The build command

The build command takes a script that uses shtk modules and generates an executable script with the necessary boilerplate code to be able to locate said modules.

The input_file argument specifies the path to the file to process. This file should end with an ‘.sh’ extension, in which case the build command will proceed to generate a binary of the same name without any extension. If the extension is not present, then the name of the output file must be provided by specifying the -o flag. The input file can also be ‘-’, in which case the code is read from the standard input.

The following options are supported:

-m main_function

Name of the function that implements the entry point of the program. For programs using the shtk_unittest(3) module, this should be passed as -m shtk_unittest_main.

All scripts that use shtk should implement a main function and should avoid placing any code at the top level of the file. The main function is automatically executed by the generated script. If the value of main_function is empty, then no call to main is issued which may be necessary if the script already has a call to main for whatever reason.

Default: ‘main’.

-o output_file

Path to the output file to generate.

Default: if input_file ends in ‘.sh’, the output file has the same name as the input file but without the extension. Otherwise this flag must be specified.

-s shell

Path to the shell interpreter to write in the shebang line of the file. This must refer to an executable interpreter binary and not to another script.

Default: /usr/bin/bash.

The run command

The run command takes a script that uses shtk modules, builds it into a temporary directory, and executes the generated script. The temporary files are removed after execution. This command is intended to simplify running scripts during development; use the build command to generate a persistent executable.

The input_file argument specifies the path to the file to run. Unlike the build command, this argument cannot be ‘-’ because the standard input is left available to the executed script. Any arguments after the input file are passed unchanged to the script.

The following options are supported:

-m main_function

Name of the function that implements the entry point of the program.

Default: ‘main’.

-s shell

Path to the shell interpreter to write in the shebang line of the temporary script. This must refer to an executable interpreter binary and not to another script.

Default: /usr/bin/bash.

The run command exits with the exit status returned by the executed script.

The test command

The test command takes one or more scripts that use the shtk_unittest(3) module, builds them into temporary directories, and executes the generated test programs. The temporary files are removed after execution.

The input_file arguments specify the paths to the test programs to run. All input files are validated before any test program is executed. The test programs run sequentially in command-line order and execution continues if any of them fails. Unlike the build command, the input files cannot be ‘-’ because the standard input is left available to the test programs.

The following options are supported:

-m main_function

Name of the function that implements the entry point of the test programs.

Default: ‘shtk_unittest_main’.

-s shell

Path to the shell interpreter to write in the shebang line of the temporary test programs. This must refer to an executable interpreter binary and not to another script.

Default: /usr/bin/bash.

The test command exits successfully only if all test programs exit successfully.

The version command

The version command prints the name of the package and its version to the standard output. This command always exits successfully.