DSXchange

Please move this to DataStage Server section.

I assume you downloaded KgdGenHtml.zip. You need to unzip it. There is a dsx that you need to import into DataStage. There is a sequence named SeqGenHtml. You need to run this. It has a parameter called CategoryToDoc. If you type in a category then it will generate an html file for each job in this ...

We leave comment lines in Oracle and it works fine. I am not sure about DB2 or other databases.

Anupam

I am sorry. I reworded the above answer. What I was trying to say was CHashedInput really outputs to a hash file. The way ASCL named the OLETYPEs was confusing.

Class ETL 101 1. I would go to gantthead.com. They have a really nice source to target spreadsheet. I would fill this out before I start on any ETL work. 2. I then put this in a database and map source columns to target columns. I use Access. 3. Next extract system tables from all sources and all ta...

The ODBC drivers for Lotus Notes are the slowest drivers out there. You need to import and export data into Notes with Notes programs. Create sequential files and both sides. Going into and out of Notes. Save yourself the trouble of not trying the ODBC drivers.

What we think as output is OLETYPE has input on the end like CHashedInput. Here are the 2 queries above together. select DS_JOBS.NAME AS JOB_NAME FMT '35L' from DS_JOBOBJECTS, DS_JOBS where DS_JOBOBJECTS.OLETYPE in ('CHashedOutput','CHashedInput') and DS_JOBOBJECTS.OBJIDNO = DS_JOBS.JOBNO an...

Your query is wrong. The NAME field in DS_JOBOBJECTS is normally the link name. So unless you name your links and hash files the same name then this will fail. I would use this query. select DS_JOBS.NAME AS JOB_NAME FMT '35L' from DS_JOBOBJECTS, DS_JOBS where DS_JOBOBJECTS.OLETYPE = 'CHashedInput' a...

That is the trick. In the hash file you do not treat this field as multivalued. You do so only in a UV stage. This allows one job to create the multivalues and another job to use them. Most of the time I would only use this in a lookup. I would write a routine to do it. If you look at Ken Bland's SC...

You can get evaluation copies for almost free for 30 days. Ask your sales rep. There is a perfomance hit for each job running but unless you are really maxing out the system then the perfomance loss should be small.

Well, Tony they all said it did not work. You could of said something sooner.

Ken

Look at this topic:

http://www.dsxchange.com/viewtopic.php?t=88934

You can do the rollup in a job. You can also use @VM and not "|". Makes this a lot easier.

Bobby

You are changing the primary key. It has to do a table scan the way your update is written. Either delete the original record and insert the new one or create an index on order_id.

Michael It can be done. I just wrote a job which takes a source. It writes to a hash file. It does a lookup on the same hash file. If the lookup was there then the output column was lookup.col1 :@VM: In.col1. You can aggregate to multivalues to a hash file. JobsByCatHash | v UvStage -> Trans -> Jobs...

Bobby

You need to add another column for the key. This will update the whole table. An update will not update the key. It will use it in the where clause.