Error: Attempted to skip more lines than in file. Please check the import

[robsuddaby]

Hi all, I'm getting a really strange error when trying to work with a CSV file. I download a host of CSV files regularly and then need to add a column to each of the files, which is done manually. I then input them and import from line 13, wildcard union them and run the rest of my worfklow. This worked fine until this week. I've recently changed from a Windows laptop to a Parallels run copy of Windows run on my Mac and anytime I edit a file (even if I just open, save and close) I get this error message. I don't get it if I import from line 1 but still, nothing has changed in the files and there definitely is the correct amount of data.

Any ideas??

[MarqueeCrew]

@robsuddaby,

That is strange!  You manually add a column to the data?

I run via parallels too.  If you'd like to create some sample data for me, I'd be happy to try it out on my system or I could just watch you via webex and see if I can provide some assistance.  When you "wildcard union" them I imagine that you are wildcarding the the input file name and configuring all of the files to skip the first N records.

As a work-around, you might want to include the filename as a field and read all records.  You can create a record ID for each file (using multi-row formula) grouping by the filename.  After that, you can filter out all records with a recordid less than 13.

Odd.

Mark 

[robsuddaby]

Hi Mark, thanks for your reply.

Yes, counter-intuitive I know but takes 30s to do all of the files each week and I haven't been able to find a quicker solution yet!

Correct - I just tell the input to start on line 13. The issue isn't strictly what it says, however, as every one of the 220 files has identical data and there are absolutely no missing bits of data or gaps or files that are less than 13 rows so I'm starting wonder if it's actually a Mac Excel csv issue rather than Alteryx. As I said, if I open up one of the files and overwrite a number anywhere in the file, it gives me this error which is why I'm now wondering if it's an Excel issue. 

[MarqueeCrew · Accepted solution]

@robsuddaby,

This looks like a "feature" (aka bug) to me.  Having reviewed this with you, I am going to post it as an "idea" to fix it.

Using Alteryx in parallels, I read a csv file and when I skip any number of records (>1) I get the following result:

I also tried to change the dependency from UNC to relative.  No path change occurs when you change the dependency.

Cheers,

Mark

[robsuddaby]

Thanks again Mark, very bizarre huh!