Add Traject smoke tests to Tier 1 testing infrastructure

[Copilot]

Adds fast-feedback traject configuration validation to catch Ruby syntax errors and config issues before Solr indexing. Tests run in ~2s with cached gems, suitable for CI and agent iteration.

Changes

GitHub Actions workflow (.github/workflows/test.yml)

• Ruby 3.1 + Python 3.12 setup with bundler and pip caching
• Explicit minimal permissions (contents: read)

Traject smoke tests (tests/unit/test_traject_smoke.py)

• Ruby syntax validation via ruby -c
• Config load verification without input files
• XML transformation using Traject::DebugWriter (bypasses Solr)
• Graceful skip when traject config doesn't exist

Dependencies (Gemfile)

gem 'traject', '~> 3.0'
gem 'traject_plus'  # Required by example_traject_config_eac_cpf.rb

Test infrastructure

• Shared sample_eac_cpf_xml fixture in conftest.py
• Documentation in tests/README.md covering performance and skip behavior
• .gitignore entries for Ruby artifacts

Performance

• First run: ~60s (gem install)
• Cached runs: ~2s (fast enough for Tier 1)

Original prompt

Update the testing infrastructure PR to include Traject smoke tests in Tier 1 (fast feedback loop).

Additional Requirements

1. Update GitHub Actions Workflow

Add Ruby setup and bundler caching to .github/workflows/test.yml:

- name: Setup Ruby
  uses: ruby/setup-ruby@v1
  with:
    ruby-version: '3.1'
    bundler-cache: true

- name: Install Ruby dependencies
  run: |
    bundle install
    # Or if Gemfile doesn't exist yet, create minimal one:
    # gem install traject

2. Add Traject Smoke Tests

Create tests/unit/test_traject_smoke.py:

"""
Traject smoke tests - verify traject config and XML processing work.

These tests run traject without Solr to catch config errors quickly.
Goal: < 60 seconds total including Ruby setup (with caching).
"""

import pytest
import subprocess
from pathlib import Path

@pytest.fixture
def sample_eac_cpf_xml():
    """Minimal valid EAC-CPF XML for testing"""
    return '''<?xml version="1.0" encoding="UTF-8"?>
<eac-cpf xmlns="urn:isbn:1-931666-33-4">
  <control>
    <recordId>creator_people_1</recordId>
    <maintenanceStatus>new</maintenanceStatus>
    <maintenanceAgency>
      <agencyName>Test</agencyName>
    </maintenanceAgency>
  </control>
  <cpfDescription>
    <identity>
      <nameEntry>
        <part>Test Person</part>
      </nameEntry>
    </identity>
  </cpfDescription>
</eac-cpf>'''

def test_traject_config_syntax_valid():
    """Verify traject config has valid Ruby syntax"""
    # Find traject config (might be in different locations)
    possible_paths = [
        "traject_config_eac_cpf.rb",
        "example_traject_config_eac_cpf.rb",
    ]
    
    config_path = None
    for path in possible_paths:
        if Path(path).exists():
            config_path = path
            break
    
    if not config_path:
        pytest.skip("No traject config found (expected if not yet created)")
    
    # Ruby syntax check (fast, doesn't execute)
    result = subprocess.run(
        ["ruby", "-c", config_path],
        capture_output=True,
        text=True
    )
    
    assert result.returncode == 0, f"Invalid Ruby syntax: {result.stderr}"

@pytest.mark.skipif(
    not Path("example_traject_config_eac_cpf.rb").exists(),
    reason="Traject config not yet available"
)
def test_traject_loads_config():
    """Verify traject can load config without errors"""
    result = subprocess.run(
        ["bundle", "exec", "traject", "-c", "example_traject_config_eac_cpf.rb"],
        capture_output=True,
        text=True
    )
    
    # Should show usage/help, not crash
    assert "error" not in result.stderr.lower() or result.returncode == 1

@pytest.mark.skipif(
    not Path("example_traject_config_eac_cpf.rb").exists(),
    reason="Traject config not yet available"
)
def test_traject_processes_sample_xml(tmp_path, sample_eac_cpf_xml):
    """Verify traject can transform XML without Solr (smoke test)"""
    xml_file = tmp_path / "sample.xml"
    xml_file.write_text(sample_eac_cpf_xml)
    
    # Use NullWriter to process without Solr
    result = subprocess.run([
        "bundle", "exec", "traject",
        "-c", "example_traject_config_eac_cpf.rb",
        "-w", "Traject::NullWriter",
        str(xml_file)
    ], capture_output=True, text=True)
    
    # Should complete without errors
    assert result.returncode == 0, f"Traject processing failed: {result.stderr}"

3. Create Minimal Gemfile

If one doesn't exist, create Gemfile in repo root:

source 'https://rubygems.org'

gem 'traject', '~> 3.0'

# Optional: for testing
group :test do
  gem 'rspec', '~> 3.0'
end

4. Update tests/README.md

Add section about traject smoke tests:

## Traject Smoke Tests

Tests in `tests/unit/test_traject_smoke.py` verify traject configuration without requiring Solr.

### What They Test
- Ruby syntax validity of traject configs
- Traject can load and parse configs
- XML transformation logic (without indexing)

### Setup Requirements
- Ruby 3.1+
- Bundler
- Run `bundle install` to install traject gem

### Performance
- First run: ~60 seconds (includes gem install)
- Cached runs: ~40 seconds (gems cached)
- Still fast enough for CI/agent iteration

### Skipping
These tests skip gracefully if traject config doesn't exist yet.

5. Update conftest.py

Add traject-related fixtures:

@pytest.fixture
def sample_eac_cpf_xml():
    """Minimal valid EAC-CPF XML for testing"""
    return '''<?xml version="1.0" encoding="UTF-8"?>
<eac-cpf xmlns="urn:isbn:1-931666-33-4">
  <control>
    <recordId>creator_people_1</recordId>
  </control>
  <cpfDescription>
    <identity>
      <nameEntry><part>Test Person</part></nameEntry>
    </identity>
  </cpfDescription>
</eac-cpf>'''

Key Points

• Traject tests should be part of unit test suite (not integration)
• Use pytest.skipif to gracefully skip if traject config doesn't exist yet
• Use Traject::NullWriter to avoid needing Solr
• Leverage bundler-cache in Git...

This pull request was created from Copilot chat.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.