@christopherpickering, your new delimiter identification functionality looks good to me. I did a little code review and added some tests to ensure everything was running as expected. The most significant change I made was in the use of defaultdict to count the occurrences of a given delimiter.
Let me know if you have any questions or concerns, if not, this is the pull request I will merge. After the merge, I will probably push a version update to PyPI. That version update will have this fix and the addition of the '.835' file extension type.
Anyway, thank you for your contributions to this project. Definitely interested in knowing what would cause the remaining 10% fallout. That 170k EDI file count is impressive! You definitely work at a larger scale than I do.
@christopherpickering, your new delimiter identification functionality looks good to me. I did a little code review and added some tests to ensure everything was running as expected. The most significant change I made was in the use of defaultdict to count the occurrences of a given delimiter.
Let me know if you have any questions or concerns, if not, this is the pull request I will merge. After the merge, I will probably push a version update to PyPI. That version update will have this fix and the addition of the '.835' file extension type.
Anyway, thank you for your contributions to this project. Definitely interested in knowing what would cause the remaining 10% fallout. That 170k EDI file count is impressive! You definitely work at a larger scale than I do.