arize-phoenix 5.5.2__tar.gz → 5.6.0__tar.gz

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: arize-phoenix
-Version: 5.5.2
+Version: 5.6.0
 Summary: AI Observability and Evaluation
 Project-URL: Documentation, https://docs.arize.com/phoenix/
 Project-URL: Issues, https://github.com/Arize-ai/phoenix/issues
@@ -23,13 +23,11 @@ Requires-Dist: arize-phoenix-evals>=0.13.1
 Requires-Dist: arize-phoenix-otel>=0.5.1
 Requires-Dist: authlib
 Requires-Dist: cachetools
-Requires-Dist: fast-hdbscan>=0.2.0
 Requires-Dist: fastapi
 Requires-Dist: grpc-interceptor
 Requires-Dist: grpcio
 Requires-Dist: httpx
 Requires-Dist: jinja2
-Requires-Dist: numba>=0.60.0
 Requires-Dist: numpy!=2.0.0
 Requires-Dist: openinference-instrumentation>=0.1.12
 Requires-Dist: openinference-semantic-conventions>=0.1.9
@@ -51,7 +49,6 @@ Requires-Dist: starlette
 Requires-Dist: strawberry-graphql==0.243.1
 Requires-Dist: tqdm
 Requires-Dist: typing-extensions>=4.6
-Requires-Dist: umap-learn
 Requires-Dist: uvicorn
 Requires-Dist: websockets
 Requires-Dist: wrapt
@@ -86,7 +83,7 @@ Requires-Dist: jupyter; extra == 'dev'
 Requires-Dist: langchain>=0.0.334; extra == 'dev'
 Requires-Dist: litellm>=1.0.3; extra == 'dev'
 Requires-Dist: llama-index>=0.10.3; extra == 'dev'
-Requires-Dist: mypy==1.11.2; extra == 'dev'
+Requires-Dist: mypy==1.12.1; extra == 'dev'
 Requires-Dist: nbqa; extra == 'dev'
 Requires-Dist: pandas-stubs==2.2.2.240603; extra == 'dev'
 Requires-Dist: pandas>=1.0; extra == 'dev'
@@ -193,7 +190,7 @@ Phoenix container images are available via [Docker Hub](https://hub.docker.com/r
 | [Data Export](https://docs.arize.com/phoenix/tracing/how-to-tracing/extract-data-from-spans)                     | ✅             |
 | REST API                                                                                                         | ✅             |
 | GraphQL API                                                                                                      | ✅             |
-| Data Retention                                                                                                   | Costomizable   |
+| Data Retention                                                                                                   | Customizable   |
 | Authentication                                                                                                   | ✅             |
 | Social Login                                                                                                     | ✅             |
 | RBAC                                                                                                             | ✅             |

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/README.md

@@ -66,7 +66,7 @@ Phoenix container images are available via [Docker Hub](https://hub.docker.com/r
 | [Data Export](https://docs.arize.com/phoenix/tracing/how-to-tracing/extract-data-from-spans)                     | ✅             |
 | REST API                                                                                                         | ✅             |
 | GraphQL API                                                                                                      | ✅             |
-| Data Retention                                                                                                   | Costomizable   |
+| Data Retention                                                                                                   | Customizable   |
 | Authentication                                                                                                   | ✅             |
 | Social Login                                                                                                     | ✅             |
 | RBAC                                                                                                             | ✅             |

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/pyproject.toml

@@ -25,9 +25,6 @@ dependencies = [
   "numpy!=2.0.0",  # https://github.com/lmcinnes/umap/issues/1138
   "pandas>=1.0",
   "jinja2",
-  "umap-learn",
-  "fast-hdbscan>=0.2.0",
-  "numba>=0.60.0",  # https://github.com/astral-sh/uv/issues/6281
   "starlette",
   "uvicorn",
   "psutil",
@@ -77,7 +74,7 @@ dev = [
   "jupyter",
   "nbqa",
   "ruff==0.6.9",
-  "mypy==1.11.2",
+  "mypy==1.12.1",
   "pandas>=1.0",
   "tabulate",  # used by DataFrame.to_markdown()
   "types-tabulate",
@@ -215,7 +212,7 @@ dependencies = [
 dependencies = [
   "grpcio",
   "litellm>=1.0.3",
-  "mypy==1.11.2",
+  "mypy==1.12.1",
   "openai>=1.0.0",
   "opentelemetry-exporter-otlp",
   "opentelemetry-instrumentation-fastapi",
@@ -396,14 +393,14 @@ module = [
   "orjson",  # suppress fastapi internal type errors
   "pypistats",
   "authlib.*",
+  "google.*",
+  "mistralai.*",
 ]
 ignore_missing_imports = true
 
 [tool.ruff]
 exclude = [
   "api_reference",
-  "packages",
-  "src/phoenix/evals/",
   "dist/",
   ".git",
   "__pycache__",

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/src/phoenix/config.py

@@ -6,7 +6,7 @@ from dataclasses import dataclass
 from datetime import timedelta
 from enum import Enum
 from pathlib import Path
-from typing import Dict, List, Optional, Tuple, overload
+from typing import Optional, overload
 from urllib.parse import urlparse
 
 from phoenix.utilities.logging import log_a_list
@@ -273,7 +273,7 @@ def get_env_phoenix_api_key() -> Optional[str]:
     return os.environ.get(ENV_PHOENIX_API_KEY)
 
 
-def get_env_auth_settings() -> Tuple[bool, Optional[str]]:
+def get_env_auth_settings() -> tuple[bool, Optional[str]]:
     """
     Gets auth settings and performs validation.
     """
@@ -329,8 +329,8 @@ def get_env_refresh_token_expiry() -> timedelta:
     return timedelta(minutes=minutes)
 
 
-def get_env_csrf_trusted_origins() -> List[str]:
-    origins: List[str] = []
+def get_env_csrf_trusted_origins() -> list[str]:
+    origins: list[str] = []
     if not (csrf_trusted_origins := os.getenv(ENV_PHOENIX_CSRF_TRUSTED_ORIGINS)):
         return origins
     for origin in csrf_trusted_origins.split(","):
@@ -428,7 +428,7 @@ class OAuth2ClientConfig:
         )
 
 
-def get_env_oauth2_settings() -> List[OAuth2ClientConfig]:
+def get_env_oauth2_settings() -> list[OAuth2ClientConfig]:
     """
     Get OAuth2 settings from environment variables.
     """
@@ -494,7 +494,7 @@ def ensure_working_dir() -> None:
 ensure_working_dir()
 
 
-def get_exported_files(directory: Path) -> List[Path]:
+def get_exported_files(directory: Path) -> list[Path]:
     """
     Yields the list of paths of exported files.
 
@@ -505,7 +505,7 @@ def get_exported_files(directory: Path) -> List[Path]:
 
     Returns
     -------
-    list: List[Path]
+    list: list[Path]
         List of paths of the exported files.
     """
     return list(directory.glob("*.parquet"))
@@ -584,7 +584,7 @@ def get_env_enable_prometheus() -> bool:
     )
 
 
-def get_env_client_headers() -> Optional[Dict[str, str]]:
+def get_env_client_headers() -> Optional[dict[str, str]]:
     if headers_str := os.getenv(ENV_PHOENIX_CLIENT_HEADERS):
         return parse_env_headers(headers_str)
     return None

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/src/phoenix/core/model.py

@@ -1,4 +1,4 @@
-from typing import List, Optional, Union
+from typing import Optional, Union
 
 from phoenix.inferences.inferences import Inferences
 from phoenix.inferences.schema import EmbeddingColumnNames, EmbeddingFeatures
@@ -8,8 +8,8 @@ from .embedding_dimension import EmbeddingDimension
 
 def _get_embedding_dimensions(
     primary_inferences: Inferences, reference_inferences: Optional[Inferences]
-) -> List[EmbeddingDimension]:
-    embedding_dimensions: List[EmbeddingDimension] = []
+) -> list[EmbeddingDimension]:
+    embedding_dimensions: list[EmbeddingDimension] = []
     embedding_features: EmbeddingFeatures = {}
 
     primary_embedding_features: Optional[EmbeddingFeatures] = (

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/src/phoenix/core/model_schema.py

@@ -4,6 +4,7 @@ import re
 import threading
 from abc import ABC, abstractmethod
 from collections import defaultdict
+from collections.abc import Callable, Hashable, Iterable, Iterator, Mapping, Sequence
 from contextlib import contextmanager
 from dataclasses import dataclass, field, fields, replace
 from datetime import datetime, timedelta, timezone
@@ -14,19 +15,9 @@ from random import random
 from typing import (
     Any,
     BinaryIO,
-    Callable,
-    Dict,
     Generic,
-    Hashable,
-    Iterable,
-    Iterator,
-    List,
-    Mapping,
     NamedTuple,
     Optional,
-    Sequence,
-    Tuple,
-    Type,
     TypeVar,
     Union,
     cast,
@@ -265,7 +256,7 @@ class _Cache(Generic[_Key, _Value]):
     2
     """
 
-    _cache: Dict[_Key, _Value] = field(
+    _cache: dict[_Key, _Value] = field(
         init=False,
         default_factory=dict,
     )
@@ -275,7 +266,7 @@ class _Cache(Generic[_Key, _Value]):
     )
 
     @contextmanager
-    def __call__(self) -> Iterator[Dict[_Key, _Value]]:
+    def __call__(self) -> Iterator[dict[_Key, _Value]]:
         with self._lock:
             yield self._cache
 
@@ -405,14 +396,14 @@ class Dimension(Column, ABC):
 @dataclass(frozen=True)
 class ScalarDimension(Dimension):
     @property
-    def min_max(self) -> Tuple[Any, Any]:
+    def min_max(self) -> tuple[Any, Any]:
         if self._model is None:
             return np.nan, np.nan
         model = cast(Model, self._model)
         return model.dimension_min_max_from_all_df(self.name)
 
     @property
-    def categories(self) -> Tuple[str, ...]:
+    def categories(self) -> tuple[str, ...]:
         if self._model is None or self.data_type is CONTINUOUS:
             return ()
         model = cast(Model, self._model)
@@ -502,7 +493,7 @@ class RetrievalEmbeddingDimension(EmbeddingDimension):
 Name: TypeAlias = str
 ColumnKey: TypeAlias = Union[Name, Column, SingularDimensionalRole]
 MultiDimensionKey: TypeAlias = Union[MultiDimensionalRole, Sequence[DimensionRole]]
-RowNumbering: TypeAlias = Union[int, List[int]]
+RowNumbering: TypeAlias = Union[int, list[int]]
 
 
 def _is_column_key(key: Any) -> TypeGuard[ColumnKey]:
@@ -525,7 +516,7 @@ def _is_multi_dimension_key(
 
 def _is_dimension_type_filter(
     key: Any,
-) -> TypeGuard[Tuple[MultiDimensionKey, Union[Type[ScalarDimension], Type[EmbeddingDimension]]]]:
+) -> TypeGuard[tuple[MultiDimensionKey, Union[type[ScalarDimension], type[EmbeddingDimension]]]]:
     return (
         isinstance(key, tuple)
         and len(key) == 2
@@ -534,7 +525,7 @@ def _is_dimension_type_filter(
     )
 
 
-def _is_named_df(obj: Any) -> TypeGuard[Tuple[Name, pd.DataFrame]]:
+def _is_named_df(obj: Any) -> TypeGuard[tuple[Name, pd.DataFrame]]:
     return (
         isinstance(obj, tuple)
         and len(obj) == 2
@@ -664,7 +655,7 @@ class Events(ModelData):
     def __getitem__(self, key: ColumnKey) -> "pd.Series[Any]": ...
 
     @overload
-    def __getitem__(self, key: List[RowId]) -> "Events": ...
+    def __getitem__(self, key: list[RowId]) -> "Events": ...
 
     def __getitem__(self, key: Any) -> Any:
         if isinstance(key, list):
@@ -722,7 +713,7 @@ class Inferences(Events):
     def __getitem__(self, key: ColumnKey) -> "pd.Series[Any]": ...
 
     @overload
-    def __getitem__(self, key: List[RowId]) -> Events: ...
+    def __getitem__(self, key: list[RowId]) -> Events: ...
 
     def __getitem__(self, key: Any) -> Any:
         if isinstance(key, list):
@@ -746,19 +737,19 @@ class Model:
     a column of NaNs.
     """
 
-    _inference_sets: Dict[InferencesRole, Inferences]
-    _dimensions: Dict[Name, Dimension]
-    _dim_names_by_role: Dict[DimensionRole, List[Name]]
-    _original_columns_by_role: Dict[InferencesRole, "pd.Index[Any]"]
+    _inference_sets: dict[InferencesRole, Inferences]
+    _dimensions: dict[Name, Dimension]
+    _dim_names_by_role: dict[DimensionRole, list[Name]]
+    _original_columns_by_role: dict[InferencesRole, "pd.Index[Any]"]
     _default_timestamps_factory: _ConstantValueSeriesFactory
     _nan_series_factory: _ConstantValueSeriesFactory
-    _dimension_categories_from_all_inferences: _Cache[Name, Tuple[str, ...]]
-    _dimension_min_max_from_all_inferences: _Cache[Name, Tuple[float, float]]
+    _dimension_categories_from_all_inferences: _Cache[Name, tuple[str, ...]]
+    _dimension_min_max_from_all_inferences: _Cache[Name, tuple[float, float]]
 
     def __init__(
         self,
         dimensions: Iterable[Dimension],
-        dataframes: Iterable[Union[pd.DataFrame, Tuple[Name, pd.DataFrame]]],
+        dataframes: Iterable[Union[pd.DataFrame, tuple[Name, pd.DataFrame]]],
         /,
         treat_omitted_columns_as_features: bool = True,
         timestamps_already_normalized: bool = False,
@@ -775,11 +766,11 @@ class Model:
         object.__setattr__(
             self,
             "_dimension_min_max_from_all_inferences",
-            _Cache[Name, Tuple[float, float]](),
+            _Cache[Name, tuple[float, float]](),
         )
 
         df_names, dfs = cast(
-            Tuple[Iterable[Name], Iterable[pd.DataFrame]],
+            tuple[Iterable[Name], Iterable[pd.DataFrame]],
             zip(*_coerce_tuple(dataframes)),
         )
         str_col_dfs = _coerce_str_column_names(dfs)
@@ -956,7 +947,7 @@ class Model:
         )
 
     @cached_property
-    def scalar_dimensions(self) -> Tuple[ScalarDimension, ...]:
+    def scalar_dimensions(self) -> tuple[ScalarDimension, ...]:
         """Put these in a cached sequence because currently their positions
         in this list also determine their node IDs in graphql.
         """
@@ -969,7 +960,7 @@ class Model:
         )
 
     @cached_property
-    def embedding_dimensions(self) -> Tuple[EmbeddingDimension, ...]:
+    def embedding_dimensions(self) -> tuple[EmbeddingDimension, ...]:
         """Put these in a cached sequence because currently their positions
         in this list also determine their node IDs in graphql.
         """
@@ -982,10 +973,10 @@ class Model:
     def dimension_categories_from_all_inferences(
         self,
         dimension_name: Name,
-    ) -> Tuple[str, ...]:
+    ) -> tuple[str, ...]:
         dim = self[dimension_name]
         if dim.data_type is CONTINUOUS:
-            return cast(Tuple[str, ...], ())
+            return cast(tuple[str, ...], ())
         with self._dimension_categories_from_all_inferences() as cache:
             try:
                 return cache[dimension_name]
@@ -1003,7 +994,7 @@ class Model:
     def dimension_min_max_from_all_df(
         self,
         dimension_name: Name,
-    ) -> Tuple[float, float]:
+    ) -> tuple[float, float]:
         dim = self[dimension_name]
         if dim.data_type is not CONTINUOUS:
             return (np.nan, np.nan)
@@ -1021,7 +1012,7 @@ class Model:
         return ans
 
     @overload
-    def __getitem__(self, key: Type[Inferences]) -> Iterator[Inferences]: ...
+    def __getitem__(self, key: type[Inferences]) -> Iterator[Inferences]: ...
 
     @overload
     def __getitem__(self, key: InferencesRole) -> Inferences: ...
@@ -1033,20 +1024,20 @@ class Model:
     def __getitem__(self, key: MultiDimensionKey) -> Iterator[Dimension]: ...
 
     @overload
-    def __getitem__(self, key: Type[ScalarDimension]) -> Iterator[ScalarDimension]: ...
+    def __getitem__(self, key: type[ScalarDimension]) -> Iterator[ScalarDimension]: ...
 
     @overload
-    def __getitem__(self, key: Type[EmbeddingDimension]) -> Iterator[EmbeddingDimension]: ...
+    def __getitem__(self, key: type[EmbeddingDimension]) -> Iterator[EmbeddingDimension]: ...
 
     @overload
-    def __getitem__(self, key: Type[Dimension]) -> Iterator[Dimension]: ...
+    def __getitem__(self, key: type[Dimension]) -> Iterator[Dimension]: ...
 
     @overload
     def __getitem__(
         self,
-        key: Tuple[
+        key: tuple[
             MultiDimensionKey,
-            Union[Type[ScalarDimension], Type[EmbeddingDimension]],
+            Union[type[ScalarDimension], type[EmbeddingDimension]],
         ],
     ) -> Iterator[Dimension]: ...
 
@@ -1094,9 +1085,9 @@ class Model:
 
     def _get_multi_dims_by_type(
         self,
-        key: Tuple[
+        key: tuple[
             MultiDimensionKey,
-            Union[Type[ScalarDimension], Type[EmbeddingDimension]],
+            Union[type[ScalarDimension], type[EmbeddingDimension]],
         ],
     ) -> Iterator[Dimension]:
         return filter(lambda dim: type(dim) is key[1], self[key[0]])
@@ -1105,7 +1096,7 @@ class Model:
     def _new_dimension(
         self,
         obj: DimensionRole,
-        cls: Type[Dimension] = ScalarDimension,
+        cls: type[Dimension] = ScalarDimension,
         **kwargs: Any,
     ) -> Dimension: ...
 
@@ -1113,7 +1104,7 @@ class Model:
     def _new_dimension(
         self,
         obj: Name,
-        cls: Type[Dimension] = ScalarDimension,
+        cls: type[Dimension] = ScalarDimension,
         **kwargs: Any,
     ) -> Dimension: ...
 
@@ -1125,7 +1116,7 @@ class Model:
     ) -> Dimension: ...
 
     def _new_dimension(
-        self, obj: Any, cls: Type[Dimension] = ScalarDimension, **kwargs: Any
+        self, obj: Any, cls: type[Dimension] = ScalarDimension, **kwargs: Any
     ) -> Dimension:
         """Creates a new Dimension or copies an existing one, setting the
         model weak reference to the `self` Model instance, and sharing the
@@ -1175,13 +1166,13 @@ class Schema(SchemaSpec):
     prediction_score: Optional[str] = None
     actual_label: Optional[str] = None
     actual_score: Optional[str] = None
-    prompt: Optional[RetrievalEmbedding] = None
+    prompt: Optional[Embedding] = None
     response: Optional[Union[str, Embedding]] = None
     features: Iterable[Union[str, CompositeDimensionSpec]] = field(default_factory=list)
     tags: Iterable[Union[str, CompositeDimensionSpec]] = field(default_factory=list)
 
     # internal attribute not exposed to users
-    _dimensions: List[Dimension] = field(
+    _dimensions: list[Dimension] = field(
         init=False, repr=False, hash=False, compare=False, default_factory=list
     )
 
@@ -1248,7 +1239,7 @@ class Schema(SchemaSpec):
 
     def __call__(
         self,
-        *dataframes: Union[pd.DataFrame, Tuple[Name, pd.DataFrame]],
+        *dataframes: Union[pd.DataFrame, tuple[Name, pd.DataFrame]],
         **kwargs: Any,
     ) -> Model:
         """Dimensions are the "baton" that Schema hands over to Model."""
@@ -1302,7 +1293,7 @@ def _get_omitted_column_names(
 
 def _group_names_by_dim_role(
     dimensions: Iterable[Dimension],
-) -> Iterator[Tuple[DimensionRole, List[str]]]:
+) -> Iterator[tuple[DimensionRole, list[str]]]:
     return (
         (role, [dim.name for dim in dims])
         for role, dims in groupby(
@@ -1356,8 +1347,8 @@ def _coerce_str(obj: Optional[str]) -> str:
 
 
 def _coerce_tuple(
-    dataframes: Iterable[Union[pd.DataFrame, Tuple[Name, pd.DataFrame]]],
-) -> Iterator[Tuple[Name, pd.DataFrame]]:
+    dataframes: Iterable[Union[pd.DataFrame, tuple[Name, pd.DataFrame]]],
+) -> Iterator[tuple[Name, pd.DataFrame]]:
     for dataframe in dataframes:
         if isinstance(dataframe, pd.DataFrame):
             yield (_rand_str(), dataframe)

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/src/phoenix/core/model_schema_adapter.py

@@ -1,6 +1,7 @@
+from collections.abc import Iterable, Sized
 from itertools import chain
 from operator import itemgetter
-from typing import Dict, Iterable, List, Optional, Sized, Tuple, Union
+from typing import Optional, Union
 
 import pandas as pd
 from pandas.api.types import is_object_dtype
@@ -24,18 +25,18 @@ def create_model_from_inferences(*inference_sets: Optional[Inferences]) -> Model
         # have the same length between inferences.
         _ = _get_embedding_dimensions(inference_sets[0], inference_sets[1])
 
-    named_dataframes: List[Tuple[InferencesName, pd.DataFrame]] = []
-    prediction_ids: List[ColumnName] = []
-    timestamps: List[ColumnName] = []
-    prediction_labels: List[ColumnName] = []
-    prediction_scores: List[ColumnName] = []
-    actual_labels: List[ColumnName] = []
-    actual_scores: List[ColumnName] = []
-    features: List[ColumnName] = []
-    tags: List[ColumnName] = []
-    embeddings: Dict[DisplayName, EmbeddingColumnNames] = {}
-    prompts: List[EmbeddingColumnNames] = []
-    responses: List[Union[str, EmbeddingColumnNames]] = []
+    named_dataframes: list[tuple[InferencesName, pd.DataFrame]] = []
+    prediction_ids: list[ColumnName] = []
+    timestamps: list[ColumnName] = []
+    prediction_labels: list[ColumnName] = []
+    prediction_scores: list[ColumnName] = []
+    actual_labels: list[ColumnName] = []
+    actual_scores: list[ColumnName] = []
+    features: list[ColumnName] = []
+    tags: list[ColumnName] = []
+    embeddings: dict[DisplayName, EmbeddingColumnNames] = {}
+    prompts: list[EmbeddingColumnNames] = []
+    responses: list[Union[str, EmbeddingColumnNames]] = []
 
     for inferences in filter(_is_inferences, inference_sets):
         df = inferences.dataframe
@@ -184,14 +185,14 @@ def _translate_prompt_embedding(
 def _split_vectors_vs_scalars(
     names: Iterable[str],
     *dataframes: pd.DataFrame,
-) -> Tuple[List[str], List[Embedding]]:
+) -> tuple[list[str], list[Embedding]]:
     """A best-effort attempt at separating vector columns from scalar columns
     by examining the first non-null item of the column from each dataframe. If
     any item is `Iterable` and `Sized`, but not `str`, then the column is
     returned as `Embedding`, else it's returned as scalar.
     """
-    scalars: List[str] = []
-    vectors: List[Embedding] = []
+    scalars: list[str] = []
+    vectors: list[Embedding] = []
     # convert to sets for a speedier lookup
     column_names = [set(df.columns) for df in dataframes]
     for name in names:

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/src/phoenix/datetime_utils.py

@@ -1,5 +1,5 @@
 from datetime import datetime, timedelta, timezone, tzinfo
-from typing import Any, Optional, Tuple, cast
+from typing import Any, Optional, cast
 
 import pandas as pd
 import pytz
@@ -96,7 +96,7 @@ MINUTE_DATETIME_FORMAT = "%Y-%m-%dT%H:%M:00%z"
 def right_open_time_range(
     min_time: Optional[datetime],
     max_time: Optional[datetime],
-) -> Tuple[Optional[datetime], Optional[datetime]]:
+) -> tuple[Optional[datetime], Optional[datetime]]:
     """
     First adds one minute to `max_time`, because time intervals are right
     open and one minute is the smallest interval allowed, then rounds down

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/src/phoenix/db/bulk_inserter.py

@@ -1,22 +1,12 @@
 import asyncio
 import logging
 from asyncio import Queue, as_completed
+from collections.abc import AsyncIterator, Awaitable, Callable, Iterable
 from dataclasses import dataclass, field
 from functools import singledispatchmethod
 from itertools import islice
 from time import perf_counter
-from typing import (
-    Any,
-    AsyncIterator,
-    Awaitable,
-    Callable,
-    Iterable,
-    List,
-    Optional,
-    Set,
-    Tuple,
-    cast,
-)
+from typing import Any, Optional, cast
 
 from typing_extensions import TypeAlias
 
@@ -43,7 +33,7 @@ ProjectRowId: TypeAlias = int
 
 @dataclass(frozen=True)
 class TransactionResult:
-    updated_project_rowids: Set[ProjectRowId] = field(default_factory=set)
+    updated_project_rowids: set[ProjectRowId] = field(default_factory=set)
 
 
 class BulkInserter:
@@ -52,7 +42,7 @@ class BulkInserter:
         db: DbSessionFactory,
         *,
         event_queue: CanPutItem[DmlEvent],
-        initial_batch_of_spans: Optional[Iterable[Tuple[Span, str]]] = None,
+        initial_batch_of_spans: Optional[Iterable[tuple[Span, str]]] = None,
         initial_batch_of_evaluations: Optional[Iterable[pb.Evaluation]] = None,
         sleep: float = 0.1,
         max_ops_per_transaction: int = 1000,
@@ -76,10 +66,10 @@ class BulkInserter:
         self._max_ops_per_transaction = max_ops_per_transaction
         self._operations: Optional[Queue[DataManipulation]] = None
         self._max_queue_size = max_queue_size
-        self._spans: List[Tuple[Span, str]] = (
+        self._spans: list[tuple[Span, str]] = (
             [] if initial_batch_of_spans is None else list(initial_batch_of_spans)
         )
-        self._evaluations: List[pb.Evaluation] = (
+        self._evaluations: list[pb.Evaluation] = (
             [] if initial_batch_of_evaluations is None else list(initial_batch_of_evaluations)
         )
         self._task: Optional[asyncio.Task[None]] = None
@@ -91,7 +81,7 @@ class BulkInserter:
 
     async def __aenter__(
         self,
-    ) -> Tuple[
+    ) -> tuple[
         Callable[[Any], Awaitable[None]],
         Callable[[Span, str], Awaitable[None]],
         Callable[[pb.Evaluation], Awaitable[None]],
@@ -183,7 +173,7 @@ class BulkInserter:
                 self._event_queue.put(event)
             await asyncio.sleep(self._sleep)
 
-    async def _insert_spans(self, spans: List[Tuple[Span, str]]) -> None:
+    async def _insert_spans(self, spans: list[tuple[Span, str]]) -> None:
         project_ids = set()
         for i in range(0, len(spans), self._max_ops_per_transaction):
             try:
@@ -220,7 +210,7 @@ class BulkInserter:
                 logger.exception("Failed to insert spans")
         self._event_queue.put(SpanInsertEvent(tuple(project_ids)))
 
-    async def _insert_evaluations(self, evaluations: List[pb.Evaluation]) -> None:
+    async def _insert_evaluations(self, evaluations: list[pb.Evaluation]) -> None:
         for i in range(0, len(evaluations), self._max_ops_per_transaction):
             try:
                 start = perf_counter()
@@ -273,7 +263,7 @@ class _QueueInserters:
         if self.empty:
             return
         for coro in as_completed([q.insert() for q in self._queues if not q.empty]):
-            if events := cast(Optional[List[DmlEvent]], await coro):
+            if events := cast(Optional[list[DmlEvent]], await coro):
                 for event in events:
                     yield event
 

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/src/phoenix/db/engines.py

@@ -1,9 +1,10 @@
 import asyncio
 import json
+from collections.abc import Callable
 from datetime import datetime
 from enum import Enum
 from sqlite3 import Connection
-from typing import Any, Callable
+from typing import Any
 
 import aiosqlite
 import numpy as np

{arize_phoenix-5.5.2 → arize_phoenix-5.6.0}/src/phoenix/db/enums.py

@@ -1,5 +1,5 @@
+from collections.abc import Mapping
 from enum import Enum
-from typing import Mapping, Type
 
 from sqlalchemy.orm import InstrumentedAttribute
 
@@ -15,6 +15,6 @@ class UserRole(Enum):
     MEMBER = "MEMBER"
 
 
-COLUMN_ENUMS: Mapping[InstrumentedAttribute[str], Type[Enum]] = {
+COLUMN_ENUMS: Mapping[InstrumentedAttribute[str], type[Enum]] = {
     models.UserRole.name: UserRole,
 }