Re: Review: Row-level Locks & SERIALIZABLE transactions, postgres vs. Oracle

Поиск
Список
Период
Сортировка
От Kevin Grittner
Тема Re: Review: Row-level Locks & SERIALIZABLE transactions, postgres vs. Oracle
Дата
Msg-id 4C42F0FF020000250003381E@gw.wicourts.gov
обсуждение исходный текст
Ответ на Re: Review: Row-level Locks & SERIALIZABLE transactions, postgres vs. Oracle  (Florian Pflug <fgp@phlo.org>)
Ответы Re: Review: Row-level Locks & SERIALIZABLE transactions, postgres vs. Oracle  (Andrew Dunstan <andrew@dunslane.net>)
Re: Review: Row-level Locks & SERIALIZABLE transactions, postgres vs. Oracle  (Florian Pflug <fgp@phlo.org>)
Список pgsql-hackers
Florian Pflug <fgp@phlo.org> wrote:
> On Jul17, 2010, at 18:25 , Kevin Grittner wrote:
>> * Does it follow the project coding guidelines?
>> 
>> Comments are not all in standard style.
> Does that refer to the language used, or to the formatting?
Formatting.  Comment style seems to be defined here:
http://developer.postgresql.org/pgdocs/postgres/source-format.html
as being:
/** comment text begins here* and continues here*/
You have these formats in your patch:
/* comment text begins here* and continues here*/
/* comment text begins here  and continues here */
/* One line comment like this. */
That last one is actually pretty common in PostgreSQL source, so I'm
not sure that its omission from the style page isn't accidental.
> Btw, while the patch obsoletes the crosscheck snapshot, it
> currently doesn't remove its traces of it throughout the executor
> and the ri triggers. Mainly because I felt doing so would make
> forward-porting and reviewing harder without any gain. But
> ultimately, those traces should probably all go, unless someone
> feels that for some #ifdef NOT_USED is preferable.
My view is that we have a revision control system which makes the
code easy to restore, should it be found to be useful again.  If it
has no use at the point of applying this patch, my inclination would
be to delete it.  If you're particularly concerned that it could
later be useful, you might want to do that deletion in as a separate
patch to facilitate later resurrection of the code.
-Kevin


В списке pgsql-hackers по дате отправления:

Предыдущее
От: David Christensen
Дата:
Сообщение: Re: Patch: psql \whoami option
Следующее
От: Tom Lane
Дата:
Сообщение: Re: ALTER TABLE SET STATISTICS requires AccessExclusiveLock