I’m reproducing the issue using two simple modules. In my workflow, I receive text that appears to contain a normal numeric value, but it actually includes an invisible character (U+200A). I’ve tried several cleaning methods, but none have worked except the last one, which is more of a temporary workaround. I’m now looking for a proper, reliable solution that can handle other invisible characters as well.
This seems like a common scenario that others may have encountered before. If anyone has a formal or proven approach to handle this, I’d really appreciate your help.
