Command Line Runner
Introduction
This page describes how to execute a specification using the command line interface that comes with GreenPepper.
Features
GreenPepper provides a command line interface to execute specifications captured in HTML files. The command line interface has the following features:
- Execution of a single specification file
- Execution of a suite of specifications stored in a directory tree
Running the command line application
Running GreenPepper requires Java JDK 6.0+.
The following libraries are required on the classpath to run the command line interface:
- greenpepper-core-xxx-complete.jar
- your fixture code
- your system under test
- all the external libraries your system under test depends upon
The main class for the command line interface is com.greenpepper.runner.Main.
Arguments
Invoking the command line interface with the --help option brings up the usual help message:
Usage: greenpepper [options] input [ouput] Run the input specification and produce a report in output file or in directory specified by -o Options: --lazy Execute document in lazy mode -l, --locale LANG Set application language (en, fr, ...) -f, --sud CLASS;ARGS Use CLASS as the system under development and instantiate it with ARGS -o DIRECTORY Produce reports in DIRECTORY (defaults to current directory) -r CLASS;ARGS Use CLASS as the document repository and instantiate it with ARGS (defaults to com.greenpepper.repository.FileSystemRepository) --selector CLASS Use CLASS as the interpreter selector (defaults to com.greenpepper.document.GreenPepperInterpreterSelector) -s, --suite Run a suite rather than a single test (output must refer to a directory) -t SECTIONS Filter input specification to only execute SECTIONS (comma separated list of sections) --xml Generate XML report (defaults to plain) --help Display this help and exit --version Output version information and exit --debug Enable debug mode --stop Stop the execution of the specification on the first failure
Default options
When no option is given, GreenPepper defaults to the following rules:
- Input is a single specification, not a suite
- If output is not specified, reports are produced in the current directory
- Report is in plain format
- The operating system locale is used
- The input specification is not filtered
- The specification file must be located on the file system
- Fixtures are created using the default built-in mechanism
- Tables use HTML formalism
- Lazy mode is disabled
- Debug mode is disabled
- Will not stop the execution on first failure
Any of the rules above can be overridden using command line switches.
Running with the default options
By default GreenPepper executes the specification file given as an input and produces a report in the current directory. The name of the report is constructed from the name of the specification.
The specification file has to be an HTML file (with the .html extension) accessible on the file system. GreenPepper will execute the specification and produce both:
- an execution summary displayed in the terminal window
- a report file, which contains the executed specification annotated with test results
Here's an example:
Running with default options |
---|
Executing the command: java -cp ... com.greenpepper.runner.Main example.html might display the following in the terminal window: Running example.html 14 example(s): 9 right, 3 wrong, 1 ignored, 1 exception(s) <<< FAILURE! |
Specifying an output directory -o
The destination directory for report files can be specified using the -o option.
Alternatively, the output argument can be used to specify the absolute path of the report file.
Specifying an output directory |
---|
Executing the command: java -cp ... com.greenpepper.runner.Main -o reports example.html might display the following in the terminal window: Running example.html 14 test(s): 9 right, 3 wrong, 1 ignored, 1 exception(s) and produce a report file named example.html in the reports directory. |
Executing a set of specification files (-s, --suite)
GreenPepper can execute all specification files found in a given directory (recursively) and produce result files in the specified output directory. The suite mode is activated using the -s option.
If the output directory does not exist, GreenPepper will try to create it. Input files need to be in HTML form and have the .html extension.
Running a suite of specifications |
---|
To execute all specification files found in the /home/myuser/workspace/myprojects/specs directory and produce output in the /home/myuser/workspace/myprojects/reports directory, use the following command from the /home/myuser/workspace/myprojects directory: java -cp ... com.greenpepper.runner.Main -s -o reports specs If the specs directory looks like this: home The reports directory would then contain the following files:
home |
Changing the default Locale (--locale, -l)
It is possible to change the current locale with the --locale or -l option, followed by a valid java 2 letters locale code.
Changing the locale |
---|
Running the application in french is done like this: java -cp ... com.greenpepper.runner.Main --locale fr example.html result.html |
Filtering out parts of the specification (-t)
When the specification file contains sections identified by tags, it is possible to execute only those parts of the specification that are relevant using the -t option. The -t switch has to be followed by a comma separated list of tags to consider for execution.
Filtering the specification |
---|
We can execute only the sections of the specification marked with the authentication and authorization tags like this: java -cp ... com.greenpepper.runner.Main -t authentication,authorization example.html result.html |
Producing XML reports (--xml)
It is possible to produce the report file in XML format using the --xml option. The XML format produces an output file that contains both execution statistics and the annotated HTML specification.
Producing XML reports |
---|
We can execute only the sections of the specification marked with the authentication and authorization tags like this: java -cp ... com.greenpepper.runner.Main --xml example.html result.html |
Using a custom specification repository (-r)
The default is for GreenPepper to look for specifications on the file system. When you want to use a different specification repository, use the -r option followed by the fully qualified class name of the Repository class and a semi-colon separated list of constructor parameters for that class.
Changing the specification repository |
---|
To use the VFSRepository issue the command: java -cp ... com.greenpepper.runner.Main -r com.greenpepper.runner.repository.VFSRepository example.html result.html |
Specifying a custom system under development (--sud)
GreenPepper will use the DefaultSystemUnderDevelopment unless told otherwise. To specify a different system under development class, use the --sud option followed by the fully qualified class name and a semi-colon separated list of constructor parameters for that class.
Changing the system under development |
---|
To use the SpringSystemUnderDevelopment with the application-context.xml file located in the conf directory, use the command: java -cp ... com.greenpepper.runner.Main --sud com.greenpepper.systemunderdevelopment.SpringSystemUnderDevelopment;conf/application-context.xml example.html result.html |
Changing table formalism (e.g. for Fit compliance)
It is possible to change GreenPepper table formalism by specifying an InterpreterSelector class with the --selector option. By default GreenPepper will use its own formalism.
Enabling Fit compliance mode |
---|
As an example, to enable Fit compliance, use the class com.greenpepper.fit.FitInterpreterSelector available in the greenpepper-extensions-fit.jar. This will cause GreenPepper to use Fit formalism for tables and support Fit basic interpreters: java -cp greenpepper-extensions-fit.jar;... com.greenpepper.runner.Main --selector com.greenpepper.fit.FitInterpreterSelector example.html result.html |
Setting lazy mode (--lazy)
Enabling lazy mode changes GreenPepper behavior :
- Lazy mode : On a greenpepperized document, all tables that are contained in documents will be skipped until you specify a greenpepper-example macro for the table or group of tables you want to be not skipped.
- Not lazy mode (default) : this mode is an Eager mode, it means that all tables will be interpreted.
Enabling lazy mode |
---|
To enable lazy mode, use the command: java -cp ... com.greenpepper.runner.Main --lazy |
Setting debug mode (--debug)
Enabling debug mode allows you to have a full view of exceptions stack trace in your GreenPepper specifications documents. Note that in the normal mode the stack trace are limited.
Enabling debug mode |
---|
To enable debug mode, use the command: java -cp ... com.greenpepper.runner.Main --debug |
Stop on first failure (--stop)
Stop on first failure |
---|
To stop the execution of a specification when a failure is detected, use the command: java -cp ... com.greenpepper.runner.Main --stop |