2010/3/7 Tom Lane <tgl@sss.pgh.pa.us>:
> Magnus Hagander <magnus@hagander.net> writes:
>> 2010/3/7 Tom Lane <tgl@sss.pgh.pa.us>:
>>> If we were going to support multiple -f options, it would be sensible
>>> to interpret "-f -" as "read from stdin until EOF".
>
>> Right, that would work. Though it would be a lot more user-unfriendly
>> for such a simple thing, IMHO.
>
> If the issue had come up even once before in psql's existence, I might
> think that user-friendliness would be a good argument. As things stand,
> I don't believe the average user will care about it in the least. I'd
> be willing to lay long odds that the average user doesn't even have a
> .psqlrc file, much less feel the need to override it. I'd rather see
> "use a substitute psqlrc" be a behavior you can build out of existing
> general-purpose switches than still another option that has to be
> documented and remembered.
I've heard if a couple of times before, but I agree it's certainly not
a much asked-for one. Most if it has been in off-list scenarios and
people have probabliy just thought it's not a big enough feature to
bother emailing about.
>> Also, "-f -" and just "psql" behaves different today (for example, in
>> the showing of startup banners).
>
> Yes, there would be some things to think about there, which is why it's
> a topic for a new devel cycle rather than something to shoehorn in
> after the close of the last CF.
Fair enough. I expected it to be a small and noncontroversial thing,
but since there are objections, I'll go revert it.
-- Magnus HaganderMe: http://www.hagander.net/Work: http://www.redpill-linpro.com/