Walkthrough with sample data
Follow a sample test report from upload to a finished run, without uploading anything. The walkthrough shows that TestLookup accepting an upload is a different step from the run being finished.
InstallTry it
State 1 of 5: Accepted
Nothing plays on its own, and nothing is uploaded or sent to TestLookup or any other service.
The pretend upload gets an accepted reply (HTTP 202) with a task ID and a run ID. TestLookup has accepted the report but hasn't processed it yet.
The buttons appear once the page has finished loading. All the steps are already written out below.
Step by step
Example with made-up data. Nothing on this page contacts a server.
You've just uploaded the sample file. The page shows that TestLookup accepted it, with a separate task ID and run ID, and that the task is pending. Accepted doesn't mean finished.
Choose Observe parsing. The task starts reading the file, saving the results comes next, and the run is still in progress.
Choose Observe finalization. The upload task succeeded, but the run is FAILED. The page shows the sample counts and both pass rates. The one over tests that ran matches the pass rate TestLookup shows.
Choose invalid file. The task fails after the upload was accepted, and the page doesn't guess whether a run exists.
Choose all skipped. The run finishes as STOPPED because no test ran.
Example with made-up data. Nothing on this page contacts a server.
The sample has 100 made-up test results: 70 passed, 15 failed, 5 broken and 10 skipped. Its task ID and run ID are fake. It contains no passwords or keys, personal data, production logs or signed download links.
TestLookup’s pass rate counts only the tests that ran. It divides the passed tests by the passed, failed and broken tests together. For this sample, that’s 70 of 90, or 77.8%, and that’s the figure TestLookup would show.
The walkthrough also shows 70.0%, which is 70 out of all 100 with the skipped tests included. That figure is there only for comparison.
The walkthrough follows the steps in the workflow guide on GitHub.
The walkthrough starts at Accepted, as if you had just uploaded the sample file. It shows an example reply from TestLookup: accepted (HTTP 202), with a task ID and a run ID. Nothing is actually sent.
Accepted means TestLookup has the file. It doesn’t yet mean the file has been read, saved, analyzed or reviewed.
Choose Observe parsing, then Observe finalization. The upload task and the run stay separate the whole time.
In TestLookup, the upload task goes through these stages:
The walkthrough shows the task as pending at Accepted. At Parsing, it notes that saving the results comes next. At Saved and finished, it shows that the task succeeded while the run itself is FAILED.
Details: in the API, the task states are pending, parsing, ingesting, succeeded and failed.
A task can fail even after TestLookup accepted the upload. TestLookup keeps the task record for 24 hours, so a missing task record doesn’t prove that there’s no run. The upload guide on GitHub describes these steps.
Choose invalid file shows a task that fails after it was accepted. The walkthrough leaves open whether a run exists. It never retries a real upload.
The sample run finishes as FAILED, because some tests failed or broke. Choose all skipped, offered at the Parsing step, shows a run that finishes as STOPPED, because no test ran.
A run with no results stays in progress until TestLookup marks it as finished. The test run guide on GitHub sets out these rules.
The last step lists the sample’s counts and both pass rates. It labels the task’s state, the run’s state and any missing information separately.
The run ID shown is the one TestLookup sends back. In TestLookup, use that ID to open a run, rather than a display name.
Back and Reset walkthrough only change what this page shows. Nothing is saved, and they touch no data or activity record in TestLookup.
Reset walkthrough goes back to Accepted. Reloading the page does the same, unless the page address names a step the page recognizes.
The step-by-step text at the end of this page lists every step. It covers the accepted upload, reading the file, the finished FAILED run, the invalid file and the all-skipped STOPPED run. It’s complete even with JavaScript turned off.
When the steps make sense, try them for real. Follow the Quick start on a spare machine that you can wipe afterwards, then import your first report.
