Dbfread

^^^^^^^^^^^^^^^^^^

* Sometimes numeric (N) and float fields (F) are padded with
'*'. These are now stripped. (Reported by sgiangola and Matungos,
issue 10.)

* added ``char_decode_errors`` option which lets you choose how to
handle characters that can't be decoded. (Implemented by ZHU Enwei,
pull request 16.)

* added ``--char-decode-errors`` option to ``dbf2sqlite``.

* added ``dbfread.version_info``.

^^^^^^^^^^^^^^^^^^

* Added support for long character (C) fields (up to 65535 bytes).
(Requested by Eric Mertens and Marcelo Manzano.)

* Added support for Visual FoxPro varchar fields (V). (Thanks to Roman
Kharin for reporting and bobintetley for providing a solution.)

* Bugfix (dbf2sqlite): some table or field names might actually collide with
sql reserved words. (Fix by vthriller, pull request 15.)

* Documented how to convert records to Pandas data frames. (Thanks to
Roman Yurchak for suggesting this.)

^^^^^^^^^^^^^^^^^^

* Bugfix: memo field parser used str instead of bytes. (Fix submitted
independently by Sebastian Setzer (via email) and by Artem Vlasov,
pull request 11.)

* Bugfix: some field parsers called self._get_memo() instead of
self.get_memo(). (Fix by Yu Feng, pull request 9.)

^^^^^^^^^^^^^^^^^^

* DBF header and field headers are no longer read-only. For example
you can now change field names by doing ``table.fields[0].name = 'price'``
or read from files where field sizes in the header don't match those
in the actual records by doing ``table.fields[0].length = 500``.

* fixed some examples that didn't work with Python 3.

^^^^^^^^^^^^^^^^^^

* added currency field (Y). (Patch by Stack-of-Pancakes.)

^^^^^^^^^^^^^^^^^^

* bugfix: a date with all zeroes in the DBF header resulted in
'ValueError: month must be in 1..12'. (Reported by Andrew Myers.)
The ``date`` attribute is now set to ``None`` for any value that is
not a valid date.