Playback is what turns the Recorder from a capture tool into a development environment. It executes your statements directly on the live application, in the recorder browser, on the connected device, or in the open document, and shows you the result of every step as it happens. You use it to prove each step works as you build it, to rerun a flow after a change, and to troubleshoot failures with the application right in front of you. No formal test job is needed, and nothing about your project's executions is affected.
Playback supports all of your automation logic, including non-UI statements such as database and API steps that appear in the same logic.
Starting a Playback
You can start playback in two ways:
- The Playback button in the Logic Editor toolbar. Clicking it runs all statements, the dropdown arrow offers the other scopes.
- The right-click menu on a statement or a selection of statements in the Logic Editor, for targeted playback.
Both offer the same scope options:
| Option | What Runs |
|---|---|
| Play All | Every statement, top to bottom |
| Play Selected | Only the selected statement(s) |
| Play Up To Here | From the beginning through the selected statement |
| Play From Here | From the selected statement to the end |
Make sure the application is in the right state for the statements you are about to run. If your selection starts with signing in, the application should be on the login screen. Use the recorder in OFF mode to navigate there first if needed.
Watching It Run
As playback proceeds:
- The application responds live. You see each step performed in the recorder browser or on the mirrored device. In multi-modal logic, the recorder view follows the statement being executed, switching to show whichever browser or device is being driven at that moment.
- The margin tells the story. Icons beside each statement move from pending to in-progress to success or failure, so you can see exactly where the flow stands.
-
The Playback Console logs the detail. Status, messages, errors, and timing for every statement appear at the bottom of the workspace. When a step fails, this is where you find out why.
The recorder mode does not matter for starting playback; it works whether the recorder is OFF, Fluid, Intent, or ON. While playback is in progress, recording is temporarily unavailable so that stray interactions cannot get captured mid-run. If you need to intervene, for example to dismiss an unexpected dialog, switch the recorder to OFF and interact manually; this neither records anything nor disturbs the playback queue.
Auto Playback: Validate as You Build
Auto Playback executes each statement automatically as you record or type it, so your logic is continuously proven against the live application. It is enabled by default and can be toggled from the Playback Console toolbar.
How it plays with the recording modes:
- In Fluid mode, the interaction already happened in the application, so Auto Playback serves as confirmation that the step was captured correctly.
- In Intent mode, the statement is created first; Auto Playback runs it on the application as soon as you provide the required argument values, so the application catches up to your logic.
Either way, the Logic Editor and the application stay in sync without you triggering anything.
Requests are queued: keep recording without waiting for earlier statements to finish. The Playback Progress indicator in the console toolbar shows the queue, and you can stop pending requests from there; the statement currently running completes before stopping.
Note: Auto Playback applies when recording from the live application or typing statements directly. It is not available when recording statements from a View (offline).
When an Element Cannot Be Found: Playback Pauses
Not every failure needs the manual loop. When playback cannot find an element on the live target, it does not simply fail the step: it pauses and helps you resolve it on the spot.
- With an Autopilot subscription, Autopilot analyzes the screen and proposes the element it believes is the match, highlighted on the target. Accept the suggestion and playback records the corrected identification and moves on; if the suggestion is wrong, double-click the correct element yourself.
- Without Autopilot, the pause asks you directly: double-click the element on the screen to identify it, or continue with the error.
Either way, you then choose to continue playback from that point or stop. The same pause is how cross-platform instances get created when running mobile logic on the other platform; see Cross-Platform Portability: iOS and Android.
Troubleshooting a Failing Step
For failures beyond element identification (wrong application state, a genuine defect, a timing issue), everything you need is already open:
- Read the failure message in the Playback Console.
- Check the application state on the recorder side; often the application simply is not on the expected screen.
- Fix the statement in the Logic Editor, or turn the recorder OFF and bring the application to the right state manually.
- Right-click the statement and Play From Here (or Play Selected) to confirm the fix.
This tight loop, fail, inspect, fix, replay, is the everyday rhythm of developing with the Recorder.
Playback Settings
Open the settings from the gear icon in the Playback Console toolbar. The ones you will touch most:
| Setting | What It Does |
|---|---|
| Application Environment | Which environment (global properties) playback uses |
| Element Timeout | Maximum seconds to wait for an element |
| Page Timeout | Maximum seconds to wait for a page or screen to load |
| Enable Self-Healing | For Smart Locator elements, auto-recover when the primary locator fails |
| Enable Auto Playback | Toggle automatic statement execution |
| Start Playback Job on Login | Start the playback job automatically when you log in to ACCELQ (requires a running Agent) |
| Playback Job Inactivity Timeout | Idle period after which the playback job terminates |
| Execution Parameters | Values for the project's execution parameters during playback |
Tip: Shorter timeouts are appropriate during recording sessions, because the application is expected to already be in the right state; you want failures to surface fast. Formal executions may warrant longer timeouts.
Playback vs. Formal Execution
Playback is for development. When your logic is ready, formal execution through a Scenario or Test Suite gives you the full run with reporting.
| Aspect | Playback (Recorder) | Formal Run (Scenario/Suite) |
|---|---|---|
| Purpose | Real-time validation while developing | Full execution with reporting |
| Trigger | Playback button or Auto Playback | Run from a Scenario or Test Suite |
| Scope | The Action or Scenario open in the editor | Full Scenario with all test cases |
| Agent | Locally connected Agent | Local or remote Agent |
| Reporting | Console logs only | Full test report with analytics |
| Element troubleshooting | Pauses interactively to locate or map the element | Fails the step (Self-Healing recovers Smart Locator elements if enabled) |
Comments
0 comments
Please sign in to leave a comment.