Replace pydantic_plugin in logfire.configure() with `logfire.inst…

…rument_pydantic()` (#453)

docs/integrations/pydantic.md

@@ -1,6 +1,6 @@
 # Pydantic
 
-Logfire has a [Pydantic plugin][pydantic-plugin] to instrument [Pydantic][pydantic] models.
+Logfire has a Pydantic plugin to instrument [Pydantic][pydantic] models.
 The plugin provides logs and metrics about model validation.
 
 To enable the plugin, do one of the following:
@@ -13,35 +13,36 @@ To enable the plugin, do one of the following:
 pydantic_plugin_record = "all"
 ```
 
-- Use the [`pydantic_plugin`][logfire.configure(pydantic_plugin)] parameter in `logfire.configure`, e.g:
+- Call [`logfire.instrument_pydantic`][logfire.Logfire.instrument_pydantic] with the desired configuration, e.g:
 
 ```py
 import logfire
 
-logfire.configure(pydantic_plugin=logfire.PydanticPlugin(record='all'))
+logfire.instrument_pydantic()  # Defaults to record='all'
 ```
 
-Note that if you only use the last option then only models defined and imported *after* calling `logfire.configure`
+Note that if you only use the last option then only model classes defined and imported *after* calling `logfire.instrument_pydantic`
 will be instrumented.
 
+!!! note
+    Remember to call [`logfire.configure()`][logfire.configure] at some point, whether before or after
+    calling `logfire.instrument_pydantic` and defining model classes.
+    Model validations will only start being logged after calling `logfire.configure()`.
+
 ## Third party modules
 
 By default, third party modules are not instrumented by the plugin to avoid noise. You can enable instrumentation for those
 using the [`include`][logfire.PydanticPlugin.include] configuration.
 
 ```py
-import logfire
-
-logfire.configure(pydantic_plugin=logfire.PydanticPlugin(record='all', include={'openai'}))
+logfire.instrument_pydantic(include={'openai'})
 ```
 
 You can also disable instrumentation for your own modules using the
 [`exclude`][logfire.PydanticPlugin.exclude] configuration.
 
 ```py
-import logfire
-
-logfire.configure(pydantic_plugin=logfire.PydanticPlugin(record='all', exclude={'app.api.v1'}))
+logfire.instrument_pydantic(exclude={'app.api.v1'})
 ```
 
 ## Model configuration
@@ -60,13 +61,13 @@ class Foo(BaseModel, plugin_settings=PluginSettings(logfire={'record': 'failure'
 
 ### Record
 
-The [`record`][logfire.integrations.pydantic.LogfireSettings.record] is used to configure what to record.
+The [`record`][logfire.integrations.pydantic.LogfireSettings.record] argument is used to configure what to record.
 It can be one of the following values:
 
-  * `off`: Disable instrumentation. This is default value.
-  * `all`: Send traces and metrics for all events.
+  * `all`: Send traces and metrics for all events. This is default value for `logfire.instrument_pydantic`.
   * `failure`: Send metrics for all validations and traces only for validation failures.
   * `metrics`: Send only metrics.
+  * `off`: Disable instrumentation.
 
 <!--
 [Sampling](../usage/sampling.md) can be configured by `trace_sample_rate` key in
@@ -98,4 +99,3 @@ class Foo(
 ```
 
 [pydantic]: https://docs.pydantic.dev/latest/
-[pydantic-plugin]: https://docs.pydantic.dev/latest/concepts/plugins/

docs/integrations/third-party/mirascope.md

@@ -32,7 +32,7 @@ This will give you:
   <figcaption>Mirascope Anthropic call span and Anthropic span and conversation</figcaption>
 </figure>
 
-Since Mirascope is built on top of [Pydantic][pydantic], you can use the [Pydantic plugin][pydantic-plugin] to track additional logs and metrics about model validation, which you can enable using the [`pydantic_plugin`][logfire.configure(pydantic_plugin)] configuration.
+Since Mirascope is built on top of [Pydantic][pydantic], you can use the [Pydantic plugin](../pydantic.md) to track additional logs and metrics about model validation.
 
 This can be particularly useful when [extracting structured information][mirascope-extracting-structured-information] using LLMs:
 
@@ -44,7 +44,8 @@ from mirascope.core import openai, prompt_template
 from mirascope.integrations.logfire import with_logfire
 from pydantic import BaseModel
 
-logfire.configure(pydantic_plugin=logfire.PydanticPlugin(record="all"))
+logfire.configure()
+logfire.instrument_pydantic()
 
 
 class TaskDetails(BaseModel):

docs/why-logfire/pydantic.md

@@ -31,7 +31,8 @@ from datetime import date
 import logfire
 from pydantic import BaseModel
 
-logfire.configure(pydantic_plugin=logfire.PydanticPlugin(record='all'))  # (1)!
+logfire.configure()
+logfire.instrument_pydantic()  # (1)!
 
 class User(BaseModel):
     name: str
@@ -43,7 +44,7 @@ User(name='Ben', country_code='USA', dob='2000-02-02')
 User(name='Charlie', country_code='GBR', dob='1990-03-03')
 ```
 
-1. This configuration means details about all Pydantic model validations will be recorded. You can also record details about validation failures only, or just metrics; see the [pydantic plugin docs][logfire.PydanticPlugin].
+1. This configuration means details about all Pydantic model validations will be recorded. You can also record details about validation failures only, or just metrics; see the [pydantic plugin docs](../integrations/pydantic.md).
 2. Since we've enabled the Pydantic Plugin, all Pydantic validations will be recorded in Logfire.
 
 Learn more about the [Pydantic Plugin here](../integrations/pydantic.md).

logfire-api/logfire_api/__init__.py

@@ -95,6 +95,8 @@ def decorator(func):
             def instrument_fastapi(self, *args, **kwargs) -> ContextManager[None]:
                 return nullcontext()
 
+            def instrument_pydantic(self, *args, **kwargs) -> None: ...
+
             def instrument_pymongo(self, *args, **kwargs) -> None: ...
 
             def instrument_sqlalchemy(self, *args, **kwargs) -> None: ...
@@ -144,6 +146,7 @@ def shutdown(self, *args, **kwargs) -> None: ...
         log_slow_async_callbacks = DEFAULT_LOGFIRE_INSTANCE.log_slow_async_callbacks
         install_auto_tracing = DEFAULT_LOGFIRE_INSTANCE.install_auto_tracing
         instrument = DEFAULT_LOGFIRE_INSTANCE.instrument
+        instrument_pydantic = DEFAULT_LOGFIRE_INSTANCE.instrument_pydantic
         instrument_fastapi = DEFAULT_LOGFIRE_INSTANCE.instrument_fastapi
         instrument_openai = DEFAULT_LOGFIRE_INSTANCE.instrument_openai
         instrument_anthropic = DEFAULT_LOGFIRE_INSTANCE.instrument_anthropic

logfire-api/logfire_api/__init__.pyi

@@ -11,14 +11,15 @@ from .integrations.structlog import LogfireProcessor as StructlogProcessor
 from .version import VERSION as VERSION
 from logfire.sampling import SamplingOptions as SamplingOptions
 
-__all__ = ['Logfire', 'LogfireSpan', 'LevelName', 'AdvancedOptions', 'ConsoleOptions', 'PydanticPlugin', 'configure', 'span', 'instrument', 'log', 'trace', 'debug', 'notice', 'info', 'warn', 'error', 'exception', 'fatal', 'force_flush', 'log_slow_async_callbacks', 'install_auto_tracing', 'instrument_fastapi', 'instrument_openai', 'instrument_anthropic', 'instrument_asyncpg', 'instrument_httpx', 'instrument_celery', 'instrument_requests', 'instrument_psycopg', 'instrument_django', 'instrument_flask', 'instrument_starlette', 'instrument_aiohttp_client', 'instrument_sqlalchemy', 'instrument_redis', 'instrument_pymongo', 'instrument_mysql', 'instrument_system_metrics', 'AutoTraceModule', 'with_tags', 'with_settings', 'shutdown', 'load_spans_from_file', 'no_auto_trace', 'ScrubMatch', 'ScrubbingOptions', 'VERSION', 'suppress_instrumentation', 'StructlogProcessor', 'LogfireLoggingHandler', 'SamplingOptions', 'MetricsOptions']
+__all__ = ['Logfire', 'LogfireSpan', 'LevelName', 'AdvancedOptions', 'ConsoleOptions', 'PydanticPlugin', 'configure', 'span', 'instrument', 'log', 'trace', 'debug', 'notice', 'info', 'warn', 'error', 'exception', 'fatal', 'force_flush', 'log_slow_async_callbacks', 'install_auto_tracing', 'instrument_pydantic', 'instrument_fastapi', 'instrument_openai', 'instrument_anthropic', 'instrument_asyncpg', 'instrument_httpx', 'instrument_celery', 'instrument_requests', 'instrument_psycopg', 'instrument_django', 'instrument_flask', 'instrument_starlette', 'instrument_aiohttp_client', 'instrument_sqlalchemy', 'instrument_redis', 'instrument_pymongo', 'instrument_mysql', 'instrument_system_metrics', 'AutoTraceModule', 'with_tags', 'with_settings', 'shutdown', 'load_spans_from_file', 'no_auto_trace', 'ScrubMatch', 'ScrubbingOptions', 'VERSION', 'suppress_instrumentation', 'StructlogProcessor', 'LogfireLoggingHandler', 'SamplingOptions', 'MetricsOptions']
 
 DEFAULT_LOGFIRE_INSTANCE = Logfire()
 span = DEFAULT_LOGFIRE_INSTANCE.span
 instrument = DEFAULT_LOGFIRE_INSTANCE.instrument
 force_flush = DEFAULT_LOGFIRE_INSTANCE.force_flush
 log_slow_async_callbacks = DEFAULT_LOGFIRE_INSTANCE.log_slow_async_callbacks
 install_auto_tracing = DEFAULT_LOGFIRE_INSTANCE.install_auto_tracing
+instrument_pydantic = DEFAULT_LOGFIRE_INSTANCE.instrument_pydantic
 instrument_fastapi = DEFAULT_LOGFIRE_INSTANCE.instrument_fastapi
 instrument_openai = DEFAULT_LOGFIRE_INSTANCE.instrument_openai
 instrument_anthropic = DEFAULT_LOGFIRE_INSTANCE.instrument_anthropic

logfire-api/logfire_api/_internal/config.pyi

@@ -17,7 +17,7 @@ from .metrics import ProxyMeterProvider as ProxyMeterProvider
 from .scrubbing import BaseScrubber as BaseScrubber, NOOP_SCRUBBER as NOOP_SCRUBBER, Scrubber as Scrubber, ScrubbingOptions as ScrubbingOptions
 from .stack_info import warn_at_user_stacklevel as warn_at_user_stacklevel
 from .tracer import PendingSpanProcessor as PendingSpanProcessor, ProxyTracerProvider as ProxyTracerProvider
-from .utils import UnexpectedResponse as UnexpectedResponse, ensure_data_dir_exists as ensure_data_dir_exists, get_version as get_version, read_toml_file as read_toml_file, suppress_instrumentation as suppress_instrumentation
+from .utils import SeededRandomIdGenerator as SeededRandomIdGenerator, UnexpectedResponse as UnexpectedResponse, ensure_data_dir_exists as ensure_data_dir_exists, read_toml_file as read_toml_file, suppress_instrumentation as suppress_instrumentation
 from _typeshed import Incomplete
 from dataclasses import dataclass
 from functools import cached_property
@@ -59,7 +59,10 @@ class AdvancedOptions:
 
 @dataclass
 class PydanticPlugin:
-    """Options for the Pydantic plugin."""
+    """Options for the Pydantic plugin.
+
+    This class is deprecated for external use. Use `logfire.instrument_pydantic()` instead.
+    """
     record: PydanticPluginRecordValues = ...
     include: set[str] = ...
     exclude: set[str] = ...
@@ -74,7 +77,7 @@ class MetricsOptions:
 
 class DeprecatedKwargs(TypedDict): ...
 
-def configure(*, send_to_logfire: bool | Literal['if-token-present'] | None = None, token: str | None = None, service_name: str | None = None, service_version: str | None = None, console: ConsoleOptions | Literal[False] | None = None, config_dir: Path | str | None = None, data_dir: Path | str | None = None, additional_span_processors: Sequence[SpanProcessor] | None = None, metrics: MetricsOptions | Literal[False] | None = None, pydantic_plugin: PydanticPlugin | None = None, scrubbing: ScrubbingOptions | Literal[False] | None = None, inspect_arguments: bool | None = None, sampling: SamplingOptions | None = None, advanced: AdvancedOptions | None = None, **deprecated_kwargs: Unpack[DeprecatedKwargs]) -> None:
+def configure(*, send_to_logfire: bool | Literal['if-token-present'] | None = None, token: str | None = None, service_name: str | None = None, service_version: str | None = None, console: ConsoleOptions | Literal[False] | None = None, config_dir: Path | str | None = None, data_dir: Path | str | None = None, additional_span_processors: Sequence[SpanProcessor] | None = None, metrics: MetricsOptions | Literal[False] | None = None, scrubbing: ScrubbingOptions | Literal[False] | None = None, inspect_arguments: bool | None = None, sampling: SamplingOptions | None = None, advanced: AdvancedOptions | None = None, **deprecated_kwargs: Unpack[DeprecatedKwargs]) -> None:
     """Configure the logfire SDK.
 
     Args:
@@ -94,8 +97,6 @@ def configure(*, send_to_logfire: bool | Literal['if-token-present'] | None = No
         additional_span_processors: Span processors to use in addition to the default processor which exports spans to Logfire's API.
         metrics: Set to `False` to disable sending all metrics,
             or provide a `MetricsOptions` object to configure metrics, e.g. additional metric readers.
-        pydantic_plugin: Configuration for the Pydantic plugin. If `None` uses the `LOGFIRE_PYDANTIC_PLUGIN_*` environment
-            variables, otherwise defaults to `PydanticPlugin(record='off')`.
         scrubbing: Options for scrubbing sensitive data. Set to `False` to disable.
         inspect_arguments: Whether to enable
             [f-string magic](https://logfire.pydantic.dev/docs/guides/onboarding-checklist/add-manual-tracing/#f-strings).
@@ -123,21 +124,20 @@ class _LogfireConfigData:
     console: ConsoleOptions | Literal[False] | None
     data_dir: Path
     additional_span_processors: Sequence[SpanProcessor] | None
-    pydantic_plugin: PydanticPlugin
     scrubbing: ScrubbingOptions | Literal[False]
     inspect_arguments: bool
     sampling: SamplingOptions
     advanced: AdvancedOptions
 
 class LogfireConfig(_LogfireConfigData):
-    def __init__(self, send_to_logfire: bool | None = None, token: str | None = None, service_name: str | None = None, service_version: str | None = None, console: ConsoleOptions | Literal[False] | None = None, config_dir: Path | None = None, data_dir: Path | None = None, additional_span_processors: Sequence[SpanProcessor] | None = None, metrics: MetricsOptions | Literal[False] | None = None, pydantic_plugin: PydanticPlugin | None = None, scrubbing: ScrubbingOptions | Literal[False] | None = None, inspect_arguments: bool | None = None, sampling: SamplingOptions | None = None, advanced: AdvancedOptions | None = None) -> None:
+    def __init__(self, send_to_logfire: bool | None = None, token: str | None = None, service_name: str | None = None, service_version: str | None = None, console: ConsoleOptions | Literal[False] | None = None, config_dir: Path | None = None, data_dir: Path | None = None, additional_span_processors: Sequence[SpanProcessor] | None = None, metrics: MetricsOptions | Literal[False] | None = None, scrubbing: ScrubbingOptions | Literal[False] | None = None, inspect_arguments: bool | None = None, sampling: SamplingOptions | None = None, advanced: AdvancedOptions | None = None) -> None:
         """Create a new LogfireConfig.
 
         Users should never need to call this directly, instead use `logfire.configure`.
 
         See `_LogfireConfigData` for parameter documentation.
         """
-    def configure(self, send_to_logfire: bool | Literal['if-token-present'] | None, token: str | None, service_name: str | None, service_version: str | None, console: ConsoleOptions | Literal[False] | None, config_dir: Path | None, data_dir: Path | None, additional_span_processors: Sequence[SpanProcessor] | None, metrics: MetricsOptions | Literal[False] | None, pydantic_plugin: PydanticPlugin | None, scrubbing: ScrubbingOptions | Literal[False] | None, inspect_arguments: bool | None, sampling: SamplingOptions | None, advanced: AdvancedOptions | None) -> None: ...
+    def configure(self, send_to_logfire: bool | Literal['if-token-present'] | None, token: str | None, service_name: str | None, service_version: str | None, console: ConsoleOptions | Literal[False] | None, config_dir: Path | None, data_dir: Path | None, additional_span_processors: Sequence[SpanProcessor] | None, metrics: MetricsOptions | Literal[False] | None, scrubbing: ScrubbingOptions | Literal[False] | None, inspect_arguments: bool | None, sampling: SamplingOptions | None, advanced: AdvancedOptions | None) -> None: ...
     def initialize(self) -> ProxyTracerProvider:
         """Configure internals to start exporting traces and metrics."""
     def force_flush(self, timeout_millis: int = 30000) -> bool:

logfire-api/logfire_api/_internal/config_params.pyi

@@ -75,5 +75,3 @@ class ParamManager:
         """
     @cached_property
     def pydantic_plugin(self): ...
-
-def default_param_manager(): ...

logfire-api/logfire_api/_internal/main.pyi

@@ -5,6 +5,7 @@ from . import async_ as async_
 from ..version import VERSION as VERSION
 from .auto_trace import AutoTraceModule as AutoTraceModule, install_auto_tracing as install_auto_tracing
 from .config import GLOBAL_CONFIG as GLOBAL_CONFIG, LogfireConfig as LogfireConfig, OPEN_SPANS as OPEN_SPANS
+from .config_params import PydanticPluginRecordValues as PydanticPluginRecordValues
 from .constants import ATTRIBUTES_JSON_SCHEMA_KEY as ATTRIBUTES_JSON_SCHEMA_KEY, ATTRIBUTES_LOG_LEVEL_NUM_KEY as ATTRIBUTES_LOG_LEVEL_NUM_KEY, ATTRIBUTES_MESSAGE_KEY as ATTRIBUTES_MESSAGE_KEY, ATTRIBUTES_MESSAGE_TEMPLATE_KEY as ATTRIBUTES_MESSAGE_TEMPLATE_KEY, ATTRIBUTES_SAMPLE_RATE_KEY as ATTRIBUTES_SAMPLE_RATE_KEY, ATTRIBUTES_SPAN_TYPE_KEY as ATTRIBUTES_SPAN_TYPE_KEY, ATTRIBUTES_TAGS_KEY as ATTRIBUTES_TAGS_KEY, ATTRIBUTES_VALIDATION_ERROR_KEY as ATTRIBUTES_VALIDATION_ERROR_KEY, DISABLE_CONSOLE_KEY as DISABLE_CONSOLE_KEY, LEVEL_NUMBERS as LEVEL_NUMBERS, LevelName as LevelName, NULL_ARGS_KEY as NULL_ARGS_KEY, OTLP_MAX_INT_SIZE as OTLP_MAX_INT_SIZE, log_level_attributes as log_level_attributes
 from .formatter import logfire_format as logfire_format, logfire_format_with_magic as logfire_format_with_magic
 from .instrument import LogfireArgs as LogfireArgs, instrument as instrument
@@ -24,7 +25,7 @@ from .json_schema import JsonSchemaProperties as JsonSchemaProperties, attribute
 from .metrics import ProxyMeterProvider as ProxyMeterProvider
 from .stack_info import get_user_stack_info as get_user_stack_info
 from .tracer import ProxyTracerProvider as ProxyTracerProvider
-from .utils import SysExcInfo as SysExcInfo, handle_internal_errors as handle_internal_errors, log_internal_error as log_internal_error, uniquify_sequence as uniquify_sequence
+from .utils import SysExcInfo as SysExcInfo, get_version as get_version, handle_internal_errors as handle_internal_errors, log_internal_error as log_internal_error, uniquify_sequence as uniquify_sequence
 from django.http import HttpRequest as HttpRequest, HttpResponse as HttpResponse
 from fastapi import FastAPI
 from flask.app import Flask
@@ -367,6 +368,24 @@ class Logfire:
                 modules in `sys.modules` (i.e. modules that have already been imported) match the modules to trace.
                 Set to `'warn'` to issue a warning instead, or `'ignore'` to skip the check.
         """
+    def instrument_pydantic(self, record: PydanticPluginRecordValues = 'all', include: Iterable[str] = (), exclude: Iterable[str] = ()):
+        """Instrument Pydantic model validations.
+
+        This must be called before defining and importing the model classes you want to instrument.
+        See the [Pydantic integration guide](https://logfire.pydantic.dev/docs/integrations/pydantic/) for more info.
+
+        Args:
+            record: The record mode for the Pydantic plugin. It can be one of the following values:
+
+                - `all`: Send traces and metrics for all events. This is default value.
+                - `failure`: Send metrics for all validations and traces only for validation failures.
+                - `metrics`: Send only metrics.
+                - `off`: Disable instrumentation.
+            include:
+                By default, third party modules are not instrumented. This option allows you to include specific modules.
+            exclude:
+                Exclude specific modules from instrumentation.
+        """
     def instrument_fastapi(self, app: FastAPI, *, capture_headers: bool = False, request_attributes_mapper: Callable[[Request | WebSocket, dict[str, Any]], dict[str, Any] | None] | None = None, use_opentelemetry_instrumentation: bool = True, excluded_urls: str | Iterable[str] | None = None, record_send_receive: bool = False, **opentelemetry_kwargs: Any) -> ContextManager[None]:
         """Instrument a FastAPI app so that spans and logs are automatically created for each request.