Snowplow is maintained by the pipeline team at Snowplow Analytics and improved on by external contributors for which we are extremely grateful.
First and foremost, please do not log an issue if you are asking for support, all of our community support requests go through our Discourse forum: https://discourse.snowplowanalytics.com/.
Posting your problem there ensures more people will see it and you should get support faster than creating a new issue on GitHub. Please do create a new issue on GitHub if you think you've found a bug though!
If you want to discuss already created issues, potential bugs, new features you would like to work on or any kind of developer chat, you can head over to our Gitter room.
Being an open source company, transparency is very important to us, that's why we try to share as much as possible regarding what we will be working on next so that you can:
- see how your contributions fit into our roadmap
- help us design new features
- share your opinions on the technical direction of the Snowplow pipeline
You can peek into what the pipeline team is working on by looking at the open GitHub projects.
For insights into what we will be working on next, you can look at the RFC category in our Discourse.
The snowplow/snowplow project is split into different Scala projects:
2-collectors/scala-stream-collectorwhich contains the code to collect events as HTTP requests and output raw events to a streaming platform (Kafka, Kinesis, NSQ or PubSub)3-enrich/scala-common-enrich, a library common to all the enrichers listed below which turns the raw events outputted by a collector into validated and enriched events3-enrich/stream-enrich, the pipeline which turns a stream of raw events into a stream of validated and enriched events and pushes them to a streaming platform (Kafka, Kinesis, NSQ or PubSub)
All of these projects can be built and tested with SBT.
The project contains an issue template which should help guiding you through the process. However, please keep in mind that support requests should go to our Discourse forum: https://discourse.snowplowanalytics.com/ and not GitHub issues.
It's also a good idea to log an issue before starting to work on a pull request to discuss it with the maintainers.
If you see an issue you would like to work on, please let us know in the issue! That will help us in terms of scheduling and not doubling the amount of work.
If you don't know where to start contributing, you can look at
the issues labeled good first issue.
These are a few guidelines to keep in mind when opening pull requests, there is a GitHub template that reiterates most of the points described here.
We keep a strict 1-to-1 correspondance between commits and issues, as such our commit messages are formatted in the following fashion:
Component: add issues description (closes #1234)
for example:
Scala Common Enrich: add Vero adapter (closes #1234)
Whenever necessary, it's good practice to add the corresponding unit tests to whichever feature you are working on.
Reviews should happen fairly quickly during weekdays. If you feel your pull request has been forgotten, please ping one or more maintainers in the pull request.
If your pull request is fairly chunky, there might be a non-trivial delay between the moment the pull request is approved and the moment it gets merged. This is because your pull request will have been scheduled for a specific milestone which might or might not be actively worked on by a maintainer at the moment.
We require outside contributors to sign a Contributor license agreement (or CLA) before we can merge their pull requests. You can find more information on the topic in the dedicated wiki page. The @snowplowcla bot will guide you through the process.