File Reader node - read row ID

KNIME Analytics Platform
1

Hi there!

Today I held Knime Analytics Platform education/presentation and observed two things:

  • in node File Reader after choosing file (.csv in my case) basic settings are automatically populated. That present an issue if first column is recognized as row ID and actually it is not. There is a preview to check it but if you are not familiar with data (or careless) you might miss it. Here is a beginning of column which is recognized as Row ID.

image

  • customer was impressed with Knime :slight_smile:

Br,
Ivan

2

Hi izapin,
we decided to do this on purpose. Mainly because most users do not have a row id number in their data. And the node will fail if a wrongly chosen row id is selected as rowID.

Best, Iris

3

Hi Iris!

ok but then you can leave the option empty i.e. do not populate it automatically. In this case if a user doesn’t have row id it is ok. If he does have row id he needs to check it.

In current case you populate it automatically and make a mistake. And node did not fail although wrongly chosen row id is selected as row ID if I got you right.

Br,
Ivan

4

Dear @Iris I agree with @ipazin in case of default configuration problems.
Why? Here I explain:
There are three nodes which I think I use more often to read structured data from a file: “File Reader”, “Excel Reader (XLS)” and “CSV Reader”.
All these three nodes have different default options to read row IDs and column headers from data:
In “File Reader” node, row ID is unchecked and column header is checked.
In “Excel Reader (XLS)” both are unchecked.
In “CSV Reader” both are checked.
I think the best case is that all options (to read column headers and row IDs) be unchecked and user decides whether to check the options or not.

5

Hi Both,

thank you for this explanation. I opened a ticket for this and will discuss this with our developers.

Cheers, Iris

1 Like