DSXchange

The simplest way to explain it is to set up a partitioning key ahead of where you want to partition the process flow in a transform - perhaps called PartionKey and set to 0 or 1 (could also be used to partition at higher granularity if necessary by using more values and using a modulus partitioning ...

Thanks Mike, Im not familiar with the link partitioner yet, I only saw ae brief few minutes on it at the launch - does this mean you can effectively put a constraint in the link partitioner for when to use a particular transformer? Not haveing looked at the stage my impression was that it was only r...

With DataStage V6 server jobs you could use the link partition/link collector stages to split the process based on whether to do the lookup or not, do the lookup on one leg and nothing on the other and rejoin the flow. It may not be as elegant as the XE/390 solution, but it should provide the functi...

You can use the basic function SYSTEM(32) to return the UV/DS home directory so for something generic c.UVSep = / ;* or for windows c.Exe = ;* or .exe for windows mkdbpath = SYSTEM(32):c.UVSep:bin:c.UVSep:mkdbfile:c.Exe Phil Walker Altis Consulting Limited Auckland New Zealand. > -----Original Messa...

I attended the DS6.0 Viper launch last wednesday and asked the very same thing there after they had mentioned theyd implemented conditional lookups for mainframe jobs. I believe the answer was that this is where they percieved the need most so thats where they implemented it first. I had the same *a...

If the number of columns you need transferring to rows is fixed then you can also achieve this via the pivot stage. Dave. > -----Original Message----- > From: Victoria.A.Oliver-Macklin@gsk.com > [mailto:Victoria.A.Oliver-Macklin@gsk.com] > Sent: 28 October 2002 15:11 > To: datastage-users@oliver.com...

that works fine. Thanks -------------------------------------------------------- Tom P. Nielsen Masterdata Reg.nr. 4853 / HH55.1.3 Phone: +45 33335223 / Plus nr. 435223 wolfgang.huerter@gerling.de 30-10-2002 10:53 Please respond to datastage-users To: datastage-users@oliver.com@SMTP@Meriposti cc: Su...

Hi Tom, with the Unix-command `cat /.dshome` you get the "/usr/dstage/dsadm/Ascential/DataStage/DSEngine"-Part of your directory. Then you only have to add the bin and you are there! Wolfgang > -----Original Message----- > From: tom.peter.nielsen@nordea.com > [mailto:tom.peter.nielsen@nordea.com] > ...

Hi, Ive made a routine that generates Hashedfiles in an optimized structure, but i could need a little tweak to improve it. atm. im using a predefined path when running mkdbfile - is there a better (more dynamic) way to do this. so far: sExecuteString = "/usr/dstage/dsadm/Ascential/DataStage/DSEngin...

Dear Sirs, While trying to read data from FileMaker Pro 6 through ODBC we get the following error: DSR.MetaGeta(GET.TABLES)(SQLTables): BCI Error: SQLSTATE=S1000,CODE=16394,[DataStage][SQL Client][ODBC][FileMaker][ODBC FileMaker Pro driver][FileMaker Pro]Connect failed SQLSTATE=S1000,CODE=0,[DataSta...

Started to play with DataStage 6.0 by installing it on our test cluster and was delighted to see that it finally supports Conditional Lookups. (oh, and the new Designer with its spanky icons are very cool) However, on digging deeper, the rub seems to be that they are only supported on a *Mainframe* ...

Ray Thanks for the advice. As suspected, yes the upgrade caused cat to behave differently. However by changing the command to "cat -s file2 >> file1" we get the same result without error, which to the UNIX gurus may have been obvious. Something I am continually learning about. I guess the frustratin...

We have just upgraded from 4.2.2r2 to 5.2.1 on Windows 2000. We did not see any of these during testing, but they are happening in some of our frequently run jobs (ie run every 5 minutes). I would see at least one or 2 of these every day. Here is a sample error message: DataStage Job 826 Phantom 246...

Hi Vicky, an other way is to write a simple routine or even transform: ----------------------------------------- Convert "," TO CHAR(10) IN Arg1 Ans=Arg1 ----------------------------------------- With this a list like: item1,item2,item3,item4.........itemz you get as output: item1 item2 item3 item4 ...

Hi Vicky, you can concatinate any Chars you want. First you have to check which carriage return you need. CHAR(10) and CHAR(13) or only CHAR(13) or only CHAR(10) If it is CHAR(10) and CHAR(13) you write for item1 item1:CHAR(10):CHAR(13) and you get a carriage return after Item1 because of this add-o...