CobolParser is already filtering those out (line 374) so the parser doesn't have to see them. My guess is that in your example there is something going wrong with the basic tokenization. It's probably treating ',.' as a single token...
If you play with the log4j properties a bit to expose the tokenizer stages you should be able to verify this assumption.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
So where should we look for a fix? I guess it's somewhere in the preprocessing where ',' has to be ingnored?
Simon
CobolParser is already filtering those out (line 374) so the parser doesn't have to see them. My guess is that in your example there is something going wrong with the basic tokenization. It's probably treating ',.' as a single token...
If you play with the log4j properties a bit to expose the tokenizer stages you should be able to verify this assumption.