On 10/06/12 3:45 AM, Heikki Linnakangas wrote:
> At the SQL level, there's the convert(bytea, name, name) function.
ahhh, right. (forehead slap)
>
>> a 2nd function would do the same, but replace
>> errors with the substitution character in the target charset and not
>> error.
>
> Hmm, I don't think we have that.
me thinks this would be extremely useful for importing 'dirty' data.
that or a per-connection flag (or option on the COPY command?) that
said "substitute-on-error" for the likes of UTF-8 imports from CSV.
--
john r pierce N 37, W 122
santa cruz ca mid-left coast