Question

I am reading about Direct-Path INSERT in oracle documentation Loading Tables

It is written that :
During direct-path INSERT operations, the database appends the inserted data after existing data in the table. Data is written directly into datafiles, bypassing the buffer cache. Free space in the table is not reused, and referential integrity constraints are ignored. Direct-path INSERT can perform significantly better than conventional insert.

Can anyone explain me ,how referential integrity constraints is been ignored,According to my understanding it will load the data into the table ignoring the referential constraint .and after insert it will check for referential constraint.

If this is so ,if i use like this .

  FORALL i IN v_temp.first..v_temp.last save exceptions
  INSERT /*+ APPEND_VALUES */ INTO orderdata 
  VALUES(v_temp(i).id,v_temp(i).name);
 COMMIT;

Will this will gave me correct index ,in case of any exceptions and how ?.

Sorry to ask so many questions in one ,but they are releated to each other.

1. How refrential constraint is been ignored
2. What is Free Space in table above
3. How it will give correct Index in case of any exceptions.

Answer 1

The first question should really be (Do I want/need to use direct path insert?”, and the second should be “Did my query use direct path insert?”

If you need referential integrity checks, then you do not use direct path insert.

If you do not want the table to be exclusively locked for modifications, then do not use direct path insert.

If you remove data by deletion and only insert with this code, then do not use direct path insert.

One quick and easy check on whether direct path insert was used is to immediately, before committing the insert, issue a select of one row from the table. If it succeeds then direct path insert was not used — you will receive an error message if it was because your change has to be commited before your session can read the table.