Add integration tests and CI workflow

.github/workflows/ci.yml

@@ -0,0 +1,49 @@
+name: Integration Tests
+
+on:
+  pull_request:
+    branches:
+      - main
+  push:
+    branches:
+      - main
+
+jobs:
+  integration-tests:
+    name: Run Integration Tests
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    permissions:
+      contents: read
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.11"
+          cache: "pip"
+
+      - name: Install dependencies
+        run: pip install -r requirements.txt pytest pytest-asyncio
+
+      - name: Generate config.yaml from secrets
+        env:
+          COSMOS_URI: ${{ secrets.COSMOS_URI }}
+          COSMOS_KEY: ${{ secrets.COSMOS_KEY }}
+          COSMOS_DATABASE_NAME: ${{ secrets.COSMOS_DATABASE_NAME }}
+          COSMOS_SOURCE1_CONTAINER: ${{ secrets.COSMOS_SOURCE1_CONTAINER }}
+          COSMOS_SOURCE2_CONTAINER: ${{ secrets.COSMOS_SOURCE2_CONTAINER }}
+          LLM_ENDPOINT: ${{ secrets.LLM_ENDPOINT }}
+          LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
+          LLM_MODEL: ${{ secrets.LLM_MODEL }}
+          EMBED_ENDPOINT: ${{ secrets.EMBED_ENDPOINT }}
+          EMBED_API_KEY: ${{ secrets.EMBED_API_KEY }}
+          EMBED_MODEL: ${{ secrets.EMBED_MODEL }}
+          EMBED_DIMENSIONS: ${{ secrets.EMBED_DIMENSIONS }}
+        run: python scripts/generate_config.py
+
+      - name: Run integration tests
+        run: pytest tests/ -v --tb=short

scripts/generate_config.py

@@ -0,0 +1,89 @@
+#!/usr/bin/env python
+"""Generate config.yaml from config.yaml.example by substituting environment variables.
+
+Environment variables consumed
+-------------------------------
+COSMOS_URI                   – Cosmos DB account URI
+COSMOS_KEY                   – Cosmos DB primary key
+COSMOS_DATABASE_NAME         – database name (default: divdet)
+COSMOS_SOURCE1_CONTAINER     – container name for source_1 (default: container_1)
+COSMOS_SOURCE2_CONTAINER     – container name for source_2 (default: container_2)
+LLM_ENDPOINT                 – Azure OpenAI LLM endpoint URL
+LLM_API_KEY                  – Azure OpenAI LLM API key
+LLM_MODEL                    – LLM deployment/model name
+EMBED_ENDPOINT               – Azure OpenAI embedding endpoint URL
+EMBED_API_KEY                – Azure OpenAI embedding API key
+EMBED_MODEL                  – Embedding deployment/model name
+EMBED_DIMENSIONS             – Embedding output dimensions (default: 1024)
+"""
+import os
+import sys
+from pathlib import Path
+
+import yaml
+
+REPO_ROOT = Path(__file__).parent.parent
+EXAMPLE_PATH = REPO_ROOT / "config.yaml.example"
+OUTPUT_PATH = REPO_ROOT / "config.yaml"
+
+
+def _require(name: str) -> str:
+    value = os.environ.get(name, "").strip()
+    if not value:
+        print(f"ERROR: required environment variable '{name}' is not set or empty.", file=sys.stderr)
+        sys.exit(1)
+    return value
+
+
+def _optional(name: str, default: str = "") -> str:
+    return os.environ.get(name, default).strip() or default
+
+
+def main():
+    if not EXAMPLE_PATH.exists():
+        print(f"ERROR: {EXAMPLE_PATH} not found.", file=sys.stderr)
+        sys.exit(1)
+
+    with open(EXAMPLE_PATH) as fh:
+        cfg = yaml.safe_load(fh)
+
+    # --- Cosmos DB -----------------------------------------------------------
+    cfg["cosmos"]["uri"] = _require("COSMOS_URI")
+    cfg["cosmos"]["key"] = _require("COSMOS_KEY")
+    cfg["cosmos"]["database_name"] = _optional("COSMOS_DATABASE_NAME", "divdet")
+
+    # Patch container names inside sources list
+    container_overrides = {
+        "source_1": _optional("COSMOS_SOURCE1_CONTAINER", "container_1"),
+        "source_2": _optional("COSMOS_SOURCE2_CONTAINER", "container_2"),
+    }
+    for source in cfg.get("cosmos", {}).get("sources", []):
+        source_id = source.get("id", "")
+        if source_id in container_overrides:
+            source["container_name"] = container_overrides[source_id]
+
+    # --- LLM -----------------------------------------------------------------
+    cfg["llm"]["llm_endpoint"] = _require("LLM_ENDPOINT")
+    cfg["llm"]["llm_api_key"] = _require("LLM_API_KEY")
+    cfg["llm"]["llm_model"] = _optional("LLM_MODEL", cfg["llm"].get("llm_model", "model-router"))
+
+    # --- Embedding -----------------------------------------------------------
+    if "embedding" not in cfg:
+        cfg["embedding"] = {}
+    cfg["embedding"]["embed_endpoint"] = _require("EMBED_ENDPOINT")
+    cfg["embedding"]["embed_api_key"] = _require("EMBED_API_KEY")
+    cfg["embedding"]["embed_model"] = _require("EMBED_MODEL")
+    cfg["embedding"]["embed_dimensions"] = int(_optional("EMBED_DIMENSIONS", "1024"))
+
+    # Disable local LLM fallback in CI to avoid noise
+    cfg.setdefault("local_llm", {})["use_local_fallback_for_subtasks"] = False
+
+    # Write output
+    with open(OUTPUT_PATH, "w") as fh:
+        yaml.dump(cfg, fh, allow_unicode=True, sort_keys=False)
+
+    print(f"config.yaml written to {OUTPUT_PATH}")
+
+
+if __name__ == "__main__":
+    main()

tests/conftest.py

@@ -0,0 +1,67 @@
+"""Shared pytest fixtures for integration tests."""
+import sys
+import asyncio
+from pathlib import Path
+
+import pytest
+import yaml
+
+# Ensure the repo root is on the path so rag_divdet can be imported
+REPO_ROOT = Path(__file__).parent.parent
+sys.path.insert(0, str(REPO_ROOT))
+
+
+# ---------------------------------------------------------------------------
+# Event-loop fixture (module-scoped so all fixtures share one loop)
+# ---------------------------------------------------------------------------
+
+@pytest.fixture(scope="module")
+def event_loop():
+    """Create a module-scoped event loop shared by all fixtures and tests."""
+    loop = asyncio.new_event_loop()
+    asyncio.set_event_loop(loop)
+    yield loop
+    loop.close()
+
+
+# ---------------------------------------------------------------------------
+# Config fixture: load config.yaml and expose as a dict
+# ---------------------------------------------------------------------------
+
+@pytest.fixture(scope="module")
+def config():
+    config_path = REPO_ROOT / "config.yaml"
+    if not config_path.exists():
+        pytest.skip("config.yaml not found – skipping integration tests")
+    with open(config_path) as fh:
+        return yaml.safe_load(fh)
+
+
+# ---------------------------------------------------------------------------
+# LLMClient fixture
+# ---------------------------------------------------------------------------
+
+@pytest.fixture(scope="module")
+def llm_client(config, event_loop):
+    """Return an initialised LLMClient (module-scoped to avoid repeated setup)."""
+    import rag_divdet as rd
+    client = rd.LLMClient()
+    yield client
+    event_loop.run_until_complete(client.close())
+
+
+# ---------------------------------------------------------------------------
+# CombinedRetriever fixture
+# ---------------------------------------------------------------------------
+
+@pytest.fixture(scope="module")
+def retriever(config, event_loop):
+    """Return an initialised CombinedRetriever (module-scoped)."""
+    import rag_divdet as rd
+    sources = rd._build_retrieval_sources(config)
+    if not sources:
+        pytest.skip("No retrieval sources configured – skipping retriever tests")
+    ret = rd.CombinedRetriever(retrieval_sources=sources)
+    event_loop.run_until_complete(ret.initialize())
+    yield ret
+    event_loop.run_until_complete(ret.close())