Skip to content

Conversation

@saskwoch
Copy link

@saskwoch saskwoch commented Sep 7, 2015

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

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'.

[I've written these notes in retrospect, so apologies if they aren't 100% accurate]

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
Inserted "codecs.register_error("replace_spc", replace_spc_error_handler)"
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant