Re: Converting from scientific to integer/decimal

[Scott Klement <rpg400-l@xxxxxxxxxxxxxxxx>]

> 2)  The data is originally keyed into Excel, but gets saved as tab
> delimited text.  So if they forget to reformat the column (that Excel
> automatically displays as scientic notation) and save the file, the real
> number does not get saved.  Very annoying behavior.

Aha! I thought you were reading an Excel spreadsheet, but you're actually
reading a tab-delimited file. That's where you lost me.

See, if it were an actual Excel spreadsheet, each cell has a type
associated with it: numeric, text, formula, or null.  So, you wouldn't
have to worry about invalid character data in a numeric cell.

In tab delimited, of course, everything is text.

If you're interested, I wrote a program that demonstrates parsing the
actual Excel spreadsheet, you can find it at the following link:
http://www.iseriesnetwork.com/resources/clubtech/index.cfm?fuseaction=ShowNewsletterIssue&ID=17839

Though, the instructions for how to get the POI classes are out of date,
in the above article.  You'd be better off with the instructions from the
articles I wrote for getting these same classes in my article about
creating spreadsheets:
http://www.iseriesnetwork.com/resources/clubtech/index.cfm?fuseaction=ShowNewsletterIssue&ID=19277

> 3)  I was hoping to avoid an MI built-in, but if all else fails I guess I
> could use LBCPYNV.

Carsten Flensburg wrote a sample program demonstrating LBCPYNV and CVTEFN
to do number conversions that you might find helpful if you're not already
familiar with these functions.  His article can be found at the following
link:
http://www.iseriesnetwork.com/resources/clubtech/index.cfm?fuseaction=ShowNewsletterIssue&ID=17495

All of the links above are to the iSeries Network and require a membership
to view.  A free "Associate" membership will work, however.