[GRASS-user] Finding why postgres database cannot be opened

[Rich Shepard]

On Thu, 8 Nov 2018, Markus Metz wrote:

> It's not resolved, and it is not a valid connection:

>>> v.db.connect map=stations -p
>> Vector map <stations> is connected by:
>> layer <1/stations> table <stations> in database
>> </home/rshepard/data/grassdata/willamette_basin/willamette-river-hg>
> through
>> driver <pg> with key <cat>

> for driver=pg, the database must not be a path, it must be
> database=willamette-river-hg

Markus M.

   I now understand the differences between db.connect and v.db.connnect and
did not recognize the full path returend by v.db.connect.

> Also, be aware of the differences between db.connect and v.db.connect. The
> pg database willamette-river-hg must have been established first with
> db.connect + db.login, only after that you can use v.db.connect which is
> not needed for newly created vectors.

   Each time I run db.connect -p it shows the postgres database and the pg
driver. When I saw the results of v.db.connect I added this line to the top
of the script
db.connect driver=pg database=willamette-river-hg
(I did not re-login because that always returned the message that I had been
logged in so the command overwrote the existing login with a new one.)

   Re-running the script produced the same error.

   If db.connect returns the proper database name (without the full path),
while v.db.connect has the full path, how do I correct this? I've looked at
db.* and v.db.* and not seen what I should do to have the v.db.connect see
the database without a path.

Thanks,

Rich