I have a table that is stored in a qvd file. The setup is as follows:
[UpdatedClients]:
LOAD
[ClientID],
[ClientName],
[CorporateID],
[CorporateName],
[BusinessClassID],
[BusinessClass]
FROM [lib://Operations:DataFiles/UpdatedClients.qvd](qvd);
Upon using this table, with several others associated by ClientID, i created filter panes for ClientID and ClientName as some people use the ID while others use the name. However, when ClientName is selected all data goes blank, and the corresponding ClientID filter pane doesn't reference any ClientID correlated to the client name. Using the ClientID filter pane, everything works as expected with the exception of correlating to the client name filter pane. This is very confusing as they are fields on the same table, and all ClientIDs have a corresponding ClientName. Anyone have any thoughts on why this may be happening? Thanks.
Thank you for the input. I have checked this and i am not pulling any ClientID's from other tables in this instance. I have also checked the field types and have found no issues with non identical data types in ClientID field. Again, the ClientID field works perfectly. The ClientName field, which is on the same table somehow does not seem associated with the ClientID once the other tables have been associated. I appreciate all the help so far, thank you!
Hey, do you guys ever speak too soon? Apparently i do! This whole process has been a conversion from an old system to a new one. The previous ClientID field used to have a three letter prefix followed by a dash, like so CRF-(ClientID). It appears that when we did away with the prefix, the DBA did not change the field type from a varchar to an int. This is the root of all the issues and once corrected the filter panes now work as expected. Sorry for the wild goose chase all! Thank you for all the assistance you have provided and i know i will be checking field types first from here on!
Even if synthetic keys are working in many scenarios they are IMHO an error and should be resolved - especially if "strange" things happens ...
It seems that your placement and payment tables contain the same information and could be therefore concatenated - by harmonizing all field-names and data-structures as much as possible and adding an extra field which contained the source-information which could be later used as selection and/or within the set analysis and/or as dimension to differentiate between the tables.
