indexify 0.0.25__py3-none-any.whl → 0.0.27__py3-none-any.whl

indexify/client.py

@@ -4,7 +4,7 @@ import uuid
 import hashlib
 import json
 from collections import namedtuple
-from .settings import DEFAULT_SERVICE_URL
+from .settings import DEFAULT_SERVICE_URL, DEFAULT_SERVICE_URL_HTTPS
 from .extractor import Extractor
 from .extraction_policy import ExtractionPolicy, ExtractionGraph
 from .index import Index
@@ -108,7 +108,7 @@ class IndexifyClient:
         cert_path: str,
         key_path: str,
         ca_bundle_path: Optional[str] = None,
-        service_url: str = DEFAULT_SERVICE_URL,
+        service_url: str = DEFAULT_SERVICE_URL_HTTPS,
         *args,
         **kwargs,
     ) -> "IndexifyClient":
@@ -560,7 +560,7 @@ class IndexifyClient:
         )
         return response.json()
 
-    def get_extracted_content(self, content_id: str, level: int = 0):
+    def get_extracted_content(self, content_id: str, graph_name: str, policy_name: str):
         """
         Get list of child for a given content id and their content up to the specified level.
 
@@ -570,20 +570,13 @@ class IndexifyClient:
         """
         content_tree = self.get_content_tree(content_id)
         child_list = []
-
-        def traverse_content(parent_id, current_level):
-            if current_level > level:
-                return
-
-            for item in content_tree["content_tree_metadata"]:
-                if item["parent_id"] == parent_id:
-                    child_id = item["id"]
-                    content = self.download_content(child_id)
-                    child_list.append({"id": child_id, "content": content})
-
-                    traverse_content(child_id, current_level + 1)
-
-        traverse_content(content_id, 0)
+        for item in content_tree["content_tree_metadata"]:
+            if (
+                graph_name in item["extraction_graph_names"]
+                and item["source"] == policy_name
+            ):
+                content = self.download_content(item["id"])
+                child_list.append({"id": item["id"], "content": content})
 
         return child_list
 
@@ -631,14 +624,19 @@ class IndexifyClient:
         )
         return response.json()
 
-    def wait_for_extraction(self, content_id: str):
+    def wait_for_extraction(self, content_ids: Union[str, List[str]]):
         """
         Wait for extraction to complete for a given content id
 
         Args:
             - content_id (str): id of content
         """
-        response = self.get(f"namespaces/{self.namespace}/content/{content_id}/wait")
+        if type(content_ids) == str:
+            content_ids = [content_ids]
+        print("Waiting for extraction to complete for content id: ", ",".join(content_ids))
+        for content_id in content_ids:
+            response = self.get(f"namespaces/{self.namespace}/content/{content_id}/wait")
+            print("Extraction completed for content id: ", content_id)
         response.raise_for_status()
 
     def generate_unique_hex_id(self):

{indexify-0.0.25.dist-info → indexify-0.0.27.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: indexify
-Version: 0.0.25
+Version: 0.0.27
 Summary: Python Client for Indexify
 Home-page: https://github.com/tensorlakeai/indexify
 License: Apache 2.0

{indexify-0.0.25.dist-info → indexify-0.0.27.dist-info}/RECORD

@@ -1,5 +1,5 @@
 indexify/__init__.py,sha256=Y40-Ur_tL7kGGs-reh9BTfEYGe-KyGxgdg-CmoFsXRQ,473
-indexify/client.py,sha256=ArCbwqozqAN6dTNhoD9hWSCZOPkWD0f5ecwwU9-mqYE,21940
+indexify/client.py,sha256=NwEvU-9FU2U-UkhlKjfVmIUJ9E9DcSIjIJYE4AEqwJA,22109
 indexify/data_containers.py,sha256=r1wxJPtsmXbyKvb17fqxm-dPjKz51oZ62f8A8Zxls1c,361
 indexify/error.py,sha256=3umTeYb0ugtUyehV1ibfvaeACxAONPyWPc-1HRN4d1M,856
 indexify/exceptions.py,sha256=vjd5SPPNFIEW35GorSIodsqvm9RKHQm9kdp8t9gv-WM,111
@@ -8,7 +8,7 @@ indexify/extractor.py,sha256=sWFLlXHgEfWlmiKAXN6ytUt_uG7th-XGNHqz-TG39gs,1216
 indexify/index.py,sha256=RvxYhJXEth-GKvqzlMiz5PuN1eIbZk84pt20piA1Gsw,504
 indexify/settings.py,sha256=LSaWZ0ADIVmUv6o6dHWRC3-Ry5uLbCw2sBSg1e_U7UM,99
 indexify/utils.py,sha256=rDN2lrsAs9noJEIjfx6ukmC2SAIyrlUt7QU-kaBjujM,125
-indexify-0.0.25.dist-info/LICENSE.txt,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-indexify-0.0.25.dist-info/METADATA,sha256=RShV5jaxLXTgo85SnTEr-jJrq0Uo7YmQiXpdEevbmvE,1798
-indexify-0.0.25.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-indexify-0.0.25.dist-info/RECORD,,
+indexify-0.0.27.dist-info/LICENSE.txt,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+indexify-0.0.27.dist-info/METADATA,sha256=Hz-3CZFY5Nz99brOcraBKR8moLhghxgoHOugofABP_c,1798
+indexify-0.0.27.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+indexify-0.0.27.dist-info/RECORD,,