Loading WormBehavior database file in Python without Matlab (iPython Notebook)

gistfile1.txt

{
 "metadata": {
  "name": "Playing with WormBehavior Database"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "First, we need to grab a .mat file from the server:"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "import shutil\nimport mimetypes\nimport os\nimport urllib2\nimport urlparse\n\ndef download_file_locally(url, dest):\n    req, filename, content_type = download_file(url)        \n    if dest.endswith('/'):\n        dest = os.path.join(dest, filename)\n    with open(dest, 'wb') as f:\n        shutil.copyfileobj(req, f)\n    req.close()\n\ndef filename_from_url(url):\n    return os.path.basename(urlparse.urlsplit(url)[2])\n\ndef download_file(url):\n    \"\"\"Create an urllib2 request and return the request plus some useful info\"\"\"\n    name = filename_from_url(url)\n    r = urllib2.urlopen(urllib2.Request(url))\n    info = r.info()\n    if 'Content-Disposition' in info:\n        # If the response has Content-Disposition, we take filename from it\n        name = info['Content-Disposition'].split('filename=')[1]\n        if name[0] == '\"' or name[0] == \"'\":\n            name = name[1:-1]\n    elif r.geturl() != url:\n        # if we were redirected, take the filename from the final url\n        name = filename_from_url(r.geturl())\n    content_type = None\n    if 'Content-Type' in info:\n        content_type = info['Content-Type'].split(';')[0]\n    # Try to guess missing info\n    if not name and not content_type:\n        name = 'unknown'\n    elif not name:\n        name = 'unknown' + mimetypes.guess_extension(content_type) or ''\n    elif not content_type:\n        content_type = mimetypes.guess_type(name)[0]\n    return r, name, content_type\n \nurl = \"ftp://[email protected]/pub/tjucikas/wormdatabase/results-12-06-08/Laura%20Grundy/unc-8/n491n1192/MT2611/on_food/XX/30m_wait/L/tracker_2/2010-03-19___09_14_57/unc-8%20(rev)%20on%20food%20R_2010_03_19__09_14_57___2___2_features.mat\"\n\ndownload_file_locally(url, \"./\")",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Then, we need to open the file up.  Since this is a Matlab file from version greater than 7.3, we need to turn to the [h5py](http://www.h5py.org/docs/intro/quick.html#what-is-hdf5) library to open these .mat files without using Matlab.\n\nLet's confirm we have the right tools:"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "import h5py\nf = h5py.File('unc-8%20(rev)%20on%20food%20R_2010_03_19__09_14_57___2___2_features.mat')\nprint f",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "<HDF5 file \"unc-8%20(rev)%20on%20food%20R_2010_03_19__09_14_57___2___2_features.mat\" (mode r+)>"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Check!"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "import h5py\nf = h5py.File('unc-8%20(rev)%20on%20food%20R_2010_03_19__09_14_57___2___2_features.mat')\n\nprint list(f)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "[u'#refs#', u'info', u'worm']\n"
      }
     ],
     "prompt_number": 12
    }
   ],
   "metadata": {}
  }
 ]
}