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_time timestamp NOT NULL,
end_time timestamp
)
For each feature_id there may be multiple rows with time ranges specified by begin_time/end_time. They may overlap, but this is relatively rare. I'm looking for a fast way to find all feature_ids that have/don't have any overlaps.
I tried to do this using window functions, like this:
SELECT feature_id, bool_or(end_time > lead(begin_time) OVER ts_win) OVER ts_win AS overlaps_any
FROM my_features
WINDOW ts_win AS (PARTITION BY feature_id ORDER BY begin_time)
... but this doesn't work:
ERROR: window function calls cannot be nested
The algorithm is simple: order the rows for a given feature_id by begin_time and check if any end_time > the next begin_time (if any). I suspect there must be an easy way to do this, perhaps with tsrange functions, but can't seem to find it just now.
See Question&Answers more detail:
os 与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…