L2 Coverage Generation Using Docker Script

.github/workflows/L2-tests.yml

@@ -26,6 +26,13 @@ jobs:
           token: ${{ secrets.GITHUB_TOKEN }}
           ref: ${{ github.head_ref || github.ref }}
 
+      - name: Check out shared L2 coverage tooling
+        uses: actions/checkout@v4
+        with:
+          repository: 'rdkcentral/docker-device-mgt-service-test'
+          ref: 'feature/l2docupdate'
+          path: docker-device-mgt-service-test
+
       - name: Check out dependent repostiories
         uses: actions/checkout@v4
         with:
@@ -105,7 +112,8 @@ jobs:
       - name: Generate L2_Coverage.md
         if: (github.event_name == 'push' || github.event_name == 'pull_request') && steps.copy_tracefile.outcome == 'success'
         run: |
-          python3 remote_debugger/test/functional-tests/generate_l2_coverage_report.py \
+          python3 docker-device-mgt-service-test/generate_l2_coverage_report.py \
+            --component    "Remote Debugger" \
             --tracefile    /tmp/coverage.info \
             --src-dir      remote_debugger/src \
             --features-dir remote_debugger/test/functional-tests/features \

docs/rrd_l2_function_coverage_HLD.md

@@ -183,7 +183,10 @@ The L2 suite terminates the daemon by signal during teardown in coverage mode. W
 
 ### 5.5 Coverage Report Generation
 
-**Module:** `test/functional-tests/generate_l2_coverage_report.py` (as referenced by the workflow)
+**Module:** `generate_l2_coverage_report.py` in the shared
+`rdkcentral/docker-device-mgt-service-test` repository. The L2 workflow checks
+out that repository alongside `remote_debugger` and invokes the shared script
+with `--component "Remote Debugger"`. There is no module-local copy.
 
 **Purpose:** Convert the raw coverage trace and the feature/test inventory into a human-readable summary and actionable gap analysis.
 
@@ -203,6 +206,22 @@ The L2 suite terminates the daemon by signal during teardown in coverage mode. W
 **Outputs:**
 - Markdown report with summary metrics, tables, and scenario coverage status
 
+For local use, check out both repositories into the same parent directory and
+run from that directory after collecting the tracefile:
+
+```bash
+python3 docker-device-mgt-service-test/generate_l2_coverage_report.py \
+   --component "Remote Debugger" \
+   --tracefile /tmp/coverage.info \
+   --src-dir remote_debugger/src \
+   --features-dir remote_debugger/test/functional-tests/features \
+   --tests-dir remote_debugger/test/functional-tests/tests \
+   --output remote_debugger/test/functional-tests/L2_Coverage.md
+```
+
+The shared generator uses Python's standard library. See the shared repository's
+README for reuse by other modules and report limitations.
+
 ## 6. Functional Workflow
 
 ### 6.1 Build Flow
@@ -225,10 +244,10 @@ The L2 suite terminates the daemon by signal during teardown in coverage mode. W
 2. The Python report generator reads the trace file plus the feature/test directory metadata.
 3. A Markdown report is produced with sections like:
    - coverage overview
-   - direct and indirect coverage counts
+   - source function hit and miss counts
    - mapped vs unmapped features
    - orphan features and orphan tests
-   - gap analysis by behavior group
+   - scenario-to-test count gaps
 4. On push-based workflow execution, the report is committed to the branch for review.
 
 ## 7. Data and Interface Model
@@ -256,13 +275,17 @@ The generator reads both and builds mappings to support consistent coverage narr
 The generated report documents the following:
 
 - total source-function estimate
-- functions with direct L2 coverage
-- functions with indirect coverage
+- functions called at least once according to lcov
 - functions with no L2 coverage
-- active vs disabled tests
-- active feature scenarios and expected gaps
-- per-behavior coverage tables
-- feature/test gap summary by area such as static profile handling, dynamic profile handling, upload flow, and harmful command detection
+- test-definition counts and disabled-test estimates
+- feature scenario counts and expected count gaps
+- per-source-file line and function coverage
+- unmatched features/tests and scenario-to-test count gaps
+
+Static per-behavior coverage claims, RRD-specific infrastructure notes, and
+hand-written priority recommendations are not generated by the shared script.
+Keep such assessments in module-owned documentation and review them separately
+from measured lcov coverage. Counts alone do not prove behavior coverage.
 
 ## 8. Error Handling and Boundary Conditions