mimecat
A catalogue for working with MIME types and extensions.
Introduction
I started the mimecat package in order to solve two problems at work: determine if a given MIME type was supported by our product and determine potential extensions given a MIME type.
My initial solution used the mimetypes package included with default Python
implementations. mimetypes worked fine, but I eventually encountered a problem
where I had two MIME types that could potentially have the same file
extensions. The implementation of mimetypes only permits an extension to be
associated with one type. I had a need to know all potential types, so I created
mimecat.
There are also other issues with mimetypes, but I don't want to bag on
it. Other
people,
more talented than I, have done that already.
My goals for mimecat are to keep track of MIME types, their associated
extensions, and allow inspection of their associations.
This module can be found at the Python Package Index.
Installation
The easiest way to install is:
$ sudo pip install mimecatIf you want to install as part of a virtualenv:
(venv)$ pip install mimecatIf you want to install from the source tarball, download the source at the Python Package Index and execute:
$ sudo pip install mimecat-<version>.tar.gzUsage
Using mimecat is straightforward:
>>> from mimecat import Catalogue
>>> cat = Catalogue() # this will search in a number of common locations for a
# mime.types file. Loading will stop on first successful
# load.
>>> "text" in cat.known_mediatypes # Media types are the first part of a
# MIME type,
# (see http://www.ietf.org/rfc/rfc2046.txt)
True
>>> "garbage" in cat.known_mediatypes
False
>>> "text/plain" in cat.known_mimetypes
True
>>> cat.get_extensions("text/plain")
['.txt', '.text', '.conf', '.def', '.list', '.log', '.in']
>>> cat.get_types("txt")
['text/plain']
>>> cat.get_types(".txt")
['text/plain']
>>> cat.get_extensions("text/garbage") # KeyError is raised for unknown types.
Traceback (most recent call last): # This is also the case for get_types(...)
File "<stdin>", line 1, in <module>
File "mimecat.py", line 150, in get_extensions
return self._types_to_exts[typename]
KeyError: 'text/garbage'
>>> cat = Catalogue("/path/to/mime.types") # Catalogues can be intialized with
# a custom mime.types file
>>> cat = Catalogue(["/path/to/mime.types", # A list of filenames
"/path/to/additional/mime.types"]) # can also be supplied.
# This will cause Catalogue
# to load all of them.
>>> cat.add_type("text/not-so-plain", [".special_text"]) # Add custom types
>>> "text/not-so-plain" in cat.known_mimetypes
True
>>> cat.get_types(".special_text")
['text/not-so-plain']
>>> cat.add_type("text/not-so-plain2", [".special_text"]) # types can share extensions
>>> cat.get_types(".special_text")
['text/not-so-plain', 'text/not-so-plain2']Caveats
One caveat really. mimecat currently has no support for obtaining MIME type
information from the Windows registry. I don't dispute the utility of having
this support, but I currently lack both the business need and the Windows setup
for testing. If anyone wants to take a crack at coding this up and sending me a
pull request on github, go for it.
