sunholo 0.132.0__py3-none-any.whl → 0.134.1__py3-none-any.whl

sunholo/custom_logging.py

@@ -193,6 +193,26 @@ class GoogleCloudLogging:
             except Exception as text_err:
                 print(f"Even fallback text logging failed: {text_err}")
 
+    def log(self, message, *args, **kwargs):
+        """
+        Some weird bug keeps calling this method - do not use normally
+
+        A catch-all method to handle unexpected .log() calls on this class.
+        Routes the call to the appropriate logging method based on severity level.
+        """
+        severity = kwargs.get('severity', 'INFO')
+        # Remove severity from kwargs if it exists to avoid passing it twice
+        if 'severity' in kwargs:
+            del kwargs['severity']
+        
+        # Determine if this is a structured log or simple message
+        if isinstance(message, dict):
+            # Assume this is a structured log
+            return self.structured_log(log_struct=message, severity=severity)
+        else:
+            # Assume this is a text log
+            return self.structured_log(log_text=message, severity=severity)
+
     def debug(self, log_text=None, log_struct=None):
 
         """

sunholo/database/alloydb_client.py

@@ -941,6 +941,65 @@ class AlloyDBClient:
         log.info(f"Updated row in {table_name} with {primary_key_column}={primary_key_value}")
         
         return result
+
+    async def check_row(self, table_name: str, primary_key_column: str, primary_key_value: str, 
+                    columns: list = None, condition: str = None):
+        """
+        Retrieves a row from the specified table based on the primary key.
+        
+        Args:
+            table_name (str): Name of the table to query
+            primary_key_column (str): Name of the primary key column (e.g., 'id')
+            primary_key_value (str): Value of the primary key for the row to retrieve
+            columns (list, optional): List of column names to retrieve. If None, retrieves all columns
+            condition (str, optional): Additional condition for the WHERE clause
+            
+        Returns:
+            The row data if found, None otherwise
+        """
+        # Determine which columns to select
+        if columns and isinstance(columns, list):
+            columns_str = ", ".join([f'"{col}"' for col in columns])
+        else:
+            columns_str = "*"  # Select all columns if none specified
+        
+        # Create the WHERE clause
+        where_clause = f'"{primary_key_column}" = :pk_value'
+        values = {'pk_value': primary_key_value}
+        
+        if condition:
+            where_clause += f" AND ({condition})"
+        
+        # Construct the SQL statement
+        sql = f'SELECT {columns_str} FROM "{table_name}" WHERE {where_clause} LIMIT 1'
+        
+        log.info(f"Checking row in {table_name} with {primary_key_column}={primary_key_value}")
+        
+        # Execute SQL based on engine type
+        try:
+            if self.engine_type == "pg8000":
+                # Use the synchronous method for pg8000
+                result = self._execute_sql_pg8000(sql, values)
+                # Extract the row data from the result
+                if result and hasattr(result, 'fetchone'):
+                    row = result.fetchone()
+                    if row:
+                        # If we have column names, convert to dictionary
+                        if hasattr(result, 'keys'):
+                            column_names = result.keys()
+                            return dict(zip(column_names, row))
+                        return row
+                return None
+            else:
+                # Use the async method for langchain
+                result = await self._execute_sql_async_langchain(sql, values)
+                # For langchain engine, check result format and return first row if exists
+                if result and len(result) > 0:
+                    return result[0]
+                return None
+        except Exception as e:
+            log.error(f"Error checking row: {e}")
+            return None
     
     async def get_table_columns(self, table_name, schema="public"):
         """

sunholo/discovery_engine/cli.py

@@ -277,11 +277,25 @@ def search_by_id_and_or_date_command(args):
         elif results_data:
             console.print("\n[bold magenta]--- Individual Chunks (Filtered) ---[/bold magenta]")
             chunk_count = 0
-            # ... (pager iteration identical to search_command) ...
             try:
+                 # Iterate through the pager returned by get_chunks
                  for page in results_data.pages:
-                     # ... iterate results and chunks ...
-                      pass # Replace with actual iteration and printing
+                     if not hasattr(page, 'results') or not page.results: continue
+                     for result in page.results:
+                          # Ensure the result structure is as expected by get_chunks
+                          if hasattr(result, 'chunk'):
+                               chunk_count += 1
+                               console.print(f"\n[bold]Chunk {chunk_count}:[/bold]")
+                               # Use the client's formatter if available
+                               console.print(client.chunk_format(result.chunk))
+                          elif hasattr(result, 'document') and hasattr(result.document, 'chunks'):
+                               # Fallback if structure is different (e.g., document with chunks)
+                               for chunk in result.document.chunks:
+                                    chunk_count += 1
+                                    console.print(f"\n[bold]Chunk {chunk_count} (from doc {result.document.id}):[/bold]")
+                                    console.print(f"  Content: {getattr(chunk, 'content', 'N/A')}")
+                                    console.print(f"  Doc Name: {getattr(chunk, 'document_metadata', {}).get('name', 'N/A')}") # Example access
+
                  if chunk_count == 0:
                      console.print("[yellow]No chunks found in the filtered results.[/yellow]")
             except Exception as page_err:
@@ -306,9 +320,7 @@ def search_engine_command(args):
     try:
         client = DiscoveryEngineClient(
             project_id=args.project,
-            # data_store_id is required by __init__ but less relevant here.
-            # Provide a default or the primary one associated with the project/engine.
-            data_store_id=args.data_store_id_for_init,
+            engine_id=args.engine_id,
             location=args.location
         )
 
@@ -404,8 +416,8 @@ def search_engine_command(args):
                                 try:
                                     struct_dict = convert_composite_to_native(doc.struct_data)
                                     metadata_output = struct_dict.get("structData", {})
-                                    console.print(f"  Title: {struct_dict.get("title", "")}")
-                                    console.print(f"  Content: {struct_dict.get("content", "")}")
+                                    console.print(f'  Title: {struct_dict.get("title", "")}')
+                                    console.print(f'  Content: {struct_dict.get("content", "")}')
                                 except Exception as json_err:
                                     console.print(f"[yellow]  Warning: Could not convert metadata Struct to JSON: {json_err}[/yellow]")
                                     metadata_output = doc.struct_data
@@ -525,8 +537,6 @@ def setup_discovery_engine_subparser(subparsers):
     search_engine_parser = discovery_engine_subparsers.add_parser('search-engine', help='Search a Discovery Engine (fetches documents/summary)')
     search_engine_parser.add_argument('--query', required=True, help='The search query')
     search_engine_parser.add_argument('--engine-id', required=True, help='Engine ID to search')
-    # Add data_store_id needed for client init, maybe make it optional if client handles it?
-    search_engine_parser.add_argument('--data-store-id-for-init', required=True, help='A primary data store ID associated with the project/engine (for client init)')
     search_engine_parser.add_argument('--serving-config-id', default='default_config', help='Serving config ID for the engine')
     search_engine_parser.add_argument('--collection-id', default='default_collection', help='Collection ID for the engine path')
     search_engine_parser.add_argument('--page-size', type=int, default=10, help='Max results per page')

sunholo/discovery_engine/discovery_engine_client.py

@@ -67,12 +67,19 @@ class DiscoveryEngineClient:
                 print(f"Document Name: {chunk_document_name}")
         ```
     """
-    def __init__(self, data_store_id, project_id, location="eu"):
+    def __init__(self, data_store_id=None, engine_id=None, project_id=None, location="eu"):
         if not discoveryengine:
             raise ImportError("Google Cloud Discovery Engine not available, install via `pip install sunholo[gcp]`")
+        
+        if project_id is None:
+            raise ValueError("Must specify project_id")
+        
+        if data_store_id is None and engine_id is None:
+            raise ValueError("Must specify at least one of data_store_id or engine_id")
 
         self.project_id = project_id
         self.data_store_id = data_store_id
+        self.engine_id = engine_id
         self.location = location
         client_options = (
             ClientOptions(api_endpoint=f"{location}-discoveryengine.googleapis.com")
@@ -469,7 +476,7 @@ class DiscoveryEngineClient:
             operation = self.engine_client.create_engine(request=request)
         except AlreadyExists as err:
             log.info(f"Engine already exists: - {str(err)}")
-
+            self.engine_id = engine_id
             return engine_id
         
         log.info(f"Waiting for create vertex ai search operation to complete: {operation.operation.name}")
@@ -481,7 +488,7 @@ class DiscoveryEngineClient:
 
         # Handle the response
         log.info(f"{response=} {metadata=}")
-
+        self.engine_id = engine_id
         return operation.operation.name
 
     def _import_document_request(self,
@@ -862,7 +869,7 @@ class DiscoveryEngineClient:
     def search_engine(
         self,
         search_query: str,
-        engine_id: str,
+        engine_id: str = None,
         serving_config_id: str = "default_config",
         page_size: int = 10,
         return_snippet: bool = True,
@@ -890,7 +897,7 @@ class DiscoveryEngineClient:
 
         Args:
             search_query: The user's search query string.
-            engine_id: The ID of the search engine to query.
+            engine_id: The ID of the search engine to query or uses class engine_id it init with.
             serving_config_id: The ID of the specific serving config for the engine.
             page_size: Maximum number of results per page.
             return_snippet: Whether to request snippets in the results.
@@ -961,13 +968,20 @@ class DiscoveryEngineClient:
             log.error("Discovery Engine library not available at runtime.")
             return None
         
+        if engine_id:
+            self.engine_id = engine_id
+        
+        if engine_id is None and self.engine_id is None:
+            raise ValueError("Could not find self.engine_id")
+        
+        
         try:
             # Construct the serving config path for an ENGINE
             # Note: The client library path helper is for data stores/serving configs within them.
             # We need the path for an engine's serving config.
             serving_config_path = (
                 f"projects/{self.project_id}/locations/{self.location}/"
-                f"collections/{collection_id}/engines/{engine_id}/"
+                f"collections/{collection_id}/engines/{self.engine_id}/"
                 f"servingConfigs/{serving_config_id}"
             )
             log.info(f"Using Engine Serving Config Path: {serving_config_path}")
@@ -1029,22 +1043,22 @@ class DiscoveryEngineClient:
                 # Add other relevant fields like facet_specs if needed
             )
 
-            log.info(f"Searching engine '{engine_id}' with request: {request}")
+            log.info(f"Searching engine '{self.engine_id}' with request: {request}")
             response_pager = self.search_client.search(request)
-            log.info(f"Search successful for query '{search_query}' against engine '{engine_id}'.")
+            log.info(f"Search successful for query '{search_query}' against engine '{self.engine_id}'.")
             return response_pager
 
         except GoogleAPIError as e:
-            log.error(f"API error searching engine '{engine_id}': {e}")
+            log.error(f"API error searching engine '{self.engine_id}': {e}")
             return None
         except Exception as e:
-            log.error(f"Unexpected error searching engine '{engine_id}': {e}\n{traceback.format_exc()}")
+            log.error(f"Unexpected error searching engine '{self.engine_id}': {e}\n{traceback.format_exc()}")
             return None
 
     async def async_search_engine(
         self,
         search_query: str,
-        engine_id: str,
+        engine_id: str=None,
         serving_config_id: str = "default_config",
         page_size: int = 10,
         return_snippet: bool = True,
@@ -1077,6 +1091,12 @@ class DiscoveryEngineClient:
             An SearchAsyncPager object to iterate through results asynchronously,
             or None if an error occurs or the async client is not available.
         """
+        if engine_id:
+            self.engine_id = engine_id
+        
+        if engine_id is None and self.engine_id is None:
+            raise ValueError("Could not find self.engine_id")
+        
         if not self.async_search_client:
              log.error("Cannot call async_search_engine: Async client not initialized.")
              raise RuntimeError("Async client not initialized. Ensure class is instantiated within an async context.")
@@ -1085,7 +1105,7 @@ class DiscoveryEngineClient:
             # Construct the serving config path for an ENGINE (same as sync)
             serving_config_path = (
                 f"projects/{self.project_id}/locations/{self.location}/"
-                f"collections/{collection_id}/engines/{engine_id}/"
+                f"collections/{collection_id}/engines/{self.engine_id}/"
                 f"servingConfigs/{serving_config_id}"
             )
             log.info(f"Using Async Engine Serving Config Path: {serving_config_path}")
@@ -1138,16 +1158,16 @@ class DiscoveryEngineClient:
                 user_pseudo_id=user_pseudo_id,
             )
 
-            log.info(f"Async searching engine '{engine_id}' with request: {request}")
+            log.info(f"Async searching engine '{self.engine_id}' with request: {request}")
             response_pager = await self.async_search_client.search(request)
-            log.info(f"Async search successful for query '{search_query}' against engine '{engine_id}'.")
+            log.info(f"Async search successful for query '{search_query}' against engine '{self.engine_id}'.")
             return response_pager
 
         except GoogleAPIError as e:
-            log.error(f"Async API error searching engine '{engine_id}': {e}")
+            log.error(f"Async API error searching engine '{self.engine_id}': {e}")
             return None
         except Exception as e:
-            log.error(f"Async unexpected error searching engine '{engine_id}': {e}\n{traceback.format_exc()}")
+            log.error(f"Async unexpected error searching engine '{self.engine_id}': {e}\n{traceback.format_exc()}")
             return None
 
 # --- End of DiscoveryEngineClient class ---

{sunholo-0.132.0.dist-info → sunholo-0.134.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sunholo
-Version: 0.132.0
+Version: 0.134.1
 Summary: AI DevOps - a package to help deploy GenAI to the Cloud.
 Author-email: Holosun ApS <multivac@sunholo.com>
 License: Apache License, Version 2.0

{sunholo-0.132.0.dist-info → sunholo-0.134.1.dist-info}/RECORD

@@ -1,5 +1,5 @@
 sunholo/__init__.py,sha256=InRbX4V0-qdNHo9zYH3GEye7ASLR6LX8-SMvPV4Jsaw,1212
-sunholo/custom_logging.py,sha256=-n7YvgqWQ8nRjCKl1bfcbeZOMTV3m-rBdZGuVyXWDR8,17433
+sunholo/custom_logging.py,sha256=PrUXreBF0nZJMfa8A8cbFhTWoaOWWB_i6qwZfnjvRlo,18296
 sunholo/langchain_types.py,sha256=uZ4zvgej_f7pLqjtu4YP7qMC_eZD5ym_5x4pyvA1Ih4,1834
 sunholo/agents/__init__.py,sha256=X2I3pPkGeKWjc3d0QgSpkTyqD8J8JtrEWqwrumf1MMc,391
 sunholo/agents/chat_history.py,sha256=Gph_CdlP2otYnNdR1q1Umyyyvcad2F6K3LxU5yBQ9l0,5387
@@ -60,7 +60,7 @@ sunholo/components/retriever.py,sha256=Wmchv3huAM4w7DIS-a5Lp9Hi7M8pE6vZdxgseiT9S
 sunholo/components/vectorstore.py,sha256=k7GS1Y5c6ZGXSDAJvyCes6dTjhDAi0fjGbVLqpyfzBc,5918
 sunholo/database/__init__.py,sha256=bpB5Nk21kwqYj-qdVnvNgXjLsbflnH4g-San7OHMqR4,283
 sunholo/database/alloydb.py,sha256=x1zUMB-EVWbE2Zvp4nAs2Z-tB_kOZmS45H2lwVHdYnk,11678
-sunholo/database/alloydb_client.py,sha256=VCnjwzY3pw75wUmlHXIfF5oGhy7iyzo7VQVEs3MT4YY,53350
+sunholo/database/alloydb_client.py,sha256=gDYlg3sD88Nd5llGRpO7maRQ1KOnFT9XBFvS7Tx1fXM,56026
 sunholo/database/database.py,sha256=VqhZdkXUNdvWn8sUcUV3YNby1JDVf7IykPVXWBtxo9U,7361
 sunholo/database/lancedb.py,sha256=DyfZntiFKBlVPaFooNN1Z6Pl-LAs4nxWKKuq8GBqN58,715
 sunholo/database/static_dbs.py,sha256=8cvcMwUK6c32AS2e_WguKXWMkFf5iN3g9WHzsh0C07Q,442
@@ -73,9 +73,9 @@ sunholo/database/sql/sb/return_sources.sql,sha256=89KAnxfK8n_qGK9jy1OQT8f9n4uYUt
 sunholo/database/sql/sb/setup.sql,sha256=CvoFvZQev2uWjmFa3aj3m3iuPFzAAJZ0S7Qi3L3-zZI,89
 sunholo/discovery_engine/__init__.py,sha256=hLgqRDJ22Aov9o2QjAEfsVgnL3kMdM-g5p8RJ9OyKdQ,130
 sunholo/discovery_engine/chunker_handler.py,sha256=wkvXl4rFtYfN6AZUKdW9_QD49Whf77BukDbO82UwlAg,7480
-sunholo/discovery_engine/cli.py,sha256=BWYNG3V1vq3mOcmkAeQItSdTw7HeFzbCpHkfYuau2BE,32382
+sunholo/discovery_engine/cli.py,sha256=ZqP6a1bgZvNdPAytkQdC2cN1U-KEc3fNtv8Ynqay0vY,33055
 sunholo/discovery_engine/create_new.py,sha256=WUi4_xh_dFaGX3xA9jkNKZhaR6LCELjMPeRb0hyj4FU,1226
-sunholo/discovery_engine/discovery_engine_client.py,sha256=nV4loBBnOqGPRlEbFIFRJEB6cXKDVgJOMRMdf4-v510,51775
+sunholo/discovery_engine/discovery_engine_client.py,sha256=lB6D05ZOXm9Avl6hM6vJZvPZD_TzNroyBl-E5cJYWAk,52661
 sunholo/discovery_engine/get_ai_search_chunks.py,sha256=I6Dt1CznqEvE7XIZ2PkLqopmjpO96iVEWJJqL5cJjOU,5554
 sunholo/embedder/__init__.py,sha256=sI4N_CqgEVcrMDxXgxKp1FsfsB4FpjoXgPGkl4N_u4I,44
 sunholo/embedder/embed_chunk.py,sha256=did2pKkWM2o0KkRcb0H9l2x_WjCq6OyuHDxGbITFKPM,6530
@@ -168,9 +168,9 @@ sunholo/vertex/init.py,sha256=1OQwcPBKZYBTDPdyU7IM4X4OmiXLdsNV30C-fee2scQ,2875
 sunholo/vertex/memory_tools.py,sha256=tBZxqVZ4InTmdBvLlOYwoSEWu4-kGquc-gxDwZCC4FA,7667
 sunholo/vertex/safety.py,sha256=S9PgQT1O_BQAkcqauWncRJaydiP8Q_Jzmu9gxYfy1VA,2482
 sunholo/vertex/type_dict_to_json.py,sha256=uTzL4o9tJRao4u-gJOFcACgWGkBOtqACmb6ihvCErL8,4694
-sunholo-0.132.0.dist-info/licenses/LICENSE.txt,sha256=SdE3QjnD3GEmqqg9EX3TM9f7WmtOzqS1KJve8rhbYmU,11345
-sunholo-0.132.0.dist-info/METADATA,sha256=9Mpo28cEMCX2kwTQQrXK7sih-dNgQ_4JwXTbB0Kicvc,10067
-sunholo-0.132.0.dist-info/WHEEL,sha256=pxyMxgL8-pra_rKaQ4drOZAegBVuX-G_4nRHjjgWbmo,91
-sunholo-0.132.0.dist-info/entry_points.txt,sha256=bZuN5AIHingMPt4Ro1b_T-FnQvZ3teBes-3OyO0asl4,49
-sunholo-0.132.0.dist-info/top_level.txt,sha256=wt5tadn5--5JrZsjJz2LceoUvcrIvxjHJe-RxuudxAk,8
-sunholo-0.132.0.dist-info/RECORD,,
+sunholo-0.134.1.dist-info/licenses/LICENSE.txt,sha256=SdE3QjnD3GEmqqg9EX3TM9f7WmtOzqS1KJve8rhbYmU,11345
+sunholo-0.134.1.dist-info/METADATA,sha256=HV6eQJwKUSgKpPz2VE9Uv01WNtoy1FU8eAMNpBpDXjA,10067
+sunholo-0.134.1.dist-info/WHEEL,sha256=pxyMxgL8-pra_rKaQ4drOZAegBVuX-G_4nRHjjgWbmo,91
+sunholo-0.134.1.dist-info/entry_points.txt,sha256=bZuN5AIHingMPt4Ro1b_T-FnQvZ3teBes-3OyO0asl4,49
+sunholo-0.134.1.dist-info/top_level.txt,sha256=wt5tadn5--5JrZsjJz2LceoUvcrIvxjHJe-RxuudxAk,8
+sunholo-0.134.1.dist-info/RECORD,,