Partial row at end of file - best way to handle?

Hi,

I have a file where there is a partial row at the end. It doesn't cause an error, but I get a "partial row" warning during execution.

What do most people do with these partial rows? Do they just ignore them as long as they don't cause errors? Or is it better to handle the partial row with a conditional split, for example?

Just wondering what other people's thoughts on this are. I tend to be of the "get rid of it" camp, but maybe that's overkill? Just looking for opinions, best practices.

Thanks

Hi Sadie,

I'm not sure I know what you mean. Can you paste the last few lines of the file up here? Also, tell us how you have the connection manager configured (i.e. delimited, fixed-length etc...)

-Jamie

|||

Data:

col1, col2, col3, col4, etc

col1, col2, col3, col4, etc

col1, col2, col3, col4, etc

col1, col2, col3, col4, etc

col1 --> partial row

This gives a warning, but doesn't cause an error.

Comma delimited conn mgr.

|||Is it always the last row, or can it be in the middle?|||

It's only the last row.

If you have partial rows in the middle of the file, this causes an error (unless you're ignoring errors). But SSIS ignores the partial row at the very end without any special error handling.

|||

sadie519590 wrote:

It's only the last row.

If you have partial rows in the middle of the file, this causes an error (unless you're ignoring errors). But SSIS ignores the partial row at the very end without any special error handling.

Then you can leave it. No harm, no foul. Why is the last row incomplete though? Can it be fixed at the source?|||

Can't be fixed at the source, these are automatically generated extracts that get automatically ftp'd and loaded.

Which goes back to my original question:

What is the best practice for handling these partial rows at the end of file?

If they're ignored, SSIS gives a warning, but no errors. So the package still runs.

But I can also add a conditional split to remove it, which is extra work, but ensures all junk rows are removed.

I was just wondering what other people do with these partial end of file rows?

|||Is the row a footer or something? What makes it "partial"?

I don't accept "bad" files for input into SSIS. Unless the mandate is given from above, I consider files like this to be bad and those who build it need to fix it.

With that said, it's up to you. I don't think there is a best practice. Use whatever you're comfortable with. If you log warnings, and get tired of sifting through them, then perhaps you want to graciously handle the bad rows.|||Yes, these are footers, not bad rows, per se.|||Not sure there is a best practice on this, as Phil said. If it is not causing a problem, I'd leave it.