Bad encoded chars in being inserted into database

Hi everybody.

I have a doubt about how postgres deal with bad encoded characters into
database.

We have several gforge application. They are using postgres as database.

If we export a database and import again, we have to deal with several bad
encoded chars. These bad chars always come from copy & paste emails from
Lotus Notes mail client. OK, I understand the Notes client people is using
is an ancient application and does not deal very well with some Unicode
chars…

What I cannot understand is why postgres accept these bad enconded
characters into database, exports them without problema but does not allow
them when importing again.

This has been happening since postgers 7.3. However, until 7.4.XX (y don’t
remember what minor version) you could import database without ERRORs.
However, since 7.4.XX it’s impossible and it’s imperative to clean bad
characters (using iconv, for example) prior importing tables.

I agree with this postgres policy, but what I don’t is that you can INSERT
them via application. That is, no bad characters should be inserted into
database. The check should be made for both import and insert procedures
so no bad chars would appear into database.

Any suggestion / appreciation about this?

We are using php4/5-pgsql module from several distros (CentOS 4/5, debian
4/5 and Ubuntu 8.04LTS) so I discard a pgsql problem and anyway database
should deal with this…