Darrelle had a problem where removing an end-of-line at the end of an ascii file fixed a problem she saw. This shouldn't happen, and I need to study what was going on.
Interestingly, I've verified that when there are additional columns, the tailing newline has no effect. This is why the problem hasn't been noticed. See PJ78-Z-2.txt in the same directory.
Last edit: Jeremy Faden 4 days ago
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm adding a test to automatically skip any empty line, when the delim parser is used. This adds a trim operation to the record parsing, The problem is the code relied on the wrong number of fields found to skip the record, but with there is an empty line there is still one field.
This will continue to slow performance, but there are other once-per-line activities which could be cleaned up, like creating a matcher object for each record.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Yes, this is the case. Compare:
and
which doesn't end in fill (or a newline).
Interestingly, I've verified that when there are additional columns, the tailing newline has no effect. This is why the problem hasn't been noticed. See
PJ78-Z-2.txtin the same directory.Last edit: Jeremy Faden 4 days ago
I'm adding a test to automatically skip any empty line, when the delim parser is used. This adds a
trimoperation to the record parsing, The problem is the code relied on the wrong number of fields found to skip the record, but with there is an empty line there is still one field.This will continue to slow performance, but there are other once-per-line activities which could be cleaned up, like creating a matcher object for each record.
See https://sourceforge.net/p/autoplot/bugs/2904/.