HTTP resources as random-access file-like objects
httpio is a small Python library that allows you to access files
served over HTTP as file-like objects (which is to say that they
support the interface of the standard library's BufferedIOBase
class). It differs from libraries like urllib
and requests
in
that it supports seek()
(which moves an internal pointer), and
that read()
makes a request with the Range
header set. It also
supports caching of contents using a configurable block size, and will
reuse TCP connections where possible.
Use pip
to install httpio:
$ pip install httpio
import zipfile
import httpio
url = "http://some/large/file.zip"
with httpio.open(url) as fp:
zf = zipfile.ZipFile(fp)
print(zf.namelist())
Unit tests are provided for the standard behaviours implemented by the library. They can be run with
$ python -m unittest discover -s tests
or a tox.ini
file is provided which allows the tests to be run in
virtual environments using the tox
tool:
$ tox