- Started a 2-week sprint on Nov. 26 with a plan to release RNode v0.8 the week of Dec. 10. Will be updating the release plan this week.
- Continuing conversations to align on the Mercury MVP requirements.
- Defining user story's for desired outcomes of the platform
- Converting these stories to tests designed to assure platform functionality delivers defined desired outcome
- Developed a dashboard that shows each user story, describes the test, and shows the status of the test
- We anticipate changes to the published timeline for Mercury based on changes to the team last month and work on Mercury requirements this month. The priority now and in Dec. is assuring the currently implemented features work and there is adequate test coverage for all aspects of the project. Next priority will be feature implementation based on the Mercury MVP documentation. Estimate of feature complete is in March 2019.
- Work to ensure stack safety continues. PRs merged this week to make AST equals stacksafe and make AST hashCode stacksafe. Bugs found in the effect stack used in the matcher leading to wrong costs and in the StreamT transformer used in the stack. Bug investigation and resolution is the high priority this week.
- Refactoring business logic to use a generic F and limit use of Task to boundary classes PR is blocked by two failing integration tests. Priority this week is work on the tests and the PR, and to complete this refactoring effort.
- Work in progress to update tests related to the new DAG API.
- Working on improving test suite of Casper features.
- Working on bugs found in community tests related to nodes catching the state of the network.
- Investigation and resolution of bugs is a priority this week. Working with Casper and SRE teams to support.
- Work in progress to improve metrics provided by RNode. Working on a proof of concept of a StatsD implementation.
- Team making changes to the implementation of ProtoUtil.conflicts to better support debugging in Casper by allowing identification of individual channels in produces and consumes in the replay log. Currently this information is all hashed together. Plan is to store sequence of hashes per channel rather than a hash of all channels.
- Working on migration to new infrastructure. Work to be complete by end of the month.
- Investigating test failures, test improvements and implementing fixes in CI.
- Building out comprehensive tests for Mercury deliverables
- Planning to relaunch a test net on new infrastructure during this sprint.
https://developer.rchain.coop