unstructured-ingest 1.0.8__tar.gz → 1.0.11__tar.gz

{unstructured_ingest-1.0.8 → unstructured_ingest-1.0.11}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: unstructured_ingest
-Version: 1.0.8
+Version: 1.0.11
 Summary: Local ETL data pipeline to get data RAG ready
 Author-email: Unstructured Technologies <devops@unstructuredai.io>
 License-Expression: Apache-2.0

unstructured_ingest-1.0.11/unstructured_ingest/__version__.py

@@ -0,0 +1 @@
+__version__ = "1.0.11"  # pragma: no cover

{unstructured_ingest-1.0.8 → unstructured_ingest-1.0.11}/unstructured_ingest/processes/connectors/google_drive.py

@@ -153,7 +153,13 @@ class GoogleDriveIndexer(Indexer):
         """
         try:
             # A very minimal call: list 1 file from the drive.
-            client.list(spaces="drive", pageSize=1, fields="files(id)").execute()
+            client.list(
+                supportsAllDrives=True,
+                includeItemsFromAllDrives=True,
+                spaces="drive",
+                pageSize=1,
+                fields="files(id)",
+            ).execute()
         except HttpError as e:
             error_content = e.content.decode() if hasattr(e, "content") else ""
             lower_error = error_content.lower()
@@ -183,6 +189,8 @@ class GoogleDriveIndexer(Indexer):
             page_token = None
             while True:
                 response = files_client.list(
+                    supportsAllDrives=True,
+                    includeItemsFromAllDrives=True,
                     spaces="drive",
                     q=query,
                     fields="nextPageToken, files(id, mimeType, fileExtension)",
@@ -251,6 +259,8 @@ class GoogleDriveIndexer(Indexer):
                 else:
                     # Non-recursive: check for at least one immediate non-folder child.
                     response = client.list(
+                        supportsAllDrives=True,
+                        includeItemsFromAllDrives=True,
                         spaces="drive",
                         fields="files(id)",
                         pageSize=1,
@@ -348,6 +358,8 @@ class GoogleDriveIndexer(Indexer):
         files_response = []
         while not done:
             response: dict = files_client.list(
+                supportsAllDrives=True,
+                includeItemsFromAllDrives=True,
                 spaces="drive",
                 fields=fields_input,
                 corpora="user",
@@ -381,7 +393,9 @@ class GoogleDriveIndexer(Indexer):
         return files_response
 
     def get_root_info(self, files_client, object_id: str) -> dict:
-        return files_client.get(fileId=object_id, fields=",".join(self.fields)).execute()
+        return files_client.get(
+            supportsAllDrives=True, fileId=object_id, fields=",".join(self.fields)
+        ).execute()
 
     def get_files(
         self,

{unstructured_ingest-1.0.8 → unstructured_ingest-1.0.11}/unstructured_ingest/processes/connectors/ibm_watsonx/ibm_watsonx_s3.py

@@ -172,7 +172,7 @@ class IbmWatsonxUploaderConfig(UploaderConfig):
     namespace: str = Field(description="Namespace name")
     table: str = Field(description="Table name")
     max_retries: int = Field(
-        default=5, description="Maximum number of retries to upload data", ge=2, le=10
+        default=5, description="Maximum number of retries to upload data", ge=2, le=500
     )
     record_id_key: str = Field(
         default=RECORD_ID_LABEL,

{unstructured_ingest-1.0.8 → unstructured_ingest-1.0.11}/unstructured_ingest/processes/connectors/pinecone.py

@@ -1,6 +1,7 @@
 import json
 import re
 from dataclasses import dataclass, field
+from pathlib import Path
 from typing import TYPE_CHECKING, Any, Literal, Optional
 
 from pydantic import Field, Secret
@@ -18,11 +19,14 @@ from unstructured_ingest.interfaces import (
 )
 from unstructured_ingest.logger import logger
 from unstructured_ingest.processes.connector_registry import DestinationRegistryEntry
+from unstructured_ingest.utils import ndjson
 from unstructured_ingest.utils.constants import RECORD_ID_LABEL
 from unstructured_ingest.utils.data_prep import (
     flatten_dict,
     generator_batching_wbytes,
     get_enhanced_element_id,
+    get_json_data,
+    write_data,
 )
 from unstructured_ingest.utils.dep_check import requires_dependencies
 
@@ -162,6 +166,28 @@ class PineconeUploadStager(UploadStager):
             "metadata": metadata,
         }
 
+    def stream_update(self, input_file: Path, output_file: Path, file_data: FileData) -> None:
+        with input_file.open() as in_f:
+            reader = ndjson.reader(in_f)
+            with output_file.open("w") as out_f:
+                writer = ndjson.writer(out_f)
+                for element in reader:
+                    if "embeddings" not in element:
+                        continue
+                    conformed_element = self.conform_dict(element_dict=element, file_data=file_data)
+                    writer.write(row=conformed_element)
+                    writer.f.flush()
+
+    def process_whole(self, input_file: Path, output_file: Path, file_data: FileData) -> None:
+        elements_contents = get_json_data(path=input_file)
+
+        conformed_elements = [
+            self.conform_dict(element_dict=element, file_data=file_data)
+            for element in elements_contents
+            if "embeddings" in element
+        ]
+        write_data(path=output_file, data=conformed_elements)
+
 
 @dataclass
 class PineconeUploader(VectorDBUploader):

unstructured_ingest-1.0.8/unstructured_ingest/__version__.py

@@ -1 +0,0 @@
-__version__ = "1.0.8"  # pragma: no cover