- Changes after retest with and without Docker

- Output restructuring
- Support for POUT environment (PIN is automatic)

Author: Alan Christie

README.md

@@ -73,20 +73,20 @@ run the following Gradle command from here: -
     $ ./gradlew runPipelineTester
 
 ### In Docker
-You can run the pipeline tests in Docxker using their expected container
+You can run the pipeline tests in Docker using their expected container
 image (defined in the service descriptor). Doing this gives you added
 confidence that your pipeline will work wen deployed.
 
-To run in Docker you need to add the `-d` or `--indocker` command-line
-argument. To pass these variables through Gradle into the pipeline tester
-run the tests like this: -
-
-    $ ./gradlew runPipelineTester -Pptargs=-d
-
-Or by using the Docker-specific gradle command:
+You can use the docker-specific Gradle task: -
 
     $ ./gradlew runDockerPipelineTester
 
+Or, by add the `-d` or `--indocker` command-line argument into the basic
+task. To pass these variables through Gradle into the pipeline tester
+run the Docker tests like this: -
+
+    $ ./gradlew runPipelineTester -Pptargs=-d
+
 >   When you run _in docker_ only the tests that can run in Docker (those with
     a defined image name) will be executed. Tests that cannot be executed in
     Docker will be _skipped_.

pipeline.test.template

@@ -31,19 +31,12 @@
     //          file. Individual tests can define their own test-specific files
     //          with their own 'creates' block, discussed later in this file.
     //
-    // pin:     This property can be used to over-ride the automatically
-    //          generated data input path (which is typically the repository's
-    //          own data directory. This allows you to use date outside the
-    //          repository - maybe because it's a large file or an experimental
-    //          file.
-    //
     // There can only be one setup_collection section and it must be
     // the first section in the file.
 
     setup_collection = [
         timeout: 10,
-        creates: [ 'output.txt' ],
-        pin: /Users/alan/data
+        creates: [ 'output.txt' ]
     ],
 
     // Individual tests.
@@ -68,7 +61,12 @@
     // If a test is not working an you want to keep it and avoid running it
     // then simply prefix the test section name with `ignore_`.
     //
-    // The location of any input data you provide in your test
+    // The location of any input data you provide in your test is normally
+    // located in the project's `data` directory or mounted as `/data`
+    // when running in a container. You can safely refer to data in both cases
+    // with the PIN environment variable created by the PipelineTester.
+    // To allow execution from the command-line and in a container the input
+    // data file `blob.dat` should be referred to using `${PIN}blob.dat`
 
     test_1 = [
 
@@ -80,7 +78,7 @@
         // If you provide a command you **cannot** provide parameters
         // (see the params section below).
 
-        command: ```python my_own_command
+        command: ```python my_own_command -i ${PIN}blob.dat
                  --my-own-param-1 32
                  --my-own-param-2 18.5```,
 

src/groovy/ContainerExecutor.groovy

@@ -23,10 +23,19 @@
 class ContainerExecutor {
 
     /**
-     * Executes the given command in the supplied container image.
+     * Executes the given command in the supplied container image. The data
+     * directory (pin) is mounted as `/data` in the running container
+     * and the designated output directory (pout) is mounted in the container
+     * as `/output`. Two environment variables are defined: PIN and POUT and
+     * are set to `/data` and `/output` respectively. The script is executed in
+     * the output directory in the container.
      *
-     * @param imageName The image to run the command in
      * @param command The command to run
+     * @param imageName The image to run the command in
+     * @param pin The pipeline input directory
+     *            (used to define the PIN environment variable)
+     * @param pout The pipeline output directory
+     *             (used to define the POUT environment variable)
      * @param timeoutSeconds The time to allow for the command to execute
      * @return A list containing the STDOUT and STDERR encapsulated in a
      *         StringBuilder(), an integer command exit code and a timeout

src/groovy/PipelineTester.groovy

@@ -16,37 +16,25 @@
  * limitations under the License.
  */
 
-import Tester
-
 /**
  * The PipelineTester. A groovy script for the automated testing
  * of Informatics Matters pipeline scripts. This utility searches for
- * pipeline test files and then loads and executes them.
+ * pipeline `.test` files and then loads and executes them.
  *
  *      `pipeline.test.template` is template that can be used as to create your
  *      own tests. Read its inline documentation to understand how it is used
  *      within this framework.
  *
  * Depending on your pipeline's requirements you may need to execute this
  * the PipelineTester within a suitable execution environment like Conda.
- *
- * Typical execution is from this project's root using Gradle: -
- *
- *      gradle runPipelineTester
- *
- * ...or with an abbreviated task name like:
- *
- *      gradle rPT
- *
- * This utility has been tested with groovy v2.4
  */
 
 // Version
 // Update with every change/release
 String version = '2.0.0'
 
 println "+----------------+"
-println "| PipelineTester | (v$version)"
+println "| PipelineTester | v$version"
 println "+----------------+"
 
 // Build command-line processor

src/groovy/ShellExecutor.groovy

@@ -17,13 +17,14 @@
  */
 
 /**
- * The PipelineTester Executor class. The class is responsible for executing
- * a pipeline command using the command-line.
+ * The PipelineTester Shell (command-line) Executor class. The class is
+ * responsible for executing a pipeline command using the command-line.
  */
 class ShellExecutor {
 
     /**
-     * Executes the given command in the supplied execution directory.
+     * Executes the given command using the shell in the supplied execution
+     * directory.
      *
      * @param command The command to run
      * @param edir The directory in which to run the command