From: Mark Thomas <mark@...>
Date: 2009-01-13T23:34:18+09:00
Subject: Re: Complex CSV parsing

On Jan 13, 8:42 am, Stuart Clarke <stuart.clarke1...@gmail.com> wrote:
> I am trying to parse data from a file where the values are common
> seperated, however there is slightly more to the file than just commas,
> see below
>
> for
> (;;);{"t":"msg","c":"p_114000000","ms":[{"type":"msg","msg":{"text":"you
> around"}]}

That's not a CSV file. It looks like some sort of serialized data
structure. If you know how it was serialized, you should be able to
easily restore the structure.

If not, you can use Treetop to specify the grammar including the
balanced delimiters {}, (), and [], which apparently take precedence
over the commas, and perform the parsing.