.. _capture: Capture test reports ==================== In the preceeding subsections, two ways were demonstrated to access the simplified electronic archive introduced in :ref:`introduction`. In subsection :ref:`useWebInterface`, the HTTP interface of the System Under Test was accessed. In :ref:`useTestPipeline`, the HTTP interface was bypassed such that the input message was given directly to the business logic. We saw reports of these actions in Ladybug. In this subsection, we capture these reports and demonstrate how they can be run again. In subsection :ref:`edit`, you will change captured test reports. Please do the following: #. Ensure that the Frank!Framework is set up according to :ref:`preparations` and that it is running. Go to Ladybug. Ensure that the report generator is enabled. See :ref:`useWebInterfaceWindows` or :ref:`useWebInterfaceLinux`. #. Ensure that you have a report in Ladybug for running adapter "sutGet". If it misses, repeat the steps of :ref:`useTestPipeline`. #. Select the line corresponding to the execution of "sutGet". This is shown with number 1 in the picture below: .. image:: doCapture.jpg #. Select the upper-most "Pipeline" node (number 2). #. Select stub strategy "Never" in the pull down menu with number 3. #. Press "Copy" (number 4). #. Go to tab "Test" (number 5). This changes the screen as shown: .. image:: afterCapture.jpg #. You see you are in tab "Test" (number 1). Press "Refresh" (number 2). This causes your captured test reports to appear (number 3). #. Select your test (number 4). #. Run your test by pressing "Run" (number 6). The other "Run" (button 5) can be used to run individual tests. #. Press "Refresh". A message appears showing that the test succeeded (number 1 in the figure below): .. image:: successfulRun.jpg #. Go back to tab "Debug", number 1 in the figure below. Press "Refresh" (number 2). .. image:: seeRunOfCaptured.jpg #. Numbers 3 and 4 show the impact of running a captured test. Number 3 was the run of "sutGet" that was captured. Number 4 was introduced by running the captured test. Every execution of an adapter results in a table row. .. NOTE:: Frank developers may be interested in the following. When you use the Frank!Runner, then the test reports you have in tab "Test" are stored within your instance directory. In this example, the directory is ``Frank2Manual/tests``. When you have another instance directory, then your test reports are stored in subdirectory ``tests`` of that instance directory. You can save your test reports by copying them from directory ``tests`` to some other directory. You can restore the test reports by putting the files back. If you need a different directory for storing your Ladybug test reports, you can configure it by setting property ``ibistesttool.directory``. When you set this property, the directory of Larva is not modified; you get separate directories for Ladybug and Larva by setting this property. By default, the directory of stored Ladybug test reports is shared with Larva tests. You need to distinguish Larva files from Ladybug files, because you have to keep your Ladybug files together. You can recognize your Ladybug test report files as follows: * ``metadata.xml`` is part of your Ladybug test reports. * Every file ending with ``.report.xml`` belongs to the Ladybug test reports. * All other files in directory ``tests`` are not related to Ladybug. Saving and loading test reports can also be achieved by downloading and uploading test reports as will be explained in subsection :ref:`downloadUpload`. Please do not skip that subsection, because the user interface explained there will also work when you do not use the Frank!Runner. The Frank!Runner is not intended to be used in your production environment. The advantage of using the ``tests`` directory is as follows. The Ladybug test reports in directory ``tests`` are text files, allowing them to be analyzed by version control software. Frank developers can thus capture test reports during development, and they can include these test reports in their release to the customer.