On 12/1/18 6:09 AM, Rich Shepard wrote:
I have a new database with five tables, and text files with data to
populate
the tables using the insert command.
1. Each table has a sequential primary key. Should I manually add keys to
each insert row or will postgres generate it automatically? Example, for
the
companies table:
org_id | integer | | not null |
nextval('companies_org_id_seq'::regclass)
org_name | character varying(64) | | not null |
2. Should I use sepatate files as input to each table or can I combine them
in a single file? For example, adding a company name and contact name for
someone in that company using input commands for both tables in a single
file? I will need to provide the primary key to the company table as the
foreign key in contact table and have not before pupulated multiple related
tables in a new database from .sql files.
Forgot to add that if you pre-assign the keys to the serial field you
will need to advance the sequence to a value past the last key value to
avoid a duplicate key error when you let the sequence assign numbers.
TIA,
Rich
--
Adrian Klaver
adrian.kla...@aklaver.com
