From: Patrick Roemer <sangamon@...>
Date: 2008-02-18T22:54:58+09:00
Subject: Ruby 1.9, CSV and encodings

Hi,

I'd like to process an ISO-8859-1 encoded .csv file on a UTF-8 system
using Ruby 1.9. The naive approach of using the CSV lib fails with
"invalid mbstring sequence". I don't see where to specify the encoding.
I've even tried the '--encoding' command line switch, but it doesn't
seem to know ISO-8895-1 in all spellings I could come up with. (BTW,
where is command line switch behavior like this documented in detail?)

Currently I'm reading the complete file, enforce the encoding and
process the in-memory string:

file_content = IO.readlines(file_name).join
file_content.force_encoding('ISO-8859-1')
parser = CSV.new(file_content, ...)

But that looks like a rather clumsy hack. What's the "right" way to do it?

Thanks and best regards,
Patrick