Qlik Community

New to QlikView

Discussion board where members can get started with QlikView.

Announcements
Modernize Your QlikView Deployment webinar, Nov. 3rd. REGISTER
cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted
Not applicable

Non-unique fields causing duplicate records when tables joined.

I have been given a script with the following:

CallTable:

SQL SELECT

     unique_id,

     start_time,

     region_key,

     result

FROM $(DB_NAME).$(vTableName);

join (CallTable)

Load

start_time,

maketime(hour(start_time),minute(start_time),second(start_time)) as [Call Start Time]

Resident

CallTable;

The resulting CallTable then gets stored in a QVD and later loaded by another script for reporting purposes.

This works fine except in certain cases where, I believe, two records have the same start_time value. When this happens, then we end up with duplicate records in the table/QVD which we can see by creating a list box for unique_id with the frequency option enabled.  How can I modify this script to prevent these duplicate records? Since we're working with a fairly large dataset an efficient solution is  preferable. 

Thanks

Steve

1 Solution

Accepted Solutions
Highlighted
MVP
MVP

Maybe I misunderstood what you are trying to achieve, but why not try a preceding load:

CallTable:

LOAD *,

      maketime(hour(start_time),minute(start_time),second(start_time)) as [Call Start Time];

SQL SELECT     

     unique_id,

     start_time,

     region_key,

     result

FROM $(DB_NAME).$(vTableName);


View solution in original post

2 Replies
Highlighted
MVP
MVP

Maybe I misunderstood what you are trying to achieve, but why not try a preceding load:

CallTable:

LOAD *,

      maketime(hour(start_time),minute(start_time),second(start_time)) as [Call Start Time];

SQL SELECT     

     unique_id,

     start_time,

     region_key,

     result

FROM $(DB_NAME).$(vTableName);


View solution in original post

Highlighted
Not applicable

Perfect, thanks.