Thanks for the replies guys. In the end I found an example lurking in the depths of the Internet that used a CTE and a join to that which seemed to have the desired effect
On Thursday, 14 July 2016, Begin Daniel <jfd...@hotmail.com> wrote: > *From:* pgsql-general-ow...@postgresql.org > <javascript:_e(%7B%7D,'cvml','pgsql-general-ow...@postgresql.org');> > [mailto:pgsql-general-ow...@postgresql.org > <javascript:_e(%7B%7D,'cvml','pgsql-general-ow...@postgresql.org');>] *On > Behalf Of *David G. Johnston > *Sent:* Thursday, 14 July, 2016 08:23 > *To:* Nick Babadzhanian > *Cc:* Tim Smith; pgsql-general > *Subject:* Re: [GENERAL] Merging timeseries in postgres > > > > On Thu, Jul 14, 2016 at 8:18 AM, Nick Babadzhanian <n...@cobra.ru > <javascript:_e(%7B%7D,'cvml','n...@cobra.ru');>> wrote: > > Whats exactly is wrong with the following query? > > select > dx date, > nx, > nx1 > from > test t > join test1 t1 on t.dx=t1.dx1 > ; > > > > > > Please don't top-post. > > > > test t join test1 t1 -- this is the default inner join, your query returns > no records for the given data. > > > > David J. > > > > Look at the above documentation (7.2.1.1) on full join > > https://www.postgresql.org/docs/9.3/static/queries-table-expressions.html > > > > select coalesce(dx,dx1)as dt, n, nx1 > > from test full join test1 on dx=dx1; > > > > Daniel >
