arize 8.0.0b1__py3-none-any.whl → 8.0.0b4__py3-none-any.whl

arize/client.py

@@ -5,7 +5,7 @@ from __future__ import annotations
 import logging
 import shutil
 from pathlib import Path
-from typing import TYPE_CHECKING, ClassVar
+from typing import TYPE_CHECKING, ClassVar, cast
 
 from arize._lazy import LazySubclientsMixin
 from arize.config import SDKConfiguration
@@ -20,42 +20,22 @@ if TYPE_CHECKING:
 
 logger = logging.getLogger(__name__)
 
-# TODO(Kiko): InvalidAdditionalHeadersError is unused. Have we handled extra headers?
-
-# TODO(Kiko): Clean commented lines over the SDK
-# TODO(Kiko): Implement https://github.com/Arize-ai/arize/pull/59917
-
-# TODO(Kiko): Go private connect. Need a `base_domain`, such that we get:
-# - api.<base_domain>
-# - app.<base_domain>
-# - flight.<base_domain>
-# - otlp.<base_domain>
-
-# TODO(Kiko): Enforce type checking, remove all type ignores
-
-# TODO(Kiko): Go over docstrings
-# TODO(Kiko): Missing parameter descriptions in some docstrings
-# TODO(Kiko): Missing return descriptions in some docstrings
-
 # TODO(Kiko): Go over headers on each logging call
+# TODO(Kiko): InvalidAdditionalHeadersError is unused. Have we handled extra headers?
 
 # TODO(Kiko): Need to implement 'Update existing examples in a dataset'
 
-# TODO(Kiko): why logs don't show on scripts, only on jupyter notebooks
-# TODO(Kiko): test caching in colab environment
 # TODO(Kiko): Protobuf versioning is too old
 # TODO(Kiko): Go through main APIs and add CtxAdapter where missing
 # TODO(Kiko): Search and handle other TODOs
-# TODO(Kiko): Go over **every file** and do not import anything at runtime, use `if TYPE_CHECKING`
-# with `from __future__ import annotations` (must include for Python < 3.11)
 
 
 class ArizeClient(LazySubclientsMixin):
     """Root client for the Arize SDK.
 
     The ArizeClient provides access to all Arize platform services including datasets,
-    experiments, ML models, projects, and spans. It uses SDKConfiguration internally to
-    manage configuration settings.
+    experiments, ML models, projects, and spans. It uses :class:`arize.config.SDKConfiguration`
+    internally to manage configuration settings.
 
     All parameters are optional (except api_key which must be provided via argument
     or environment variable). For each parameter, values are resolved in this order:
@@ -139,6 +119,7 @@ class ArizeClient(LazySubclientsMixin):
         enable_caching: bool | None = None,
         single_host: str | None = None,
         single_port: int | None = None,
+        base_domain: str | None = None,
     ) -> None:
         """Initialize the Arize client with configuration parameters.
 
@@ -147,44 +128,77 @@ class ArizeClient(LazySubclientsMixin):
 
         Args:
             api_key: Arize API key for authentication. Required - must be provided here
-                or via ARIZE_API_KEY environment variable. Raises MissingAPIKeyError if not set.
+                or via ARIZE_API_KEY environment variable.
+                Raises MissingAPIKeyError if not set.
             region: Arize region (e.g., Region.US_CENTRAL, Region.EU_WEST). When specified,
-                overrides individual host/port settings. ENV: ARIZE_REGION. Default: Region.UNSPECIFIED.
-            api_host: Custom API endpoint host. ENV: ARIZE_API_HOST. Default: "api.arize.com".
-            api_scheme: API endpoint scheme (http/https). ENV: ARIZE_API_SCHEME. Default: "https".
-            otlp_host: OTLP endpoint host. ENV: ARIZE_OTLP_HOST. Default: "otlp.arize.com".
-            otlp_scheme: OTLP endpoint scheme (http/https). ENV: ARIZE_OTLP_SCHEME. Default: "https".
-            flight_host: Apache Arrow Flight endpoint host. ENV: ARIZE_FLIGHT_HOST.
+                overrides individual host/port settings.
+                ENV: ARIZE_REGION.
+                Default: Region.UNSET.
+            api_host: Custom API endpoint host.
+                ENV: ARIZE_API_HOST.
+                Default: "api.arize.com".
+            api_scheme: API endpoint scheme (http/https).
+                ENV: ARIZE_API_SCHEME.
+                Default: "https".
+            otlp_host: OTLP endpoint host.
+                ENV: ARIZE_OTLP_HOST.
+                Default: "otlp.arize.com".
+            otlp_scheme: OTLP endpoint scheme (http/https).
+                ENV: ARIZE_OTLP_SCHEME.
+                Default: "https".
+            flight_host: Apache Arrow Flight endpoint host.
+                ENV: ARIZE_FLIGHT_HOST.
                 Default: "flight.arize.com".
-            flight_port: Apache Arrow Flight endpoint port (1-65535). ENV: ARIZE_FLIGHT_PORT.
+            flight_port: Apache Arrow Flight endpoint port (1-65535).
+                ENV: ARIZE_FLIGHT_PORT.
                 Default: 443.
-            flight_scheme: Apache Arrow Flight endpoint scheme. ENV: ARIZE_FLIGHT_SCHEME.
+            flight_scheme: Apache Arrow Flight endpoint scheme.
+                ENV: ARIZE_FLIGHT_SCHEME.
                 Default: "grpc+tls".
             pyarrow_max_chunksize: Maximum PyArrow chunk size (1 to MAX_CHUNKSIZE).
-                ENV: ARIZE_MAX_CHUNKSIZE. Default: 10_000.
-            request_verify: Whether to verify SSL certificates. ENV: ARIZE_REQUEST_VERIFY.
+                ENV: ARIZE_MAX_CHUNKSIZE.
+                Default: 10_000.
+            request_verify: Whether to verify SSL certificates.
+                ENV: ARIZE_REQUEST_VERIFY.
                 Default: True.
             stream_max_workers: Maximum worker threads for streaming (minimum: 1).
-                ENV: ARIZE_STREAM_MAX_WORKERS. Default: 8.
+                ENV: ARIZE_STREAM_MAX_WORKERS.
+                Default: 8.
             stream_max_queue_bound: Maximum queue size for streaming (minimum: 1).
-                ENV: ARIZE_STREAM_MAX_QUEUE_BOUND. Default: 5000.
+                ENV: ARIZE_STREAM_MAX_QUEUE_BOUND.
+                Default: 5000.
             max_http_payload_size_mb: Maximum HTTP payload size in MB (minimum: 1).
-                ENV: ARIZE_MAX_HTTP_PAYLOAD_SIZE_MB. Default: 100.
+                ENV: ARIZE_MAX_HTTP_PAYLOAD_SIZE_MB.
+                Default: 100.
             arize_directory: Directory for SDK files (cache, logs, etc.).
-                ENV: ARIZE_DIRECTORY. Default: "~/.arize".
-            enable_caching: Whether to enable local caching. ENV: ARIZE_ENABLE_CACHING.
+                ENV: ARIZE_DIRECTORY.
+                Default: "~/.arize".
+            enable_caching: Whether to enable local caching.
+                ENV: ARIZE_ENABLE_CACHING.
                 Default: True.
-            single_host: Single host for all endpoints (overrides individual hosts).
-                ENV: ARIZE_SINGLE_HOST. Default: None.
-            single_port: Single port for all endpoints (overrides individual ports).
-                ENV: ARIZE_SINGLE_PORT. Default: 0 (not set).
+            single_host: Single host for all endpoints. When specified, overrides
+                individual hosts.
+                ENV: ARIZE_SINGLE_HOST.
+                Default: None.
+            single_port: Single port for all endpoints. When specified, overrides
+                individual ports.
+                ENV: ARIZE_SINGLE_PORT.
+                Default: 0 (not set).
+            base_domain: Base domain for generating endpoint hosts as api.<base_domain>,
+                otlp.<base_domain>, flight.<base_domain>. Intended for Private Connect setups.
+                When specified, overrides individual hosts.
+                ENV: ARIZE_BASE_DOMAIN.
+                Default: None.
 
         Raises:
             MissingAPIKeyError: If api_key is not provided via argument or environment variable.
+            MultipleEndpointOverridesError: If multiple endpoint override options (region,
+                single_host/single_port, base_domain) are provided.
 
         Notes:
             Values provided to this class override environment variables, which in turn
-            override default values. See SDKConfiguration for detailed parameter documentation.
+            override default values. See :class:`arize.config.SDKConfiguration`
+            for detailed parameter documentation.
         """
         cfg_kwargs: dict = {}
         if api_key is not None:
@@ -223,6 +237,8 @@ class ArizeClient(LazySubclientsMixin):
             cfg_kwargs["single_host"] = single_host
         if single_port is not None:
             cfg_kwargs["single_port"] = single_port
+        if base_domain is not None:
+            cfg_kwargs["base_domain"] = base_domain
 
         # Only the explicitly provided fields are passed; the rest use
         # SDKConfiguration's default factories / defaults.
@@ -232,27 +248,27 @@ class ArizeClient(LazySubclientsMixin):
     @property
     def datasets(self) -> DatasetsClient:
         """Access the datasets client for dataset operations (lazy-loaded)."""
-        return self.__getattr__("datasets")
+        return cast("DatasetsClient", self.__getattr__("datasets"))
 
     @property
     def experiments(self) -> ExperimentsClient:
         """Access the experiments client for experiment operations (lazy-loaded)."""
-        return self.__getattr__("experiments")
+        return cast("ExperimentsClient", self.__getattr__("experiments"))
 
     @property
     def ml(self) -> MLModelsClient:
         """Access the ML models client for ML model operations (lazy-loaded)."""
-        return self.__getattr__("ml")
+        return cast("MLModelsClient", self.__getattr__("ml"))
 
     @property
     def projects(self) -> ProjectsClient:
         """Access the projects client for project operations (lazy-loaded)."""
-        return self.__getattr__("projects")
+        return cast("ProjectsClient", self.__getattr__("projects"))
 
     @property
     def spans(self) -> SpansClient:
         """Access the spans client for tracing and span operations (lazy-loaded)."""
-        return self.__getattr__("spans")
+        return cast("SpansClient", self.__getattr__("spans"))
 
     def __repr__(self) -> str:
         """Return a string representation of the Arize client configuration."""

arize/config.py

@@ -3,10 +3,8 @@
 import logging
 import os
 import sys
-import threading
 from dataclasses import dataclass, field, fields
 from pathlib import Path
-from typing import Any
 
 from arize.constants.config import (
     DEFAULT_API_HOST,
@@ -27,6 +25,7 @@ from arize.constants.config import (
     ENV_API_KEY,
     ENV_API_SCHEME,
     ENV_ARIZE_DIRECTORY,
+    ENV_BASE_DOMAIN,
     ENV_ENABLE_CACHING,
     ENV_FLIGHT_HOST,
     ENV_FLIGHT_PORT,
@@ -44,6 +43,7 @@ from arize.constants.config import (
 )
 from arize.constants.pyarrow import MAX_CHUNKSIZE
 from arize.exceptions.auth import MissingAPIKeyError
+from arize.exceptions.config import MultipleEndpointOverridesError
 from arize.regions import REGION_ENDPOINTS, Region
 from arize.version import __version__
 
@@ -55,18 +55,44 @@ ALLOWED_HTTP_SCHEMES = {"http", "https"}
 
 
 def _is_sensitive_field(name: str) -> bool:
+    """Check if a field name contains sensitive information markers.
+
+    Args:
+        name: The field name to check.
+
+    Returns:
+        bool: True if the field name contains 'key', 'token', or 'secret' (case-insensitive).
+    """
     n = name.lower()
     return bool(any(k in n for k in SENSITIVE_FIELD_MARKERS))
 
 
 def _mask_secret(secret: str, N: int = 4) -> str:
-    """Show first N chars then '***'; empty string if empty."""
+    """Mask a secret string by showing only the first N characters.
+
+    Args:
+        secret: The secret string to mask.
+        N: Number of characters to show before masking. Defaults to 4.
+
+    Returns:
+        str: The masked string (first N chars + '***'), or empty string if input is empty.
+    """
     if len(secret) == 0:
         return ""
     return f"{secret[:N]}***"
 
 
 def _endpoint(scheme: str, base: str, path: str = "") -> str:
+    """Construct a full endpoint URL from scheme, base, and optional path.
+
+    Args:
+        scheme: The URL scheme (e.g., "http", "https").
+        base: The base URL or hostname.
+        path: Optional path to append to the base URL. Defaults to empty string.
+
+    Returns:
+        str: The fully constructed endpoint URL.
+    """
     endpoint = scheme + "://" + base.rstrip("/")
     if path:
         endpoint += "/" + path.lstrip("/")
@@ -74,6 +100,18 @@ def _endpoint(scheme: str, base: str, path: str = "") -> str:
 
 
 def _env_http_scheme(name: str, default: str) -> str:
+    """Get an HTTP scheme from environment variable with validation.
+
+    Args:
+        name: The environment variable name.
+        default: The default value if the environment variable is not set.
+
+    Returns:
+        str: The validated HTTP scheme ('http' or 'https').
+
+    Raises:
+        ValueError: If the scheme is not 'http' or 'https'.
+    """
     v = _env_str(name, default).lower()
     if v not in ALLOWED_HTTP_SCHEMES:
         raise ValueError(
@@ -88,6 +126,20 @@ def _env_str(
     min_len: int | None = None,
     max_len: int | None = None,
 ) -> str:
+    """Get a string value from environment variable with length validation.
+
+    Args:
+        name: The environment variable name.
+        default: The default value if the environment variable is not set.
+        min_len: Optional minimum length constraint for the string.
+        max_len: Optional maximum length constraint for the string.
+
+    Returns:
+        str: The validated string value (stripped of whitespace).
+
+    Raises:
+        ValueError: If the string length violates min_len or max_len constraints.
+    """
     val = os.getenv(name, default).strip()
 
     if min_len is not None and len(val) < min_len:
@@ -109,6 +161,20 @@ def _env_int(
     min_val: int | None = None,
     max_val: int | None = None,
 ) -> int:
+    """Get an integer value from environment variable with range validation.
+
+    Args:
+        name: The environment variable name.
+        default: The default value if the environment variable is not set.
+        min_val: Optional minimum value constraint for the integer.
+        max_val: Optional maximum value constraint for the integer.
+
+    Returns:
+        int: The validated integer value.
+
+    Raises:
+        ValueError: If the value cannot be parsed as an integer or violates min_val/max_val constraints.
+    """
     raw = os.getenv(name, default)
     try:
         val = int(raw)
@@ -134,6 +200,20 @@ def _env_float(
     min_val: float | None = None,
     max_val: float | None = None,
 ) -> float:
+    """Get a float value from environment variable with range validation.
+
+    Args:
+        name: The environment variable name.
+        default: The default value if the environment variable is not set.
+        min_val: Optional minimum value constraint for the float.
+        max_val: Optional maximum value constraint for the float.
+
+    Returns:
+        float: The validated float value.
+
+    Raises:
+        ValueError: If the value cannot be parsed as a float or violates min_val/max_val constraints.
+    """
     raw = os.getenv(name, default)
     try:
         val = float(raw)
@@ -154,10 +234,28 @@ def _env_float(
 
 
 def _env_bool(name: str, default: bool) -> bool:
+    """Get a boolean value from environment variable.
+
+    Args:
+        name: The environment variable name.
+        default: The default boolean value if the environment variable is not set.
+
+    Returns:
+        bool: The parsed boolean value.
+    """
     return _parse_bool(os.getenv(name, str(default)))
 
 
 def _parse_bool(val: bool | str | None) -> bool:
+    """Parse a boolean value from various input types.
+
+    Args:
+        val: The value to parse. Can be a bool, string, or None.
+
+    Returns:
+        bool: True if the value is already True or matches one of the truthy strings
+            ('1', 'true', 'yes', 'on', case-insensitive). False otherwise.
+    """
     if isinstance(val, bool):
         return val
     return (val or "").strip().lower() in {"1", "true", "yes", "on"}
@@ -167,6 +265,9 @@ def _parse_bool(val: bool | str | None) -> bool:
 class SDKConfiguration:
     """Configuration for the Arize SDK with endpoint and authentication settings.
 
+    This class holds pure configuration data and does not manage client lifecycle.
+    Client creation and caching is handled by :class:`arize.ArizeClient`.
+
     This class is used internally by ArizeClient to manage SDK configuration. It is not
     recommended to use this class directly; users should interact with ArizeClient
     instead.
@@ -225,13 +326,28 @@ class SDKConfiguration:
         region: Arize region (e.g., US_CENTRAL, EU_WEST). When specified, overrides
             individual host/port settings.
             Environment variable: ARIZE_REGION.
-            Default: Region.UNSPECIFIED.
-        single_host: Single host to use for all endpoints. Overrides individual host settings.
+            Default: :class:`Region.UNSET`.
+        single_host: Single host to use for all endpoints. When specified, overrides
+            individual host settings.
             Environment variable: ARIZE_SINGLE_HOST.
             Default: "" (not set).
-        single_port: Single port to use for all endpoints. Overrides individual port settings (0-65535).
+        single_port: Single port to use for all endpoints. When specified, overrides
+            individual port settings (0-65535).
             Environment variable: ARIZE_SINGLE_PORT.
             Default: 0 (not set).
+        base_domain: Base domain for generating all endpoint hosts. Intended for Private Connect
+            setups. When specified, generates hosts as api.<base_domain>, otlp.<base_domain>,
+            flight.<base_domain>. When specified, overrides individual host settings.
+            Environment variable: ARIZE_BASE_DOMAIN.
+            Default: "" (not set).
+
+    Note:
+        The endpoint override options (region, single_host/single_port, base_domain) are
+        mutually exclusive. Specifying more than one will raise MultipleEndpointOverridesError.
+
+    Raises:
+        MissingAPIKeyError: If api_key is not provided via argument or environment variable.
+        MultipleEndpointOverridesError: If multiple endpoint override options are provided.
     """
 
     api_key: str = field(
@@ -322,33 +438,73 @@ class SDKConfiguration:
             ENV_SINGLE_PORT, 0, min_val=0, max_val=65535
         )
     )
-
-    # Private, excluded from comparisons & repr
-    _gen_client: Any = field(default=None, repr=False, compare=False)
-    _gen_lock: threading.Lock = field(
-        default_factory=threading.Lock, repr=False, compare=False
+    base_domain: str = field(
+        default_factory=lambda: _env_str(ENV_BASE_DOMAIN, "")
     )
 
     def __post_init__(self) -> None:
         """Validate and configure SDK endpoints after initialization.
 
+        Endpoint override options are mutually exclusive. Only one of the following
+        can be specified:
+        1. region - Overrides all via REGION_ENDPOINTS mapping
+        2. single_host/single_port - Overrides individual hosts/ports
+        3. base_domain - Generates hosts from base domain
+
+        If none are specified, per-endpoint host/port settings are used.
+
         Raises:
-            MissingAPIKeyError: If API key is not provided.
+            MissingAPIKeyError: If api_key is not provided.
+            MultipleEndpointOverridesError: If multiple endpoint override options are provided.
         """
-        # Validate Configuration
+        # Validate configuration
         if not self.api_key:
             raise MissingAPIKeyError()
 
+        # Check which override options are set
+        has_base_domain = bool(self.base_domain)
         has_single_host = bool(self.single_host)
         has_single_port = self.single_port != 0
-        has_region = self.region is not Region.UNSPECIFIED
-        if (has_single_host or has_single_port) and has_region:
+        has_region = self.region is not Region.UNSET
+
+        # Ensure only one override method is used (mutually exclusive)
+        override_count = sum(
+            [has_base_domain, has_single_host or has_single_port, has_region]
+        )
+        if override_count > 1:
+            # Determine which overrides were provided
+            provided_overrides = []
+            if has_region:
+                provided_overrides.append(f"region={self.region.value}")
+            if has_single_host or has_single_port:
+                if has_single_host:
+                    provided_overrides.append(
+                        f"single_host={self.single_host!r}"
+                    )
+                if has_single_port:
+                    provided_overrides.append(f"single_port={self.single_port}")
+            if has_base_domain:
+                provided_overrides.append(f"base_domain={self.base_domain!r}")
+
+            error_message = (
+                f"Multiple endpoint override options provided: {', '.join(provided_overrides)}. "
+                "Only one of the following can be specified: 'region', "
+                "'single_host'/'single_port', or 'base_domain'."
+            )
+            logger.error(error_message)
+            raise MultipleEndpointOverridesError(error_message)
+
+        if has_base_domain:
             logger.info(
-                "Multiple endpoint override options provided. Preference order is: "
-                "region > single_host/single_port > per-endpoint host/port."
+                "Base domain %r provided; generating hosts from base domain.",
+                self.base_domain,
+            )
+            object.__setattr__(self, "api_host", f"api.{self.base_domain}")
+            object.__setattr__(self, "otlp_host", f"otlp.{self.base_domain}")
+            object.__setattr__(
+                self, "flight_host", f"flight.{self.base_domain}"
             )
 
-        # Single host override: if single_host is set, it overrides hosts
         if has_single_host:
             logger.info(
                 "Single host %r provided; overriding hosts configuration with single host.",
@@ -358,7 +514,6 @@ class SDKConfiguration:
             object.__setattr__(self, "otlp_host", self.single_host)
             object.__setattr__(self, "flight_host", self.single_host)
 
-        # Single port override: if single_port is set, it overrides ports
         if has_single_port:
             logger.info(
                 "Single port %s provided; overriding ports configuration with single port.",
@@ -366,15 +521,12 @@ class SDKConfiguration:
             )
             object.__setattr__(self, "flight_port", self.single_port)
 
-        # Region override: if region is set, it *always* wins over host/port fields
         if has_region:
-            endpoints = REGION_ENDPOINTS[self.region]
-
-            # Override config (region trumps everything)
             logger.info(
                 "Region %s provided; overriding hosts & ports configuration with region defaults.",
                 self.region.value,
             )
+            endpoints = REGION_ENDPOINTS[self.region]
             object.__setattr__(self, "api_host", endpoints.api_host)
             object.__setattr__(self, "otlp_host", endpoints.otlp_host)
             object.__setattr__(self, "flight_host", endpoints.flight_host)
@@ -444,28 +596,3 @@ class SDKConfiguration:
             lines.append(f"  {f.name}={val!r},")
         lines.append(")")
         return "\n".join(lines)
-
-    # TODO(Kiko): This may not be well placed in this class
-    def get_generated_client(self) -> object:
-        """Get or create the generated OpenAPI client instance."""
-        # If already cached, return immediately
-        if self._gen_client is not None:
-            return self._gen_client
-
-        # Thread-safe initialization
-        with self._gen_lock:
-            if self._gen_client is not None:
-                return self._gen_client
-
-            # Import lazily so extra dependencies can be
-            # enforced outside the configuration class
-            from arize._generated import api_client as gen
-
-            cfg = gen.Configuration(host=self.api_url)
-            if self.api_key:
-                cfg.access_token = self.api_key
-            client = gen.ApiClient(cfg)
-
-            # Bypass frozen to set the cache once
-            object.__setattr__(self, "_gen_client", client)
-            return client

arize/constants/config.py

@@ -14,6 +14,7 @@ ENV_FLIGHT_PORT = "ARIZE_FLIGHT_PORT"
 ENV_FLIGHT_SCHEME = "ARIZE_FLIGHT_SCHEME"
 ENV_SINGLE_HOST = "ARIZE_SINGLE_HOST"
 ENV_SINGLE_PORT = "ARIZE_SINGLE_PORT"
+ENV_BASE_DOMAIN = "ARIZE_BASE_DOMAIN"
 ENV_PYARROW_MAX_CHUNKSIZE = "ARIZE_MAX_CHUNKSIZE"
 ENV_REQUEST_VERIFY = "ARIZE_REQUEST_VERIFY"
 ENV_MAX_HTTP_PAYLOAD_SIZE_MB = "ARIZE_MAX_HTTP_PAYLOAD_SIZE_MB"

arize/constants/ml.py

@@ -3,50 +3,43 @@
 import json
 from pathlib import Path
 
-# MAX_BYTES_PER_BULK_RECORD = 100000
-# MAX_DAYS_WITHIN_RANGE = 365
 MIN_PREDICTION_ID_LEN = 1
 MAX_PREDICTION_ID_LEN = 512
 MIN_DOCUMENT_ID_LEN = 1
 MAX_DOCUMENT_ID_LEN = 128
-# # The maximum number of character for tag values
+# The maximum number of character for tag values
 MAX_TAG_LENGTH = 20_000
 MAX_TAG_LENGTH_TRUNCATION = 1_000
-# # The maximum number of character for embedding raw data
+# The maximum number of character for embedding raw data
 MAX_RAW_DATA_CHARACTERS = 2_000_000
 MAX_RAW_DATA_CHARACTERS_TRUNCATION = 5_000
 # The maximum number of acceptable years in the past from current time for prediction_timestamps
 MAX_PAST_YEARS_FROM_CURRENT_TIME = 5
 # The maximum number of acceptable years in the future from current time for prediction_timestamps
 MAX_FUTURE_YEARS_FROM_CURRENT_TIME = 1
-# # The maximum number of character for llm model name
+# The maximum number of character for llm model name
 MAX_LLM_MODEL_NAME_LENGTH = 20_000
 MAX_LLM_MODEL_NAME_LENGTH_TRUNCATION = 50
-# # The maximum number of character for prompt template
+# The maximum number of character for prompt template
 MAX_PROMPT_TEMPLATE_LENGTH = 50_000
 MAX_PROMPT_TEMPLATE_LENGTH_TRUNCATION = 5_000
-# # The maximum number of character for prompt template version
+# The maximum number of character for prompt template version
 MAX_PROMPT_TEMPLATE_VERSION_LENGTH = 20_000
 MAX_PROMPT_TEMPLATE_VERSION_LENGTH_TRUNCATION = 50
-# # The maximum number of embeddings
+# The maximum number of embeddings
 MAX_NUMBER_OF_EMBEDDINGS = 30
 MAX_EMBEDDING_DIMENSIONALITY = 20_000
-# # The maximum number of classes for multi class
+# The maximum number of classes for multi class
 MAX_NUMBER_OF_MULTI_CLASS_CLASSES = 500
 MAX_MULTI_CLASS_NAME_LENGTH = 100
 # The maximum number of references in embedding similarity search params
 MAX_NUMBER_OF_SIMILARITY_REFERENCES = 10
-#
-# # Arize generated columns
-# GENERATED_PREDICTION_LABEL_COL = "arize_generated_prediction_label"
-# GENERATED_LLM_PARAMS_JSON_COL = "arize_generated_llm_params_json"
-#
-# # reserved columns for LLM run metadata
+# reserved columns for LLM run metadata
 LLM_RUN_METADATA_TOTAL_TOKEN_COUNT_TAG_NAME = "total_token_count"  # noqa: S105
 LLM_RUN_METADATA_PROMPT_TOKEN_COUNT_TAG_NAME = "prompt_token_count"  # noqa: S105
 LLM_RUN_METADATA_RESPONSE_TOKEN_COUNT_TAG_NAME = "response_token_count"  # noqa: S105
 LLM_RUN_METADATA_RESPONSE_LATENCY_MS_TAG_NAME = "response_latency_ms"
-#
+
 # all reserved tags
 RESERVED_TAG_COLS = [
     LLM_RUN_METADATA_TOTAL_TOKEN_COUNT_TAG_NAME,

arize/constants/spans.py

@@ -5,19 +5,15 @@ DEFAULT_DATETIME_FMT = "%Y-%m-%dT%H:%M:%S.%f+00:00"
 # Minumum/Maximum number of characters for span/trace/parent ids in spans
 SPAN_ID_MIN_STR_LENGTH = 12
 SPAN_ID_MAX_STR_LENGTH = 128
-# # Minumum/Maximum number of characters for span name
+# Minumum/Maximum number of characters for span name
 SPAN_NAME_MIN_STR_LENGTH = 0
 SPAN_NAME_MAX_STR_LENGTH = 50
-# # Minumum/Maximum number of characters for span status message
+# Minumum/Maximum number of characters for span status message
 SPAN_STATUS_MSG_MIN_STR_LENGTH = 0
 SPAN_STATUS_MSG_MAX_STR_LENGTH = 10_000
-# # Minumum/Maximum number of characters for span event name
+# Minumum/Maximum number of characters for span event name
 SPAN_EVENT_NAME_MAX_STR_LENGTH = 100
-# # Minumum/Maximum number of characters for span event attributes
-# SPAN_EVENT_ATTRS_MAX_STR_LENGTH = 10_000
-# # Maximum number of characters for span kind
-# SPAN_KIND_MAX_STR_LENGTH = 100
-# SPAN_EXCEPTION_TYPE_MAX_STR_LENGTH = 100
+# Minumum/Maximum number of characters for span event attributes
 SPAN_EXCEPTION_MESSAGE_MAX_STR_LENGTH = 100
 SPAN_EXCEPTION_STACK_TRACE_MAX_STR_LENGTH = 10_000
 SPAN_IO_VALUE_MAX_STR_LENGTH = 4_000_000
@@ -29,7 +25,6 @@ SPAN_LLM_MESSAGE_ROLE_MAX_STR_LENGTH = 100
 SPAN_LLM_MESSAGE_CONTENT_MAX_STR_LENGTH = 4_000_000
 SPAN_LLM_TOOL_CALL_FUNCTION_NAME_MAX_STR_LENGTH = 500
 SPAN_LLM_PROMPT_TEMPLATE_MAX_STR_LENGTH = 4_000_000
-# SPAN_LLM_PROMPT_TEMPLATE_VARIABLES_MAX_STR_LENGTH = 10_000
 SPAN_LLM_PROMPT_TEMPLATE_VERSION_MAX_STR_LENGTH = 100
 SPAN_TOOL_NAME_MAX_STR_LENGTH = 100
 SPAN_TOOL_DESCRIPTION_MAX_STR_LENGTH = 1_000
@@ -43,7 +38,7 @@ JSON_STRING_MAX_STR_LENGTH = 4_000_000
 EVAL_LABEL_MIN_STR_LENGTH = 1  # we do not accept empty strings
 EVAL_LABEL_MAX_STR_LENGTH = 100
 EVAL_EXPLANATION_MAX_STR_LENGTH = 10_000
-#
+
 # # Annotation related constants
 ANNOTATION_LABEL_MIN_STR_LENGTH = 1
 ANNOTATION_LABEL_MAX_STR_LENGTH = 100  # Max length for annotation label string