From: | Dave Page <dpage(at)pgadmin(dot)org> |
---|---|
To: | Magnus Hagander <magnus(at)hagander(dot)net> |
Cc: | Guillaume Lelarge <guillaume(at)lelarge(dot)info>, pgadmin-support(at)postgresql(dot)org, DUPREZ Cédric <cedric(dot)duprez(at)ifn(dot)fr> |
Subject: | Re: Savepoints in PgAdmin |
Date: | 2009-09-03 12:02:24 |
Message-ID: | 937d27e10909030502q602ac29nd259413a4b55ff2d@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgadmin-support |
On Thu, Sep 3, 2009 at 12:57 PM, Magnus Hagander<magnus(at)hagander(dot)net> wrote:
> Good question. In general we shouldn't change the default behavior in
> a backpatch, but I think the use-case for the current behavior is
> pretty limited.
It's not being back-patched. That code has been there for years, with
only one complaint that I'm aware of. The complaint is a valid one
though, which is why I think we should add the option for 1.12.
> AIUI, nothing at all changes for single-statement queries, right? It
> would only affect those that explicitly open a transaction. In which
> case I think it's a reasonable default to behave the same way as psql,
> which means change the default for pgadmin.
>
> Specifically on the patch I don't like an option named "disabled auto
> rollback". Negations in options suck. It shuold be "Enable auto
> rollback", and then whatever the default is can be argued. I'd also
> find it very useful if this was available as a checkbox on the query
> menu, not just in the options dialog. Thoughts?
Which would do what - override the default for that window only, or
control the default as well? I prefer the former, but some of the
other options are stored for all future query windows, so there may be
some inconsistency there.
--
Dave Page
EnterpriseDB UK: http://www.enterprisedb.com
From | Date | Subject | |
---|---|---|---|
Next Message | Magnus Hagander | 2009-09-03 12:05:33 | Re: Savepoints in PgAdmin |
Previous Message | Magnus Hagander | 2009-09-03 11:57:49 | Re: Savepoints in PgAdmin |