feat: sampling feature for logger

.gitignore

@@ -291,5 +291,5 @@ $RECYCLE.BIN/
 # End of https://www.gitignore.io/api/osx,linux,python,windows,pycharm,visualstudiocode
 
 # Misc
-
-test_report
+test_report
+/.idea/*

python/aws_lambda_powertools/logging/logger.py

@@ -2,19 +2,22 @@
 import itertools
 import logging
 import os
+import random
 from distutils.util import strtobool
 from typing import Any, Callable, Dict
 
-from . import aws_lambda_logging
 from ..helper.models import MetricUnit, build_lambda_context_model, build_metric_unit_from_str
+from . import aws_lambda_logging
 
 logger = logging.getLogger(__name__)
 logger.setLevel(os.getenv("LOG_LEVEL", "INFO"))
 
 is_cold_start = True
 
 
-def logger_setup(service: str = "service_undefined", level: str = "INFO", **kwargs):
+def logger_setup(
+    service: str = "service_undefined", level: str = "INFO", sampling_rate: float = 0.0, **kwargs
+):
     """Setups root logger to format statements in JSON.
 
     Includes service name and any additional key=value into logs
@@ -26,13 +29,17 @@ def logger_setup(service: str = "service_undefined", level: str = "INFO", **kwar
         service name
     LOG_LEVEL: str
         logging level (e.g. INFO, DEBUG)
+    POWERTOOLS_LOGGER_SAMPLE_RATE: float
+        samping rate ranging from 0 to 1, 1 being 100% sampling
 
     Parameters
     ----------
     service : str, optional
         service name to be appended in logs, by default "service_undefined"
     level : str, optional
         logging.level, by default "INFO"
+    sample_rate: float, optional
+        sample rate for debug calls within execution context defaults to 0
 
     Example
     -------
@@ -47,6 +54,7 @@ def logger_setup(service: str = "service_undefined", level: str = "INFO", **kwar
     Setups structured logging in JSON for Lambda functions using env vars
 
         $ export POWERTOOLS_SERVICE_NAME="payment"
+        $ export POWERTOOLS_LOGGER_SAMPLE_RATE=0.01 # 1% debug sampling
         >>> from aws_lambda_powertools.logging import logger_setup
         >>> logger = logger_setup()
         >>>
@@ -55,12 +63,24 @@ def logger_setup(service: str = "service_undefined", level: str = "INFO", **kwar
 
     """
     service = os.getenv("POWERTOOLS_SERVICE_NAME") or service
+    sampling_rate = os.getenv("POWERTOOLS_LOGGER_SAMPLE_RATE") or sampling_rate
     log_level = os.getenv("LOG_LEVEL") or level
     logger = logging.getLogger(name=service)
+
+    try:
+        if sampling_rate and random.random() <= float(sampling_rate):
+            log_level = logging.DEBUG
+    except ValueError:
+        raise ValueError(
+            f"Expected a float value ranging 0 to 1, but received {sampling_rate} instead. Please review POWERTOOLS_LOGGER_SAMPLE_RATE environment variable."
+        )
+
     logger.setLevel(log_level)
 
     # Patch logger by structuring its outputs as JSON
-    aws_lambda_logging.setup(level=log_level, service=service, **kwargs)
+    aws_lambda_logging.setup(
+        level=log_level, service=service, sampling_rate=sampling_rate, **kwargs
+    )
 
     return logger
 

python/tests/functional/test_aws_lambda_logging.py

@@ -112,4 +112,4 @@ class X:
 
     log_dict = json.loads(stdout.getvalue())
 
-    assert log_dict["message"]["x"].startswith("<")
+    assert log_dict["message"]["x"].startswith("<")

python/tests/functional/test_logger.py

@@ -80,6 +80,24 @@ def test_setup_service_env_var(monkeypatch, root_logger, stdout):
     assert service_name == log["service"]
 
 
+def test_setup_sampling_rate(monkeypatch, root_logger, stdout):
+    # GIVEN samping rate is explicitly defined via POWERTOOLS_LOGGER_SAMPLE_RATE env
+    # WHEN logger is setup
+    # THEN sampling rate should be equals POWERTOOLS_LOGGER_SAMPLE_RATE value and should sample debug logs
+
+    sampling_rate = "1"
+    monkeypatch.setenv("POWERTOOLS_LOGGER_SAMPLE_RATE", sampling_rate)
+    monkeypatch.setenv("LOG_LEVEL", "INFO")
+
+    logger = logger_setup()
+    logger.debug("I am being sampled")
+    log = json.loads(stdout.getvalue())
+
+    assert sampling_rate == log["sampling_rate"]
+    assert "DEBUG" == log["level"]
+    assert "I am being sampled" == log["message"]
+
+
 def test_inject_lambda_context(root_logger, stdout, lambda_context):
     # GIVEN a lambda function is decorated with logger
     # WHEN logger is setup
@@ -130,7 +148,6 @@ def handler(event, context):
 def test_inject_lambda_context_log_event_request_env_var(
     monkeypatch, root_logger, stdout, lambda_context
 ):
-
     # GIVEN a lambda function is decorated with logger instructed to log event
     # via POWERTOOLS_LOGGER_LOG_EVENT env
     # WHEN logger is setup