let the parser read content either from a file or from a file-like object

[omerholz]

reading and loading the WebVTT content are coupled. As a result, if the webvtt content to parse is not on the file system, we need to put it to the file system for the parser to be able to load it.
This PR decouples these 2 functions.
It keeps full backward compatibility, including for "internal" methods prefixed with _.
With the addition of read_buffer method to the WebVTT class, this PR adds support for parsing any kind of file-like object:

import webvtt
import requests
from io import StringIO

payload = requests.get('http://subtitles.com/1234.vtt').text()
buffer = StringIO(payload)
for caption in webvtt.read_buffer(buffer):
    print(caption.start)
    print(caption.end)
    print(caption.text)

[glut23]

Thanks for that @omerholz !

[sdavids13]

Any plans on publishing a new version of this library to pypi? Having this particular PR integrated would be extremely helpful for me... Thanks!

[glut23]

Hi @sdavids13 I was planning to do that. I will publish it soon. Thanks!

[glut23]

Hi @sdavids13 I just released 0.4.3.