[prev in list] [next in list] [prev in thread] [next in thread] 

List:       postgresql-general
Subject:    \COPY to accept non UTF-8 chars in CHAR columns
From:       Matthias Apitz <guru () unixarea ! de>
Date:       2020-03-27 15:36:23
Message-ID: 20200327153623.GA32266 () sh4-5 ! 1blu ! de
[Download RAW message or body]

Hello,

We're moving databases from Sybase/Oracle to PostgreSQL 11.4 by
unloading them with our own tool into a CSV like format and loading the
rows with \COPY. One can imagine that in old databases, in use for
a longer time, you will find any sort of dust which should not be there, for
example ISO-8859-1 chars while the CHAR column should be in UTF-8. Needless
to say, this data should not be there, it is just a fact, in parts
caused by foreign data loaded into the Sybase/Oracle database, sometimes
our own Java- or Perl-written software, breaks a longer CHAR column into
two sequential rows (with an INT column to note the order), but does
the break in the middle of an UTF-8 multi-byte, clumsy. :-(

In short, it there a way to let \COPY accept such broken ISO bytes, just
complaining about, but not stopping the insert of the row?

Thanks

	matthias

-- 
Matthias Apitz, ✉ guru@unixarea.de, http://www.unixarea.de/ +49-176-38902045
Public GnuPG key: http://www.unixarea.de/key.pub
May, 9: Спаси́бо освободители! Thank you very much, Russian liberators!


[prev in list] [next in list] [prev in thread] [next in thread]