I've used the Palladian nodes to retrieve data from a web server, and parse the returned html to an xml column. All good.
Using the XPath node, I can extract the values from each node into separate columns in a table so long as the column header is specified in the New Column Name field (that is, a fixed value for all columns, followed by a number: COLUMNnnn).
But I can't work out how to extract the node names as column names.
My XPath Value query is //*
XPath query for column name: /name()
Xpath data type: String cell
Multiple tag options: Multiple columns
This gives in the summary section Column Name = Value of //*/name()
But on executing the (knime) node:
Error Xpath 2:23 Execute failed: Xpath expression cannot be compiled
Instead of /name(), I have also tried /local-name(), /node-name().
with the Selenium Nodes, you can check the option "Append additional WebElement information", which will give you a dedicated column with the extracted node names. You can use the same XPath query //* to get all elements on the page. However, the elements will be extracted row-wise (i.e. for each found element, a row is created). I guess it should be easy to transform that into a column-based table.