Skip to content

Test Results Ingestion

Test results ingestion is a 100% deterministic, script-only command — no AI generation needed. The script parses JUnit XML files, matches test case names to V-Model scenario IDs, and updates matrix statuses in-place.

Command

/speckit.v-model.test-results

# Basic: ingest JUnit XML results
scripts/bash/ingest-test-results.sh \
  --input test-results.xml \
  specs/<feature>/v-model

# With code coverage (adds Coverage column to Matrix D)
scripts/bash/ingest-test-results.sh \
  --input results.xml \
  --coverage coverage.xml \
  specs/<feature>/v-model

# With coverage mapping override
scripts/bash/ingest-test-results.sh \
  --input results.xml \
  --coverage coverage.xml \
  --coverage-map coverage-map.yml \
  specs/<feature>/v-model

# With explicit commit SHA
scripts/bash/ingest-test-results.sh \
  --input results.xml \
  --commit-sha abc1234 \
  specs/<feature>/v-model

# JSON output for CI
scripts/bash/ingest-test-results.sh \
  --input results.xml --json \
  specs/<feature>/v-model

# Basic: ingest JUnit XML results
scripts/powershell/Ingest-Test-Results.ps1 `
  -InputFile test-results.xml `
  -VModelDir specs/<feature>/v-model

# With code coverage
scripts/powershell/Ingest-Test-Results.ps1 `
  -InputFile results.xml `
  -Coverage coverage.xml `
  -VModelDir specs/<feature>/v-model

# JSON output
scripts/powershell/Ingest-Test-Results.ps1 `
  -InputFile results.xml -Json `
  -VModelDir specs/<feature>/v-model

Arguments

Argument Required Description
--input <path> Yes Path to JUnit XML file containing test results
--coverage <path> No Path to Cobertura XML file containing code coverage data
--matrix <path> No Path to traceability-matrix.md (defaults to auto-detect)
--coverage-map <path> No Path to coverage-map.yml for explicit MOD→files mapping
--commit-sha <sha> No Explicit commit SHA (defaults to git rev-parse --short=7 HEAD)
--json No Output structured JSON to stdout

How It Works

JUnit XML Parsing

The script reads JUnit XML test results (the standard format produced by most test frameworks) and matches test case names to V-Model scenario IDs:

JUnit Test Name Pattern Matched V-Model ID
test_SCN_001_A1_* SCN-001-A1
STS_002_B1_* STS-002-B1
ITS_003_A1_* ITS-003-A1
UTS_005_A2_* UTS-005-A2

Matrix Status Updates

The traceability matrix is updated in-place with test results:

Before JUnit Result After
⬜ Untested <testcase> passes ✅ Passed
⬜ Untested <testcase> fails ❌ Failed
⬜ Untested <testcase> skipped ⏭️ Skipped
✅ Passed <testcase> fails ❌ Failed

Audit Metadata

Each matrix update includes audit metadata columns:

Column Source Example
Date Ingestion timestamp 2025-01-15
Commit SHA --commit-sha or git rev-parse abc1234
Coverage Cobertura XML (if provided) 87.3%

Coverage Mapping

When --coverage is provided, the script maps Cobertura coverage data to MOD-NNN modules using one of two strategies:

Convention-Based Mapping (Default)

The script infers file-to-module mapping from naming conventions:

src/parser/validate_schema.py  →  MOD-002 (validate_schema)
src/api/handle_create.py       →  MOD-005 (handle_create)

Explicit Mapping via coverage-map.yml

For projects where naming conventions don't match, provide an explicit mapping:

# coverage-map.yml
modules:
  MOD-001:
    files:
      - src/parser/parse_input.py
      - src/parser/tokenizer.py
  MOD-002:
    files:
      - src/parser/validate_schema.py
  MOD-005:
    files:
      - src/api/handlers/create.py
      - src/api/handlers/create_helpers.py

Use the --coverage-map argument to specify this file.