Why the various JPEG Extensions?

Question

While working on a downloader, I've encountered the following with Python's mimetypes.guess_extension function:

In [2]: mimetypes.guess_extension('image/jpeg', strict=False)
Out[2]: '.jpe'

I knew that jpeg and jpg are valid JPEG extensions, but I didn't know about jpe. So looking at the wikipedia site did reveal the following:

The most common filename extensions for files employing JPEG compression are .jpg and .jpeg, though .jpe, .jfif and .jif are also used

Even more extensions I didn't know of.

So the main question: Why does JPEG have so many (valid) extensions associated with it?

On a related note I'd like to know why Python does return 'jpe' and not 'jpg' or 'jpeg' since I see these used the most.

Answer 1

It appears mimetypes.guess_extension returns the first of all possible extensions:

def guess_extension(self, type, strict=True):
    # ...
    extensions = self.guess_all_extensions(type, strict)
    if not extensions:
        return None
    return extensions[0]

So you'll get whichever is first in the list returned by mimetypes.guess_all_extensions, which turns out to be:

>>> mimetypes.guess_all_extensions('image/jpeg', strict=False)
['.jpe', '.jpg', '.jpeg']

My guess as to why .jpe is also valid:

In DOS and early Windows versions filenames could only have 8 characters and 3 characters for the extension (see the article 8.3 filename on Wikipedia for more info). It could be that they abbreviated JPEG to .JPE or .jpe - which is why we now have .jpe, .jpeg and .jpg.

It's indeed true that .jpeg and .jpg are more common.