Feat: remove username/password (can be part of URI)

actors/milvus/.actor/input_schema.json

@@ -3,16 +3,17 @@
   "type": "object",
   "schemaVersion": 1,
   "properties": {
-    "milvusUrl": {
-      "title": "Milvus URL",
+    "milvusUri": {
+      "title": "Milvus URI",
       "type": "string",
-      "description": "REST URL of the Milvus instance to connect to",
+      "description": "The URI of the Milvus instance to connect to. You can include the username and password in the URI, for example: `https://username:password@****.serverless.gcp-us-west1.cloud.zilliz.com`.",
       "editor": "textfield",
-      "sectionCaption": "Milvus settings"
+      "sectionCaption": "Milvus settings",
+      "isSecret": true
     },
-    "milvusApiKey": {
-      "title": "Milvus API KEY",
-      "description": "Milvus API KEY",
+    "milvusToken": {
+      "title": "Milvus Token",
+      "description": "Milvus Token",
       "type": "string",
       "editor": "textfield",
       "isSecret": true
@@ -23,19 +24,6 @@
       "description": "Name of the Milvus collection where the data will be stored",
       "editor": "textfield"
     },
-    "milvusUser": {
-      "title": "Milvus user name",
-      "type": "string",
-      "description": "User name for the Milvus cluster",
-      "editor": "textfield"
-    },
-    "milvusPassword": {
-      "title": "Milvus user password",
-      "type": "string",
-      "description": "Password for the Milvus cluster user",
-      "editor": "textfield",
-      "isSecret": true
-    },
     "embeddingsProvider": {
       "title": "Embeddings provider (as defined in the langchain API)",
       "description": "Choose the embeddings provider to use for generating embeddings",
@@ -152,8 +140,8 @@
     }
   },
   "required": [
-    "milvusUrl",
-    "milvusApiKey",
+    "milvusUri",
+    "milvusToken",
     "milvusCollectionName",
     "embeddingsProvider",
     "embeddingsApiKey",

actors/milvus/README.md

@@ -33,7 +33,7 @@ It uses [LangChain](https://www.langchain.com/) to compute embeddings and intera
 
 To utilize this integration, ensure you have:
 
-- Created or existing `Milvus` database. You need to know `milvusUrl`, `milvusApiKey`, and `milvusCollectionName`.
+- Created or existing `Milvus` database. You need to know `milvusUri`, `milvusToken`, and `milvusCollectionName`.
 - If the collection does not exist, it will be created automatically.
 - An account to compute embeddings using one of the providers, e.g., [OpenAI](https://platform.openai.com/docs/guides/embeddings) or [Cohere](https://docs.cohere.com/docs/cohere-embed).
 
@@ -53,11 +53,13 @@ For detailed input information refer to the [Input page](https://apify.com/apify
 #### Database: Milvus
 ```json
 {
-  "milvusUrl": "YOUR-MILVUS-URL",
-  "milvusApiKey": "YOUR-MILVUS-API-KEY",
+  "milvusUri": "YOUR-MILVUS-URI",
+  "milvusToken": "YOUR-MILVUS-TOKEN",
   "milvusCollectionName": "YOUR-MILVUS-COLLECTION-NAME"
 }
 ```
+If you're using a username and password for authentication, you can include them in the `milvusUri` as follows:
+`"milvusUri": "https://username:password@YOUR-MILVUS-URI"`.``
 
 #### Embeddings provider: OpenAI
 ```json 
@@ -172,8 +174,8 @@ This integration will save the selected fields from your Actor to Milvus and sto
 
 ```json
 {
-  "milvusUrl": "YOUR-MILVUS-URL",
-  "milvusApiKey": "YOUR-MILVUS-API-KEY",
+  "milvusUri": "YOUR-MILVUS-URI",
+  "milvusToken": "YOUR-MILVUS-TOKEN",
   "milvusCollectionName": "YOUR-MILVUS-COLLECTION-NAME",
   "embeddingsApiKey": "YOUR-OPENAI-API-KEY",
   "embeddingsConfig": {
@@ -195,17 +197,17 @@ This integration will save the selected fields from your Actor to Milvus and sto
 #### Milvus
 ```json
 {
-  "milvusUrl": "YOUR-MILVUS-URL",
-  "milvusApiKey": "YOUR-MILVUS-API-KEY",
+  "milvusUri": "YOUR-MILVUS-URI",
+  "milvusToken": "YOUR-MILVUS-TOKEN",
   "milvusCollectionName": "YOUR-MILVUS-COLLECTION-NAME"
 }
 ```
 
 #### Managed Milvus service at [Zilliz](https://zilliz.com/)
 ```json
 {
-  "milvusUrl": "https://in03-***********.api.gcp-us-west1.zillizcloud.com",
-  "milvusApiKey": "d46**********b4b",
+  "milvusUri": "https://in03-***********.api.gcp-us-west1.zillizcloud.com",
+  "milvusToken": "d46**********b4b",
   "milvusCollectionName": "YOUR-MILVUS-COLLECTION-NAME"
 }
 ```

code/pyproject.toml

@@ -4,7 +4,7 @@ authors = ["[email protected]"]
 description = ""
 name = "store-vector-db"
 readme = "README.md"
-version = "0.0.1"
+version = "0.1.4"
 package-mode = false
 
 [tool.poetry.dependencies]
@@ -13,10 +13,10 @@ apify-client = "^1.6.4"
 openai = "^1.17.0"
 python = ">=3.11,<3.12"
 python-dotenv = "^1.0.1"
-langchain-openai = "^0.1.6"
-langchain-cohere = "^0.1.4"
-langchain-community = "^0.2.0"
-langchain-core = "0.2.10"
+langchain-openai = "^0.2.0"
+langchain-cohere = "^0.3.0"
+langchain-community = "^0.3.0"
+langchain-core = "0.3.5"
 
 [tool.poetry.group.dev.dependencies]
 coverage = "^7.5.4"
@@ -35,7 +35,7 @@ ruff = "^0.3.5"
 optional = true
 
 [tool.poetry.group.pinecone.dependencies]
-langchain-pinecone = "^0.1.3"
+langchain-pinecone = "^0.2.0"
 
 [tool.poetry.group.chroma]
 optional = true
@@ -54,21 +54,21 @@ langchain-qdrant = "^0.1.3"
 optional = true
 
 [tool.poetry.group.pgvector.dependencies]
-langchain-postgres = "^0.0.8"
+langchain-postgres = "^0.0.12"
 psycopg = {extras = ["binary", "pool"], version = "^3.1.19"}
 psycopg2-binary = "^2.9.9"
 
 [tool.poetry.group.weaviate]
 optional = true
 
 [tool.poetry.group.weaviate.dependencies]
-langchain-weaviate = "^0.0.2"
+langchain-weaviate = "^0.0.3"
 
 [tool.poetry.group.milvus]
 optional = true
 
 [tool.poetry.group.milvus.dependencies]
-langchain-milvus = "^0.1.1"
+langchain-milvus = "^0.1.5"
 
 [tool.ruff]
 line-length = 150

code/src/examples/2024-07-08-milvus.py

@@ -35,8 +35,8 @@
 
 db = MilvusDatabase(
     actor_input=MilvusIntegration(
-        milvusUrl=os.getenv("MILVUS_URL"),
-        milvusApiKey=os.getenv("MILVUS_API_KEY"),
+        milvusUri=os.getenv("MILVUS_URI"),
+        milvusToken=os.getenv("MILVUS_TOKEN"),
         milvusCollectionName=MILVUS_COLLECTION_NAME,
         embeddingsProvider=EmbeddingsProvider.OpenAI.value,
         embeddingsApiKey=os.getenv("OPENAI_API_KEY"),

code/src/examples/2024-09-09-milvus-actor.py

@@ -30,11 +30,8 @@
 OPENAI_API_KEY = os.getenv("OPENAI_API_KEY") or "YOUR-OPENAI-API-KEY"
 
 MILVUS_COLLECTION_NAME = "apify"
-MILVUS_URL = os.getenv("MILVUS_URL") or "YOUR-MILVUS-URL"
-MILVUS_API_KEY = os.getenv("MILVUS_API_KEY") or "YOUR-MILVUS-API-KEY"
-MILVUS_USER = os.getenv("MILVUS_USER") or "YOUR-MILVUS-USER"
-MILVUS_PASSWORD = os.getenv("MILVUS_PASSWORD") or "YOUR-MILVUS-PASSWORD"
-
+MILVUS_URI = os.getenv("MILVUS_URI") or "YOUR-MILVUS-URI"
+MILVUS_TOKEN = os.getenv("MILVUS_TOKEN") or "YOUR-MILVUS-TOKEN"
 
 client = ApifyClient(APIFY_API_TOKEN)
 
@@ -51,11 +48,9 @@
 print(actor_call)
 
 milvus_integration_inputs = {
-    "milvusUrl": MILVUS_URL,
-    "milvusApiKey": MILVUS_API_KEY,
+    "milvusUri": MILVUS_URI,
+    "milvusToken": MILVUS_TOKEN,
     "milvusCollectionName": MILVUS_COLLECTION_NAME,
-    "milvusUser": MILVUS_USER,
-    "milvusPassword": MILVUS_PASSWORD,
     "datasetFields": ["text"],
     "datasetId": actor_call["defaultDatasetId"],
     "deltaUpdatesPrimaryDatasetFields": ["url"],
@@ -71,7 +66,7 @@
 
 print("Question answering using Milvus/Zilliz database")
 vectorstore = Milvus(
-    connection_args={"uri": MILVUS_URL, "token": MILVUS_API_KEY, "user": MILVUS_USER, "password": MILVUS_PASSWORD},
+    connection_args={"uri": MILVUS_URI, "token": MILVUS_TOKEN},
     embedding_function=embeddings,
     collection_name=MILVUS_COLLECTION_NAME,
 )

code/src/models/milvus_input_model.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  input_schema.json
-#   timestamp: 2024-07-23T09:53:48+00:00
+#   timestamp: 2024-09-26T09:34:48+00:00
 
 from __future__ import annotations
 
@@ -16,25 +16,17 @@ class EmbeddingsProvider(Enum):
 
 
 class MilvusIntegration(BaseModel):
-    milvusUrl: str = Field(
+    milvusUri: str = Field(
         ...,
-        description='REST URL of the Milvus instance to connect to',
-        title='Milvus URL',
+        description='The URI of the Milvus instance to connect to. You can include the username and password in the URI, for example: `https://username:password@****.serverless.gcp-us-west1.cloud.zilliz.com`.',
+        title='Milvus URI',
     )
-    milvusApiKey: str = Field(..., description='Milvus API KEY', title='Milvus API KEY')
+    milvusToken: str = Field(..., description='Milvus Token', title='Milvus Token')
     milvusCollectionName: str = Field(
         ...,
         description='Name of the Milvus collection where the data will be stored',
         title='Milvus collection name',
     )
-    milvusUser: Optional[str] = Field(
-        None, description='User name for the Milvus cluster', title='Milvus user name'
-    )
-    milvusPassword: Optional[str] = Field(
-        None,
-        description='Password for the Milvus cluster user',
-        title='Milvus user password',
-    )
     embeddingsProvider: EmbeddingsProvider = Field(
         ...,
         description='Choose the embeddings provider to use for generating embeddings',
@@ -92,12 +84,12 @@ class MilvusIntegration(BaseModel):
         title='Delete expired objects from the database after a specified number of days',
     )
     performChunking: Optional[bool] = Field(
-        False,
+        True,
         description='When set to true, the text will be divided into smaller chunks based on the settings provided below. Proper chunking helps optimize retrieval and ensures accurate and efficient responses.',
         title='Enable text chunking',
     )
     chunkSize: Optional[int] = Field(
-        1000,
+        2000,
         description='Defines the maximum number of characters in each text chunk. Choosing the right size balances between detailed context and system performance. Optimal sizes ensure high relevancy and minimal response time.',
         ge=1,
         title='Maximum chunk size',

code/src/utils.py

@@ -141,7 +141,7 @@ def add_item_checksum(items: list[Document], dataset_fields_to_item_id: list[str
 def add_chunk_id(chunks: list[Document]) -> list[Document]:
     """For every chunk (document stored in vector db) add chunk_id to metadata.
 
-    The chunk_id is a unique identifier for each chunk and is not required but it is better to keep it in metadata.
+    The chunk_id is a unique identifier for each chunk and is not required, but it is better to keep it in metadata.
     """
     for d in chunks:
         d.metadata["chunk_id"] = d.metadata.get("chunk_id", str(uuid4()))

code/src/vector_stores/milvus.py

@@ -20,20 +20,16 @@ class MilvusDatabase(Milvus, VectorDbBase):
     def __init__(self, actor_input: MilvusIntegration, embeddings: Embeddings) -> None:
         self.collection_name = actor_input.milvusCollectionName
 
-        connection_args = {"uri": actor_input.milvusUrl, "token": actor_input.milvusApiKey}
-
-        if actor_input.milvusUser and actor_input.milvusPassword:
-            connection_args |= {"user": actor_input.milvusUser, "password": actor_input.milvusPassword}
-
+        connection_args = {"uri": actor_input.milvusUri, "token": actor_input.milvusToken}
         super().__init__(connection_args=connection_args, embedding_function=embeddings, collection_name=self.collection_name)
         self.client = MilvusClient(**connection_args)
         self._dummy_vector: list[float] = []
 
     @property
     def dummy_vector(self) -> list[float]:
         if not self._dummy_vector and self.embeddings:
-            self._dummy_vector = self.embeddings.embed_query("dummy")
-        return self._dummy_vector
+            self._dummy_vector = self.embeddings.embed_query("dummy")  # type: ignore
+        return self._dummy_vector  # type: ignore
 
     async def is_connected(self) -> bool:
         raise NotImplementedError