Add Skip Straight to SQL Query Execution Option and Include REsults (#34)

* Add option to pre-run the first option in the query cache.

* Speed up existing approach for vector and cache

* Add experimentation data

* Improve write up

Author: BenConstable9

.devcontainer/devcontainer.json

@@ -9,7 +9,8 @@
         "ms-python.python",
         "ms-python.vscode-pylance",
         "ms-toolsai.jupyter",
-        "ms-azuretools.vscode-docker"
+        "ms-azuretools.vscode-docker",
+        "ms-azuretools.vscode-azurefunctions"
       ],
       "settings": {
         "editor.formatOnSave": true

deploy_ai_search/deploy.py

@@ -4,6 +4,9 @@
 from rag_documents import RagDocumentsAISearch
 from text_2_sql import Text2SqlAISearch
 from text_2_sql_query_cache import Text2SqlQueryCacheAISearch
+import logging
+
+logging.basicConfig(level=logging.INFO)
 
 
 def deploy_config(arguments: argparse.Namespace):

deploy_ai_search/text_2_sql.py

@@ -93,7 +93,8 @@ def get_index_fields(self) -> list[SearchableField]:
                 type=SearchFieldDataType.String,
                 collection=True,
                 hidden=True,
-            ),  # This is needed to enable semantic searching against the column names as complex field types are not used.
+                # This is needed to enable semantic searching against the column names as complex field types are not used.
+            ),
             SimpleField(
                 name="DateLastModified",
                 type=SearchFieldDataType.DateTimeOffset,
@@ -213,7 +214,8 @@ def get_indexer(self) -> SearchIndexer:
                     target_field_name="ColumnNames",
                 ),
                 FieldMapping(
-                    name="DateLastModified", source="/document/DateLastModified"
+                    source_field_name="/document/DateLastModified",
+                    target_field_name="DateLastModified",
                 ),
             ],
             parameters=indexer_parameters,

deploy_ai_search/text_2_sql_query_cache.py

@@ -55,12 +55,6 @@ def get_index_fields(self) -> list[SearchableField]:
             SearchableField(
                 name="Query", type=SearchFieldDataType.String, filterable=True
             ),
-            SearchField(
-                name="QueryEmbedding",
-                type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
-                vector_search_dimensions=self.environment.open_ai_embedding_dimensions,
-                vector_search_profile_name=self.vector_search_profile_name,
-            ),
             ComplexField(
                 name="Schemas",
                 collection=True,

text_2_sql/.env

@@ -5,6 +5,7 @@ OpenAI__ApiKey=<openAIKey if using non managed identity>
 OpenAI__ApiVersion=<openAIApiVersion>
 Text2Sql__DatabaseEngine=<databaseEngine>
 Text2Sql__UseQueryCache=<whether to use the query cache first or not>
+Text2Sql__PreRunQueryCache=<whether to pre-run the top result from the query cache or not>
 Text2Sql__DatabaseName=<databaseName>
 Text2Sql__DatabaseConnectionString=<databaseConnectionString>
 AIService__AzureSearchOptions__Endpoint=<searchServiceEndpoint>

text_2_sql/README.md

@@ -5,7 +5,7 @@
 import os
 import json
 import logging
-from ai_search import run_ai_search_query
+from utils.ai_search import run_ai_search_query
 
 
 class AISearchPlugin: