chore: add destination name format for weaviate (#390)

* add name formatting function; update test

* tidy and version

Author: shreyanid

CHANGELOG.md

@@ -1,3 +1,9 @@
+## 0.5.6-dev0
+
+### Enhancements
+
+* Add name formatting to Weaviate destination uploader
+
 ## 0.5.5
 
 ### Enhancements

test/integration/connectors/weaviate/test_local.py

@@ -78,7 +78,6 @@ def run_uploader_and_validate(
     validate_count(expected_count=expected_count)
 
 
-@pytest.mark.asyncio
 @pytest.mark.tags(CONNECTOR_TYPE, DESTINATION_TAG, VECTOR_DB_TAG)
 def test_weaviate_local_destination(upload_file: Path, collection: str, tmp_path: Path):
     file_data = FileData(
@@ -142,11 +141,12 @@ def test_weaviate_local_create_destination(weaviate_instance):
         upload_config=LocalWeaviateUploaderConfig(),
         connection_config=LocalWeaviateConnectionConfig(),
     )
-    collection_name = "system_created"
+    collection_name = "system_created-123"
+    formatted_collection_name = "System_created_123"
     created = uploader.create_destination(destination_name=collection_name)
     assert created
     with uploader.connection_config.get_client() as weaviate_client:
-        assert weaviate_client.collections.exists(name=collection_name)
+        assert weaviate_client.collections.exists(name=formatted_collection_name)
 
     created = uploader.create_destination(destination_name=collection_name)
     assert not created

unstructured_ingest/__version__.py

@@ -1 +1 @@
-__version__ = "0.5.5"  # pragma: no cover
+__version__ = "0.5.6-dev0"  # pragma: no cover

unstructured_ingest/v2/processes/connectors/weaviate/weaviate.py

@@ -1,4 +1,5 @@
 import json
+import re
 from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from dataclasses import dataclass, field
@@ -232,16 +233,26 @@ def precheck(self) -> None:
     def init(self, *kwargs: Any) -> None:
         self.create_destination()
 
+    def format_destination_name(self, destination_name: str) -> str:
+        # Weaviate naming requirements:
+        # must be alphanumeric and underscores only
+        formatted = re.sub(r"[^a-zA-Z0-9]", "_", destination_name)
+        # must begin with capital letter
+        return formatted.capitalize()
+
     def create_destination(
         self, destination_name: str = "elements", vector_length: Optional[int] = None, **kwargs: Any
     ) -> bool:
+        destination_name = self.format_destination_name(destination_name)
         collection_name = self.upload_config.collection or destination_name
         self.upload_config.collection = collection_name
+
         connectors_dir = Path(__file__).parents[1]
         collection_config_file = connectors_dir / "assets" / "weaviate_collection_config.json"
         with collection_config_file.open() as f:
             collection_config = json.load(f)
         collection_config["class"] = collection_name
+
         if not self._collection_exists():
             logger.info(
                 f"creating default weaviate collection '{collection_name}' with default configs"