DSXchange

This is, unfortunately, currently still a restriction in Px; you can view data on oracle when it is being used as a source, but not when it is a target.

Prashant,

looking at your code, you would need to quote 'JAN' and if you only have dates in January and July then you will be set. But what happens with 'MAR'?

It looks like you might have specified Windows-type termination when writing and then you are reading it with Unix-type termination.

Nripendra,

I'm not going to supply you with your code or solution; but why can you not use a derivation in your transform somewhat like "IF In.Line[1,9]='Migration' THEN 'NewL1':CHAR(13):'NewL2' ELSE In.Line"

The output shows that the conversion wasn't successful, either because of a bad format or a bad input type date. If you are getting a Date type field from a database, you would first need to DateToString and then a StringToDate with the new format.

In order to replace lines in a sequential file you will need to use a temporary interim file, you cannot easily do this type of selective overwriting in a sequential file - in fact the only way to do this would be if the records are all of identical length. Write a job that does a Sequential -> Tran...

Take a look in the Documentation and also in this forum at the StringToDate and DateToString functions.

Hello Joszef, The message itself means that a process received an interrupt signal, and this occurred and was processed from within a system call - normally the interrupt is disabled during system calls but this is not always the case. In order to find the cause we would need more information. What ...

Ray - Yes, I do I can't think of a non-trivial job in either Server or Px where I didn't at least once have to do some sort of a TRIM - even if it was just a prophylactic measure.

VasanthRM,

I hadn't used that option so I did something really exceptional - I tried it.

I received a list of all BATCH generated messages written to my logfile for a sequencer job. It would seem to include all messages generated by the system when controlling other job calls.

Ravi, declare the record length as 106 by adding a char(18) dummy column to the end of your existing Pl/1 imported declaration. Run the job and output to a delimited sequential file. Then go through the columns one-by-one until you find the location where your imported declaration and the data begin...

Nripendra,

you will need to narrow down your question quite a bit. What do you mean by "clean"? What kind of file, what kind of additional data?

The Px Conversion to use in your case would be StringToDate (page B-12 in the Parallel Job Developer's Guide ). I'm away from my Px connection right now, but I think it returns an empty string or null when unsuccessful; you will have to try this on your own. I still have more trust in the server sid...

If you wish to take a date and see if it is a valid one, you have some Px conversions that will return an empty or null value when converting invalid inputs; this can be used as a type of validation. The Server variant of DataStage has two very powerful functions called ICONV/OCONV which will do thi...

srekant,

that's nice, so your input format is different from your initial query. Now you need to use a date conversion algorithm instead of a string function. Have you looked at the date conversions that Px offers?