DSXchange

Could that be done with input sorted by PR_CODE (ascending) and running through an Aggregator stage, grouping by the other columns (which seem to be constant for variations in PR_CODE), and using Last as the aggregate function on the PR_CODE column? If my assumption holds, this should offer a simple...

New stages are announced in the readme documents that accompany the software, and usually in marketing announcements on the Ascential web site. Please don't demand "ASAP" response from the volunteers who post on this site; some of us find it offensive. Others of us would be happy to sell you "urgent...

Row count can be had from interrogating the job itself, using DSGetStageInfo or DSGetLinkInfo function. It would help if you have a standard naming convention for your stages and links. How you determine "do not match" would have to be a function of your job design logic (unless mismatched rows are ...

It's not a DataStage limitation; DataStage can handle VarChar up to 64K characters. Are there any warning messages in the log about data being truncated? Create a job that writes to a textfile, so you can determine whether the truncation is occurring during SELECT or INSERT/UPDATE operations. Verify...

(From Administering UniVerse , Appendix E): Fault type 6 is a SIGIOT (a signal generated by an IOT instruction). This might, as Ken suggests, result from an attempt to extend a hashed file to 2GB, or it may be something else. The IDENT1 stage is not, in fact, a hashed file stage; it's an invisible T...

chulett wrote:First solution Ken mentioned, Ray - attention to detail, lad!

Ken's post wasn't there when I posted - sometimes it takes a long time for my posts to get through.

Hawk? (8.x?)
Rhapsody? (9.x?)
Another forum for Sorcerer? (the new UI for all Ascl products)

You might be able to use a UNIX command, such as sort -u (if my memory serves), to remove duplicates.

Thank you for posting the solution. Next time someone searches the forum looking for an answer to this problem, they will be saved some time and effort. Yay!

Yes, but it's not totally reliable. By its very nature CRC32 introduces a 1 in 2**32 risk of missing identifying a duplicate. A hashed file or some form of searchable list (in a Routine) are the more common ways of removing duplicates. Or you could invest in PX and use the RemoveDuplicates stage. Or...

Sorry I couldn't give you the answer right away; I was already at work (in New Zealand this week). But, hey, you now know another place to get UniVerse answers. This was the other half of the celebrated Clif Oliver mailing lists.

But did you check the actual hashed file's metadata (not what's in the DataStage repository)? Assuming you have a VOC pointer (search for SETFILE) execute the command LIST.DICT hashedfilename @ SQLTYPE 'UPDATE_STATE' or, if you prefer SQL, SELECT ID,CODE,LOC,CONV,NAME,FORMAT,SM,ASSOC,SQLTYPE FROM DI...

In the current release the DataStage repository is closed; that is, its table definitions are not in the public domain. The Reporting Assistant can dump information from the DataStage repository into a first-normal-form, with your choice of target database. Within DataStage, the repository makes a l...

DSR_RECORD is usually invoked when refreshing something in the client. Check for timed-out connections and invalidated repository indexing.

How are you performing the search? Yes, it can be done in a Server job, in a number of ways. The obvious one, the Folder stage, is only good for a small number of small files, due to a known bug. The search can also be performed in a Routine (perhaps invoked from a Routine activity), which is probab...