Run, stop & re-run
Ways to run
Section titled “Ways to run”| Action | Typical use |
|---|---|
| Play (gutter left of a test name) / Run → Run Test at Cursor | Only the test (or task) that owns that line — no extra run configuration |
Run (toolbar) / Run → Run File (F5) |
Tests in the current open .robot file — not a text selection |
| Project (toolbar) / Run → Run Project | Entire project |
| Tests (activity bar) | One test, one suite, all, current file, or failed |
| Re-run Test | A single failure from the last run (Failed Tests) |
Play controls appear on .robot suites after the outline is ready. They use the same run configuration as toolbar Run. While a run is in progress the buttons stay disabled. .resource files and Python modules do not get play controls — use the Tests tree for those workflows.
A file or single-test run still executes parent *** Settings *** in __init__.robot (Suite Setup / Teardown). Robot Studio starts Robot at that parent folder and filters with --suite, so a session created in tests/__init__.robot is available when you run tests/posts/posts_api.robot or one test inside it. Run Project is unchanged.
Run configurations
Section titled “Run configurations”The toolbar selector immediately before Run / Project / Stop chooses a run configuration — a named execution context for the next run.
| Selector | Meaning |
|---|---|
| Default | No extra tags, variables, or environment pin. Uses the active toolbar environment. |
| Smoke - Staging (example) | Applies that configuration’s include/exclude tags, variables, variable files, and optional environment pin |
Run and Project keep their usual meaning (current file vs whole project). The selector only supplies execution context. Tests-tree Run, Run Failed, Re-run Test, and the editor gutter play control use the same selected configuration.
Selecting a configuration does not change the toolbar environment. Packages, Libraries, and language intelligence stay on the active environment. A configuration may pin an environment for that run only; the environment actually used is recorded on the run in Reports.
Create and manage
Section titled “Create and manage”From the selector: New Configuration… or Manage Configurations… (dialogs — not a new sidebar or Settings page). In Manage, Duplicate is a first-class action so you can clone Smoke - Dev into Smoke - Staging and change a few fields.
In Manage Configurations…, the active configuration shows a green, disabled In Use button. Choose Use on another row to make that configuration active for the next run.
Configurations are stored with the project at .robotstudio/run-configurations.json (same Studio metadata folder as environments and reports).
Advanced Robot arguments are an escape hatch: one argv token per row, not a shell command.
| Allowed | Blocked (Studio-owned) |
|---|---|
--listener + class/module (two rows), --pythonpath, --loglevel, … |
--outputdir, --output, --log, --report (and short forms) |
Studio always attaches its own progress listener first. Your --listener rows run in addition. Put the flag and value on separate rows — a single cell like --listener helper.Foo is rejected.
Logs stream into Live Output while a run is in progress (Robot is started with unbuffered stdout so lines are not held until the process exits). The Now Running panel on the right shows the live call stack — suite, test, and current keyword — plus elapsed time while the run is active. After the run ends, it keeps the last location until the next start.
Stop cancels the run (with a confirmation that shows what is currently executing). The toolbar and status switch to Stopping until Robot exits — that can take a few seconds if a keyword is blocked. Output so far is kept; the HTML report may be incomplete.
Large runs
Section titled “Large runs”If a project-wide run would execute more than the configured threshold (default 100 tests), Robot Studio asks for confirmation. Change the threshold in Settings → Execution → Large Run Threshold.
After you stop or finish
Section titled “After you stop or finish”- Finishing a run does not yank you into another view; use the View Report toast when you want results.
- The toolbar Last: chip is the outcome of that run: Passed, Failed, No tests, or Error. A keywords-only file, or a tag/filter that matches nothing, is No tests — not Failed.
- Failed Tests lists failures with Jump to Source and Re-run Test. It only appears when tests failed. A pass, or a Robot error with no tests (for example no cases matching a tag), stays on Live Output. Long failure lists scroll so Live Output stays visible below.
- Typical fix loop: Failed Tests → Jump to Source → edit → save → Re-run Test (or Run /
F5) → Reports if you need the HTML log. - Run history is available from the reports flow — see Reports & failed tests.
- Opening or creating another project clears the live console and failed-tests list for the previous project so you never see the wrong run’s output.