cbdocloader can't handle UTF-16 input files created on Windows

When using cbdocloader on Windows, and passing in JSON input files created on Windows in UTF-16 format, cbdocloader fails with "No JSON object could be decoded", as seen below.

Input files inside the zip are:
$ file venue_2.json
venue_2.json: Little-endian UTF-16 Unicode text, with very long lines, with no line terminators

When running cbdocloader:
cbdocloader.exe -n localhost:8091 -u Administrator -p [password] -b timbre H:\json\venues.zip

output

[2014-04-10 15:02:09,473] - [rest_client] [9308] - INFO - existing buckets : [u'beer-sample', u'gamesim-sample', u'timbre']
[2014-04-10 15:02:09,473] - [rest_client] [9308] - INFO - found bucket timbre
No JSON object could be decoded
No JSON object could be decoded
No JSON object could be decoded
done

If the input file is unzipped, passed through dos2unix, zipped back up again, cbdocloader will process successfully.

After dos2unix, the input files are:
$ file venue_1.json
venue_1.json: ASCII text, with very long lines, with no line terminators

When running cbdocloader with reformatted input, it works successfully:

./cbdocloader -n localhost:8091 -u Administrator -p [password] -b timbre venues_dos2unix.zip
[2014-04-10 10:24:25,393] - [rest_client] [140735166104336] - INFO - existing buckets : [u'default', u'timbre']
[2014-04-10 10:24:25,393] - [rest_client] [140735166104336] - INFO - found bucket timbre
done