Join query query
Hi,
I'd like to do something which I think should be quite easy - that is join
2 tables and create a new table.
Table A postcode_input has columns which include postcode, eastings,
northings. there are 1,687,605 rows.
Table B bng_lat_long has columns lat, lon, e, n. There are 1,687,605 rows.
eastings = e and northings = n so there should be a 1 to 1 match. The
eastings northings pair should be unique in aggregate.
So I tried doing this:
SELECT A.postcode, A.eastings, A.northings, B.lat, B.lon INTO
postcode_lat_long
FROM postcode_input AS A
LEFT JOIN bng_lat_long AS B On A.eastings = B.e AND A.northings = B.n
And ended up with a table 13,708,233 rows long with what looks like plenty
of duplicated rows. Some but not all are duplicated. What can I do to sort
this out?
Thanks,
Andy
On Wed, Feb 13, 2013 at 3:01 PM, Andrew Taylor <andydtaylor@gmail.com> wrote:
And ended up with a table 13,708,233 rows long with what looks like plenty
of duplicated rows. Some but not all are duplicated. What can I do to sort
this out?
It means that (e, n) pairs are not unique in A and B and you got a
superposition of them. If you have 5 equal pairs in A and 7 same pairs
with in B you will get 35 combinations as a result.
And BTW when you use LEFT JOIN if there are rows in A that have no
matching pairs in B you will get one row for each of them where lan
and lon are NULLs.
See the join_type section here
http://www.postgresql.org/docs/9.2/static/sql-select.html.
--
Sergey Konoplev
Database and Software Architect
http://www.linkedin.com/in/grayhemp
Phones:
USA +1 415 867 9984
Russia, Moscow +7 901 903 0499
Russia, Krasnodar +7 988 888 1979
Skype: gray-hemp
Jabber: gray.ru@gmail.com
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
On Thursday, February 14, 2013 4:31 AM Andrew Taylor wrote:
Hi,
I'd like to do something which I think should be quite easy - that is join
2 tables and create a new table.
Table A postcode_input has columns which include postcode, eastings,
northings. there are 1,687,605 rows.
Table B bng_lat_long has columns lat, lon, e, n. There are 1,687,605 rows.
eastings = e and northings = n so there should be a 1 to 1 match. The
eastings northings pair should be unique in
aggregate.
I think mapping is m to n.
So I tried doing this:
SELECT A.postcode, A.eastings, A.northings, B.lat, B.lon INTO
postcode_lat_long
FROM postcode_input AS A
LEFT JOIN bng_lat_long AS B On A.eastings = B.e AND A.northings = B.n
And ended up with a table 13,708,233 rows long with what looks like plenty
of duplicated rows. Some but not all are
duplicated. What can I do to sort this out?
What is you exact expection of data in postcode_lat_long?
From the above it seems you want distinct rows which match between
postcode_input and bng_lat_long, if there is no match then take the values
of postcode_input and NULL for bng_lat_long.
If I am right, then you can try with using DISTINCT operator:
http://www.postgresql.org/docs/9.2/static/sql-select.html#SQL-DISTINCT
With Regards,
Amit Kapila.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general