| From: | Célestin Matte <celestin(dot)matte(at)cmatte(dot)me> |
|---|---|
| To: | Magnus Hagander <magnus(at)hagander(dot)net> |
| Cc: | pgsql-www(at)lists(dot)postgresql(dot)org |
| Subject: | Re: [PATCH] pgarchives: Bugfix: missing ids in pglister_sync |
| Date: | 2022-02-03 13:51:59 |
| Message-ID: | a5b19a46-2069-f3ec-d177-35aa46c12a8f@cmatte.me |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-www |
>> I think the attached patch is more correct (I've backpatched it into
>> the 0001 migration because making it a separate migration makes it
>> somewhat destructive -- it's easier to just change it manually on
>> existing installs if you have one)
>
> Thanks! Can it be pushed into the git repository, or does it need testing or something?
>
I understand that changing id fields to AutoFields can be destructive for existing install, but on the other hand, the current state of the repository is problematic as it causes pglister_sync to fail by default.
Another solution could be to add sequences directly in postgres using RunSQL() in a migration to turn these fields into de facto serial fields. This is a bit hacky, but non-destructive. What do you think?
--
Célestin Matte
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Magnus Hagander | 2022-02-05 17:20:35 | Re: [PATCH] pgarchives: Bugfix: missing ids in pglister_sync |
| Previous Message | Célestin Matte | 2022-02-03 13:30:22 | [PATCHES] pgarchives: merge schema.sql into Django's model |