Re: Process for populating tables in new database [RESOLVED]
| От | Rich Shepard |
|---|---|
| Тема | Re: Process for populating tables in new database [RESOLVED] |
| Дата | |
| Msg-id | alpine.LNX.2.20.1812010729220.8748@salmo.appl-ecosys.com обсуждение исходный текст |
| Ответ на | Re: Process for populating tables in new database (Adrian Klaver <adrian.klaver@aklaver.com>) |
| Список | pgsql-general |
On Sat, 1 Dec 2018, Adrian Klaver wrote: > If you don't supply the key it will be generated as the default for the PK > column is a sequence. Thanks, Adrian. I thought this to be the case and did not find confirmation in the manual (perhaps I just missed seeing it.) > So are the tables you are INSERTing into currently unpopulated? Yes. > In any case you will need to do this in sequence, where you populate the > company table and then the contact table. The question is whether you want > to pre-assign the company id's in the company data and the company_id_fk > in the contacts data or not. If not then you will need to grab the company > id's after populating the company table and match those to the contacts > data before inserting it. This clears up everything. I'll let pg assign company id's then use them as you write to relate the other tables to the proper company/contact. Much appreciated, Rich
В списке pgsql-general по дате отправления: