Peter Eisentraut <peter_e@gmx.net> writes:
> On fre, 2012-03-16 at 13:47 -0400, Tom Lane wrote:
>> I'm a bit concerned about whether that's actually going to be useful.
>> A quick check shows that in the regression database, the proposed patch
>> produces 3246 possible completions, which suggests that by the time you
>> get down to a unique match you're going to have typed most of the name
>> anyway.
> Well, the regression test database is not really an example of real-life
> object naming, I think.
Perhaps not, but a solid 2000 of those names are from the system-created
entries in pg_proc, and the regression DB doesn't have an especially
large number of tables either. I doubt that real DBs are likely to have
materially fewer completions.
This connects somewhat to the discussions we've had in the past about
trying to get not-intended-for-public-use functions out of the standard
search path. Not that you want to put a full visibility check into the
tab-completion query, but if it could simply exclude a "pg_private"
namespace, that probably wouldn't be too expensive.
regards, tom lane
