DSXchange

That is a known gap in the functionality, which is why others have developed "tables used in this job" tools. Search the forum to find reference to these.
It can be thwarted by ill-disciplined metadata management and by parameterized table/file names.

Welcome aboard. :D

This is most easily done using a sorted file and stage variables to detect changed values (and reset the sequence) or increment the sequence.

Search the forum for actual examples.

You may need to trim the command output of extraneous field marks and leading and trailing spaces.

Check out the mnemonics for error codes (DSJE...) in JOBCONTROL.H. Also in this file you will observe that the mnemonics for function arguments must be all upper case, so that DSJ.JobStatus is not a valid argument; you need DSJ.JOBSTATUS.

It's called "DataStage Engine". It was derived from UniVerse which, in turn, emulated Prime INFORMATION. Both UniVerse and Prime INFORMATION are now owned by IBM.

Names are case sensitive, so DSLinkrowcount does not exist and that is one big difference.

Otherwise the previous answer is good, but incomplete. DSJ.LINKROWCOUNT can be used as the fourth (InfoType) argument in the DSGetLinkInfo() function.

Do you have a consistent naming convention that allows you to identify where the "input", "output" and "reject" links occur? If so, you can use the DSGetLinkInfo() function to retrieve the link row counts.

There is no "default path". You must use a Data Set stage to create a Data Set. The File property must be a pathname with the file name ending in ".ds". If you do not provide a pathname, the ".ds" file is created in the project directory. This file is the Data Set control file; the actual data files...

Lookup File Set pretty much locks you in to its own, stored, schema. This is also why you don't get the opportunity to specify a key in the Lookup stage - the key in the Lookup File Set is already known.

You create a routine to harvest what you require. Everything you need is captured by DataStage. You can download ETLstats from Kim Duke's website which will provide some prototypes. There are two sets of functions in the DataStage API which map onto options in the dsjob command. One set retrieves a ...

Please do not double post. It wastes everyone's time.

Responses should be made to this post

I think you will find that there's no such thing as a Surrogate Key Generator stage (however spelled) in DataStage TX. Please post any question you may have in the correct forum. You also need to be alert to the implications of running a job in parallel.

Thank you for sharing that. However I think you will find that there's no such thing as a Surrogate Key Generator stage (however spelled) in DataStage TX. Please post any question you may have in the correct forum.

There is no limit on the size of a Sequential File that DataStage can read. From the other posters' experiences I would suggest that the delay is not in reading the text file, but elsewhere in your job design. You can prove this easily enough. Construct a job as follows. Sequential File -----> Trans...

You might also contemplate indexing the search fields in the hashed file (the UV table).