Re: [HACKERS] Re: tab completion in psql - Mailing list pgsql-hackers

From Brian E Gallew
Subject Re: [HACKERS] Re: tab completion in psql
Date
Msg-id emacs-smtp-20280-14404-7220-662833@export.andrew.cmu.edu
Whole thread Raw
In response to Re: tab completion in psql  (Peter Eisentraut <e99re41@DoCS.UU.SE>)
List pgsql-hackers
Then Peter Eisentraut <e99re41@DoCS.UU.SE> spoke up and said:
> The tab completion is not very smart, it only covers the really obvious
> cases. Doing FROM shouldn't be so hard, but once you get into WHERE you
> almost end up writing a complete SQL parser just for this. Not that
> there's anything fundamentally wrong with that. It's a very evolving piece
> of code, however; it can only get better. I bet those readline authors
> never had this one in mind. Otherwise readline would play nicer with it.

I've used Python's readline support, and I must say it's really nice
when the readline engine can complete more dynamic names.  It would be
very nice if "\d table" or "\dt" populated a dictionary of some kind
which could then be used for completion.  While context sensitivity
would be even better, it might be worthwhile to simply have a dynamic
dictionary. 

-- 
=====================================================================
| JAVA must have been developed in the wilds of West Virginia.      |
| After all, why else would it support only single inheritance??    |
=====================================================================
| Finger geek@cmu.edu for my public key.                            |
=====================================================================

pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: tab completion in psql
Next
From: Peter Eisentraut
Date:
Subject: Re: Ownership/protection (was Re: [HACKERS] Portability)