docs: add quick-start commands for testing in tests/README.md#1171
docs: add quick-start commands for testing in tests/README.md#1171VaibhavUpreti merged 1 commit intoTracer-Cloud:mainfrom
Conversation
Greptile SummaryThis PR adds a "Quick-start commands" table to Confidence Score: 5/5Safe to merge — documentation-only change with all referenced commands verified against the Makefile. All commands exist in the Makefile and the table adds genuine value. The only finding is a P2 description inaccuracy for test-cov, which does not block merge. tests/README.md — minor description accuracy for the test-cov row. Important Files Changed
Reviews (1): Last reviewed commit: "docs: add quick-start commands for testi..." | Re-trigger Greptile |
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
|
|
||
| | Goal | Command | When to use it | | ||
| |---|---|---| | ||
| | Run the default unit suite with coverage | `make test-cov` | First thing to run locally; no live infrastructure required. | |
There was a problem hiding this comment.
test-cov
The test-cov target runs pytest -n auto … --ignore=tests/e2e/kubernetes_local_alert_simulation --ignore=tests/synthetic -m "not synthetic", which still collects and executes the rest of tests/e2e/ (Grafana, RDS, upstream_lambda, etc.). Those e2e paths can require live services. Describing this row as "no live infrastructure required" may mislead a developer who runs make test-cov expecting a fully offline run, in contrast with the test-full row which correctly flags "requires live infrastructure" despite running the very same e2e tests.
|
@retr0-kernel take a look at review comments |
|
|
||
| | Goal | Command | When to use it | | ||
| |---|---|---| | ||
| | Run the default unit suite with coverage | `make test-cov` | First thing to run locally; no live infrastructure required. | |
There was a problem hiding this comment.
test-cov runs pytest with coverage over the whole default tree minus synthetic ignores and -m "not synthetic"; it is not limited to unit tests. Prefer wording like default CI-aligned suite or standard pytest suite with coverage.
| | Run a live RCA end-to-end test | `make test-rca` | When you need to validate a full investigation against real services. | | ||
| | Run a single RCA fixture | `make test-rca FILE=<name>` | When iterating on one specific alert scenario. | | ||
| | Run the full suite including e2e | `make test-full` | Pre-release or CI; requires live infrastructure. | | ||
| | Run synthetic scenarios (no live infra) | `make test-synthetic` | When testing scenario logic without external service dependencies. | |
There was a problem hiding this comment.
Synthetic tests are LLM-driven RCA scenarios per repo docs; they still need configured LLM/API access even when customer infra is absent. Drop external service dependencies or qualify it as no customer live infra.
VaibhavUpreti
left a comment
VaibhavUpreti
left a comment
There was a problem hiding this comment.
Thanks a lot for fixing this @retr0-kernel , and for the review @muddlebee ! Merging this for velocity, will fix in a separate patch
|
🏆 @retr0-kernel did not come to play. PR opened. Review survived. Merged clean. Retire the jersey. 🎽
👋 Join us on Discord - OpenSRE : hang out, contribute, or hunt for features and issues. Everyone's welcome. |
3 participants
Fixes
Resolved issue number : #1130
Describe the changes you have made in this PR -
Added a table with test commands in
tests/README.mdwith three columnsGoal,Command, andWhen to use it. The table includes commands for running different types of tests, such as unit tests with coverage, integration verification, live RCA end-to-end tests, single RCA fixtures, full suite including e2e, and synthetic scenarios. Each command is accompanied by a brief description of when to use it.Code Understanding and AI Usage
Did you use AI assistance (ChatGPT, Claude, Copilot, etc.) to write any part of this code?
NOTE → Used AI to rummage through the existing
tests/README.mdand extract relevant information to create a concise table of test commands. The AI helped in structuring the information in a clear and organized manner.If you used AI assistance:
Checklist before requesting a review