DSXchange

doing it with xslt will not require a lookup to happen - unless 2 different xmls are being combined on element or attribute data( liketeh first couple of possts by OP) . If an XML transformer is used , even an xsd is not absolutely necessary if the later posted example by the OP is what they are loo...

Hi There , you cannot use XML input stage for this in my opinion - especially if attribute names can have many other 'undefined' values. You would be better off defining a small xslt and use an XML transformer stage. Try This <xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/...

I not sure if im helping you , how are you invoking the first job in the control ? Are you using dsjob -run or dsjob -run -wait ? The latter will return a status only after completion of the job being invoked and would be the preferred method. Polling with a while loop may cost you more cpu cycles!

We have successfully set up Ops Console with v 8.7 with SQL Server hosting the ops db on a separate database for it. Do not use an existing database. The same server also hosts xmeta and iadb. Did not install Quality Stage - so no Match Designer DB

Good Work. The DS Basic syntax highlighter is available for Text Pad already and I have been using it to debug code that is very long.

JOBNO =nnn is an internal identifier that maps to respective RT_BPnnn [ Basic programs - like Basic code for your server transfomer stages ] , RT_BPnnn.O [ Object code ??? ] , RT_CONFIGnnn ( Stage & Links list + other info ) and RT_LOGnnn(Job Log ) files for that job in the UV Metadata reportory...

What are the permissions on the object file ?

We have a remote Dataflux server hosted on Windows Server 2003. Since Dataflux is a SAS product , Im wondering if we could call Dataflux jobs using the SAS stage in Parallel Extender ( as a PROC step) . While I'm personally not to confident of this approach , I may get some answers posting on a wide...

Depending on the version of SQL server you have , Change tracking can be enabled at the database level . Once you have done that . I think its an easy task to get only newly modified records( Adds/Updates)

Here is a good link for
Change Tracking in SQL Server 2008

I agree with whatever Craig says.Additionally , When running in sequential mode , try using a sort-merge collector ( on table keys) if its applicable to your stage. Duplicate records for table keys can cause issues also.

The TFM 1 in the logic you use needs read ahead buffering. the Output link is much faster as your are doing row splits and incresing number of rows due to logic . If performance is an issue , you can still go with promote subrecord stage - trust me it works. Use a tr -d command to remove your operni...

I remember reading somewhere that only one Multi row lookup was allowed PER lookup stage - so you may need 2 cascaded lookup stages with a Mandatory Copy operator ( FORCE=TRUE) in between them so that both lookup stages will not undergo operator combination. Draw an output link from copy to see your...

Which machine ? Client machine or Server machines ? The Install guide , incidentally also guides you through the uninstall process - some files may be left behind and you need to manually clean them. For your server , you will have to use the automated uninstaller and do some manual clean up steps p...

Yes, I heard you on the range lookup issue also. Another place where a compiler is a must would be the custom compile flags in the lookup stage. The compiler license is being checked with IBM. (point to ponder : Looks like the compiler licensing is also based on number of individual cores within a s...

An Update : Yes i found much to my dismay it does use a transform operator . I tried compiling a job with SCD stage and here's the error I got ##W IIS-DSEE-TBLD-00000 17:29:56(012) <main_program> Error when checking composite operator: Output from subprocess: sh: /usr/vacpp/bin/xlC_r: not found. ##I...