---
sidebar_position: 2
---

# L2 — Containers

> **C4 Level 2**: Zooms into Stock Market Professional and shows its deployable units — services, databases, and schedulers.

:::info C4 Model — Level 2
The Container diagram answers: *What are the high-level technical building blocks, and how do they talk to each other?*
:::

## Container Diagram

```mermaid
C4Container
    title Container Diagram — Stock Market Professional

    Person(trader, "Trader", "Reviews signals, executes manually")

    System_Boundary(smp, "Stock Market Professional") {
        Container(ingestor, "Data Ingestor", "Python / APScheduler", "Polls market data APIs on schedule, normalises OHLCV candles, detects gaps.")
        Container(analyser, "Analysis Engine", "Python / pandas-ta", "Applies technical indicators and pattern recognition on latest candle data.")
        Container(signalGen, "Signal Generator", "Python", "Scores setups by confluence, filters by risk rules, produces ≤3 daily signals.")
        Container(scheduler, "Job Scheduler", "APScheduler / cron", "Orchestrates pipeline runs: ingest every minute, analyse every 5 min, generate plan pre-market.")
        ContainerDb(tsdb, "Market Data Store", "SQLite / TimescaleDB", "Stores normalised OHLCV candles, indicator snapshots, and signal history.")
        Container(notifier, "Notifier", "Python / Discord Webhook", "Formats and delivers the daily trading plan and intraday alerts.")
    }

    System_Ext(marketData, "Market Data APIs", "Yahoo Finance, Alpha Vantage, Polygon")
    System_Ext(discord, "Discord", "Signal delivery")
    System_Ext(newsApi, "Economic Calendar API", "Forex Factory, Investing.com")

    Rel(scheduler, ingestor, "Triggers", "Internal call")
    Rel(scheduler, analyser, "Triggers", "Internal call")
    Rel(scheduler, signalGen, "Triggers", "Internal call")
    Rel(ingestor, marketData, "Fetches OHLCV", "REST / WS")
    Rel(ingestor, tsdb, "Writes candles", "SQL")
    Rel(analyser, tsdb, "Reads candles", "SQL")
    Rel(analyser, tsdb, "Writes indicator snapshots", "SQL")
    Rel(signalGen, tsdb, "Reads snapshots", "SQL")
    Rel(signalGen, newsApi, "Checks economic calendar", "REST")
    Rel(signalGen, notifier, "Passes signals", "In-process")
    Rel(notifier, discord, "Posts daily plan", "Webhook")
    Rel(trader, discord, "Reads signals", "Discord UI")
```

## Containers Inventory

| Container | Technology | Responsibility |
|---|---|---|
| **Data Ingestor** | Python, requests, websockets | Poll & normalise market data |
| **Analysis Engine** | Python, pandas, pandas-ta, ta-lib | Technical indicators + pattern recognition |
| **Signal Generator** | Python | Confluence scoring, risk rules, signal assembly |
| **Job Scheduler** | APScheduler | Pipeline orchestration & timing |
| **Market Data Store** | SQLite (dev) / TimescaleDB (prod) | Persist OHLCV, indicators, signals |
| **Notifier** | Python, Discord Webhook | Format and deliver signals |

## Deployment

All containers run as a single **Python process** in development (scheduler + all modules), and can be split into separate Docker containers in production:

```
docker/
├── ingestor/     # Data poller container
├── analyser/     # Analysis worker container
├── signal-gen/   # Signal generation + notifier
└── timescaledb/  # Database container
```

The CI pipeline (Gitea Actions) builds images and pushes to the Zot container registry at `lego-cloud.eu:30264`.