I used 'IST' in a query like this - (timestamp_hour) at time zone 'IST' time_ist and did not get the expected output - timestamp in Indian Standard Time. So I queried the 2 views that provide timezone info and did not really understand the abbrev column.
select name, abbrev, utc_offset from pg_timezone_names where abbrev = 'IST' ;
Since the S and T are non-location specific you get 26 different timezone abbreviations to choose from. That wasn't enough for the world. So IST is non-unique; and for historical reasons Ireland (Eire, which contains Dublin) is given default priority.
name | abbrev | utc_offset ---------------+--------+------------ Eire | IST | 01:00:00 Asia/Kolkata | IST | 05:30:00 Asia/Calcutta | IST | 05:30:00 Europe/Dublin | IST | 01:00:00
Suggest you adapt to using ISO names (the name column above) for timezones; which are long enough and location-specific enough to be unique. In your case, pick your preferred spelling of Calcutta I suppose.
There is a way to get a different interpretation for IST to be recognized but I'd have to find it or wait for others to chime in.
