How do I extract the numeric parts of a message?

Note that if the message looks like a date, you can use parse_timestamp on it only if the message is consistently structured. There is no parse_duration function.

To extract the numeric parts of a message, you can use the get_regex_all function. For example, assuming you have a column called message in your Dataset containing the string you want to parse, you can use the following:

This will create a new column called numeric_parts that contains an array of all the numeric parts found in the message column.

If you want to extract the numeric parts of a specific string, you can do that as well: