Hi there I've taken the liberty of updating the archiva-consumer-plugin project to compile against the latest 1.1 codebase. I was going to create an archetype for use in creating consumers but I wanted to ask a few questions first and fill in some gaps in my understanding.
* I'm presuming the target JDK for consumers is 1.5? The plugin does not specify this or extend from the archiva-base parent, so defaults to 1.3 * There currently is no closure for the CompleteScan Case though I see comments about adding it. I'd be happy to add this (Seeing as most of the work and testing is done already ;)). Should the completeScan Case be consistent with the begin case (take a repo arg and throw a ConsumerEx)? * I've created a MockRepositorySearch object to mock out the search interaction. It arguably might have been neater to do this with EasyMock, though I was not sure how to install the mock in the PlexusContainer as a component so it will get autowired into the consumer correctly. Not knowing enough about plexus, I went for the soft option, though if someone could point me in the right direction I'll update it. * Is there an express reason the project uses an old(er) version of EasyMock? I've found the newer 2.X syntax to be much more intuitive and concise * Looking at use cases for consumers, simple CRUD like events for artifacts might be high on the list (I'm massively biased, they are high on mine at least ;) ). Might this DiscoverNewArtifactConsumer save a lot of wheel reinvention if promoted (along with Updated an Removed counterparts) to the core consumers project? * there are lots of reusable test fixtures available throughout the project, the repo layer has a few choice candidates; do you see any worth in aggregating these into an artifact to facilitate testing * How do you see complex consumers being configured? might they possibly sometime be given their own configuration web pages? * Currently the Do you see a situation where certain (read-only?) consumers could be configured to run at different intervals (i.e. have different cron specs) A lot of this is driven by a need to receive notifications when artifacts are released to certain repos, email and IM notifications being first on the list. Are there any plans for a notifications api. I have some ideas for making this happen and would be happy to share if there is interest. Anyhow I've created a patch for the above its attached to the issue MRM-651. rgds, stephen -- View this message in context: http://www.nabble.com/Updating-consumer-plugin-to-build-against-the-1.1-apis-tp14682491p14682491.html Sent from the archiva-dev mailing list archive at Nabble.com.
