joiner

joiner

1. active

2. conected

3. matches 1 or more pairs of columns between 2 sources

4. limitations

   • cant use when either input pipeline contains and update strategy
   • Sequence Generator cant be directly before it

5. caching

   • joiner data cache size
     • default is 2M
   • joiner index cache size
     • default 1M

6. join condition

   • Join type
     • normal (FASTER)
       • Discards all rows of data from the master and detail sources that do not match
     • master outer (FASTER)
       • Keeps all rows of data from detail source and matching rows from master; discards unmatched rows from master
     • detail outer
       • Keeps all rows of data from master source and matching rows from detail; discards unmatched detail
     • full outer
       • keeps all rows of data from both the master and detail
   • compares each MASTER source against detail source
     • sorted input
       • to improve performance user source with fewer duplicate key values as the master
       • using source qualifier or sorter;
         • use transformations that maintain order
           • configure joiner to use sorted date
             • join condition to use sort origin psort
           • do NOT place between sort origin & joiner: Custom, Unsorted Agg, Normalizer, Rank, Union, XML P or G, or Mapplet with any of them
     • unsorted
       • to improve performance, use source with fewer rows as MASTER
   • sources
     • master pipeline or master branch
       • 2nd source added
     • detail pipeline or detail branch
       • first source added
     • can change these by clicking the M column on the Ports tab
   • must have same datatypes
   • doesn't match NULL