Changes¶

This change log is used to track all major changes to comics after the first versioned release.

v2.4.3 (2017-04-01)¶

Update all crawlers based on HeltNormaltCrawlerBase after site change.

Crawlers

New: businesscat
New: donthitsave
New: firekanta
New: mollybeans
New: truthfactsgo
New: wyyrd
Update: lunchtu after site change.
Update: thegamercat after feed change.

v2.4.2 (2016-01-19)¶

Get larger images for all GoComics.com comics.

Crawlers

New: bloomcounty2015
New: marydeath
New: sarahcandersen
New: thedanemen
New: unsounded
Update: hjalmar to fix crash.
Update: leasticoulddo to fix image format.
Update: mutts to use new domain.
Update: All HeltNormalt.no comic crawlers to set a User-Agent header.

v2.4.1 (2015-09-04)¶

Fix references to comics list “below”. It was moved to its own page in v2.3.0.

Crawlers

New: criticalmiss
New: dumbingofage
New: fowllanguage
New: jesusandmo
New: ssss
Update: adam4d after site change.
Update: asofterworld after site change.
Update: babyblues after site change.
Update: beetlebailey after site change.
Update: blasternation after feed change.
Update: cyanideandhappiness after site change.
Update: dilbert after site change.
Update: dustin after site change.
Update: hijinksensue after feed change.
Update: hjalmar after file format change.
Update: joyoftech after feed change.
Update: kiwiblitz after site change.
Update: lizclimo after feed change.
Update: lookingforgroup after feed change.
Update: nerfnow to include text.
Update: pidjin after feed change.
Update: satw after site change.
Update: scenesfromamultiverse to include User-Agent in request.
Update: smbc after feed change.
Update: thegamercat after feed change.
Update: threepanelsoul after feed change.
Update: wumovg after file format change.
Update: yafgc after feed change.

v2.4.0 (2015-01-03)¶

Dependencies with new minimum versions:
- Django >= 1.7, < 1.8
- django-tastypie >= 0.12, < 0.13
Dependencies with new maximum versions:
- Pillow >= 1.7, < 2.8
Removed dependencies:
- South
Switched the database migrations from using South to the builtin tool in Django 1.7. Due to this, all old database migrations have been thrown away. If you’re running an old version of comics, please upgrade to the latest v2.3.x release first to get your database entirely up to date, then upgrade to v2.4.x. There are no database changes between v2.3.x and v2.4.x.

v2.3.4 (2015-01-03)¶

Crawlers

New: awkwardzombie
New: fatawesome
New: lizclimo
New: lunarbaboon
Update zits to fetch from better source.
Update nerfnow after feed change.

v2.3.3 (2014-10-06)¶

Crawlers

New: nerdrage
Update lots of comic release schedules.
Update: drmcninja after feed change.
Update: harkavagrant after site change.
Update: hijinksensue to fetch large images.
Update: lookingforgroup to not fetch unrelated comics.
Update: nedroid after feed change.
Update: pennyarcade to work around User-Agent check.
Update: pidjin to ignore repeated image.
Update: redmeat after site change.
Update: subnormality after page change.
Update: wondermark after feed change.
Inactive: marriedtothesea
Inactive: pelsogpoter
Inactive: threewordphrase
Inactive: toothpastefordinner
Inactive: virtualshackles

v2.3.2 (2014-09-11)¶

Dependencies with new maximum versions:
- django_compressor >= 1.1, < 1.5
- Pillow >= 1.7, < 2.6
Add crawler status page to the menu.
Fix position of “Add to my comics” button.
Make an attempt at fixing downloading of image URLs with non-ASCII chars in the URL.

Crawlers

New: blasternation
New: cardboardcrack
New: commitstrip
New: lunche24
New: q2qcomics
Update: asofterworld after site change.
Update: doghouse after feed change.
Update: joyoftech after site change.
Update: kiwiblitz after site change.
Update: seemikedraw after site move.
Update: thegamercat after feed change.
Update: walkoflife is no longer published.

v2.3.1 (2014-06-11)¶

Display error message instead of crashing if a password reset link is reused.

Crawlers

New: iamarg
Update: asofterworld after feed breakage.
Update: crookedgremlins after feed removal.
Update: darklegacy after site change.
Update: hijinksensue after feed change.
Update: lookingforgroup after feed change.
Update: mutts after site change.
Update: phd after feed change.
Update: poledancingadventures after site change.
Update: pcweenies after feed change.
Update: playervsplayer after site change.
Update: sinfest after site change.
Update: stickydillybuns after delayed feed update.
Update: yamac after site change.
Inactive: antics
Inactive: boxerhockey
Inactive: dungeond
Inactive: eatthattoast
Inactive: gregcomic
Inactive: lunch
Inactive: pinkparts
Inactive: somethingofthatilk
Inactive: thechalkboardmanifesto

v2.3.0 (2014-04-07)¶

Dependencies with new minimum versions:
- Django >= 1.6, < 1.7
- django-boostrap-form >= 3.1, < 3.2
Dependencies with new maximum versions:
- cssmin >= 0.1, < 0.3
- cssselector >= 0.8, < 0.10
- django-tastypie >= 0.9.13, < 0.12
- Pillow >= 1.7, < 2.5
Upgraded to Bootstrap 3, which gives way for a refreshed and more responsive/mobile friendly design.
Upgraded to Font Awesome 4.
Upgraded to JQuery 2. IE6/7/8 are no longer supported. IE9 is the oldest IE version you can expect to work with comics.
Moved comics list from bottom of each page to its own page in the top menu. This makes it more available on mobile clients, as well as to new users that don’t know where to find the list after they’ve subscribed to their first comic.
Use Moment.js to show time since fetched timestamp in release meta data. This can’t be done on the server side due to heavy caching.

v2.2.3 (2014-03-31)¶

Make num_releases_since/:id view return 404 instead of 500 for unknown release IDs.

Crawlers

New: adam4d
New: poorlydrawnlines
Update: lookingforgroup after feed change.
Update: pennyarcade after site change.
Update: questionablecontent after site change.
Update: satw to include description text.

v2.2.2 (2013-12-21)¶

Crawlers

Update many comic schedules
Update: hijinksensue after feed change.
Update: partiallyclips to save larger image.
Update: scenesfromamultiverse after feed change.
Update: toothpastefordinner after feed change.

v2.2.1 (2013-11-08)¶

Security: Disabled the GZip middleware to help prevent the BREACH attack. See https://www.djangoproject.com/weblog/2013/aug/06/breach-and-django/ for details.
The feedback form no longer uses the logged in user’s email address as the sender address as this can cause the mail to be rejected due to sender validation and similar anti spam measures. The user’s email is still in the email signature and is now also in the mail’s Reply-To header.
The status page now sorts comics by the number of days since the last release, moving the comics most in need of maintenance to the top of the page.

Crawlers

New: pinkparts
New: poledancingadventures
New: redmeat
New: seemikedraw
Update: amazingsuperpowers after feed change.
Update: axecop after feed change.
Update: bugcomic after site change.
Update: chainsawsuit after feed change.
Update: crookedgremlins after site change.
Update: cyanideandhappiness schedule.
Update: evilinc after site change.
Update: fanboys after feed change.
Update: gregcomic after site change.
Update: gucomics after feed change.
Update: harkavagrant after feed change.
Update: heijibits after addition of User-Agent check.
Update: hjinksensue after feed change.
Update: joyoftech after feed change.
Update: icanbarelydraw after addition of User-Agent check.
Update: kalscartoon after addition of User-Agent check.
Update: kiwiblitz after feed change.
Update: lunch to use feed.
Update: marriedtothesea after feed change.
Update: menagea3 after site change and delayed feed update.
Update: mysticrevolution after site change.
Update: nedroid after feed change.
Update: optipess to add related text.
Update: pidjin after site change.
Update: questionablecontent after site change.
Update: reallife after feed change.
Update: sheldon after site change.
Update: shortpacked after site change.
Update: stickydillybuns to use site feed and include title.
Update: thegamercat after addition of User-Agent check.
Update: thegutters after feed change.
Update: wulffmorgenthaler after site change.
Inactive: picturesforsadchildren
Inactive: radiogaga
Inactive: reveland

v2.2.0 (2013-07-07)¶

New dependencies, and dependencies with new minimum versions:
- cssselect >= 0.8, < 0.9
- Django >= 1.5, < 1.6
- django-tastypie >= 0.9.13, < 0.10
- lxml >= 3, < 4
- defusedxml >= 0.4, < 0.5
Dependencies with new maximum versions:
- django_compressor >= 1.1, < 1.4
- Pillow >= 1.7, < 2.2
- South >= 0.7, < 2.0
Fix crash in importing of old comic sets, which has been broken since v2.1.0.

Crawlers

New: 20px
New: completelyseriouscomics
New: hjalmar
New: kollektivet
New: tommyogtigern
New: truthfacts
New: wumovg
Update: amazingsuperpowers after feed change.
Update: antics after feed change.
Update: beyondthetree is no longer published.
Update: chainsawsuit after feed change.
Update: choppingblock is no longer published.
Update: darylcagle after feed change.
Update: dilbert after removal of feed.
Update: dilbertvg after move to new site.
Update: eatthattoast to not throw exception when the site is broken.
Update: eon is no longer published.
Update: evilinc to be more robust.
Update: exiern to track new storyline.
Update: extralife after feed change.
Update: extraordinary after site change.
Update: fagprat after site change.
Update: geekandpoke after feed change.
Update: gws after site change.
Update: harkavagrant schedule.
Update: havet is no longer published.
Update: hejibits schedule.
Update: heltnils is no longer published.
Update: hipsterhitler is no longer published.
Update: kiwiblitz schedule.
Update: kukuburi is no longer published.
Update: leasticoulddo after site change.
Update: manalanextdoor is no longer published.
Update: manlyguys after feed change.
Update: orneryboy is no longer published.
Update: overcompensating is no longer published.
Update: perrybiblefellowship to be an active crawler again.
Update: picturesforsadchildren is no longer published.
Update: pidjin to ignore repeated non-comic image.
Update: reallife after site change.
Update: sheldon schedule.
Update: slagoon after site change.
Update: smbc after site change.
Update: somethingofthatilk schedule.
Update: subnormality to include title text.
Update: thechalkboardmanifesto schedule.
Update: thegamercat after feed change.
Update: theidlestate is no longer published.
Update: undeclaredmajor is no longer published.
Update: utensokker is published again.
Update: uvod after feed change.
Update: veslemoy is no longer published.
Update: whiteninja is no longer published.
Update: wulffmorgenthaler to work after site change.
Update: wulffmorgenthalerap is no longer active.
Update: yehudamoon is no longer published.
Update: zofiesverden is no longer published.

v2.1.1 (2013-02-26)¶

Crawlers

New: lunchtu
New: mutts. Contributed by Anders Birkenes.
New: pelsogpoter. Contributed by Anders Birkenes.
New: stickygillybuns
New: undeclaredmajor
New: yamac
Update: abstrusegoose after feed change.
Update: bizarro after feed change.
Update: joyoftech after site change.
Update: lookingforgroup after feed change.
Update: thegamercat to be more robust.

v2.1.0 (2012-10-15)¶

Added a Web service interface to the comics instance’s data to enable the development of custom frontends to comics and apps for Android and iOS.
Improved admin interface. A lot of fields on the comic, release, and image models are now read-only, as they are only intended to be changed by the comics_addcomics and comics_getreleases commands. The comics admin interface is mainly intended for browsing and deleting comics/releases/images, not changing.
Proper time zone support for comics crawling. We now calculate the current date at the location a comic is published using time zone aware datetime objects for the current time, which are converted to the comic’s local time zone using pytz.
Removed the setting COMICS_DEFAULT_TIME_ZONE.
Updated time zone data for all crawlers. A lot of releases will now be fetched an hour earlier during daylight savings time, which is now taken into consideration when crawling.

v2.0.1 (2012-10-06)¶

Add dependency on pytz.
Make conversion from publication date to epoch used by 11 crawlers aware of the time zone.
Set sender of feedback emails to the email address of the logged in user.

Crawlers

New: antics
New: beetlebailey
New: choppingblock
New: dungeond
New: dustin
New: exiern
New: pickles
Update: boxerhockey after site change.
Update: exiern after site change.
Update: gregcomic schedule.
Update: havet with better time zone handling.
Update: kiwiblitz after site change.
Update: misfile after site change.
Update: mysticrevolution to be more robust.
Update: reveland with better time zone handling.
Update: spikedmath to only fetch the correct images.
Update: tehgladiators schedule.
Update: thegamercat to fetch full size images.
Update: virtualshackles schedule.
Update: walkoflife with better time zone handling.
Update: whattheduck schedule.
Update: whiteninja schedule.
Update: wulffmorgenthaler to fetch the previous day due to releases being delayed.
Update: yehudamoon after site change.
Update: zelda schedule.
Update: zits after site change.

v2.0.0 (2012-06-11)¶

Version 2 refreshes the entire comics web interface. The aggregation part of comics is mostly unchanged since v1.1.

Design: New design based on Twitter Bootstrap.
User accounts:
- Add user account registration flow, which includes email address verification, login, logout, password change, and password reset.
- Add account management interface.
- Add user information to footer of emails sent from the feedback page.
- Require a user specific secret key to allow access to feeds. (Fixes: #25)
- Add support for requiring an invitation to register as a new user. Set the setting INVITE_MODE to True to require invitation before registration. (Fixes: #29)
“My comics”:
- Replace named comic sets with comic subscriptions associated with users, called “my comics”. An importer for converting old comics sets to “my comics” is included. (Fixes: #26, #27)
- Add buttons to all comic views for adding the comic to “my comics”.
- Extend comics list in the footer to include subscription management. (Fixes: #28, #49)
Comics browsing:
- Orders the “latest” view by fetched time instead of comic name. New content is always at the top. (Fixes: #13)
- Removes browsing of weeks or N days, with the exception of +1 days, which is kept as a “today” view.
- Reimplemented lots of crusty old code using Django’s class-based generic views.
- Reimplement feeds using regular views instead of Django’s feed abstraction to reduce the feed response time enough to not cause timeouts when using e.g. Netvibes to subscribe to feeds. (Fixes: #5)
Comics crawling:
- Try to verify that image files are valid by loading them with PIL before saving them. (Fixes: #17)
- Use PIL instead of server provided MIME types to identify the image type.
- Removed unused check_image_mime_type crawler setting.
- Whitelist GIF, JPEG, and PNG files. All other file types are rejected. (Fixes: #16)
- Blacklisted the GoComics placeholder image.
Development:
- The WSGI file is now also used when using Django’s runserver command while developing, making the development and deployment environments more alike.

v1.1 to v2.0 migration guide¶

New dependencies:
- django-registration >= 0.8, < 0.9
- django-bootstrap-form >= 2.0, < 2.1
Updated dependencies:
- Django >= 1.4, < 1.5
- django_compressor >= 1.1, < 1.2
Settings:
- Removed setting COMICS_SITE_TAGLINE.
- Replaced setting COMICS_MAX_DAYS_IN_PAGE with COMICS_MAX_RELEASES_PER_PAGE.
- Removed COMICS_MEDIA_ROOT and COMICS_MEDIA_URL. As static files now are located under STATIC_ROOT and STATIC_URL, the entire namespace under MEDIA_ROOT and MEDIA_URL are now available for downloaded media, e.g. crawled comics.
Commands:
- loadmeta is now called comics_addcomics. It no longer defaults to adding all comics to your installation, but you must now specify -c all to do so.
- getcomics is now called comics_getreleases
Remember to update your cronjobs.
Project layout:
- Moved manage.py one level higher in the directory structure, to follow the new defaults in Django 1.4. Again, remember to update your cronjobs.
- Moved file with WSGI application from wsgi/deploy.wsgi to comics/wsgi/__init__.py to follow the new default structure in Django 1.4. Remember to update your web server configuration.
As the comic sets functionality have been replaced, the app comics.sets is no longer activated by default. If you’re upgrading from comics v1.x and have existing sets in your database, you should activate the comics.sets app so that your users may import their old comic sets into their new user accounts. Add the following to your local settings file, comics/settings/local.py:
```
from comics.settings.base import INSTALLED_APPS
INSTALLED_APPS += ('comics.sets',)
```
Renamed MetaBase to ComicDataBase, and moved it to comics.core.comic_data. Remember to update any custom crawlers.
Database changes:
- The field Comic.number_of_sets have been removed as it is no longer used. If you would want to rollback from 2.x to 1.x the data in this field can be regenerated, as it’s only a denormalization of data available elsewhere.
- The datetime field Comic.added has been added. It is automatically populated with a date in the far past upon database migration.
- Added two new database indexes to the Release model, which both help a lot towards making comics browsing faster. They will be automatically created on database migration.
All of these changes can be automatically applied to your database. To do so, run:
```
python manage.py syncdb --migrate
```

v1.1.6 (2012-06-10)¶

Bugfixes

LxmlParser.text() now returns an empty list if allow_multiple is True and default is not specified. This is identical to how all other LxmlParser selector methods already work.

Crawlers

New: oatmeal
New: zelda
Update: abstrusegoose has a schedule.
Update: apokalips is no longer published.
Update: asofterworld after feed change.
Update: atheistcartoons is no longer published.
Update: axecop has a schedule.
Update: basicinstructions has a new schedule.
Update: bgobt is no longer published.
Update: boasas is no longer published.
Update: bunny is no longer published.
Update: carpediem is no longer published.
Update: countyoursheep is no longer published.
Update: crfh after site change.
Update: darklegacy does not follow a schedule.
Update: devilbear does not follow a schedule.
Update: dieselsweetiesweb to be more robust to missing elements in the feed.
Update: goblins does not follow a schedule.
Update: gunshow has a new release schedule.
Update: hijinksensue after feed change.
Update: icanbarelydraw has a new release schedule.
Update: kiwiblitz does not follow a schedule.
Update: littlegamers does not follow a schedule.
Update: m is no longer published.
Update: magpieluck is no longer published.
Update: pcweenies does not follow a schedule.
Update: picturesforsadchildren after feed change.
Update: radiogaga has a new release schedule.
Update: rhymeswithwitch is no longer published.
Update: spaceavalanche after feed change.
Update: stuffnoonetoldme is no longer published.
Update: subnormality got a sensible history capability.
Update: tehgladiators does not follow a schedule.
Update: theidlestate does not follow a schedule.
Update: utensokker is no longer published.
Update: uvod got an updated homepage address.
Update: virtualshackles does not follow a schedule.
Update: walkoflife does not follow a schedule.

v1.1.5 (2012-05-09)¶

The regular crawler updates and a small bug fix.

Bugfixes

Handle aggregated images with MIME type image/pjpeg as JPEG images instead of rejecting them.

Crawlers

New: chainsawsuit
New: goblins
New: subnormality
Update: applegeeks was discontinued a long time ago.
Update: applegeekslite was discontinued a long time ago.
Update: calamitesofnature has been discontinued.
Update: duelinganalogs was broken due to feed changes.
Update: fagprat has a new schedule.
Update: fanboys was broken due to feed changes.
Update: heltnils has a new schedule.
Update: hijinksensure was broken due to feed changes.
Update: playervsplayer was broken due to feed changes.
Update: pondus was broken due to a site change.
Update: savagechickens has a new schedule.
Update: theidlestate after site redesign and addition of a feed.
Update: veslemoy has a new schedule.

v1.1.4 (2012-04-07)¶

The regular crawler updates and a performance improvement.

Bugfixes

Store only the name of recently used sets in the session, instead of full set objects. After applying this fix, you should either delete all existing sessions:

$ python manage.py shell
>>> from django.contrib.sessions.models import Session
>>> Session.objects.all().delete()

Or migrate the content of your existing sessions:

$ python manage.py cleanup
$ python manage.py shell

# Then run the following Python script in the Python shell:

from django.contrib.sessions.backends.db import SessionStore
from django.contrib.sessions.models import Session
store = SessionStore()
for session in Session.objects.all():
    data = session.get_decoded()
    set_names = []
    for set in data.get('recent_sets', []):
        if hasattr(set, 'name'):
            set_names.append(set.name)
        else:
            set_names.append(set)
    data['recent_sets'] = set_names
    session.session_data = store.encode(data)
    session.save()
    print '.',

Crawlers

New: kellermannen
New: manalanextdoor
New: thegamercat
New: walkoflife
Update darylcagle after feed change.
Update playervsplayer after feed change.

v1.1.3 (2012-01-29)¶

This release adds 9 new crawlers and updates 46 existing crawlers.

Crawlers

New: beyondthetree
New: dresdencodak
New: extraordinary
New: gunnerkrigg
New: icanbarelydraw. Contributed by Jim Frode Hoff.
New: manlyguys. Contributed by Jim Frode Hoff.
New: menagea3
New: sequentialarts
New: somethingofthatilk. Contributed by Jim Frode Hoff.
Update amazingsuperpowers with new release schedule.
Update billy which is no longer published.
Update bizarro with new release schedule.
Update bizarrono which is no longer published.
Update boasas after site change.
Update bgobt with new release schedule.
Update buttersafe with new release schedule.
Update calvinandhobbes after site change.
Update carpediem after site change.
Update darylcagle after site change.
Update devilbear with new release schedule.
Update eatthattoast after site change.
Update eon after site change.
Update extralife to be more robust.
Update fanboys after site change.
Update gregcomic with new release schedule.
Update gucomics after site change.
Update heltnils after site change.
Update hipsterhitler after site change.
Update kalscartoon after site change.
Update lefthandedtoons with new release schedule.
Update loku which is no longer published.
Update m with new release schedule.
Update mortenm which is no longer published.
Update mysticrevolution after site change.
Update nemibt with new release schedule.
Update nerfnow with new release schedule.
Update optipess with new release schedule.
Update orneryboy with new release schedule.
Update pidjin after site change.
Update pondusno which is no longer published.
Update questionablecontent to be more robust.
Update radiogaga after site change.
Update reallife with new release schedule.
Update reveland with new release schedule.
Update romanticallyapocalyptic to be more robust.
Update savagechickens with new release schedule.
Update sheldon with new release schedule.
Update somethingpositive after site change.
Update stickycomics after site change.
Update tehgladiators after site change.
Update thedreamer with new release schedule.
Update threewordphrase to be more robust.
Update utensokker with new release schedule.
Update wulffmorgenthalerap after site change.
Update yehudamoon with new release schedule.

v1.1.2 (2011-09-18)¶

A couple of bugfixes easing the transition from 1.0.x to 1.1.x by jwyllie83, and some new crawlers.

Bugfixes

Updated South requirement to v0.7, which is needed to support the last migration introduced by comics v1.1.0.
If you use WSGI, you can now add a file wsgi/local.py based off of wsgi/local.py.template to set local settings for WSGI, like the use of virtualenv and debugging settings. This removes the need for changing Git tracked files, like deploy.wsgi for adding e.g. virtualenv support.

Crawlers

New: buttersafe
New: doghouse
New: eatthattoast
New: hejibits
New: optipess
New: savagechickens
New: threewordphrase
New: timetrabble
Update pennyarcade after site change.

v1.1.1 (2011-08-22)¶

Some fixes a week after the v1.1 feature release.

Bugfixes

Fix missing whitespaces on about page after HTML minification.
Add missing CSRF token to feedback form.

Crawlers

Update asofterworld to work after feed change.

v1.1.0 (2011-08-15)¶

New/upgraded requirements:
- Django 1.3
- django_compressor
- cssmin
- jsmin
Page speed improvements:
- CSS and JavaScript is now merged and minified.
- HTML is minified.
- Optional Google Analytics code is upgraded to the asynchronous version.
- All icons have been replaced with sprites to reduce number of elements that must be downloaded.
Slightly refreshed graphical design.
The “all comics” list at the bottom of the page have been changed from a cloud to lists in four columns.
The optional comic meta data search functionality have been removed.
Better handling of inactive comics:
- Add active flag to comics.
- Marked no longer published comics as inactive.
- Inactive comics are no longer loaded by the loadmeta command unless explicitly specified by name or they have been previously loaded. In other words, inactive comics will not automatically be included in new installations.
- Inactive comics are no longer included in the top 10 on the front page.
- Inactive comics are now marked in the comics list on the bottom of all pages.
- Inactive comics are now marked in the comics list page.
- Inactive comics are now excluded from the set edit form, effectively removing them from the set on save.

v1.0.x to v1.1.x migration guide¶

Ordered steps for syncronizing your v1.0.x installation with v1.1.0. You should perform them in order.

Using virtualenv

If you choose to use virtualenv keeping all of comics’ dependencies sandboxed, be sure to activate the environment both in your cronjob and when manually executing manage.py:

source <path_to_virtualenv>/bin/activate
python manage.py getcomics

If you use WSGI, the WSGI file must be modified to support virtualenv. See the source code for how the bundled WSGI file solves this.

New dependencies

There are several new dependencies. All of them are listed in the file requirements.txt and may be installed using pip, optionally inside a virtualenv:

pip install -r requirements.txt

To avoid compiling dependencies which are not pure Python and thus requires the installation of various C libraries and Python’s development packages, it may be wise to use your distribution’s package manger for some packages, like lxml and PIL. E.g. on Ubuntu I would install the dependencies like this:

sudo apt-get install python-lxml python-imaging
pip install -r requirements.txt

This way, lxml and PIL are installed from APT, and pip installs the remaining pure Python dependencies.

Settings changes

Database settings now use the new Django 1.2 format. See comics/settings/base.py for the new default setting and use it as an example for porting your comics/settings/local.py settings file.

Database migration

A new database field has been added. To migrate your database to work with v1.1.0, run:

python manage.py migrate

Warning

You need South v0.7 or later to perform the database migration.

comics v1.1.0’s requirements.txt file only require South v0.6 or later. This is a bug, and the migration will not work if you’re using South v0.6.x.

Static files collection

We now use Django’s new static files system. After installing you need to “collect” your static files. See Collecting static files for how to do this.

v1.0.8 (2011-08-10)¶

Just new and updated crawlers.

Crawlers

New: mysticrevolution
New: theidlestate
Update havet to work after feed removal.
Update reveland to work after feed removal.
Update thechalkboardmanifesto to work after feed change.
Update utensokker to work after feed removal.
Update whattheduck schedule.

v1.0.7 (2011-07-13)¶

Just new and updated crawlers.

Crawlers

New: fagprat
New: gregcomic
New: satw
New: shortpacked
New: stickycomics
New: tehgladiators
Update betty which has moved from comics.com to gocomics.com.
Update bizarro which moved to a new site.
Update brandondraws which is no longer published.
Update countyoursheep after URL changes.
Update darylcagle after change from GIF to JPEG.
Update faktafraverden which is no longer published.
Update fminus which has moved from comics.com to gocomics.com.
Update getfuzzy which has moved from comics.com to gocomics.com.
Update lookingforgroup after feed change.
Update m as it moved from start.no to dagbladet.no.
Update nemibt to work after site change.
Update nerfnow which crashed when it did not find an image URL.
Update peanuts which has moved from comics.com to gocomics.com.
Update pearlsbeforeswine which has moved from comics.com to gocomics.com.
Update pondusbt after URL changes.
Update rockybt to work after site change.
Update romanticallyapocalyptic to use web page instead of feed.
Update roseisrose which has moved from comics.com to gocomics.com.
Update treadingground to not crash if URL is not found.
Update threadingground which is no longer published.
Update yehudamoon which was broken by addition of new images.
Update zits with new feed URL.
Update generic GoComics.com crawler to also support larger Sunday issues.

v1.0.6 (2011-02-19)¶

The 1.0.6 release includes two bugfixes, five new crawlers, and many updated crawlers. Also, most crawler schedules have been updated to make the status page more useful.

Bugfixes

comics.aggregator.lxmlparser.LxmlParser methods now returns an empty list if allow_multiple is True and no value is given for default. This ensures that using the return value in for loops will not fail if no matches are found.
comics.aggregator.crawler.CrawlerBase.get_crawler_release() does no longer create empty releases if the do_crawl method returns false values, like empty lists. It previously only stopped processing if do_crawl returned None.
Remove safe filter from title text, and explicitly use escape, even though they should be implicitly escaped. Thanks to XKCD #859.

Crawlers

A lot of comic release schedule updates.
New: nerfnow
New: romanticallyapocalyptic
New: schlockmercenary
New: spaceavalanche
New: treadingground
Update butternutsquash which is no longer published.
Update charliehorse which is no longer published.
Update garfield to include Sunday editions.
Update hipsterhitler to work after feed change.
Update idiotcomics which is no longer published.
Update inktank which is no longer published.
Update intelsinsides which is no longer published.
Update kiwiblitz to work after feed change.
Update lifewithrippy which is no longer published.
Update pcweenies to work after feed change.
Update petpeevy which is no longer published.
Update smbc to work after feed change.
Update superpoop which is no longer published.
Update thegutters to use feed instead of broken page parser.
Update threepanelsoul to work after feed change.
Update userfriendly to support reruns.
Update wulffmorgenthaler to work after site change.

v1.0.5 (2010-12-29)¶

A couple of bugfixes, and new and updated crawlers.

Bugfixes

Do not throw ParserError in comics.aggregator.lxmlparser when the XML document is a all-space string.
Catch socket.error in comics.aggregator.downloader, like we already do in comics.aggregator.crawler.

Crawlers

New: brandondraws
New: crookedgremlins
New: faktafraverden
New: lunchdb
New: orneryboy
New: reveland
Update foxtrot crawler to work after site change.
Update gws to work again, and add text parsing.
Update havet meta data.
Update lookingforgroup to ignore non-comic releases and fetch multiple pages released on the same day.
Update magpieluck to handle titles without a dash.
Update questionablecontent to not check if the page contains the expected date, as that make us lose some releases.
Update utensokker to use RSS feed.

v1.0.4 (2010-10-23)¶

Yet another minor release bringing a bug fix, four new and five updated crawlers.

Bugfixes

Catch socket.error in CrawlerBase.get_crawler_release().

Crawlers

New: hipsterhitler
New: marriedtothesea
New: stuffnoonetoldme
New: utensokker
Update boxerhockey to use feed instead of site.
Update bugcomic to not fail if URL is not empty, and to work after source site changes.
Update extralife to work after source site changes.
Update gunshow to work after source site changes.
Update questionablecontent to use site instead of feed, since it lacks some releases.

v1.0.3 (2010-07-26)¶

Another minor release bringing 17 new and 11 updated crawlers.

Bugfixes

Make crawlers handle httplib.BadStatusLine exception raised when HTTP responses are empty.
Make crawlers convert lxml.etree._ElementUnicodeResult to unicode objects before saving to the database, to avoid DatabaseError: can't adapt errors.
Handle MIME types like image/jpeg, image/jpeg returned by mimetools.Message.gettype.
Use Crawler.headers for image requests, and not just page requests.

Crawlers

New: apokalips
New: axecop (fixes GH-8)
New: boxerhockey
New: bugcomic (fixes GH-11)
New: carpediem
New: crfh
New: darylcagle
New: havet (fixes GH-7)
New: heltnils
New: intelsinsides (fixes GH-9)
New: misfile (fixes GH-3)
New: notinventedhere (fixes GH-4)
New: pondusno
New: radiogaga
New: scenesfromamultiverse (fixes GH-10)
New: sheldon
New: thegutters
Update 8bittheater which is no longer published.
Update brinkerhoff which is no longer published.
Update ctrlaltdelete to work after source site changes.
Update ctrlaltdeletesillies to work after source site changes.
Update dieselsweetiesweb to work after source site changes.
Update eon with new source site.
Update lunch with new source site.
Update sometingpositive to get all releases.
Update supereffective to work after source site changes.
Update vgcats to work after source site changes.
Update yafgc to work after source site changes.

v1.0.2 (2010-04-11)¶

A minor release to get crawler updates out there. Two new cool but partly immature features are included, as they do not affect existing features or change database schema.

Features

Add status page which for each comic shows when releases are fetched compared to the comic’s release schedule. Contributed by Thomas Adamcik.
Add support for search in comic’s title and text fields, using Haystack. Contributed by Thomas Adamcik.

Crawlers

New: atheistcartoons
New: petpeevy
Update evilinc to work again.
Update uvod to fetch comment too.
Update gunshow to fetch multiple images per release.
Update questionablecontent to work again.
Update basicinstructions to ignore QR Code.
Update partiallyclips with new feed URL.
Update somethingpositive with new image URL.
Update spikedmath to fetch multiple images per release.

v1.0.1 (2010-02-23)¶

This release features 17 new crawlers and three updated crawlers, most by comic’s new contributor Jim Wyllie. Let’s get more of those!

Features

Add links to official sites via redirect page.
Add comics.aggregator.crawler.GoComicsComCrawlerBase for fast gocomics.com crawler creation.
Add headers argument to comics.aggregator.lxmlparser.LxmlParser for adding HTTP headers to requests it makes.
Add time since last release to release-list and comic-list.

Crawlers

New: babyblues
New: calamitiesofnature
New: charliehorse
New: fminus
New: forbetterorforworse
New: girlgenius
New: hijinksensue
New: joelovescrappymovies
New: magpieluck
New: nonsequitur
New: overcompensating
New: pluggers
New: tankmcnamara
New: theboondocks
New: thedreamer
New: wondermark
New: yehudamoon
Update playervsplayer to not miss comics on days with blog posts.
Update questionablecontent to include text below image.
Update kalscartoon after target site change.
Update butternutsquash after target site change.

v1.0.0 (2010-01-27)¶

A week has gone, and here is the 1.0.0 final release. Enjoy :-)

Crawlers

Update uvod crawler to use new feed.

v1.0.0.rc2 (2010-01-19)¶

Second release candidate for 1.0.0. Again, I will bump to 1.0.0 in a week if no new issues arises.

Bugfixes

Make core/0006 migration work on the sqlite3 backend.

v1.0.0.rc1 (2010-01-17)¶

First release, so no list of changes. Will bump to 1.0.0 in a week if no issues arise. Please report any problems at http://github.com/jodal/comics/issues.

Development on comics as a Python/Django project started in February 2007, so this release has been almost three years in the making. Far too long, and I promise it won’t be three years until the next release.