how to set it up

The only way I have found to circumvent this annoyance is to play a little trick on iMovie. And that trick is to move the original ~/Movies folder under a different name and put a symlink in its place (to the network folder of your choice).

Here are the steps I took from the Terminal to do this (you will need to have Administrator privileges and you will need to know where your network folder is attached):

• sudo mv ~/Movies/ ~/Movies-original
• ln -s /Volumes/my-network-volume/Movies/ ~/Movies

Now, when I load up iMovie it will use the networked volume. If you don’t like this, you can always move your ~/Movies-original folder back where it was.

remaining issues

Unfortunately, this isn’t a perfect fix. There are two caveats or issues I have found.

One, is that now my hard drive (called Papa Bear) is listed twice in the Project Library:

And two, is that when you delete an event or project from within iMovie, it moves the project into a temporary folder rather than actually deleting it. The next time you open iMovie, it will show up again. I’m sure this is because it thinks it is working with a bona fide Trash system but its not (because it is network storage. See this example:

In order to get around this annoyance, I created a Python script which removes the temporary files for me and created a cron job to run the script from time to time (cleaning up the mess iMovie leaves behind).

#!/usr/bin/env python
'''
Simple script to remove all directories that match the name:
    /iMovie Temporary Items */  [note the star!]

'''
import os
import fnmatch
import shutil

# tuple of directories to search through (you can add more than one)
DIRS = ('/Volumes/leaker-damon/tps/Videos/',)

for directory in DIRS:
    for root, dirnames, filenames in os.walk(directory):
        for directory in dirnames:
            if fnmatch.fnmatch(directory,'iMovie Temporary Items *'):
                shutil.rmtree(os.path.join(root,directory))

make this more better

Am interested to hear how other people have been handling this — this approach certainly isn’t perfect, but it seems to work (for now).

As a self-taught Python user I am still looking for insight on the most pythonic and programmatically-friendly way of accomplishing a given task. In this case, I have written a script that will perform a “clean bzip2″ of a directory (or directories). Mac OS X (via AFP and netatalk, in my case) tends leaves a bunch of ugly files/directories hanging around and I would rather not include them in my compressed tar file.

In writing the script, though, I ran into some questions and I am not sure what the recommended approach would be. The script works, as it is, but I feel its a little hacked together and also a little limited in its application. There is something to be said for programs that “just work” (this does) but I want to take it a little further as an educational endeavor and would like it to appear robust, future-thinking, and pythonic.

My initial questions are:

1. Is there a better way to implement a --quiet flag? [using logging]
2. I am not very clear on the use of Exceptions (or even if I am using it in a good way here) — is what I have done the right approach?
3. Finally, in general: any feedback on how to improve this? (I am thinking, just now, that the script is only suitable for a command line usage, and couldn’t be imported by another script, for example.)

Any feedback is greatly appreciated. Writing a script like this is a good learning tool (for me, at least).

#! /usr/bin/env python

'''Script to perform a "clean" bzip2 on a directory (or directories).  Removes
extraneous files that are created by Apple/AFP/netatalk before compressing.
'''

import os
import tarfile
import logging
from optparse import OptionParser

# Default files and directories to exclude from the bzip tar
IGNORE_DIRS = ('.AppleDouble',)
IGNORE_FILES = ('.DS_Store',)

class DestinationTarFileExists(Exception):
    '''If the destination tar.bz2 file already exists.'''

def ignore_walk(directory, ignore_dirs=None, ignore_files=None):
    '''Ignore defined files and directories when doing the walk.'''

    # TODO: this does not currently take wild cards into account.  For example,
    # if you wanted to exclude *.pyc files ... should fix that.  Perhaps
    # consider moving this entirely into the below function (or making it more
    # reusable for other apps).
    for dirpath, dirnames, filenames in os.walk(directory):
        if ignore_dirs:
            dirnames[:] = [dn for dn in dirnames if dn not in ignore_dirs]
        if ignore_files:
            filenames[:] = [fn for fn in filenames if fn not in ignore_files]
        yield dirpath, dirnames, filenames

def tar_bzip2_directory(directory, ignore_dirs=IGNORE_DIRS,
                                   ignore_files=IGNORE_FILES ):
    '''Takes a directory and creates a tar.bz2 file (based on the directory
    name).  You can exclude files and sub-directories as desired.'''

    file_name = '-'.join(directory.split(' '))
    tar_name = file_name.replace('/','').lower() + ".tar.bz2"

    if os.path.exists(tar_name):
        msg = ("The file %s already exists. " +
                "Please move or rename it and try again.") % tar_name
        raise DestinationTarFileExists(msg)

    tar = tarfile.open(tar_name, 'w:bz2')

    for dirpath, dirnames, filenames in ignore_walk(directory, ignore_dirs,
            ignore_files):
        for file in filenames:
            logging.info(os.path.join(dirpath, file))
            tar.add(os.path.join(dirpath, file))

    tar.close()

def main(args=None, callback=None):
    directories = []

    for arg in args:
        if os.path.isdir(arg):
            directories.append(arg)
        else:
            logging.ERROR("Ingoring: %s (it's not a directory)." % arg)

    for dir in directories:
        try:
            tar_bzip2_directory(dir)
        except DestinationTarFileExists, e:
            print e

if __name__ == "__main__":

    parser = OptionParser(usage="%prog [options: -q ] [dir1] [...dir2]")
    parser.add_option("-q", "--quiet", action="store_true", dest="quiet")
    options, args = parser.parse_args()

    if options.quiet:
        logging.basicConfig(level=logging.ERROR, format='%(message)s')
    else:
        logging.basicConfig(level=logging.INFO, format='%(message)s')

    main(args)

This isn’t meant to be a full-fledged tracking application (see django-tracking) or a real analytic tool (try Google Analytics); rather, it’s meant to simply count the number of hits/view on an object-per-object basis.

How to install:

I find that the easiest way to work with django apps is to symbolically link them to my site-packages directory. It’s easier to update the apps with svn, git, or hg than it is to manually download the files and install by hand.

For me, this is what it looks like (you can cut and paste to make it easy):

• cd ~/src
• git clone git://github.com/thornomad/django-hitcount.git
• sudo ln -s `pwd`/django-hitcount/hitcount `python -c "from distutils.sysconfig import get_python_lib; print get_python_lib()"`/hitcount

Test it works by loading python and checking the version (should not get an error):

>>> import hitcount
>>> hitcount.__version__
'0.1 alpha'
>>>

Adding to your django project:

Add the hitcount app to your INSTALLED_APPS tuple and run syncdb.

There are two additional settings you can add to your settings.py file:

HITCOUNT_KEEP_HIT_ACTIVE = { 'days': 7 }
HITCOUNT_HITS_PER_IP_LIMIT = 0
HITCOUNT_EXCLUDE_USER_GROUP = ( 'Editor', )

HITCOUNT_KEEP_HIT_ACTIVE: is the number of days, weeks, months, hours, etc (timedelta kwargs), that an Hit is kept ‘active’. If a Hit is ‘active’ a repeat viewing will not be counted. After the active period ends, however, a new Hit will be recorded. You can decide how long you want this period to last …

HITCOUNT_HITS_PER_IP_LIMIT: limit the number of ‘active’ hits from a single IP address. 0 means that it is unlimited. You may want to set this, or not.

HITCOUNT_EXCLUDE_USER_GROUP: don’t count any hits from certain logged in users. In the example above, I don’t want any of my editors inflating the total Hit count.

Adding to your urls.py:

You need to add one line to your urls.py file.

You can have this url, itself, point to anywhere you like, but you need to keep the name='hitcount_update_ajax' constant.

from django.conf.urls.defaults import *
from django.views.generic.list_detail import object_detail
from hitcount.views import update_hit_count_ajax

urlpatterns = patterns('',
    url(r'^ajax/hit/$', # you can change this url if you would like
        update_hit_count_ajax,
        name='hitcount_update_ajax'), # keep this name the same

    # other views, for example my object view is:

    url(r'^/video/(?P<object_id>\d+)$', object_detail,
        {   'queryset': Video.objects.all(),
            'template_name': "video/view.html"},
            name='video_detail_view'),

)

Edit your templates

Add the javascript to your object_detail templates (or any template that handles a single object) so that our hit counter is called after the document loads.

Here is what my head includes:

{% load hitcount_tags %}
<script src="/media/js/jquery-latest.js" type="text/javascript"></script>
<script type="text/javascript"><!--
    $(document).ready(function() {
        {% get_hit_count_javascript for object %}
    });
--></script>

When the template is rendered, it should turn into something like this:

$(document).ready(function() {

$.post( '/ajax/hit/',
	{ hitcount_pk : '3' },
	function(data, status) {
		if (data.status == 'error') {
			// do something for error?
		}
	},
	'json');

});

Display the hits!

The most exciting part, is actually displaying your hits. There are four different ways to do it:

    - Return total hits for an object:
      {% get_hit_count for [object] %}

    - Get total hits for an object as a specified variable:
      {% get_hit_count for [object] as [var] %}

    - Get total hits for an object over a certain time period:
      {% get_hit_count for [object] within ["days=1,minutes=30"] %}

    - Get total hits for an object over a certain time period as a variable:
      {% get_hit_count for [object] within ["days=1,minutes=30"] as [var] %}