I have an SSIS package where I have directed the error output to a Flat File Destination. The issue is that there are some bad entried in a set of log files, where the source file reads on more delimited column than there are actual columns. (As in there are 26 column headers, and one row will have 27 commas, or delimiters.) I am trying to redirect the row output to put the bad rows into a Flat File for debugging purposes. Although, the package is not able to continue past the error. As soon as it hits the bad row, it fails despite the error output.
Any ideas?
This wouldn't cause a problem exactly as you stated it. If you had a row that contained more delimiters than defined columns, the Flat File Source adapter will just ignore the extra delimiter on the last column. [Value of last column would be "ABCDEF,12345, FEDCBA"]. However, if the extra delimiter is in the middle of the row and shifts columns causing a defined text column to move to a defined numeric or date column, the input source will fail validation and cause the package to fail validation.
Had a similar issue with .csv files and had to resort to using a Script component as a source component instead of a Flat File Source but you have know some .NET to use that approach. It's incredibly powerful.
|||Thanks for the reply!
I believe I have isolated the issue to a particular value in the csv I'm importing. It breaks with a quote, semi-colon, and then some more data, which is what I believe throws the error. Is there a way to have multiple text qualifiers, or tell the Flat File Source to only look for the qualifier if it is next to a comma(delimiter)?
No comments:
Post a Comment