DSXchange

Hi Ray, It seems the string_trim function only removes leading or trailing characters. It has no effect on the characters in the middle of the string. As for avoiding a Transformer - The files that we are loading do not all contain the same columns. By using schema files and by passing modify stage ...

This is essentially what we ended up doing. The jobs were already set up to allow sed commands to be passed in for pre-processing. We used a more complicated regular expression though, to avoid destroying the comma delimiters in the csv files. We also tried specifying a c_format string, taking from ...

I was hoping to avoid using a transformer. I was also hoping that since they allow the use of date format masks for date conversions in the modify spec, that they would also allow the use of number format masks. The files that we are loading do not all contain the same columns. By using the schema f...

We have a DataStage job that reads multiple delimited-file formats using a sequential file stage and various schema files. We then use a modify stage to rename columns so that they line up with an interface table in our database. We are having problems reading certain numeric fields from the delimit...

Is there a way to automate the provisioning of Standardization classification files and lookup tables, or classification overrides? Could these be generated at runtime based on a database table of word replacements? Our users need to be able to add new tokens for classification and abbreviation repl...

Thanks Ernie, this was mostly a learning exercise for myself to get familiar with the Assembly. I was able to move past the error by changing from Strict to Minimal Validation in the Parser step. Somewhere in the documentation I saw a blurb that indicated minimal validation treats all data types as ...

I'm hoping this is a simple question, I couldn't find anything in the forum, documentation or the almighty Google though... I'm trying to use the XMLStage to transform some rather complex xml documents. The document has several lists that are related by ID and IDREF typed elements. I can't get past ...

We're running IIS8.7 with IADB on our services/metadata tier DB2 LUW instance. DB2 is giving us headaches, so we're considering setting up IADB on Oracle. The planning and installation guide talks about running scripts to create the metadata repository database (XMETA), but says nothing about buildi...

I would like to take the concept of the MNS country specific address standardization rules and apply it to Customer specific product standardization rules. I could supply a customer code, just as a country code is passed for MNS addresses, and have a different standardization rule executed based on ...

That billion row question is the exact one I am asking myself now. I've read back over the documentation and can see how we can restructure the billion rows by combining records to have vector columns. Has anyone ever actually used a vector column in a match command, and what were the benefits (perf...

thanks Ray. Our data is in relational tables, so there's a Company table and a Contacts table. In reality, they're not company and contacts, those are just simple examples. We also have other multi-valued attributes of the business objects that we need to match on. Some of these need to be used in b...

I'm not sure if the post title is accurate. Maybe that is why I'm not getting any search hits. We have a requirement to match on attributes that can have multiple values, e.g., the multiple contacts below Reference Set - Company A, Contacts {Person A, Person B, Person C} Company B, Contacts {Person ...

Bingo! Thanks

The documentation for QS says that there are several predefined rule sets located on the "InfoSphere Information Server file directory." This includes a GENPROD rule set that demonstrates processing product description data. I have searched my client machine and both engine and services ti...

Point taken. Thanks. We still have a lot to learn about this platform.