Development of a Musical Discord Bot: Functional Architecture and Log-Driven Development

@startuml
!theme plain

title Functional Data Flow
participant "Discord Command" as DC
participant "Validator" as V
participant "Business Logic" as BL
participant "API Client" as AC
participant "Logger" as L

DC -> V: Raw Input
activate V
V -> V: Pydantic Validation
V -> L: Log Validation
V --> DC: Maybe[ValidData]
deactivate V

DC -> BL: ValidData
activate BL
BL -> BL: Pure Computation
BL -> L: Log Business Logic
BL -> AC: API Request
activate AC
AC -> AC: IO Operation
AC -> L: Log API Call
AC --> BL: Maybe[Result]
deactivate AC
BL --> DC: Either[Error, Success]
deactivate BL

DC -> L: Log Final Result

@enduml

@startuml
!theme spacelab

title Log-driven development flow
start

:Define Expected Behavior;
note right: Spécification des logs attendus

:Write Log Assertions;
note right: Tests basés sur les logs

:Implement Minimal Code;
note right: Code juste suffisant

:Run & Observe Logs;
note right: Validation comportementale

if (Logs Match Expectations?) then (yes)
  :Refactor & Optimize;
  note right: Amélioration continue
else (no)
  :Debug via Logs;
  note right: Analyse des écarts
  :Fix Implementation;
endif

:Integration Tests;
note right: Validation end-to-end

stop

@enduml

INFO: spotify.search.start query="bohemian rhapsody" user_id=123456
DEBUG: spotify.search.validation query_length=16 safe_chars=true
DEBUG: spotify.search.api_call endpoint="/search" params={...}
INFO: spotify.search.success results_count=15 duration_ms=340

@startuml
!theme cerulean-outline

class SpotifyTrack {
  +id: str
  +name: str
  +artists: List[str]
  +duration_ms: int
  +external_urls: Dict[str, str]
  --
  +validate_duration() : bool
  +to_discord_embed() : Embed
}

class YouTubeVideo {
  +id: str
  +title: str
  +duration: timedelta
  +available: bool
  --
  +validate_availability() : bool
  +extract_audio_url() : Optional[str]
}

class DiscordCommand {
  +command: str
  +args: List[str]
  +user: User
  +channel: Channel
  --
  +validate_permissions() : bool
  +log_execution() : None
}

SpotifyTrack --|> BaseModel
YouTubeVideo --|> BaseModel
DiscordCommand --|> BaseModel

@enduml

@startuml
!theme toy

title Error Handling Flow

participant "Order" as C
participant "Maybe Monad" as M
participant "Either Monad" as E
participant "Logger" as L

C -> M: search_query
activate M

alt Valid Query
  M -> E: Success(query)
  activate E
  E -> E: api_call()

  alt API Success
    E -> L: log_success()
    E --> C: Right(result)
  else API Error
    E -> L: log_api_error()
    E --> C: Left(api_error)
  end
  deactivate E

else Invalid Query
  M -> L: log_validation_error()
  M --> C: Nothing
end

deactivate M

@enduml

@startuml
!theme mars

title Multi-Source Strategy

start

:User Request;

:Primary Source\n(Spotify);

if (Available?) then (yes)
  :Return Spotify Data;
  stop
else (no)
  :Log Fallback;
  :Secondary Source\n(YouTube Music);

  if (Available?) then (yes)
    :Return YouTube Data;
    stop
  else (no)
    :Tertiary Source\n(Local Cache);

    if (Available?) then (yes)
      :Return Cached Data;
      :Log Cache Hit;
      stop
    else (no)
      :Return Error;
      :Log Complete Failure;
      stop
    end
  end
end

@enduml

@startuml
!theme vibrant

cloud "Discord Servers" {
  [User Commands]
}

node "Production Environment" {
  [Discord Bot]
  [Log Aggregator]
  [Metrics Collector]
  [Health Monitor]
}

database "Log Storage" {
  [Structured Logs]
  [Error Traces]
  [Performance Metrics]
}

cloud "External APIs" {
  [Spotify API]
  [YouTube API]
}

[User Commands] --> [Discord Bot]
[Discord Bot] --> [Log Aggregator]
[Discord Bot] --> [Spotify API]
[Discord Bot] --> [YouTube API]
[Log Aggregator] --> [Structured Logs]
[Metrics Collector] --> [Performance Metrics]
[Health Monitor] --> [Error Traces]

@enduml