chromadb-client 0.3.26.dev10__tar.gz → 0.3.27.dev0__tar.gz

chromadb-client-0.3.27.dev0/.github/workflows/chroma-client-integration-test.yml

@@ -0,0 +1,29 @@
+name: Chroma Client Integration Tests
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+    branches:
+      - main
+
+jobs:
+  test:
+    timeout-minutes: 90
+    strategy:
+      matrix:
+        python: ['3.7', '3.8', '3.9', '3.10']
+        platform: [ubuntu-latest]
+    runs-on: ${{ matrix.platform }}
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v3
+    - name: Set up Python ${{ matrix.python }}
+      uses: actions/setup-python@v4
+      with:
+        python-version: ${{ matrix.python }}
+    - name: Install test dependencies
+      run: python -m pip install -r requirements.txt && python -m pip install -r requirements_dev.txt
+    - name: Test
+      run: clients/python/integration-test.sh

{chromadb-client-0.3.26.dev10 → chromadb-client-0.3.27.dev0}/.pre-commit-config.yaml

@@ -13,15 +13,12 @@ repos:
       - id: check-case-conflict
       - id: check-docstring-first
 
-
   - repo: https://github.com/psf/black
     # https://github.com/psf/black/issues/2493
-    rev: 'refs/tags/23.3.0:refs/tags/23.3.0'
+    rev: "refs/tags/23.3.0:refs/tags/23.3.0"
     hooks:
       - id: black
 
-
-
   - repo: https://github.com/PyCQA/flake8
     rev: 6.0.0
     hooks:
@@ -31,8 +28,16 @@ repos:
           - "--max-line-length=88"
 
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: 'v1.2.0'
+    rev: "v1.2.0"
     hooks:
-    -   id: mypy
-        args: [--strict, --ignore-missing-imports, --follow-imports=silent]
-        additional_dependencies: ["types-requests", "pydantic", "overrides", "hypothesis", "pytest"]
+      - id: mypy
+        args: [--strict, --ignore-missing-imports, --follow-imports=silent, --disable-error-code=type-abstract]
+        additional_dependencies:
+          [
+            "types-requests",
+            "pydantic",
+            "overrides",
+            "hypothesis",
+            "pytest",
+            "numpy",
+          ]

{chromadb-client-0.3.26.dev10 → chromadb-client-0.3.27.dev0}/.vscode/settings.json

@@ -33,6 +33,7 @@
     "--ignore-missing-imports",
     "--show-column-numbers",
     "--no-pretty",
-    "--strict"
+    "--strict",
+    "--disable-error-code=type-abstract"
   ]
 }

{chromadb-client-0.3.26.dev10 → chromadb-client-0.3.27.dev0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: chromadb-client
-Version: 0.3.26.dev10
+Version: 0.3.27.dev0
 Summary: Chroma Client.
 Author-email: Jeff Huber <jeff@trychroma.com>, Anton Troynikov <anton@trychroma.com>
 Project-URL: Homepage, https://github.com/chroma-core/chroma

{chromadb-client-0.3.26.dev10 → chromadb-client-0.3.27.dev0}/chromadb/__init__.py

@@ -1,6 +1,7 @@
 import chromadb.config
 import logging
 from chromadb.telemetry.events import ClientStartEvent
+from chromadb.telemetry import Telemetry
 from chromadb.config import Settings, System
 from chromadb.api import API
 
@@ -8,7 +9,7 @@ logger = logging.getLogger(__name__)
 
 __settings = Settings()
 
-__version__ = "0.3.25"
+__version__ = "0.3.26"
 
 
 def configure(**kwargs) -> None:  # type: ignore
@@ -22,14 +23,16 @@ def get_settings() -> Settings:
 
 
 def Client(settings: Settings = __settings) -> API:
-    """Return a chroma.API instance based on the provided or environmental
-    settings, optionally overriding the DB instance."""
+    """Return a running chroma.API instance"""
 
     system = System(settings)
 
-    telemetry_client = system.get_telemetry()
+    telemetry_client = system.instance(Telemetry)
+    api = system.instance(API)
+
+    system.start()
 
     # Submit event for client start
     telemetry_client.capture(ClientStartEvent())
 
-    return system.get_api()
+    return api

{chromadb-client-0.3.26.dev10 → chromadb-client-0.3.27.dev0}/chromadb/api/__init__.py

@@ -16,15 +16,12 @@ from chromadb.api.types import (
     GetResult,
     WhereDocument,
 )
+from chromadb.config import Component
 import chromadb.utils.embedding_functions as ef
-from chromadb.telemetry import Telemetry
+from overrides import override
 
 
-class API(ABC):
-    @abstractmethod
-    def __init__(self, telemetry_client: Telemetry):
-        pass
-
+class API(Component, ABC):
     @abstractmethod
     def heartbeat(self) -> int:
         """Returns the current server time in nanoseconds to check if the server is alive
@@ -289,15 +286,16 @@ class API(ABC):
         """
         pass
 
+    @override
     @abstractmethod
-    def reset(self) -> bool:
+    def reset(self) -> None:
         """Resets the database
         ⚠️ This is destructive and will delete all data in the database.
         Args:
             None
 
         Returns:
-            True if the reset was successful
+            None
         """
         pass
 

{chromadb-client-0.3.26.dev10 → chromadb-client-0.3.27.dev0}/chromadb/api/fastapi.py

@@ -22,22 +22,27 @@ from typing import Sequence
 from chromadb.api.models.Collection import Collection
 import chromadb.errors as errors
 from uuid import UUID
+from chromadb.telemetry import Telemetry
+from overrides import override
 
 
 class FastAPI(API):
     def __init__(self, system: System):
+        super().__init__(system)
         url_prefix = "https" if system.settings.chroma_server_ssl_enabled else "http"
         system.settings.require("chroma_server_host")
         system.settings.require("chroma_server_http_port")
         self._api_url = f"{url_prefix}://{system.settings.chroma_server_host}:{system.settings.chroma_server_http_port}/api/v1"
-        self._telemetry_client = system.get_telemetry()
+        self._telemetry_client = self.require(Telemetry)
 
+    @override
     def heartbeat(self) -> int:
         """Returns the current server time in nanoseconds to check if the server is alive"""
         resp = requests.get(self._api_url)
         raise_chroma_error(resp)
         return int(resp.json()["nanosecond heartbeat"])
 
+    @override
     def list_collections(self) -> Sequence[Collection]:
         """Returns a list of all collections"""
         resp = requests.get(self._api_url + "/collections")
@@ -49,6 +54,7 @@ class FastAPI(API):
 
         return collections
 
+    @override
     def create_collection(
         self,
         name: str,
@@ -73,6 +79,7 @@ class FastAPI(API):
             metadata=resp_json["metadata"],
         )
 
+    @override
     def get_collection(
         self,
         name: str,
@@ -90,6 +97,7 @@ class FastAPI(API):
             metadata=resp_json["metadata"],
         )
 
+    @override
     def get_or_create_collection(
         self,
         name: str,
@@ -102,6 +110,7 @@ class FastAPI(API):
             name, metadata, embedding_function, get_or_create=True
         )
 
+    @override
     def _modify(
         self,
         id: UUID,
@@ -115,11 +124,13 @@ class FastAPI(API):
         )
         raise_chroma_error(resp)
 
+    @override
     def delete_collection(self, name: str) -> None:
         """Deletes a collection"""
         resp = requests.delete(self._api_url + "/collections/" + name)
         raise_chroma_error(resp)
 
+    @override
     def _count(self, collection_id: UUID) -> int:
         """Returns the number of embeddings in the database"""
         resp = requests.get(
@@ -128,13 +139,15 @@ class FastAPI(API):
         raise_chroma_error(resp)
         return cast(int, resp.json())
 
-    def _peek(self, collection_id: UUID, limit: int = 10) -> GetResult:
+    @override
+    def _peek(self, collection_id: UUID, n: int = 10) -> GetResult:
         return self._get(
             collection_id,
-            limit=limit,
+            limit=n,
             include=["embeddings", "documents", "metadatas"],
         )
 
+    @override
     def _get(
         self,
         collection_id: UUID,
@@ -177,6 +190,7 @@ class FastAPI(API):
             documents=body.get("documents", None),
         )
 
+    @override
     def _delete(
         self,
         collection_id: UUID,
@@ -196,6 +210,7 @@ class FastAPI(API):
         raise_chroma_error(resp)
         return cast(IDs, resp.json())
 
+    @override
     def _add(
         self,
         ids: IDs,
@@ -227,6 +242,7 @@ class FastAPI(API):
         raise_chroma_error(resp)
         return True
 
+    @override
     def _update(
         self,
         collection_id: UUID,
@@ -255,6 +271,7 @@ class FastAPI(API):
         resp.raise_for_status()
         return True
 
+    @override
     def _upsert(
         self,
         collection_id: UUID,
@@ -285,6 +302,7 @@ class FastAPI(API):
         resp.raise_for_status()
         return True
 
+    @override
     def _query(
         self,
         collection_id: UUID,
@@ -320,18 +338,20 @@ class FastAPI(API):
             documents=body.get("documents", None),
         )
 
-    def reset(self) -> bool:
+    @override
+    def reset(self) -> None:
         """Resets the database"""
         resp = requests.post(self._api_url + "/reset")
         raise_chroma_error(resp)
-        return cast(bool, resp.json())
 
+    @override
     def persist(self) -> bool:
         """Persists the database"""
         resp = requests.post(self._api_url + "/persist")
         raise_chroma_error(resp)
         return cast(bool, resp.json())
 
+    @override
     def raw_sql(self, sql: str) -> pd.DataFrame:
         """Runs a raw SQL query against the database"""
         resp = requests.post(
@@ -340,6 +360,7 @@ class FastAPI(API):
         raise_chroma_error(resp)
         return pd.DataFrame.from_dict(resp.json())
 
+    @override
     def create_index(self, collection_name: str) -> bool:
         """Creates an index for the given space key"""
         resp = requests.post(
@@ -348,6 +369,7 @@ class FastAPI(API):
         raise_chroma_error(resp)
         return cast(bool, resp.json())
 
+    @override
     def get_version(self) -> str:
         """Returns the version of the server"""
         resp = requests.get(self._api_url + "/version")

{chromadb-client-0.3.26.dev10 → chromadb-client-0.3.27.dev0}/chromadb/api/local.py

@@ -27,6 +27,8 @@ import re
 
 from chromadb.telemetry import Telemetry
 from chromadb.telemetry.events import CollectionAddEvent, CollectionDeleteEvent
+from overrides import override
+import pandas as pd
 
 
 # mimics s3 bucket requirements for naming
@@ -55,9 +57,11 @@ class LocalAPI(API):
     _telemetry_client: Telemetry
 
     def __init__(self, system: System):
-        self._db = system.get_db()
-        self._telemetry_client = system.get_telemetry()
+        super().__init__(system)
+        self._db = self.require(DB)
+        self._telemetry_client = self.require(Telemetry)
 
+    @override
     def heartbeat(self) -> int:
         """Ping the database to ensure it is alive
 
@@ -70,6 +74,7 @@ class LocalAPI(API):
     #
     # COLLECTION METHODS
     #
+    @override
     def create_collection(
         self,
         name: str,
@@ -111,6 +116,7 @@ class LocalAPI(API):
             metadata=res[0][2],
         )
 
+    @override
     def get_or_create_collection(
         self,
         name: str,
@@ -136,6 +142,7 @@ class LocalAPI(API):
             name, metadata, embedding_function, get_or_create=True
         )
 
+    @override
     def get_collection(
         self,
         name: str,
@@ -169,6 +176,7 @@ class LocalAPI(API):
             metadata=res[0][2],
         )
 
+    @override
     def list_collections(self) -> Sequence[Collection]:
         """List all collections.
         Returns:
@@ -193,17 +201,19 @@ class LocalAPI(API):
             )
         return collections
 
+    @override
     def _modify(
         self,
         id: UUID,
         new_name: Optional[str] = None,
-        new_metadata: Optional[Metadata] = None,
+        new_metadata: Optional[CollectionMetadata] = None,
     ) -> None:
         if new_name is not None:
             check_index_name(new_name)
 
         self._db.update_collection(id, new_name, new_metadata)
 
+    @override
     def delete_collection(self, name: str) -> None:
         """Delete a collection with the given name.
         Args:
@@ -222,6 +232,7 @@ class LocalAPI(API):
     #
     # ITEM METHODS
     #
+    @override
     def _add(
         self,
         ids: IDs,
@@ -251,6 +262,7 @@ class LocalAPI(API):
         self._telemetry_client.capture(CollectionAddEvent(str(collection_id), len(ids)))
         return True  # NIT: should this return the ids of the succesfully added items?
 
+    @override
     def _update(
         self,
         collection_id: UUID,
@@ -262,6 +274,7 @@ class LocalAPI(API):
         self._db.update(collection_id, ids, embeddings, metadatas, documents)
         return True
 
+    @override
     def _upsert(
         self,
         collection_id: UUID,
@@ -323,6 +336,7 @@ class LocalAPI(API):
 
         return True
 
+    @override
     def _get(
         self,
         collection_id: UUID,
@@ -390,6 +404,7 @@ class LocalAPI(API):
             get_result["ids"].append(entry[column_index["id"]])
         return get_result
 
+    @override
     def _delete(
         self,
         collection_id: UUID,
@@ -415,10 +430,12 @@ class LocalAPI(API):
 
         return deleted_uuids
 
+    @override
     def _count(self, collection_id: UUID) -> int:
         return self._db.count(collection_id)
 
-    def reset(self) -> bool:
+    @override
+    def reset(self) -> None:
         """Reset the database. This will delete all collections and items.
 
         Returns:
@@ -426,8 +443,8 @@ class LocalAPI(API):
 
         """
         self._db.reset()
-        return True
 
+    @override
     def _query(
         self,
         collection_id: UUID,
@@ -498,14 +515,17 @@ class LocalAPI(API):
 
         return query_result
 
-    def raw_sql(self, raw_sql: str):  # type: ignore
-        return self._db.raw_sql(raw_sql)  # type: ignore
+    @override
+    def raw_sql(self, sql: str) -> pd.DataFrame:
+        return self._db.raw_sql(sql)  # type: ignore
 
+    @override
     def create_index(self, collection_name: str) -> bool:
         collection_uuid = self._db.get_collection_uuid_from_name(collection_name)
         self._db.create_index(collection_uuid=collection_uuid)
         return True
 
+    @override
     def _peek(self, collection_id: UUID, n: int = 10) -> GetResult:
         return self._get(
             collection_id=collection_id,
@@ -513,6 +533,7 @@ class LocalAPI(API):
             include=["embeddings", "documents", "metadatas"],
         )
 
+    @override
     def persist(self) -> bool:
         """Persist the database to disk.
 
@@ -523,6 +544,7 @@ class LocalAPI(API):
         self._db.persist()
         return True
 
+    @override
     def get_version(self) -> str:
         """Get the version of Chroma.
 

{chromadb-client-0.3.26.dev10 → chromadb-client-0.3.27.dev0}/chromadb/api/types.py

@@ -1,16 +1,27 @@
 from typing import Any, Optional, Union, Dict, Sequence, TypeVar, List
 from typing_extensions import Literal, TypedDict, Protocol
 import chromadb.errors as errors
+from chromadb.types import (
+    Metadata,
+    Vector,
+    LiteralValue,
+    LogicalOperator,
+    WhereOperator,
+    OperatorExpression,
+    Where,
+    WhereDocumentOperator,
+    WhereDocument,
+)
+
+# Re-export types from chromadb.types
+__all__ = ["Metadata", "Where", "WhereDocument"]
 
 ID = str
 IDs = List[ID]
 
-Number = Union[int, float]
-Embedding = List[Number]
+Embedding = Vector
 Embeddings = List[Embedding]
 
-
-Metadata = Dict[str, Union[str, int, float]]
 Metadatas = List[Metadata]
 
 CollectionMetadata = Dict[Any, Any]
@@ -22,20 +33,24 @@ Parameter = TypeVar("Parameter", Embedding, Document, Metadata, ID)
 T = TypeVar("T")
 OneOrMany = Union[T, List[T]]
 
-Include = List[Literal["documents", "embeddings", "metadatas", "distances"]]
-
-# Grammar for where expressions
-LiteralValue = Union[str, int, float]
-LogicalOperator = Literal["$and", "$or"]
-WhereOperator = Literal["$gt", "$gte", "$lt", "$lte", "$ne", "$eq"]
-OperatorExpression = Dict[Union[WhereOperator, LogicalOperator], LiteralValue]
-
-Where = Dict[
-    Union[str, LogicalOperator], Union[LiteralValue, OperatorExpression, List["Where"]]
+# This should ust be List[Literal["documents", "embeddings", "metadatas", "distances"]]
+# However, this provokes an incompatibility with the Overrides library and Python 3.7
+Include = List[
+    Union[
+        Literal["documents"],
+        Literal["embeddings"],
+        Literal["metadatas"],
+        Literal["distances"],
+    ]
 ]
 
-WhereDocumentOperator = Literal["$contains", LogicalOperator]
-WhereDocument = Dict[WhereDocumentOperator, Union[str, List["WhereDocument"]]]
+# Re-export types from chromadb.types
+LiteralValue = LiteralValue
+LogicalOperator = LogicalOperator
+WhereOperator = WhereOperator
+OperatorExpression = OperatorExpression
+Where = Where
+WhereDocumentOperator = WhereDocumentOperator
 
 
 class GetResult(TypedDict):
@@ -135,6 +150,8 @@ def validate_where(where: Where) -> Where:
     """
     if not isinstance(where, dict):
         raise ValueError(f"Expected where to be a dict, got {where}")
+    if len(where) != 1:
+        raise ValueError(f"Expected where to have exactly one operator, got {where}")
     for key, value in where.items():
         if not isinstance(key, str):
             raise ValueError(f"Expected where key to be a str, got {key}")