Support any string based ID

* Previously string based IDs that weren't convertable to hex would
  throw an exception. This fixes that issue by converting strings
  to 128 bit integer and converting the md5 hex into a UUID
* Update Weaviate to 1.17.3

airbyte-integrations/connectors/destination-weaviate/Dockerfile

@@ -34,5 +34,5 @@ COPY destination_weaviate ./destination_weaviate
 ENV AIRBYTE_ENTRYPOINT "python /airbyte/integration_code/main.py"
 ENTRYPOINT ["python", "/airbyte/integration_code/main.py"]
 
-LABEL io.airbyte.version=0.1.0
+LABEL io.airbyte.version=0.1.1
 LABEL io.airbyte.name=airbyte/destination-weaviate

airbyte-integrations/connectors/destination-weaviate/destination_weaviate/utils.py

@@ -4,6 +4,7 @@
 
 import re
 import uuid
+import hashlib
 from typing import Any, Mapping
 
 from airbyte_cdk.models import ConfiguredAirbyteCatalog
@@ -40,16 +41,23 @@ def hex_to_int(hex_str: str) -> int:
         return 0
 
 
+def is_uuid_string(uuid_string):
+    uuid_pattern = "^[0-9a-f]{8}-[0-9a-f]{4}-[0-5][0-9a-f]{3}-[089ab][0-9a-f]{3}-[0-9a-f]{12}$"
+    return re.match(uuid_pattern, uuid_string)
+
+
 def generate_id(record_id: Any) -> uuid.UUID:
     if isinstance(record_id, int):
         return uuid.UUID(int=record_id)
 
     if isinstance(record_id, str):
+        if is_uuid_string(record_id):
+            return uuid.UUID(record_id)
         id_int = hex_to_int(record_id)
-        if hex_to_int(record_id) > 0:
+        if id_int > 0:
             return uuid.UUID(int=id_int)
-
-    return uuid.UUID(record_id)
+        hex_string = hashlib.md5(record_id.encode("UTF-8")).hexdigest()
+        return uuid.UUID(hex=hex_string)
 
 
 def get_schema_from_catalog(configured_catalog: ConfiguredAirbyteCatalog) -> Mapping[str, Mapping[str, str]]:

airbyte-integrations/connectors/destination-weaviate/integration_tests/integration_test.py

@@ -95,7 +95,7 @@ def setup_teardown(config: Mapping):
         pass
 
     docker_client.containers.run(
-        "semitechnologies/weaviate:1.16.1", detach=True, environment=env_vars, name=name,
+        "semitechnologies/weaviate:1.17.3", detach=True, environment=env_vars, name=name,
         ports={8080: ('127.0.0.1', 8081)}
     )
     time.sleep(0.5)
@@ -403,6 +403,33 @@ def test_id_starting_with_underscore(config: Mapping, client: Client):
     assert actual.get("id") == str(uuid.UUID(int=int(data.get("_id"), 16))), "UUID should be created for _id field"
 
 
+def test_id_with_text_string(config: Mapping, client: Client):
+    stream_name = "article"
+    stream_schema = {"type": "object", "properties": {
+        "title": {"type": "string"},
+        "id": {"type": "string"}
+    }}
+    catalog = create_catalog(stream_name, stream_schema)
+    first_state_message = _state({"state": "1"})
+    data = {"title": "test1", "id": "not a real id"}
+    first_record_chunk = [_record(stream_name, data)]
+
+    destination = DestinationWeaviate()
+
+    expected_states = [first_state_message]
+    output_states = list(
+        destination.write(
+            config, catalog, [*first_record_chunk, first_state_message]
+        )
+    )
+    assert expected_states == output_states, "Checkpoint state messages were expected from the destination"
+
+    class_name = stream_to_class_name(stream_name)
+    assert count_objects(client, class_name) == 1, "There should be only 1 object of in Weaviate"
+    actual = get_objects(client, class_name)[0]
+    assert actual.get("id")
+
+
 def test_id_custom_field_name(config: Mapping, client: Client):
     # This is common scenario from mongoDB
     stream_name = "article"

airbyte-integrations/connectors/destination-weaviate/setup.py

@@ -5,7 +5,7 @@
 
 from setuptools import find_packages, setup
 
-MAIN_REQUIREMENTS = ["airbyte-cdk", "weaviate-client==3.9.0"]
+MAIN_REQUIREMENTS = ["airbyte-cdk", "weaviate-client==3.11.0"]
 
 TEST_REQUIREMENTS = ["pytest~=6.2", "docker"]
 

airbyte-integrations/connectors/destination-weaviate/unit_tests/unit_test.py

@@ -44,3 +44,6 @@ def test_generate_id():
     assert generate_id("0x1") == uuid.UUID(int=1)
     assert generate_id(1) == uuid.UUID(int=1)
     assert generate_id("123e4567-e89b-12d3-a456-426614174000") == uuid.UUID("123e4567-e89b-12d3-a456-426614174000")
+    assert generate_id("123e4567e89b12d3a456426614174000") == uuid.UUID("123e4567-e89b-12d3-a456-426614174000")
+    for i in range(10):
+        assert generate_id("this should be using md5") == uuid.UUID("802a479a-190e-92c8-8340-d687c860f53d")