Re: Boolean operators without commutators vs. ALL/ANY

Andrew Dunstan <andrew@dunslane.net>

From: Andrew Dunstan <andrew@dunslane.net>
To: Alvaro Herrera <alvherre@commandprompt.com>
Cc: Florian Pflug <fgp@phlo.org>, Tom Lane <tgl@sss.pgh.pa.us>, Peter Eisentraut <peter_e@gmx.net>, Robert Haas <robertmhaas@gmail.com>, PG Hackers <pgsql-hackers@postgresql.org>
Date: 2011-06-17T14:47:19Z
Lists: pgsql-hackers

On 06/17/2011 10:20 AM, Alvaro Herrera wrote:
> alvherre=# \doS ~
>
>                                                Listado de operadores
>    Esquema   | Nombre | Tipo arg izq | Tipo arg der | Tipo resultado |                Descripción
> ------------+--------+--------------+--------------+----------------+--------------------------------------------
> ...
>   pg_catalog | ~      | text         | text         | boolean        | matches regular expression, case-sensitive
>
> Note that there's no way to tell which is the regex here.  It'd be a lot
> better if the description was explicit about it.  (Or, alternatively,
> use a different data type for regexes than plain text ... but that has
> been in the Todo list for years ...)


+1 for improving the description.

>
> Have ~ keep its existing semantics, use ~= for the commutator?  There
> are a lot more chars allowed in operator names anyway, it doesn't seem
> to me like we need to limit ourselves to ~, = and @.


Yeah, maybe something like ~< for the commutator. (I know, we're 
bikeshedding somewhat.)


> I *do* like the idea of having commutate-ability for ANY/ALL, having
> needed it a couple of times in the past.
>

Indeed. me too.

cheers

andrew