category specifies something other than Sunday as the first part of the week, then you will need to adjust his algorithm appropriately. Second, since DataStage follows the convention of D_filename being the file dic...

Open the doc_tool.mdb in Access, get past the startup macro, and inspect the standard queries and reports that report on Tables. This will pretty much give you all you need by following this approach. It is also possible, with knowledge of how things are stored in the repository, to use DataStage BA...