Re: Problems with question marks in operators (JDBC, ECPG, ...) - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Problems with question marks in operators (JDBC, ECPG, ...)
Date
Msg-id 2705.1432150440@sss.pgh.pa.us
Whole thread Raw
In response to Re: Problems with question marks in operators (JDBC, ECPG, ...)  (Dave Cramer <pg@fastcrypt.com>)
Responses Re: Problems with question marks in operators (JDBC, ECPG, ...)  (Andrew Dunstan <andrew@dunslane.net>)
List pgsql-hackers
Dave Cramer <pg@fastcrypt.com> writes:
> Notably absent from the discussion is ODBC upon which JDBC was modelled and
> probably predates any use of ? as an operator

<historical-nitpicking>

It would be a mistake to imagine that operators containing '?' are some
johnny-come-lately.  The <?> operator for tintervals can be traced back
at least to Postgres v4r2 (1994), which is the oldest tarball I have at
hand.  Most of the current list are geometric operators that were added
by Tom Lockhart in 1997.  The only ones that aren't old enough to vote
are the JSONB ones we added last year.

Not that the problem's not real, but these operators predate any attempt
to make Postgres work with ODBC or JDBC or any other connector.  Otherwise
we might've thought better of using '?'.

</historical-nitpicking>
        regards, tom lane



pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: RFC: Non-user-resettable SET SESSION AUTHORISATION
Next
From: Tom Lane
Date:
Subject: Re: jsonb concatenate operator's semantics seem questionable