Very often I hear people wondering if it’s better the EXISTS or the IN operator. Let’s start by saying that the two operators behave differently when NULL values are involved (see this post). If you are sure that the field used to join the tables does not include NULL values then the IN operator “generally” performs better than EXISTS. I say “generally” because this is not always the case. For example consider the following query:
NOT IN does not behave as you would expect when NULL values are involved. Suppose you have the following tables: CD_FIELD DS_FIELD 1 ALPHA 2 BETA TABLE_A CD_FIELD DS_FIELD 1 GAMMA NULL DELTA TABLE_B Now try the following query: SELECT*FROMtable_aaWHEREa.cd_fieldNOTIN(SELECTb.cd_fieldFROMtable_bb); Does the previous query return any row? The answer is NO it doesn’t, even if there is a value (2) in table_a.
Many sources use the term timestamp to refer specifically to Unix time, the number of seconds since 00:00:00 UTC on January 1, 1970. In Oracle you can compute this number very easily. For example, the following query computes the number of seconds between 00:00:00 UTC on January 1, 1970 and October 9, 2007. SELECT(to_date('09-10-2007','DD-MM-YYYY')-to_date('01-01-1970','DD-MM-YYYY'))*60*60*24FROMdual The result of the preceding query should be 1191888000. How it works: The subtraction between two dates returns the number of days between them.