I've been working with Postgres for about 2 years now. I've never had to
do anything real advanced until now and I'm having a huge problem with
some dates. A long time ago, I thought it would be a good idea to have
dates in 4 fields:
date = 5/2/2002
date_month = 5
date_day = 2
date_year = 2002
This was so I could view everyone in May, or everyone between the 10th and
15th, or everyone in 2002, etc... This was a huge mistake. I have since
learned how to use date fields. The problem is, I created a database a
while back that has over 800 records in it that uses this style of dates.
I want to convert this format to the ISO date format (YYYY-MM-DD) and I'm
having lots of problems. I tried using this update:
update table set date_iso = (date_year || '-' || date_month || '-' ||
date_day) where table_id='838
That formats my 3 unnecesary fields to YYYY-MM-DD format. The problem is,
the month/days are entered as 1, 2, 3, 4, etc... Not 01, 02, 03, 04, etc.
I get the error message:
Attribute 'intake_date_raw' is of type 'date' but expression is of type
'varchar' You will need to rewrite or cast the expression.
Yes, I was stupid and made the date_month, date_day, and date_year
varchars. How can I update the database and use the current varchar
fields to create a new field formatted for dates?
-----------
Also... it looks like the search utility on the archives aren't working...
An error occured!
could not connect to server: Connection refused Is the server running on
host db.postgresql.org and accepting TCP/IP connections on port 5433?
