John R Pierce <pierce@hogranch.com> writes:
> Sydney Puente wrote:
>> The first isssue that occurs to me is that CP1252 is used throughout
>> the data and there is a lot of european special characters, e acute
>> for example. But the column names etc are regular chars [a-zA-Z].
> CP1252 aka Windows-1252 is actually pretty close to ISO-8859-1 aka
> LATIN1. The differences are mostly that CP1252 uses the 80-9F section
> for additional characters, this is unused in LATIN1.
> Personally, I'd probably make the Postgres database UTF-8, then use
> Windows-1252 as the client_encoding during the import process.
FWIW, we do support win1252 as a database encoding. I tend to agree
that switching to something better-standardized would be a good idea
though.
regards, tom lane
