DSXchange

we have an outstanding ticket with IBM about the H-stage dropping a record here and there at random. We can run the same data 3 times and get {one record missed, all the data, different record missed} as the result. Whether you are seeing something similar or not, I can't say from your problem state...

If the data isn't suited to partitions "as is", run some of it through a Checksum stage and use that field as the partition key. It should give solid results.

do you need some sort of logic to round up? That is doable but I have never heard of anyone doing that for your age in years. you can add a +1 if the # of days until next birthday is within N days, not hard to do...

Answering the subject and not the body... You can drop your extract to a dataset, count the # of records in the dataset with a routine, and use that to gracefully exit or proceed depending on having data or not. This is nice for efficiency as well for complex jobs that do a whole lot of nothing lik...

there is something that works for many small simple DS too: orchadmin dump -delim ',' dsname.ds outputname.csv but this is a quick fix for simple problems; there are times when it won't be sufficient, and its terrible for anything big. it will also lose the schema. But you can export a table def for...

a mix of substring operations should get you there, eg newvalue = left(oldvalue, 2):'S':right(oldvalue,4) newvalue2 = left(oldvalue, 2):'SS':right(oldvalue,3) ..etc do that in a stage variable is one way, eg if you wanted each value in a new column. use a loop if you wanted each one as a new record.

it is better if you do not do this. In an emergency, you can move the dataset AND its underlying files AND change the 'header' file (the one you interface to in datastage when you open or write a dataset) to point to the new locations. The header file is in some sort of unicodeish partial text parti...

to just remove the records entirely you can do it with a single convert, or possibly even more efficiently with 'alpha' if that fits the data need? A single convert, you convert all the valid chars to nothing and then check the length of the remainder. double convert fixes the data, which is a diffe...

if you do anything manually more than 2 or 3 times, its time to automate it. If none of the datastage ideas works for you, write something that will do the surgery for you. And its not always that much code. The one I did to transfer job params from one job to another took like 2 hours or less from ...

We do it that way. I can't comment if its the most secure way or not. I just checked 11.5 and you can extract the password with the simplest imaginable attack. Nothing even interesting, just pure datastage jobs with the right settings will print the password in the logs. So I would offer that encryp...

dsx hacking is really only useful for very simple things like copying job parameters from one job to another ( a feature that is sadly lacking), or the occasional overarching search and replace of 1 word or something. I don't understand enough of your design, but it seems to me you could use a share...

do you need to handle leap years? that aside, transformer loops create a record per loop, and are handy but you already have all your records as you want them (right?) and are just grouping them so it may not be a clean solution (you can toss extra records, but its unnecessary work best avoided to g...

so you split to two, one of those goes into another seq, that one is set to any (?) and the outer one is set to all? This would mean that the any one can finish one thing, pass that to the all outer, and the all would be set to true for that side, so if the other side finishes as well, some of the '...

Heh... about that... Look in your repository under stage types. There are a number of stages in there that are not in the palette. Its under Parallel/Processing/BasicTransformer Be careful with it. I know for sure that it cannot handle timestamps with microseconds, it loses the extended data. I don'...

you can do it in a parallel job too if you use vb transformer or one of the others, I think there is a java one etc? most of the time this sounds like a parallel job (with a parameter) invoked from a sequence that has either a routine or user-variable stage to cook up the text. For file name that ju...