Re: Removing thousands separator from CSV quoted string

[Justin Dearing <zippy1981@xxxxxxxxx>]

Roger,

You need to load the file first in its original format in a loading table.
Then transform it. Then put it in the final table. Whatever UDF you have
that turns money into decimals should be a scalar UDF that takes a string
as input and outputs a decimal.

If you load the raw CSV or tab delimited file into a loading table (can be
a temp table) and then transform it and load it into the real Physical
file, you won't have to worry about the file being comma delimted. You will
end up with small reusable components for other file transformations. You
will also easily be able to add logging and reporting at different steps.
You can start recording line by line error messages, and sending them back
to your trading partnet via whatever FTP site they send the files to.

Justin

On Tue, Sep 6, 2016 at 2:52 PM Roger Harman <roger.harman@xxxxxxxxxxx>
wrote:

> I have a CSV file coming from a trading partner that uses formatted dollar
> amounts - "$2,123.45".  I need to remove the thousands separator comma
> before I import the file into a real PF.
>
>
> I receive a number of other formats from trading partners and have SQL
> routines to remove the $ sign, trailer records, etc. from the data stream.
> I remove commas in another file but it is tab-delimited.
>
>
> Looking for suggestions to remove just the thousands separator if it
> exists without killing the comma field separators.  Preferably SQL as a lot
> of this upfront preprocess happens in a generic CL which has the RUNSQL
> routines I need.
>
>
> I've only come up with brute force - search for $1,.. $2,..  etc.
>
>
> Thanks.
>
>
>
> Roger Harman
>
> COMMON Certified Application Developer - ILE RPG on IBM i on Power
>
>
>
>
>
> --
> This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
> To post a message email: MIDRANGE-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/midrange-l
> or email: MIDRANGE-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/midrange-l.
>
> Please contact support@xxxxxxxxxxxx for any subscription related
> questions.