Switch to use astropy's get_readable_fileobj

[astrofrog]

We recently put a lot of effort into developing a robust I/O routine in astropy.utils.data that can read local or remote data, supports caching, and shows a progress bar for downloads, e.g.:

In [5]: with get_readable_fileobj('http://www.eso.org/public/archives/images/original/eso0932a.tif') as f:
            results = f.read()
   ...:     
Downloading http://www.eso.org/public/archives/images/original/eso0932a.tif
|==========================================|  26/ 26M (100.00%)        10s

It supports on-the-fly decompression of gzip and bzip2 too. We should probably make use of this instead of chunk_read.

[keflavich]

This is great! High on the to-do list. Will this be part of the astropy 0.2 release?

[astrofrog]

Yes, it's already in the developer version, so we can already update technically. I would do it, but I'm short on time, so feel free to try it out. The only trick to be aware of is that you have to use it as a context manager, not a normal function.

[keflavich]

@astrofrog - I started translating over some of the downloads. It would be nice if get_readable_fileobj had a verbose option, or some way to turn off the progressbar. I'm going to leave this in my feature branch until the next astropy release to avoid introducing a new dependency on development code.

[eteq]

@keflavich - why is it that you want to turn off the progress bar? It's supposed to figure that out automatically (e.g. it only appears when you're at a terminal or the like)

[eteq]

oh, ignore that comment - better to discuss this on #28