"""Shared test fixtures for the saiten-mcp test suite."""
from __future__ import annotations
import json
from pathlib import Path
from typing import Any
import pytest
# ---------------------------------------------------------------------------
# Path fixtures
# ---------------------------------------------------------------------------
@pytest.fixture
def project_root() -> Path:
"""Return the project root directory."""
return Path(__file__).resolve().parent.parent
@pytest.fixture
def data_dir(project_root: Path) -> Path:
"""Return the data/ directory path."""
return project_root / "data"
@pytest.fixture
def rubrics_dir(data_dir: Path) -> Path:
"""Return the data/rubrics/ directory path."""
return data_dir / "rubrics"
# ---------------------------------------------------------------------------
# Sample data fixtures
# ---------------------------------------------------------------------------
@pytest.fixture
def sample_issue_body() -> str:
"""Return a realistic Issue body matching the submission template."""
return """### Track
Creative Apps - GitHub Copilot
### Project Name
TestProject
### Microsoft Alias
_No response_
### GitHub Username
testuser
### Repository URL
https://github.com/testuser/test-project
### Project Description
A test project for validating the scoring pipeline.
### Demo Video or Screenshots
### Primary Programming Language
Python
### Key Technologies Used
Python, FastMCP, Pydantic, GitHub Copilot
### Submission Requirements
- [x] My project meets the track-specific challenge requirements
- [x] My repository includes a README.md with setup instructions
- [x] My code does not contain hardcoded API keys or secrets
- [ ] I have included demo materials (video or screenshots)
- [x] My project is my own work created during this hackathon
### Technical Highlights
Built a multi-agent scoring system with MCP integration.
### Quick Setup Summary
pip install -e . && python -m saiten_mcp.server
### Team Members (if any)
_No response_
"""
@pytest.fixture
def sample_issue_raw() -> dict[str, Any]:
"""Return a raw Issue dict as returned by GitHub API."""
return {
"number": 99,
"title": "TestProject Submission",
"body": """### Track\n\nCreative Apps - GitHub Copilot\n\n### Project Name\n\nTestProject\n\n### Repository URL\n\nhttps://github.com/testuser/test-project\n\n### Project Description\n\nA test project.\n\n### Demo Video or Screenshots\n\n\n\n### Key Technologies Used\n\nPython, FastMCP\n\n### Submission Requirements\n\n- [x] Required 1\n- [ ] Required 2\n\n### Technical Highlights\n\nMulti-agent system.\n\n### Quick Setup Summary\n\npip install .""",
"labels": [{"name": "🎨 Creative Apps"}],
"created_at": "2026-02-01T10:00:00Z",
}
@pytest.fixture
def sample_scores() -> list[dict[str, Any]]:
"""Return sample scoring results for testing."""
return [
{
"issue_number": 100,
"project_name": "AlphaProject",
"track": "creative-apps",
"criteria_scores": {
"Accuracy & Relevance": 7,
"Reasoning & Multi-step Thinking": 6,
"Creativity & Originality": 8,
"UX & Presentation": 7,
"Reliability & Safety": 6,
},
"evidence": {
"Accuracy & Relevance": "MCP server with 3 tools implemented.",
"Reasoning & Multi-step Thinking": "Linear flow with error handling.",
"Creativity & Originality": "Novel combination of MCP + CLI.",
"UX & Presentation": "README with architecture diagram.",
"Reliability & Safety": "Try/catch in main handler, .env.example provided.",
},
"confidence": "high",
"red_flags_detected": [],
"bonus_signals_detected": ["MCP server implementation found in code"],
"weighted_total": 67.8,
"strengths": ["MCP server with 3 tools", "Clean architecture"],
"improvements": ["No tests", "No rate limiting"],
"summary": "A solid MCP-based scoring tool.",
},
{
"issue_number": 101,
"project_name": "BetaProject",
"track": "reasoning-agents",
"criteria_scores": {
"Accuracy & Relevance": 8,
"Reasoning & Multi-step Thinking": 9,
"Creativity & Originality": 7,
"User Experience & Presentation": 6,
"Technical Implementation": 7,
},
"evidence": {
"Accuracy & Relevance": "Foundry model deployed with grounding.",
"Reasoning & Multi-step Thinking": "ReAct loop with self-correction.",
"Creativity & Originality": "Novel content pipeline approach.",
"User Experience & Presentation": "Screenshots only, no video.",
"Technical Implementation": "Clean code, missing tests.",
},
"confidence": "medium",
"red_flags_detected": [],
"bonus_signals_detected": ["Self-reflection loop implemented"],
"weighted_total": 75.5,
"strengths": ["ReAct loop", "Grounding with vector store"],
"improvements": ["No video demo", "Missing tests"],
"summary": "Strong reasoning pipeline with self-correction.",
},
]
@pytest.fixture
def sample_rubric_data() -> dict[str, Any]:
"""Return a minimal rubric data structure for testing."""
return {
"track": "creative-apps",
"track_display_name": "🎨 Creative Apps",
"notes": "Test rubric",
"scoring_policy": {
"evidence_required": True,
"min_evidence_per_criterion": 1,
"differentiation_rules": [
"Score 8+ requires specific technical evidence",
],
"red_flags": [
{"signal": "No README", "max_ux_score": 4},
],
"bonus_signals": [
{"signal": "MCP server implementation", "bonus_criteria": "Accuracy & Relevance", "min_score": 7},
],
},
"criteria": [
{
"name": "Accuracy & Relevance",
"weight": 0.5,
"description": "Test criterion",
"scoring_guide": {"1-3": "Bad", "4-6": "Mid", "7-9": "Good", "10": "Exceptional"},
"evidence_signals": {
"positive": ["MCP server found"],
"negative": ["No MCP integration"],
},
},
{
"name": "Creativity & Originality",
"weight": 0.5,
"description": "Test criterion 2",
"scoring_guide": {"1-3": "Bad", "4-6": "Mid", "7-9": "Good", "10": "Exceptional"},
"evidence_signals": {
"positive": ["Novel approach"],
"negative": ["Tutorial clone"],
},
},
],
}
@pytest.fixture
def tmp_scores_file(tmp_path: Path, sample_scores: list) -> Path:
"""Create a temporary scores.json file for testing."""
scores_file = tmp_path / "scores.json"
data = {
"metadata": {
"last_updated": "2026-02-13T00:00:00+00:00",
"version": "1.0",
"total_submissions": 2,
"scored_count": 2,
},
"scores": sample_scores,
}
scores_file.write_text(json.dumps(data, indent=2), encoding="utf-8")
return scores_file
