Postgres Pro
Downloads
Home   >   mailing lists

Thread: 9.3 RC1 psql encoding reporting inconsistently?

9.3 RC1 psql encoding reporting inconsistently?

From
Michael Nolan
Date:
This is 9.3 RC1 on a Fedora 7 system. Why does \l report the encoding
as SQL_ASCII
and \set report it as UTF8?

psql (9.3rc1)
Type "help" for help.

postgres=# \l                       List of databases Name     Owner   Encoding  Collate Ctype   Access privileges
--------- -------- --------- ------- ----- ---------------------
postgres  postgres SQL_ASCII C       C
template0 postgres SQL_ASCII C       C     =c/postgres          +
postgres=CTc/postgres
template1 postgres SQL_ASCII C       C     =c/postgres          +
postgres=CTc/postgres
(3 rows)

postgres=# \set
AUTOCOMMIT = 'on'
ECHO = 'queries'
PROMPT1 = '%/%R%# '
PROMPT2 = '%/%R%# '
PROMPT3 = '>> '
VERBOSITY = 'default'
VERSION = 'PostgreSQL 9.3rc1 on i686-pc-linux-gnu, compiled by gcc
(GCC) 4.1.2 20070925 (Red Hat 4.1.2-27), 32-bit'
DBNAME = 'postgres'
USER = 'postgres'
PORT = '5432'
ENCODING = 'UTF8'
postgres=#
--
Mike Nolan

Re: 9.3 RC1 psql encoding reporting inconsistently?

From
Tom Lane
Date:
Michael Nolan <htfoot@gmail.com> writes:
> This is 9.3 RC1 on a Fedora 7 system. Why does \l report the encoding
> as SQL_ASCII and \set report it as UTF8?

psql sets client_encoding based on its environment (LANG or related
variables).  That's been true for some time --- since 9.1, according
to a quick check.
        regards, tom lane

Re: 9.3 RC1 psql encoding reporting inconsistently?

From
David Johnston
Date:
Tom Lane-2 wrote
> Michael Nolan <

> htfoot@

> > writes:
>> This is 9.3 RC1 on a Fedora 7 system. Why does \l report the encoding
>> as SQL_ASCII and \set report it as UTF8?
> 
> psql sets client_encoding based on its environment (LANG or related
> variables).  That's been true for some time --- since 9.1, according
> to a quick check.
> 
>             regards, tom lane

My knowledge of encoding is minimal but to expand on the comment:

Client and server (or, more specifically, database) encodings can and often
do differ just as you are seeing here.

I'm guessing that somewhere deep inside psql and/or postgres encoding
conversion is performed if the client and server do not match.  While I
guess it is possible to try and auto-adapt the client encoding to match the
server/database the current policy is to require the user to explicitly (so
to speak) declare the encoding they are using on their client.

I guess a counter-question would be: what would you expect "\set" to report
and why?

David J.






--
View this message in context:
http://postgresql.1045698.n5.nabble.com/9-3-RC1-psql-encoding-reporting-inconsistently-tp5769334p5769339.html
Sent from the PostgreSQL - hackers mailing list archive at Nabble.com.