sunholo 0.135.0__py3-none-any.whl → 0.136.1__py3-none-any.whl

sunholo/discovery_engine/discovery_engine_client.py

@@ -374,7 +374,8 @@ class DiscoveryEngineClient:
         parse_documents_to_string: bool = True,
         serving_config: str = "default_serving_config",
         data_store_ids: Optional[List[str]] = None,
-        filter_str:str=None
+        filter_str:str=None,
+        max_limit:int=None
     ):
         """Retrieves entire documents based on a query.
 
@@ -405,7 +406,8 @@ class DiscoveryEngineClient:
             parse_chunks_to_string=parse_documents_to_string,
             serving_config=serving_config,
             data_store_ids=data_store_ids,
-            content_search_spec_type="documents"
+            content_search_spec_type="documents",
+            max_limit=max_limit
         )
 
     async def async_get_documents(
@@ -415,7 +417,9 @@ class DiscoveryEngineClient:
         parse_documents_to_string: bool = True,
         serving_config: str = "default_serving_config",
         data_store_ids: Optional[List[str]] = None,
-        filter_str:str=None
+        filter_str:str=None,
+        max_limit:int=None
+
     ):
         """Asynchronously retrieves entire documents based on a query.
 
@@ -439,7 +443,9 @@ class DiscoveryEngineClient:
             parse_chunks_to_string=parse_documents_to_string, 
             serving_config=serving_config,
             data_store_ids=data_store_ids,
-            content_search_spec_type="documents"
+            content_search_spec_type="documents",
+            max_limit=max_limit
+
         )
 
     def document_format(self, document):
@@ -476,44 +482,79 @@ class DiscoveryEngineClient:
             f"{derived_data}"
         )
 
-    def process_documents(self, response):
+    def process_documents(self, response, max_limit:int=None):
         """Process a search response containing documents into a formatted string."""
         all_documents = []
-
+        result_count = 0
         # Check if the response contains results
-        if not hasattr(response, 'results') or not response.results:
+        if not response or not hasattr(response, 'results') or not response.results:
             log.info(f'No results found in response: {response=}')
             return []
         
-        # Iterate through each result in the response
-        for result in response.results:
-            if hasattr(result, 'document'):
-                document = result.document
-                all_documents.append(self.document_format(document))
-            else:
-                log.warning("No document found in result")
+        should_break=False
+        # Process the pager properly
+        for page in response.pages:
+            if should_break:
+                break
+            if hasattr(page, 'results') and page.results:
+                for result in page.results:
+                    if result_count >= max_limit:
+                        log.info("Breaking results loop as max limit reached")
+                        should_break = True  # Set flag to break outer loop
+                        break
+
+                if hasattr(result, 'document'):
+                    document = result.document
+                    all_documents.append(self.document_format(document))
+                    result_count += 1
+                    
+                    # Check if we've reached max_limit
+                    if max_limit is not None and result_count >= max_limit:
+                        log.info(f"Reached max_limit of {max_limit} results, stopping processing")
+                        should_break = True
+                        break
+                else:
+                    log.warning("No document found in result")
         
         # Combine all documents into one long string
         result_string = "\n\n".join(all_documents)
         
         return result_string
 
-    async def async_process_documents(self, response):
+    async def async_process_documents(self, response, max_limit:int=None):
         """Process a search response containing documents into a formatted string asynchronously."""
         all_documents = []
+        result_count = 0
 
         # Check if the response contains results
-        if not hasattr(response, 'results') or not response.results:
+        if not response or not hasattr(response, 'results') or not response.results:
             log.info(f'No results found in response: {response=}')
             return []
-        
-        # Iterate through each result in the response
-        for result in response.results:
-            if hasattr(result, 'document'):
-                document = result.document
-                all_documents.append(self.document_format(document))
-            else:
-                log.warning("No document found in result")
+
+        should_break=False
+        # Process the pager properly
+        async for page in response.pages:
+            if should_break:
+                break
+            if hasattr(page, 'results') and page.results:
+                for result in page.results:
+                    if result_count >= max_limit:
+                        log.info("Breaking results loop as max limit reached")
+                        should_break = True  # Set flag to break outer loop
+                        break
+
+                if hasattr(result, 'document'):
+                    document = result.document
+                    all_documents.append(self.document_format(document))
+                    result_count += 1
+                    
+                    # Check if we've reached max_limit
+                    if max_limit is not None and result_count >= max_limit:
+                        log.info(f"Reached max_limit of {max_limit} results, stopping processing")
+                        should_break = True
+                        break
+                else:
+                    log.warning("No document found in result")
         
         # Combine all documents into one long string
         result_string = "\n\n".join(all_documents)
@@ -781,7 +822,8 @@ class DiscoveryEngineClient:
                         page_size=10, parse_chunks_to_string=True, 
                         serving_config="default_serving_config",
                         data_store_ids: Optional[List[str]] = None,
-                        content_search_spec_type="chunks"):
+                        content_search_spec_type="chunks",
+                        max_limit=None):
         """
         Searches with a generic filter string.
 
@@ -793,6 +835,8 @@ class DiscoveryEngineClient:
         Returns:
             discoveryengine.SearchResponse or str: The search response object or string of chunks.
         """
+        if max_limit is not None and max_limit < page_size:
+            page_size = max_limit
 
         serving_config_path = self.search_client.serving_config_path(
             self.project_id,
@@ -840,6 +884,33 @@ class DiscoveryEngineClient:
         except Exception as e:
             log.info(f"No results {search_request.data_store_specs=}: {str(e)}")
             return None
+
+        # Apply max_limit if needed
+        if content_search_spec_type=="documents" and max_limit is not None:
+            # For raw response objects (when parse_chunks_to_string=False)
+            if not parse_chunks_to_string:
+                # We need to limit the pager results before returning
+                limited_response = search_response
+                # Store the original pages iterator method
+                original_pages = limited_response.pages
+                
+                # Override the pages property with a custom iterator that respects max_limit
+                def limited_pages_iterator():
+                    results_count = 0
+                    for page in original_pages:
+                        yield page
+                        
+                        # Count results in this page
+                        if hasattr(page, 'results'):
+                            results_count += len(page.results)
+                        
+                        # Stop if we've reached max_limit
+                        if results_count >= max_limit:
+                            break
+                
+                # Replace the pages property with our custom iterator
+                limited_response.pages = limited_pages_iterator()
+                return limited_response
         
         if parse_chunks_to_string:
             if content_search_spec_type=="chunks":
@@ -850,7 +921,7 @@ class DiscoveryEngineClient:
                     return big_string
                 
             elif content_search_spec_type=="documents":
-                big_string = self.process_documents(search_response)
+                big_string = self.process_documents(search_response, max_limit=max_limit)
                 log.info(f"Discovery engine documents string sample: {big_string[:100]}")
 
                 return big_string
@@ -864,7 +935,8 @@ class DiscoveryEngineClient:
                             page_size=10, parse_chunks_to_string=True, 
                             serving_config="default_serving_config",
                             data_store_ids: Optional[List[str]] = None,
-                            content_search_spec_type="chunks"):
+                            content_search_spec_type="chunks",
+                            max_limit=None):
         """
         Searches with a generic filter string asynchronously.
 
@@ -876,6 +948,8 @@ class DiscoveryEngineClient:
         Returns:
             discoveryengine.SearchResponse or str: The search response object or string of chunks.
         """
+        if max_limit is not None and max_limit < page_size:
+            page_size = max_limit
 
         serving_config_path = self.async_search_client.serving_config_path(
             self.project_id,
@@ -922,6 +996,33 @@ class DiscoveryEngineClient:
             log.info(f"No results {search_request.data_store_specs=}: {str(e)}")
             return None
         
+        # Apply max_limit if needed
+        if content_search_spec_type=="documents" and max_limit is not None:
+            # For raw response objects (when parse_chunks_to_string=False)
+            if not parse_chunks_to_string:
+                # We need to limit the pager results before returning
+                limited_response = search_response
+                # Store the original pages iterator method
+                original_pages = limited_response.pages
+                
+                # Override the pages property with a custom iterator that respects max_limit
+                async def limited_pages_iterator():
+                    results_count = 0
+                    async for page in original_pages:
+                        yield page
+                        
+                        # Count results in this page
+                        if hasattr(page, 'results'):
+                            results_count += len(page.results)
+                        
+                        # Stop if we've reached max_limit
+                        if results_count >= max_limit:
+                            break
+                
+                # Replace the pages property with our custom iterator
+                limited_response.pages = limited_pages_iterator()
+                return limited_response
+            
         if parse_chunks_to_string:
             if content_search_spec_type=="chunks":
                 if parse_chunks_to_string:
@@ -931,7 +1032,7 @@ class DiscoveryEngineClient:
                     return big_string
                 
             elif content_search_spec_type=="documents":
-                big_string = await self.async_process_documents(search_response)
+                big_string = await self.async_process_documents(search_response, max_limit=max_limit)
                 log.info(f"Discovery engine documents string sample: {big_string[:100]}")
 
                 return big_string

{sunholo-0.135.0.dist-info → sunholo-0.136.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sunholo
-Version: 0.135.0
+Version: 0.136.1
 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-0.135.0.dist-info → sunholo-0.136.1.dist-info}/RECORD

@@ -75,7 +75,7 @@ sunholo/discovery_engine/__init__.py,sha256=hLgqRDJ22Aov9o2QjAEfsVgnL3kMdM-g5p8R
 sunholo/discovery_engine/chunker_handler.py,sha256=wkvXl4rFtYfN6AZUKdW9_QD49Whf77BukDbO82UwlAg,7480
 sunholo/discovery_engine/cli.py,sha256=tsKqNSDCEsDTz5-wuNwjttb3Xt35D97-KyyEiaqolMQ,35628
 sunholo/discovery_engine/create_new.py,sha256=WUi4_xh_dFaGX3xA9jkNKZhaR6LCELjMPeRb0hyj4FU,1226
-sunholo/discovery_engine/discovery_engine_client.py,sha256=Sf7Sr6FYKA_jn19Ba2ENShrB1jnZ4HgPScuytDIuK9c,58705
+sunholo/discovery_engine/discovery_engine_client.py,sha256=-0vSF4vd26ihhj7_XkxZJ6TvUCJrThLxZ8lMSTe5vqs,63448
 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-0.135.0.dist-info/licenses/LICENSE.txt,sha256=SdE3QjnD3GEmqqg9EX3TM9f7WmtOzqS1KJve8rhbYmU,11345
-sunholo-0.135.0.dist-info/METADATA,sha256=jLaY76jTW-W8S-9V7_9THZZ3-FroKO6HHoFKvloXXPI,10067
-sunholo-0.135.0.dist-info/WHEEL,sha256=SmOxYU7pzNKBqASvQJ7DjX3XGUF92lrGhMb3R6_iiqI,91
-sunholo-0.135.0.dist-info/entry_points.txt,sha256=bZuN5AIHingMPt4Ro1b_T-FnQvZ3teBes-3OyO0asl4,49
-sunholo-0.135.0.dist-info/top_level.txt,sha256=wt5tadn5--5JrZsjJz2LceoUvcrIvxjHJe-RxuudxAk,8
-sunholo-0.135.0.dist-info/RECORD,,
+sunholo-0.136.1.dist-info/licenses/LICENSE.txt,sha256=SdE3QjnD3GEmqqg9EX3TM9f7WmtOzqS1KJve8rhbYmU,11345
+sunholo-0.136.1.dist-info/METADATA,sha256=2aeqbpV_AQcKkI1c_WUgZh63q6a7ZMNaAbk5C9fSzAY,10067
+sunholo-0.136.1.dist-info/WHEEL,sha256=ooBFpIzZCPdw3uqIQsOo4qqbA4ZRPxHnOH7peeONza0,91
+sunholo-0.136.1.dist-info/entry_points.txt,sha256=bZuN5AIHingMPt4Ro1b_T-FnQvZ3teBes-3OyO0asl4,49
+sunholo-0.136.1.dist-info/top_level.txt,sha256=wt5tadn5--5JrZsjJz2LceoUvcrIvxjHJe-RxuudxAk,8
+sunholo-0.136.1.dist-info/RECORD,,

{sunholo-0.135.0.dist-info → sunholo-0.136.1.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (79.0.1)
+Generator: setuptools (80.0.1)
 Root-Is-Purelib: true
 Tag: py3-none-any