test: import no-op test into the runtime-tests

Add a new trivial runtime-test which simply runs a no-op pipeline. This
is a fast, trivial test that simply verifies osbuild is properly setup
and accessible.

Remove the explicit no-op test from the CI, now that the test-suite has
it as well.

.github/workflows/runtime-tests.yml

@@ -33,29 +33,6 @@ jobs:
       with:
         run: python3 -m unittest -v test.test_boot
 
-  noop_pipeline_tests:
-    name: "Noop-Pipeline Tests"
-    runs-on: ubuntu-latest
-    steps:
-    - name: "Clone Repository"
-      uses: actions/checkout@v2
-    - name: "Install Dependencies"
-      run: |
-        sudo apt-get update
-        sudo apt-get -y install \
-          systemd-container
-    - name: "Set up Python"
-      uses: actions/setup-python@v1
-      with:
-        python-version: 3.7
-    - name: Install Python Packages
-      run: pip install jsonschema
-    - name: "Run Noop-Pipeline Tests"
-      run: |
-        for i in {0..2} ; do
-          sudo env "PATH=$PATH" python3 -m osbuild --libdir . samples/noop.json
-        done
-
   assembler_tests:
     name: "Assembler Tests"
     runs-on: ubuntu-latest

test/run/test_noop.py

@@ -0,0 +1,30 @@
+#
+# Runtime Tests for No-op Pipelines
+#
+
+import unittest
+
+from .. import test
+
+
+class TestNoop(unittest.TestCase):
+    def setUp(self):
+        self.osbuild = test.OSBuild(self)
+
+    def test_noop(self):
+        #
+        # Run a noop Pipeline. Run twice to verify the cache does not affect
+        # the operation (we do not have checkpoints, nor any stages that could
+        # be checkpointed).
+        #
+        # Then run the entire thing again, to verify our own `osbuild` executor
+        # tears things down properly and allows to be executed multiple times.
+        #
+
+        with self.osbuild as osb:
+            osb.compile("{}")
+            osb.compile("{}")
+
+        with self.osbuild as osb:
+            osb.compile("{}")
+            osb.compile("{}")