I recently built up a node series sucessfully and the data I wanted to combine was almost exactly. Some few data I would like to separate, to have a 100 % exact result.
First, I need to replace some columns into a common date format. I don’t want to add columns, but I want to replace the data. My data has the format 31-08-2023 and I would like to replace these into a german format 31.08.2023. How does this work?
Second, I would like to work only with data, who are earlier than a specific date. Therefore it would be useful to have some kind like a user interface, where the KNIME User can type a specific date. And then the node is going to use every data when column date is earlier than the date given. How that would work in KNIME?
If you want to be able to filter based on dates, then its best to convert your date to the KNIME Date Format using the String to Date&Time node:
Its probably best to convert your dates into the same format before using this node, so you could use the String Manipulation node to replace the - with . or vice versa. This will depend on whether the String to Date&Time node can handle the . format that you prefer.
Once this is in Date&Time format, you can then filter with the following node:
Many thanks. First, I added the String Manipulation node. Is it right, that I need one node for each column I want to manipulate? It will cost a lot of loading time. I have a 6 bn rows template and 15 columns to manipulate.
Sorry, I found it already and it worked very well. Thank you.
Second question about the String to Date&Time node. I used New Type “Date” and Date Format “dd.MM.yyyy”. Locale de-DE. The result was 2017-11-07 instead of the german format 07.11.2017. What could be the solution?
You can then include the desired columns and use $$CURRENTCOLUMN$$` as wildcard.