Re: review: FDW API - Mailing list pgsql-hackers

From Heikki Linnakangas
Subject Re: review: FDW API
Date
Msg-id 4D39B423.1000203@enterprisedb.com
Whole thread Raw
In response to Re: review: FDW API  (Shigeru HANADA <hanada@metrosystems.co.jp>)
Responses Re: review: FDW API  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: review: FDW API  (Shigeru HANADA <hanada@metrosystems.co.jp>)
List pgsql-hackers
On 18.01.2011 17:26, Shigeru HANADA wrote:
> 3) 20110118-fdw_handler.patch - This patch adds support for HANDLER
> option to FOREIGN DATA WRAPPER object.

Some quick comments on that:

* I wonder if CREATE FOREIGN DATA WRAPPER should automatically create 
the handler function, if it doesn't exist yet. That's what CREATE 
LANGUAGE does, which is similar. Although it doesn't seem to be 
documented for CREATE LANGUAGE either, is it deprecated?

* The elogs in parse_func_options() should be ereports.

* pg_dump should check the version number and only try to select 
fdwhandler column if >= 9.1. See the other functions there for example 
of that.

* dumpForeignDataWrapper() in pg_dump checks if fdwhandler field is "-". 
I don't think we use that as magic value there, do we? Same with validator.

* Please check that the HANDLER and VALIDATOR options that pg_dump 
creates properly quoted.

--   Heikki Linnakangas  EnterpriseDB   http://www.enterprisedb.com


pgsql-hackers by date:

Previous
From: "Kevin Grittner"
Date:
Subject: Re: SSI and Hot Standby
Next
From: Tom Lane
Date:
Subject: Re: sepgsql contrib module