DSXchange

I'm not sure, put the path doesn't look good to me. Particularly the 3rd character

ckporter wrote:...
/E=/Ascential/DataStage/Datasets/OUTWARDS_RI_TRANSACTION_NAME.ds.marcw.APHRODITE.0000.0000.0000.10dc.c6f6873b.0000.85ce82bd failed: Not a directory
...

The first one performs only one variable assignment, the second one needs to do 2; so the first will be more efficient. As Ken has stated, they will in their current form give different results.

The INDEX() function takes 3 parameters.

Krazy used the colloquial method of giving you the bad news. On a technical level - once you have deleted a job the information is permanently removed and is not recoverable apart from restoring a backup .dsx copy, if available.

If there is never text before the "[" then it is the best solution. Good thinking!

Kris - I was referring to Joyson's response, we ended up posting at about the same time.

For some reason the transform compiled code is missing. Try to recompile your job and see if that generates an error or if it can be run afterwards.

That works, but doesn't cover rows with not "[...]" text. How about IF INDEX(In.Column,'[',1) < INDEX(In.Column,']',1) THEN In.Column[1,INDEX(In.Column,'[',1)-1]:In.Column[INDEX(In.Column,']',1)+1,999] ELSE In.Column If you want to make it handle n-occurrences in a st...

You will need to split your string into a date and time portion, then use ICONV to convert both parts seperately. The date is a number-of-days offset while the time is a seconds-since-midnight offset.

The T30FILES parameter is a DataStage internal value, not a UNIX one. I'm not at a client to ascertain the actual limit, but 500 is not a system limit to T30FILES and can be set much higher.

The concept of "client and server" is not specific to DataStage. In DataStage you have a server (typically on a UNIX or Windows) to which you attach using clients from a pc - the Designer, Director, Manager or Administrator and perhaps the telnet session into TCL might count as well. The brunt of th...

Sarven, now you might begin to realize why the recommendation is to put all hashed files used by DataStage jobs into a different directory from the projects themselves - there is not automated way to copy all the hashed files and their contents from the project directory. Do the files have some comm...

The compile has the additional advantage of clearing out log files - one thing that sometimes causes projects to "bloat"

You need to talk to the DataDirect people, as it is their driver that needs to understand the new format.

your SELECT is hanging because the job is still running and is holding a lock on this record. This will not happen if you change your job to do a HASHed file read instead of a SQL SELECT.