Visokio website     Downloads     Video tutorials     KnowledgeBase  
Function to drop new variables - Visokio Forums
Function to drop new variables
  • VoteVote Up1Vote Down     neilmerchant June 17, 2011 10:03AM
    Hi
    I have a spreadsheet I get weekly that has a time stamp in A1 - as this changes each week it adds this variable into my final dashboard.
    It would be cool if there was something to say "keep" variables x y z and any new ones automatically get deleted.
    Regards
    Neil
  • 10 Comments
  •     steve June 17, 2011 10:23AM
    So where are your headers, if A1 contains a timestamp? Can you configure the Excel source to skip the column/row?
  •     neilmerchant June 17, 2011 11:44AM
    Hi Steve
    I am using batch append (else I could do that) as we get a new file each week and we want to read them all in.
    Regards
    Neil
  •     steve June 17, 2011 11:54AM
    You should be able to change the Batch Append file type filter to XLS/XLSX files, thereby revealing the Excel specific options for skipping rows/columns - does this work?
  •     neilmerchant June 17, 2011 12:00PM
    Hi Steve
    You don't have the "select table data" option in batch append when xlsx or xls is slected in batch append - only row options - I need row 1 as the headers are in row 1.
    Regards
    Neil
  •     steve June 17, 2011 12:56PM
    If A1 contains a timestamp, not headers, then your headers aren't in row 1, they're in row 2. Unless you're counting from zero (which Omniscope doesn't)?
    Assuming your timestamp is in row 1, your headers are in row 2, and your data is in rows 3 onwards, you should be able to enter:
    Has header: [tick]
    Header rows: [2] to [2]
    Data rows: [3] to [blank]
    These options should be available for batch append, instead of the "Select table data" option you're used to for individual files.
  •     steve June 17, 2011 12:56PM
    If I've misunderstood, post a sample file here which shows your rows/headers structure.
  •     neilmerchant June 17, 2011 1:33PM
    Hi Steve
    Please see attached.
    The headers are in row 1, but A is all useless data so we need to drop it.
    N
    Attachments
    Book1.xlsx 9K
  •     steve June 17, 2011 2:38PM
    Since column A isn't blank apart from the timestamp header, you can't use "Delete empty data" to exclude blank columns.

    Your best bet then is probably to exclude the headers altogether using the "has header" and "data rows" options, and to use Field Organiser to manually set all headers.

    This will only work if all batch-appended files have the same structure.
  •     neilmerchant June 20, 2011 5:12AM
    Here's my next problem... there are 330 columns of data... I can't remap them all!
  •     steve June 20, 2011 7:59AM
    You could try importing without headers and without parsing numbers/dates, where the first (header) row is treated as a data row, then using search/replace or formulas to nuke the timestamp, then writing back out as a temporary CSV file using the same "no headers" option. Then read THAT using a separate file block.

Welcome!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Apply for Membership