Import csv file with 16 digit number give convertdata error

[TheBrenda]

Client is trying to import a csv file that has a 16 digit credit card number. Getting the following error. I do not get the error on my test site. I know that Excel has a 15 digit precision limit. I assume this is part of the problem? The file is delievered and no options for editing it first.

any ideas?

UDA-SQL-0391 A string manipulation exception occurred during the "convertData()" operation.UDA-CUR-0000 Unable to fetch the row.RSV-SRV-0042 Trace back:RSReportService.cpp(722):

[MFGF]

Hi,

Is this using the "External Data" feature in Cognos 10? What version of Cognos 10 is your client running?

Is it possible the data being imported is created in a different locale where , is used as a decimal, instead of .  ? If so. there was a PM for this in 10.1 - fixed in 10.2

http://www-01.ibm.com/support/docview.wss?uid=swg1PM45862

Cheers!

MF.

[TheBrenda]

The number is a 16 digit credit card number. No commas, hyphens, decimals etc. Just straight 16 digits. they are running 10.1.1. The are using "Manage External Data".

[MFGF]

The number is a 16 digit credit card number. No commas, hyphens, decimals etc. Just straight 16 digits. they are running 10.1.1. The are using "Manage External Data".

What locale is your user's login set to use? Is it English? If not, does the issue go away ifthey switch to an English locale?  Just wondering if the PMR was more than the ,/. issue - this might help you to find out?

Just a thought,

MF.

[TheBrenda]

How do you check it? i see in ibm cognos confirmation -> actions -> edit global configuration there is product locales, content locales, currentcies, product locale mappings and content locale mappings. is this what it is talking about?

[TheBrenda]

Well ... finally got the screen print of the error and it is a 'sqlScrollBulkFetch' status='-232'. Which seems to be a SQL error because a SQL int cannot be greater than 2,147,483,647. During the External Data definition the field is defined as Text. but somehow when SQL goes to read the csv file directly it thinks that it is an INT. The client is useing SQL 2005 and the Cognos Data Connection is using OLE DB. Thinking about using SQL Native instead.