CodeWithNK nadeem4

Feature	Normal Agent	AI Agent
Learning	No learning or adaptation	Learns and improves over time
Decision	Follows explicit "if-then" rules	Reasons, adapts, makes smart choices
Environment	Static, predictable	Dynamic, unpredictable
Complexity	Simple, repetitive tasks	Complex, adaptive challenges
Cost	Generally lower development costs	Higher initial development costs

	To Override This host.json Setting	Create This Environment Variable
	"functionTimeout": "00:10:00"	AzureFunctionsJobHost__functionTimeout = 00:05:00
	"default": "Warning" (under logging.logLevel)	AzureFunctionsJobHost__logging__logLevel__default = Information

Plan Type	Startup Mechanism	Key Technical Detail	Performance Implication
Consumption	Event-driven via Scale Controller	Host is instantiated on-demand.	Potential for cold-start latency
Premium	Event-driven + Always Ready	Always-ready and pre-warmed instances eliminate cold starts; supports VNet and long-running executions	Instant response with no cold starts; predictable baseline cost
Dedicated (App Service)	Always On	Host process continuously active; manual or rule-based scaling	No cold starts; scale based on App Service plan configuration

Characteristic	In-Process Model (Old)	Out-of-Process Model (Modern)
Process Boundary	Shared single process	Separate, isolated processes
Fault Tolerance	Low; a code crash can kill the Host	High; the Host is protected from code crashes
Dependency Mgmt.	Tightly coupled versions	Decoupled; use your desired language version
Communication	Direct in-memory calls (fast)	Inter-process communication (slightly more overhead)

	host.json Path	App Setting Key
	logging.logLevel.default	AzureFunctionsJobHost__logging__logLevel__default
	logging.logLevel.Function.MyFunction.User	AzureFunctionsJobHost__logging__logLevel__Function__MyFunction__User

	Method	Best For
	basicConfig()	Small, single-file scripts and quick debugging sessions.
	Programmatic	Highly dynamic setups where logging behavior is determined by code at runtime.
	fileConfig()	Maintaining legacy applications; dictConfig is the modern standard for new projects.
	dictConfig()	All modern libraries, microservices, and applications requiring scalable, environment-aware logging.

	import logging
	import logging.config
	import os

	LOG_DIR = 'app_logs'
	if not os.path.exists(LOG_DIR):
	os.makedirs(LOG_DIR)

	class KeywordFilter(logging.Filter):
	def __init__(self, keyword_to_allow=None, case_sensitive=True):

	import logging
	import logging.config
	import os

	LOG_DIR = 'app_logs'
	if not os.path.exists(LOG_DIR):
	os.makedirs(LOG_DIR)

	class APIMonitoringHandler(logging.Handler):
	def __init__(self, api_key, endpoint_url, failure_log_file):

	"Feature","LoggerAdapter","logging.Filter"
	"Primary Purpose","To provide a context-aware interface to a logger, automatically injecting predefined contextual information.","To exert fine-grained control over log records after creation but before handler processing; can add/modify data or decide if a record is processed."
	"When it Acts","Before the log record is fully created by the underlying logger (its `process()` method is called first).","After a LogRecord is created by a logger, but before it is processed by a handler."
	"How Context is Added","Typically injects context from its `extra` attribute (set at instantiation) into log calls made through it. Often instance-specific.","Can add context by modifying the LogRecord directly (e.g., from `contextvars`, global settings, or record attributes)."
	"Modification Capabilities","Primarily adds data to the `extra` dictionary. Can modify message/kwargs if its `process()` method is overridden.","Can modify any attribute of the LogRecord (e.g., `msg`, `levelname`,