DSXchange

We did this in a job and wrote the results to a hash file which you can read with UtilityHashRead. You could do the same with

cmd = 'select max(column) from MyHashFile'
execute cmd capturing output

You would need to parse output. The result is probably in output<2>.

We did this as a way to shorten our select on rebuilding our hash file to calculate whether or not our record is a new record or an update. You need to use a UV stage. This way you can use SQL. So we SELECT #ColumnName# FROM #HashFileName# And write the results to something like SDKsequences. This w...

Sure why not. I think a lot of people have these issues and maybe they have a better solution. I think what we did is kind of cool. Anyway here is what we did. We had a day dimension table. You could easily create one. You can join both sides to this table and create a record for each month. That wa...

Currently we have employees assigned to a team to service a customer. There is also another team of 3 assigned to manage their planning, buying and general questions. This second team gets a bonus based on their role. Example: BuyRep 50% for 2003 PlanRep 25% GenRep 25% Lets say our customer is Ford ...

Charley

I posted a few queries that you can run in Reporting Assistant to get the differences in DSProperties. I think MetaStage can also generate a report showing job changes.

Charley I have added sequences to the tables in Reporting Assistant. It works through DwNav. I noticed there is a bug in it in version 7. I am not exactly sure yet on what is going wrong. I thought about posting some of this code. The only thing I extract are the depend jobs. I think the only other ...

Thanks. I knew about REUSE and most of these but not IFS.

Ray

Didn't work. These fields are multivalued so field does not work.

Number 1 will work. Number 2 will not. The DEPENDS is not a reserve word. It is whatever we want to call our association.

There are multiple record types in the dictionary. Field 1 tells the type. If there is a PH there then it is a phrase. It is just as if you added everything on line 2 to your command. This is a special phrase required by an association. An association is how you define a nested table. A nested table...

Ray Dig it out. I tried it is ugly. I made a mistake. INSERT INTO DICT DS_METADATA (FIELD, CODE, EXP, NAME, FORMAT, SM, ASSOC) VALUES ('SQLDataTypeCode', 'D', '15', 'SQLDataTypeCode', '3R', 'M', 'MvCols'); Here is the case statement: case SQLDataTypeCode = -7 SQLType = "BIT"...

DataStage stores its metadata in a hash file named DS_METADATA. They give you very few dictionary items to list the contents of this file. Here is a way to add more dictionary items to use in SQL statements or LIST or SORT commands. INSERT INTO DICT DS_METADATA (FIELD, CODE, EXP, NAME, FORMAT, S...

Ray has explained this before but when you SELECT * then it uses the @ phrase or the @SELECT phrase to get a default list of fields. LIST.DICT HashFileName Will describe all the fields available to use in a SQL SELECT statement or a Universe LIST or SORT statement. If you do a search on @SELECT then...

Thanks Ray. I never used SQL at any Universe site. I can see several advantages with it. I know lots of Universe experts and none of them know Uv SQL. Most of them are DataStage developers now. Your SQL examples are very useful and educational.

The dictionary items need to be in DS_JOBS to get the list based on job name. The job name is the key to DS_JOBS. Field 5 is job number. The key to DS_JOBOBJECTS where the depend jobs are stored is "J\xxx\ROOT" where xxx is the job number. Now you have a complete solution. Ray how do you use a WHEN ...