Re: PG DOCS - logical replication filtering

Поиск
Список
Период
Сортировка
От Aleksander Alekseev
Тема Re: PG DOCS - logical replication filtering
Дата
Msg-id CAJ7c6TNTNLJOveKgH5v4k_V3f6e6B22bZg4xi-q=9tszZWx-Cw@mail.gmail.com
обсуждение исходный текст
Ответ на PG DOCS - logical replication filtering  (Peter Smith <smithpb2250@gmail.com>)
Ответы Re: PG DOCS - logical replication filtering  (Aleksander Alekseev <aleksander@timescale.com>)
Re: PG DOCS - logical replication filtering  (Amit Kapila <amit.kapila16@gmail.com>)
Список pgsql-hackers
Hi Peter,

> PSA a PG docs patch that is associated with the logical replication
> Row Filters feature which was recently pushed [1].

The patch looks mostly OK, but I have several nitpicks.

```
    By default, all data from all published tables will be replicated to the
    appropriate subscribers.
[...]
    By default, all operation types are replicated.
```

The second sentence seems to be redundant.

```
    (This feature is available since PostgreSQL 15)
```

Please correct me if I'm wrong, but I don't think we say that in the docs. When the user opens the documentation for version X he or she sees everything that is available in this version.

```
31.3. Filtering
[...]
There are 3 different ways to filter what data gets replicated.
31.3.1. Operation Filters
[...]
31.3.2. Row Filters
[...]
```
It looks like there are 2 different ways after all.

I see that a large part of the documentation is commented and marked as TBA (Column Filters, Combining Different Kinds of Filters). Could you please clarify if it's a work-in-progress patch? If it's not, I believe the commented part should be removed before committing.

--
Best regards,
Aleksander Alekseev

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Tatsuo Ishii
Дата:
Сообщение: Re: [HACKERS] WIP aPatch: Pgbench Serialization and deadlock errors
Следующее
От: Aleksander Alekseev
Дата:
Сообщение: Re: PG DOCS - logical replication filtering