coveralls
commented
3 years ago Pull Request Test Coverage Report for Build 1ba83332-46e3-444f-9141-effa15fe0a0b
- 19 of 34 (55.88%) changed or added relevant lines in 3 files are covered.
- 1 unchanged line in 1 file lost coverage.
- Overall coverage decreased (-0.01%) to 76.214%
| Changes Missing Coverage | Covered Lines | Changed/Added Lines | % | ||
|---|---|---|---|---|---|
| internal/internal_task_handlers.go | 18 | 20 | 90.0% | ||
| internal/session.go | 0 | 13 | 0.0% | ||
| <!-- | Total: | 19 | 34 | 55.88% | --> |
| Files with Coverage Reduction | New Missed Lines | % | ||
|---|---|---|---|---|
| internal/internal_task_handlers.go | 1 | 76.73% | ||
| <!-- | Total: | 1 | --> |
| Totals | |
|---|---|
| Change from base Build 4df57e44-e180-4974-be24-9fb3e7e1dc38: | -0.01% |
| Covered Lines: | 10327 |
| Relevant Lines: | 13550 |
What changed? Decouple
Clientfrom "worker" part of the SDK. There was only a single place where NewClient is instantiated from within worker. Several other places (related to heartbeats) instead call a common function shared between Client and workers. Do the same here as well.Why? This coupling makes it harder to make grpc changes, as switching
workflowserviceclient.Interfacein one place quickly propagates everywhere. With this change, it becomes possible to independently changeClientand workers - thus enabling smaller changes.How did you test it? Existing tests.
Potential risks