DSXchange

Just discovered the solution:

In the CXNM Name standardization, set Derivation Arguments to the number of tokens desired to be included in the comparison string. IBM indicates that the max for the CXNM standardization is 8.

It seems that when organization names are derived they produce a compare string that contains a maximum of four non-anonymous tokens. For example, let's say we have these two records: ACME AIRCRAFT PARTS CO DBA BIG DOG MECHANICS and BIG DOG MECHANICS then, after elimination of anonymous values like ...

Is there a way to remove all data from the MDM DB for a single Entity type? We have multiple Entity types loaded (e.g. Person and Employer) and one of them (e.g. Person) was loaded incorrectly and we want to remove it without having to truncate ALL data for ALL Entity types and then realoading every...

Can someone explain to me or suggest a tutorial that explains how to do a delta load with MDM ver 11? I have gone through the IBM bootcamp course but that topic was not covered. I have the bulk load process down pretty well but need to know how a delta load differs. Thanks!

I am a newbie to MDM. It appears that we may not have our configuration set up properly. When attempting to deploy a configuration project the deploy dialog shows no servers in the Server dropdown and the message "No servers are defined in the workspace" appears. Can someone help a newbie ...

Thanks Andy But even so, ALL the DATA.30 files (dozens of them in different directories) have exactly the same recent date and time. There is no way that they were all updated at the same time, especially in a Development environment where I know for certain that some of these hashed files have not ...

We are profiling some tables that are joined together with GUIDs. When we get to Key and Cross Domain Analysis IA correctly identifies the PKs but does not identify FK candidates. I must admit that we have never gone this far into profiling so we may be doing something wrong here in order to identif...

I have a DS job which has 4 different dataset files as inputs. If one of those files happens to be missing one day (it could happen) then the job blows up. Is there any way to simply ignore the non-existence of the file and continue processing with the others? I know that with Sequential files there...

As mentioned in an earlier post there are currently no overrides. However, here are the contents: .FPO file: ;;QualityStage v8.0 \FORMAT\ SORT=Y ENPREPFP TBL .FTO file: ;;QualityStage v8.0 \FORMAT\ SORT=Y ENPREPFT TBL .IPO file: ;;QualityStage v8.0 \FORMAT\ SORT=Y ENPREPIP TBL .ITO file: ;;QualitySt...

The structure is: Transformer----->USPREP Std stage------->Dataset file Columns coming into the Standardization stage are: CENSUS_GRP_ID:decimal[38,0] {delim=',', precision=38, scale=0}; SRC_TYP_CD:nullable string[max=5]; SRC_TYP_DESCR:nullable string[max=60]; CASE_ID:nullable decimal[38,0]; SHIPPIN...

Hi Ray. I have absolutely no overrides at all in the rule set. I even attempted to put in an override to see if that helped but to no avail so I removed that one.

Yes, my apologies. This error is occurring in a standardization stage that has USPREP as the Rule and "ZQMIXNZQ" ADDRESS as the Column . Examples of the input data producing the warnings are: Nick Smith AEB 8900 Burnnet St Ste 33 Houston TX 78754 Acme Mfg Llc 100 Morry Mill Rd Floral VA 24...

Getting this warning which is ultimately causing an abort since our company policy only allows a max of 50 warnings before an abort is issued. Would like to eliminate the root cause of the problem and not just remove the limit of 50 warnings. Can't find any reference to a solution in the posts or on...

Thanks Ray but...

- what is REVERSE string? Is that a DS function? I can't find that one.
- how do I match rightmost characters with QS? That's exactly what I want to do but don't see a way to do that.

Glenn

Robert!

You got it! That worked just great for me!

It's unusual that that restriction does not apply to USNAME, USADDR, etc. but it does to USPREP.

Thanks!