Thread: Granting Browser Access to pgsql database.
Hello again! I have now managed to work out where I was going wrong. However, I still get the line: NOTICE: Unrecognized variable client_encoding when executing a psql command, however it doesn't seem to affect anything. Has anyone else experienced this?
* James Snelling <fatboy2000@btinternet.com> [001115 14:09]: > Hello again! > > I have now managed to work out where I was going wrong. > > However, I still get the line: > > NOTICE: Unrecognized variable client_encoding > > when executing a psql command, however it doesn't seem to affect anything. > Has anyone else experienced this? The DB probably was built with --enable-multibyte set, but psql is finding a libpq that doesn't have it. Larry -- Larry Rosenman http://www.lerctr.org/~ler Phone: +1 972-414-9812 (voice) Internet: ler@lerctr.org US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749
Larry Rosenman <ler@lerctr.org> writes:
> * James Snelling <fatboy2000@btinternet.com> [001115 14:09]:
>> However, I still get the line:
>> NOTICE: Unrecognized variable client_encoding
> The DB probably was built with --enable-multibyte set, but psql is
> finding a libpq that doesn't have it.
Other way round, actually: libpq is sending a SET command to set the
client-side multibyte encoding, and the backend doesn't know what to
do with it. It's harmless enough, as long as you aren't expecting
multibyte characters to work ;-)
For 7.1, I've tweaked the backend so that even without
--enable-multibyte, it will accept SET client_encoding = 'SQL_ASCII'
(but *not* any other value). This should eliminate unnecessary
notices when using clients compiled with a different feature set.
regards, tom lane
* Tom Lane <tgl@sss.pgh.pa.us> [001115 23:05]: > Larry Rosenman <ler@lerctr.org> writes: > > * James Snelling <fatboy2000@btinternet.com> [001115 14:09]: > >> However, I still get the line: > >> NOTICE: Unrecognized variable client_encoding > > > The DB probably was built with --enable-multibyte set, but psql is > > finding a libpq that doesn't have it. > > Other way round, actually: libpq is sending a SET command to set the > client-side multibyte encoding, and the backend doesn't know what to > do with it. It's harmless enough, as long as you aren't expecting > multibyte characters to work ;-) > > For 7.1, I've tweaked the backend so that even without > --enable-multibyte, it will accept SET client_encoding = 'SQL_ASCII' > (but *not* any other value). This should eliminate unnecessary > notices when using clients compiled with a different feature set. Thanks, Tom. I knew I had seen this with a mis-match. -- Larry Rosenman http://www.lerctr.org/~ler Phone: +1 972-414-9812 (voice) Internet: ler@lerctr.org US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749