issue #1: ailab search endpoint and tests

.env.template

@@ -26,6 +26,9 @@ FINESSE_BACKEND_GITHUB_STATIC_FILE_URL=https://api.github.com/repos/ai-cfia/fine
 # Message for Finesse data search failures. Optional.
 # FINESSE_BACKEND_ERROR_FINESSE_DATA_FAILED="finesse-data static search failed"
 
+# Message for ailab-db search failures. Optional.
+# FINESSE_BACKEND_ERROR_AILAB_FAILED="Ailab-db search failed."
+
 # Message for unexpected errors. Optional.
 # FINESSE_BACKEND_ERROR_UNEXPECTED="Unexpected error."
 
@@ -35,3 +38,40 @@ FINESSE_BACKEND_GITHUB_STATIC_FILE_URL=https://api.github.com/repos/ai-cfia/fine
 
 # Regular expression pattern used for sanitizing input to prevent log injection. Optional.
 # FINESSE_BACKEND_SANITIZE_PATTERN="[^\w \d\"#\$%&'\(\)\*\+,-\.\/:;?@\^_`{\|}~]+|\%\w+|;|/|\(|\)"
+
+# API key for OpenAI, used for authentication when making requests.
+# Obtain from: https://portal.azure.com/#home
+OPENAI_API_KEY=<OpenAI-API-Key>
+
+# The version of the OpenAI API being used.
+# Example: 2023-05-15
+OPENAI_API_VERSION=<OpenAI-API-Version>
+
+# Deployment name for GPT-based models in Azure OpenAI.
+# Example: davinci
+AZURE_OPENAI_GPT_DEPLOYMENT=<Azure-OpenAI-GPT-Deployment>
+
+# Deployment name for ChatGPT models in Azure OpenAI.
+# Example: chat
+AZURE_OPENAI_CHATGPT_DEPLOYMENT=<Azure-OpenAI-ChatGPT-Deployment>
+
+# Data Source Name (DSN) for configuring a database connection in Louis's system.
+# Format: postgresql://PGUSER:PGPASSWORD@DB_SERVER_CONTAINER_NAME/PGBASE
+LOUIS_DSN=<Louis-DSN>
+
+# Schema within the Louis database system.
+# Example: louis_0.0.5
+LOUIS_SCHEMA=<Louis-Schema>
+
+# Endpoint URL for making requests to the OpenAI API.
+# Obtain along with the OpenAI API Key.
+OPENAI_ENDPOINT=<OpenAI-Endpoint>
+
+# File containing the weights for the search.
+# Example: finesse-weights.json
+FINESSE_WEIGHTS=<Finesse-Weights>
+
+# Specific OpenAI API model engine to be used.
+# Example: ada
+OPENAI_API_ENGINE=<OpenAI-API-Engine>
+

app/ailab_db/__init__.py

@@ -0,0 +1,26 @@
+import logging
+
+from ailab.db import DBError, connect_db
+from ailab.db.api import match_documents_from_text_query
+
+
+class AilabDbSearchException(DBError):
+    """Custom exception for errors in searching from ailab-db."""
+
+
+class EmptyQueryError(DBError):
+    "Raised when the search query is empty."
+
+
+def ailab_db_search(query):
+    if not query:
+        logging.error("Empty search query received")
+        raise EmptyQueryError("Search query cannot be empty")
+
+    try:
+        with connect_db().cursor() as cursor:
+            return match_documents_from_text_query(cursor, query)
+    # TODO: handle specific exceptions
+    except Exception as e:
+        logging.error(f"finesse-data fetch failed: {e}", exc_info=True)
+        raise AilabDbSearchException(f"finesse-data fetch failed: {e}") from e

app/blueprints/search.py

@@ -1,9 +1,9 @@
-import re
 from functools import wraps
 
 from flask import Blueprint, current_app, jsonify, request
 from index_search import AzureIndexSearchQueryError, search
 
+from app.ailab_db import DBError, ailab_db_search
 from app.finesse_data import FinesseDataFetchException, fetch_data
 from app.utils import sanitize
 
@@ -49,3 +49,17 @@ def search_static():
         return jsonify({"error": current_app.config["ERROR_FINESSE_DATA_FAILED"]}), 500
     except Exception:
         return jsonify({"error": current_app.config["ERROR_UNEXPECTED"]}), 500
+
+
+@search_blueprint.route("/ailab", methods=["POST"])
+@require_non_empty_query
+def search_ailab_db():
+    query = request.json["query"]
+    query = sanitize(query, current_app.config["SANITIZE_PATTERN"])
+    try:
+        results = ailab_db_search(query)
+        return jsonify(results)
+    except DBError:
+        return jsonify({"error": current_app.config["ERROR_AILAB_FAILED"]}), 500
+    except Exception:
+        return jsonify({"error": current_app.config["ERROR_UNEXPECTED"]}), 500

app/config.py

@@ -14,6 +14,8 @@
 DEFAULT_ERROR_FINESSE_DATA_FAILED = "finesse-data static search failed"
 DEFAULT_ERROR_UNEXPECTED = "Unexpected error."
 DEFAULT_FUZZY_MATCH_THRESHOLD = 90
+DEFAULT_ERROR_AILAB_FAILED = "Ailab-db search failed."
+DEFAULT_ERROR_UNEXPECTED = "Unexpected error."
 DEFAULT_SANITIZE_PATTERN = (
     "[^\w \d\"#\$%&'\(\)\*\+,-\.\/:;?@\^_`{\|}~]+|\%\w+|;|/|\(|\)"
 )
@@ -42,8 +44,10 @@ class Config:
         "FINESSE_BACKEND_ERROR_AZURE_FAILED", DEFAULT_ERROR_AZURE_FAILED
     )
     ERROR_FINESSE_DATA_FAILED = os.getenv(
-        "FINESSE_BACKEND_ERROR_FINESSE_DATA_FAILED",
-        DEFAULT_ERROR_FINESSE_DATA_FAILED,
+        "FINESSE_BACKEND_ERROR_FINESSE_DATA_FAILED", DEFAULT_ERROR_FINESSE_DATA_FAILED
+    )
+    ERROR_AILAB_FAILED = os.getenv(
+        "FINESSE_BACKEND_ERROR_AILAB_FAILED", DEFAULT_ERROR_AILAB_FAILED
     )
     ERROR_UNEXPECTED = os.getenv(
         "FINESSE_BACKEND_ERROR_UNEXPECTED", DEFAULT_ERROR_UNEXPECTED

finesse-weights.json

@@ -0,0 +1,7 @@
+{
+  "recency": 1,
+  "traffic": 1,
+  "current": 0.5,
+  "typicality": 0.2,
+  "similarity": 1
+}

requirements-production.txt

@@ -6,3 +6,4 @@ python-dotenv==1.0.0            # Released: 2023-02-24
 git+https://github.com/ai-cfia/azure-db.git@main#subdirectory=azure-ai-search
 fuzzywuzzy==0.18.0 
 python-Levenshtein== 0.23.0
+git+https://github.com/ai-cfia/ailab-db@main

requirements.txt

@@ -6,3 +6,4 @@ python-dotenv
 git+https://github.com/ai-cfia/azure-db.git@main#subdirectory=azure-ai-search
 fuzzywuzzy
 python-Levenshtein
+git+https://github.com/ai-cfia/ailab-db@main

tests/common.py

@@ -16,5 +16,5 @@ class TestAzureSearchConfig:
 class TestConfig(Config):
     AZURE_CONFIG = TestAzureSearchConfig()
     FINESSE_DATA_URL = ""
-    DEBUG = ""
+    DEBUG = True
     TESTING = True

tests/test_ailab_search.py

@@ -0,0 +1,36 @@
+import unittest
+from unittest.mock import patch
+
+from app.ailab_db import DBError
+from app.app_creator import create_app
+from tests.common import TestConfig
+
+
+class TestAilabSearch(unittest.TestCase):
+    def setUp(self):
+        self.config = TestConfig()
+        self.app = create_app(self.config)
+        self.client = self.app.test_client()
+
+    def test_search_ailab_success(self):
+        with patch("app.blueprints.search.ailab_db_search") as mock_search:
+            mock_search.return_value = {"some": "ailab data"}
+            response = self.client.post("/search/ailab", json={"query": "ailab query"})
+            self.assertEqual(response.status_code, 200)
+            self.assertEqual(response.json, {"some": "ailab data"})
+
+    def test_search_ailab_no_query(self):
+        response = self.client.post("/search/ailab", json={})
+        self.assertEqual(response.status_code, 400)
+
+    def test_search_ailab_db_error(self):
+        with patch("app.blueprints.search.ailab_db_search") as mock_search:
+            mock_search.side_effect = DBError("Ailab DB failed")
+            response = self.client.post("/search/ailab", json={"query": "ailab query"})
+            self.assertEqual(response.status_code, 500)
+
+    def test_search_ailab_unexpected_error(self):
+        with patch("app.blueprints.search.ailab_db_search") as mock_search:
+            mock_search.side_effect = Exception("Unexpected error")
+            response = self.client.post("/search/ailab", json={"query": "ailab query"})
+            self.assertEqual(response.status_code, 500)

tests/test_azure_search.py

@@ -0,0 +1,29 @@
+import unittest
+from unittest.mock import patch
+
+from app.app_creator import create_app
+from tests.common import TestConfig
+
+
+class TestAzureSearch(unittest.TestCase):
+    def setUp(self):
+        self.config = TestConfig()
+        self.app = create_app(self.config)
+        self.client = self.app.test_client()
+
+    def test_search_azure_success(self):
+        with patch("app.blueprints.search.search") as mock_search:
+            mock_search.return_value = {"some": "azure data"}
+            response = self.client.post("/search/azure", json={"query": "azure query"})
+            self.assertEqual(response.status_code, 200)
+            self.assertEqual(response.json, {"some": "azure data"})
+
+    def test_search_azure_no_query(self):
+        response = self.client.post("/search/azure", json={})
+        self.assertEqual(response.status_code, 400)
+
+    def test_search_azure_error(self):
+        with patch("app.blueprints.search.search") as mock_search:
+            mock_search.side_effect = Exception("Azure search failed")
+            response = self.client.post("/search/azure", json={"query": "azure query"})
+            self.assertEqual(response.status_code, 500)