John Lynn <mr.john.lynn@gmail.com> writes:
> On Windows (Windows Server 2008), I create a file using a tool (Data
> Generator for PostgreSQL), and then try to execute the script using:
> psql --file="C:\PDS\TestData\populate_companies.sql" --dbname=tst --username=tst
> I get an error, pointing to the first character in the file. So I open
> the file using pgAdmin and I see only an "I" (the first character of
> the first command in the file, which is an INSERT) in the file.
> If I load the file into a Windows editor, I see that it's a "unicode"
> file. If I re-save it as a "DOS" file, then psql is able to read it.
> Is there a way to convince psql to read a file in "unicode" format?
Most likely there is a "byte order mark" in the file (a thing that is
both useless and counter to standards in UTF-8 data, but try telling
Microsloth that :-(). There's been some talk of teaching psql to ignore
BOMs but nothing's been done about it. Saving in the plain text format
is your best bet for now.
regards, tom lane