DSXchange

I corrected the code, it seems the "LOCATE FindDate IN TimeVariantStartDates" should have been "LOCATE FindDate IN TimeVariantStartDates<1>". The choice between @VM and @FM was subtle, as the point is a delimited array of values. Somehow I mixed up the two and never posted the corrected logic. Sorry...

The correct term is a Spiral SDLC, which means a series of compacted Waterfalls that continuously feed and build on each other. Rather than a single, 18-month Waterfall SDLC that's supposed to build everything, you set small goals and in 90 days or less turn around iterative enhancements and improve...

Why not simply put the Time dimension in the warehouse? It's a relatively small, static table, what's the harm. Putting it in a hash table, do you mean hash file? If you're using PX, it would be a dataset. The PX and Server aggregators are not your preferred solution for complex aggregates. SQL is q...

Your issue is a simple one, something your architect should have solved quite easily using a Time dimension. Read Kimball's Data Warehouse Toolkit, he covers it extensively. A table that contains every date from some arbitrary point in the past, such as 1700-01-01, thru the near future, say 2099-12-...

Go to the project directory, subdirectory RT_BP113 (DataStage Job 113 Phantom message). The error message indicates that a function call in the transformer has a variable unassigned. It also appears that variable relates to a variable subroutine name (Unable to load subroutine) which is attempting t...

manteena wrote:An XML job report will give you the file name with Directory name of the source or target file stages....try to read this report for filename...

NO CHEATING

You might just have the winning suggestion! :D

Being a Horracle person and not a DB2, either use a serial assigning process that guarantees a unique identifer. In Horracle, you'd use a Sequence. If you go the stored procedure route, select the max from the batch table, add 1 to it, and try to get a lock on that row. Keep incrementing and trying ...

Why does this sound like a variable value everytime the job runs, but is hardcoded? This makes no sense. Is someone changing this filename, compiling it, and running it? Can you at least get them to write it to an empty directory and then assume anything in that directory is the file you want?

dilsere2 wrote:The job log doesn't tell what is the output file name that you have defined inside the stage....

It does if the file to be opened can't be opened, or there's some other catastrophic issue, hence my reply that the file name only shows up under tragic circumstances.

manteena wrote:Hardcode the filename as the stage name, then get the stage name with DSGetStageInfo.......

Cheater, I assumed the file is fully qualified and stage/link names can't have slashes, dashes, dots, just underscores.

Are you on a single SMP or a cluster? How are you resolving duplicate PIDs across nodes, because of Partition? You're going to have a big integer there because each field has to be fixed width and then concatenated to be the larger integer, very messy. Why not use something like a batch number inste...

What do you expect the PID to tell you? The controlling OSH script or the current process running your function?

Are you attempting to log messages to the job log? What's the underlying goal?

There is no API for requesting such information. The only thing you can do is trick the job into telling you this information thru the job log, which you can then query using the API DSGetLogSummary. Unfortunately, most of those messages tend to be fatal. I think you need to revisit why the file nam...

Is the server side a fresh install? If the server side is NOT a fresh install, then the problems are either: 1. Your new PC install is bad 2. The project is LARGE and you're not waiting long enough for it to open 3. The project is corrupted #2 is easy to figure out, can anyone else connect to the pr...

Are you asking to search the job designs for a piece of text, and return a list of jobs, maybe even down to a stage->link->column->derivation? Do you have MetaStage? If not, then you're going to have to do it yourself. The fastest method is to export all of your jobs and search the .dsx file. Otherw...