PostgreSQL query to detect overlapping time ranges

Question

I have a table in PostgreSQL 9.2 that looks like this (simplified):

CREATE TABLE my_features
(
  id integer NOT NULL,
  feature_id integer NOT NULL,
  begin_         


        

Answer 1

This can indeed be done using range types.

The following selects all those rows that do have overlapping ranges:

select f1.*
from my_features f1
where exists (select 1
              from my_features f2
              where tsrange(f2.begin_time, f2.end_time, '[]') && tsrange(f1.begin_time, f1.end_time, '[]')
                and f2.feature_id = f1.feature_id
                and f2.id <> f1.id);

When you change the condition to NOT EXISTS you'll find those that don't have any overlapping ranges.

SQLFiddle example: http://sqlfiddle.com/#!15/40b1e/1

tsrange(f2.begin_time, f2.end_time, '[]') creates a range that includes the upper and lower bounds. You can also create ranges that exclude either one or both.

More details can be found in the manual:
http://www.postgresql.org/docs/current/static/rangetypes.html#RANGETYPES-INCLUSIVITY

The && operator checks if the two ranges overlap: http://www.postgresql.org/docs/current/static/functions-range.html

^{(I just wish Oracle had something fancy like that...)}