The project grows and grows :) The next release of TYPO3-Analytics in version v0.1.0 is out! This release covers various nice features, enhancements and bugfixes.

Get the new version now! Its hot: TYPO3-Analytics v0.1.0

But what was done in the last 65 days and 109 commits? Some numbers (from the github compare view):

  • 70 changed files
  • 5,934 line additions
  • 1,297 line deletions
  • 1 contributor

Here is a list of the most important changes.

Main changes

  • Added a changelog
  • [Consumer Download\HTTP] Added error check for rename operation
  • [Consumer Analysis\PHPLoc] Parse output directly without writing a xml file to disk
  • [Consumer Analysis\CVSAnaly] Moved cvsanaly settings to seperate config file
  • [Consumer] Centralized context generation for logging
  • [Consumer Analysis\CVSAnaly] Set writable path
  • Added (useful) information to composer.json about the project
  • [Consumer] Adjusted all consumer to use the new symfony/process component
  • Added symfony/process component and created a ProcessFactory
  • Updated monolog from 1.7.0 to 1.8.0 and updated various dependencies
  • Removed sudo-usage of system calls
  • Changes auto_delete default option for queues to false
  • [Logging] No new line after every log entry. One line per log entry
  • [Consumer] Adjusted all consumer to use the new structure of DLX
  • [Consumer] Added a method to reject a message and reworked consumer to reject messages
  • [Consumer / Producer] Added support for dead lettering of RabbitMQ and reworked setup of queues and exchanges
  • [Consumer Analysis\GithubLinguist] Fixed fatal error during logging
  • Added a bunch of comments with message formats to every consumer
  • Added class comments to all Commands
  • [Producer ReviewTYPO3OrgCommand] Removed, because this is obsolete
  • [Consumer Download\HTTP] Removed TYPO3 dependency in file name
  • Added File object
  • [Consumer Download\HTTP] Replaced wget download with a curl download
  • Add missing methods to ConsumerInterface
  • Set product under MIT License
  • Added PHP >= 5.4 as requirement
  • Refactored the AMQPConnection and AMQPMessage into a factory
  • Added PSR-2 as coding guideline standard
  • Added first unit tests for several components
  • Integrated Travis-CI as continuous integration environment
  • Integrated Versioneye for dependency checks
  • Integrated scrutinizer-ci for code quality analysis
  • [Consumer Crawler\Gerrit] Add message acknowledgement if no projects are available

Did we reach the goal / roadmap from the v0.0.1 release?

In the release of version v0.0.1 we defined some goals for the next release (this one), but did we reach them? Lets check it:

  • code quality: We have refactored many lines of code and extracted them to seperate classes and introduced common design patterns like Factory. Further more we`ve introduced services like scrutinizer to get some numbers about quality. Standards like PSR-2 were applied.
  • testing: Unit tests was introduced in this release. The code coverage is not very high (18%), but 18% are better than nothing! TravisCI was introduced to check the success of the unit tests continuously.
  • stability: Many bugs were fixed (e.g. Fatal error in GithubLinguist-Logging), external dependencies updated (e.g. Monolog from 1.7 to 1.8), new components introduced to make use of proven code (e.g. symfony/process component) or more error checks added to find some failure faster.
  • flexibility: The product is now released under the MIT license as a permissive free software license. This product can now be official used in several environments (e.g. commercial). We have removed hardcoded dependencies (e.g. TYPO3-Related names in Download\HTTP consumer) or made more attributes configurable (e.g. CVSAnaly config) or support more RabbitMQ queue and exchange features on consumer / producer level.
  • documentation: There were several comments added to the source code. Every class got a class header comment which describes what the class is responsible for. Every consumer got the accepted message format in the docs.

In general this release and the goals was a success. One (important) goal was missed: Write more blog posts about the architecture and concepts of this project. This goal was missed due to the lack of time. Next to this goal during the development of TYPO3-Analytics another open source project CVSAnaly was improved a lot. See the commit history. Why? Because TYPO3-Analytics use this tool and we want to use more features of it ;)

Roadmap for the next release

Currently there is no strict roadmap for the next release. But we will try to focus on some things:

  • Find a name: We are looking for a new name for quite some time (see “TYPO3-Analytics is looking for a new name - suggestions welcome!”). We try to find a new name and apply it to the ecosystem
  • Make it more production ready: We try to make the product more production ready. How? Add some more error checks, add some checks if the system is production ready, add a proof of concept for a more easier deployment (e.g. on a amazon cloud). Ideas welcome!
  • Prove the “productioness”: We want to start some small analysis of open source projects and create reports for it that other people can see how to use this tool and what is it good for and how amazing such small piece of code can be.
  • Spread the work: We want to communicate more (on barcamps, conferences, blogs) about it. We want to try to get more / the first external contributor.

If you want to see some features in it, contact me!

We know that this are hard goals. We try to reach as much as possible of them. If you want to support us, see the next chapter ;)

Be part of the the next release

We are looking for motivated contributors for this project. You do not have to be a programmer. We are looking for designers, frontend people, people for communication, people who got new ideas or want to push some new analysis.

If you do not have (spare) time for this, we are looking for sponsors as well. If you want to push some results, we would be proud to welcome you as a sponsor :)

If you do not know how to contribute or how to get into it: I will offer workshops (for free!) for TYPO3-Analytics, how this works, about the architecture, about the goal, about the motivation or about stuff you are interested in. I can offer rooms in Düsseldorf, Germany for it with internet connection for this. If you are interested, just ping me as well.

If you are interested just contact me via Twitter (@andygrunwald) or by e-mail (github profile).