integer interpretation by xls reader node

KNIME Analytics Platform
1

Hello!

I am wondering why, sometimes, XLS reader node interprete columns of integers so bad?

Let me explain. In my xls file, there is a column that contains only integers or blanks. With Excel, each integer is saw as a number with 0 decimal (eg: 412, 535, 5467690, 45451524...) . When the node reads my file, firstly the column type is string (why not), but also integer are a little bit transformed (eg:412.0, 535.0, 5.467690E7,45451524E7...).

So, I try to convert the column in number with the String to Number node, but it fails:

WARN      StringToNumberNodeModel     Values in 410 cells could not be parsed, first error: '4124851.0' (RowKey: Row2, Position: 1)

In another xls file with a column containing also only number, I don't have the problem, xls reader recognize the integer type.

For some reasons, I can not convert this file in csv or txt, that is why I would like to know if it exists a solution.

Thanks,

Nico

2

I get around the problem by saving file with open office. I did not change anything but well, it works.

Let me just add one thing. That would be nice if, in the future, we could choose the type of a column directly in the xls reader node.

Nico