Create several functions with any contents (even empty functions will do, that do not execute any actions whatsoever) and add these functions to Test Items (this we have seen in the Creating a test plan for regular runs recipe).

In order to run tests from command line, we need to perform the following steps:

The first method of launching allows the launch of all the enabled tests from the Test Items list. The second method allows to launch a single function. Certainly, we can create just a single function that nests calls of several functions within it, thus covering the whole of the testing plan. Nonetheless, in this case, contingent upon TestComplete getting accidentally hung-up, the possibility of losing entire log entries is much likelier, since Test Items are not used. At the point of using Test Items, the contents of the TestComplete memory, related to the logs, is committed to HDD upon completion of each of the tests.

Another inconvenience lies in the fact that we would have to comment all of the function calls of the corresponding functions to disable specific tests in the second case. Therefore, in the event of using Test Items, it will be sufficient to uncheck the corresponding Test Items.

In TestComplete, there is a variety of additional parameters, allowing us in one way or another to influence the tests launching procedure: disabling the splash screen, silent mode (in which tests execution will not be stopped even in the event of a critical error), exiting upon completion of scripts processing (to continue batch-file execution), exiting codes (useful at the point of integration with the systems of continuous integration), and so on.

For running tests from command line you can use a special command-line tool called TestExecute. It uses the same parameters as TestComplete, but doesn't have IDE and can be used only for running tests.

Running scripts from command line is usually used for creating regular runs:

We can also pass additional parameters to script from command line: