Saskwoch patch 1

[saskwoch]

I needed to extract >70k image attachments from a >140k message mbox archive downloaded from Gmail.

PabloCastellano's script was a tremendous start for me, it just needed some changes to work on Python 3.4.3 (Windows) and with my particular data set.

Note: I ran my script within IDLE and used a literal filename rather than pass in the filename as an argument, hence the version of the script here is untested 'as is'.

List of changes :-

1. All print statements updated to print() function for Python 3.4.3
2. len(mb) gave me a performance hit so I used a numeric literal instead, hence this version of the script is untested 'as is'
3. I was getting "TypeError: Can't convert 'bytes' object to str implicitly" for both "subject = subject + l[0]" and "em_from = em_from + l[0]" statements so added decode to both and encode to "content = content[fh:]"
4. I was getting "TypeError: memoryview: str object does not have the buffer interface" for "extract_attachment(payl)" statement so added decode=False to preceding statement
5. I got "TypeError: must be str, not bytes" for "fp.write(content)" statement so opened file as binary
6. I was getting "UnicodeDecodeError: 'utf-8' codec can't decode byte 0xfc in position 27: invalid start byte" for "em_from = em_from + l[0].decode('utf-8')" so added the "replace_spc" taken from http://www.gossamer-threads.com/lists/python/python/780611#780611
7. A couple of messages threw "AttributeError: 'NoneType' object has no attribute 'find'" so added the exception handling to skip and didn't investigate further