A fairly flexible app that will analyze and report on links in any model that you register with it.
Links can be bare (urls or image and file fields) or embedded in HTML (linkcheck handles the parsing). It's fairly easy to override methods of the Linkcheck object should you need to do anything more complicated (like generate URLs from slug fields etc).
You should run its management command via cron or similar to check external links regularly to see if their status changes. All links are checked automatically when objects are saved. This is handled by signals.
django-linkchecks requires Python 2.6 and Django 1.6. It is Python 3 compatible.
- Install app to somewhere on your Python path (e.g.
pip install django-linkcheck).
- Add something along the lines of
examples/linklists.pyto every app that either:
- has models that contain content (e.g. url/image fields, chunks of markup or anything that gets transformed into a IMG or HREF when displayed
- can be the target of a link - i.e. is addressed by a url - in this case make sure it has an instance method named 'get_absolute_url'
./manage.py syncdbon Django 1.6).
Add to your root url config:
/admin/linkcheck/from your browser.
notifications.py is completely optional. It works with
django-admin-blocks to display a notification about broken links as
shown in the screenshot above.
We are aware that this documentation is on the brief side of things so any suggestions for elaboration or clarification would be gratefully accepted.
The following class attributes can be added to your
Linklist subclasses to
customize the extracted links:
object_filter: a dictionary which will be passed as a filter argument to the
filterapplied to the default queryset of the target class. This allows you to filter the objects from which the links will be extracted. (example:
object_exclude: a dictionary which will be passed as a filter argument to the
excludeapplied to the default queryset of the target class. As with
object_filter, this allows you to exclude objects from which the links will be extracted.
html_fields: a list of field names which will be searched for links.
url_fields: a list of
URLFieldfield names whose content will be considered as links. If the field content is empty and the field name is in
ignore_empty, the content is ignored.
ignore_empty: a list of fields from
url_fields. See the explanation above. (new in django-linkcheck 1.1)
image_fields: a list of
ImageFieldfield names whose content will be considered as links. Empty
ImageFieldcontent is always ignored.
This command goes through all registered fields and records the URLs it finds. This command does not validate anything. Typically run just after installing and configuring django-linkcheck.
For each recorded URL, check and report the validity of the URL. All internal
links are checked, but only external links that have not been checked during
LINKCHECK_EXTERNAL_RECHECK_INTERVAL minutes are checked. This
interval can be adapted per-invocation by using the
-e) command option (in minutes).
You can also limit the maximum number of links to be checked by passing a number
--l) command option.
Default: 10080 (1 week in minutes)
Will not recheck any external link that has been checked more recently than this value.
A string applied as a regex to a URL to determine whether it's internal or external.
Currently linkcheck tests whether links to internal static media are correct by wrangling the URL to be a local filesystem path.
It strips MEDIA_PREFIX off the interal link and concatenates the result onto settings.MEDIA_ROOT and tests that using os.path.exists
This 'works for me' but it is probably going to break for other people's setups. Patches welcome.
Pagination is slightly peculiar at the moment due to the way links are grouped by object.
The length of the URL field. Defaults to 255 for compatibility with MySQL (see http://docs.djangoproject.com/en/dev/ref/databases/#notes-on-specific-fields )
The timeout in seconds for each connection attempts. Sometimes it is useful to limit check time per connection in order to hold at bay the total check time.
SITE_DOMAIN and LINKCHECK_SITE_DOMAINS
Linkcheck tests external and internal using differently. Internal links use the Django test client whereas external links are tested using urllib2.
Testing internal links this as if they were external can cause errors in some circumstances so Linkcheck needs to know which external urls are to be treated as internal.
Linkcheck looks for either of the settings above. It only uses SITE_DOMAIN if LINKCHECK_SITE_DOMAINS isn't present
SITE_DOMAIN = "mysite.com"
would tell linkchecker to treat the following as internal links:
mysite.com www.mysite.com test.mysite.com
If you instead set LINKCHECK_SITE_DOMAINS to be a list or tuple then you can explicitly list the domains that should be treated as internal.
If django-filebrowser is present on your path then linkcheck will listen to the post-upload, delete and rename signals and update itself according
- Tests can be run standalone by using the runtests.py script in linkcheck root:
- $ python runtests.py
If you want to run linkcheck tests in the context of your project, you should include 'linkcheck.tests.sampleapp' in your INSTALLED_APPS setting.