Hash File ?

marc_brown98 · Post by **marc_brown98** » Mon May 10, 2004 3:07 pm

I have a job that is using 2 hash files and 1 sequential. It basically reads the 2 hash files for lookup purposes and then inserts a field from each in addition to the output of the seq. file. For the life of me I cannot get 1 of the hash files to score any hits, even though I know it has a key match on the seq. file. Has anybody seen this or have I just been staring at a dumb mistake too long?

kduke · Post by **kduke** » Mon May 10, 2004 3:39 pm

I would check your metadata on the job that built the hash file and compare it to the job reading from it. Also check to see if your keys have been trimmed. You may have spaces on the keys and not know it. Just a couple of ideas to get you started back.

marc_brown98 · Post by **marc_brown98** » Mon May 10, 2004 4:02 pm

Thanks Kim, I will check all of those things out. It has to be something like what you said. I changed the job to use the db table that the hash file is built off of and it works fine.

ray.wurlod · Post by **ray.wurlod** » Tue May 11, 2004 2:25 am

What are the data types of the key columns? What are the data types of the input columns you are using as reference key expressions? These must match.

The best approach is to make them all VarChar, Trim the key when loading the hashed file, and Trim the columns making up the reference key expression.

It's safe to make them all VarChar. The only data type in a text file is "text", and the only data type in a hashed file is "none".