django-treebeard 1.0 has been released.
Direct link: django-treebeard-1.0.tar.gz (don’t forget to read the UPDATING file).
I will be focusing now in another project: django-taggable, so stay tuned.
Share and enjoy.
Release: django-treebeard
Sunday, November 9th, 2008
django-treebeard is a library that implements efficient tree implementations for the Django Web Framework 1.0+. It includes 3 different tree implementations: Adjacency List, Materialized Path and Nested Sets. Each one has it’s own strength and weaknesses (see Benchmarks) but share the same API, so it’s easy to switch between implementations.
django-treebeard uses Django Model Inheritance with abstract classes to let you define your own models. To use django-treebeard:
- Download a release from the treebeard download page or get a development version from the treebeard subversion repository.
- Run python setup.py install
- Add ‘treebeard’ to the INSTALLED_APPS section in your django settings file.
- Create a new model that inherits from one of django-treebeard‘s abstract tree models: mp_tree.MP_Node (materialized path), ns_tree.NS_Node (nested sets) or al_tree.AL_Node (adjacency list).
- Run python manage.py syncdb
| documentation: | treebeard-docs |
| examples: | treebeard-examples (source) |
| tests: | treebeard-tests |
| benchmarks: | treebeard-benchmarks |
Share and Enjoy.
Django 1.0 was released
Wednesday, September 3rd, 2008
Finally, Django reaches 1.0 status. If you have been using 0.96.x, it will be quite a bit of work to upgrade your apps, but it’s worth the effort. According to the release notes:
Since Django 0.96, we’ve made over 4,000 code commits, fixed more than 2,000 bugs, and edited, added, or removed around 350,000 lines of code. We’ve also added 40,000 lines of new documentation, and greatly improved what was already there.
Congrats to everybody who contributed and made this happen! Now, go to the download page and try it.
Release: Feedjack 0.9.14 - Django powered Feed aggregator
Sunday, June 8th, 2008
A new version of Feedjack has been released: Feedjack 0.9.14.
Changes:
Lots of fixes/improvements in feedjack_update.py:
- better unicode handling on logging, this will prevent feeds not updating when using the –verbose flag
- Multithreaded feed fetching, needs the threadpool module, available in http://www.chrisarndt.de/projects/threadpool/. By default 10 threads will be used, can be modified with the -w option. If the threadpool module is not available, the old (SLOW) single threaded method will be used.
- Improved logging. Since multiple feeds can be processed in parallel now, all log entries go in a single line, with the feed id as a prefix.
- Stats in the logging, how many new posts, updated posts, unmodified posts and errors were found per feed and globally, and feed status (ok, unchanged, can’t parse feed, http error, some exception)
- Reporting slow feeds, if a feed takes longer than 10 seconds to process, it will appear in the logs with the tag “(SLOW FEED!)”
This is an update to the most neglected portion of feedjack: feedjack_update.py, the script that updates the sites from cron. It has remained almost untouched since I wrote it, and it was a bit problematic. This new update should solve most known problems with it (and will surely introduce new ones).
Share and Enjoy.
Bye bye twitter
Thursday, April 17th, 2008
bye bye twitter: for x in `./twitter_backup.py -U tabo -P ‘*******************’ | cut -d’ ‘ -f 1 | sort -r`; do ./twixer -D “$x”; done;