Re: The return value of SPI_connect - Mailing list pgsql-hackers

From Tom Lane
Subject Re: The return value of SPI_connect
Date
Msg-id 262208.1601301751@sss.pgh.pa.us
Whole thread Raw
In response to The return value of SPI_connect  ("Hou, Zhijie" <houzj.fnst@cn.fujitsu.com>)
Responses RE: The return value of SPI_connect
List pgsql-hackers
"Hou, Zhijie" <houzj.fnst@cn.fujitsu.com> writes:
> May be we can make "int SPI_connect()" => "void SPI_connect()" and fix the doc ?

I don't see a lot of value in changing SPI_connect() in isolation.
It's part of an API suite that generally expects functions to
return error codes.

There was a previous thread [1] about getting rid of useless SPI
error codes across the board, in favor of having those functions
throw errors for error cases.  AFAICT nobody has reviewed the
proposed patch there, but maybe you'd like to?

            regards, tom lane

[1] https://www.postgresql.org/message-id/flat/cc57ea5c-592a-6c27-aca6-7ad2a166c379%40gmail.com



pgsql-hackers by date:

Previous
From: Alvaro Herrera
Date:
Subject: Re: a problem about XLogReader callback system
Next
From: Tom Lane
Date:
Subject: Re: Skip ExecCheckRTPerms in CTAS with no data