Currency Monitoring

Overview

This project involves developing a bot using Python to monitor currency exchange rates via the Uphold API. The bot checks rates at regular intervals, compares them with previously fetched rates, and logs an alert if a specified percentage change is detected. Asynchronous programming is utilized for efficiency and scalability.

Project Components

1. Configuration and Environment Setup

   • Files:
     • config.py y_config.yaml
     • .env
   • Description:
     • Configuration Loading:
       • load_config(file_path): Reads settings from currency_config.yaml
     • Environment Variables:

2. API Uphold

   • Files:
     • api_uphold.py
   • Description:
     • APIClient Class:
       • Utilizes aiohttp for asynchronous HTTP requests to the Uphold API.
       • Example: Fetches the BTC-USD rate from https://api.uphold.com/v0/ticker/BTC-USD.
       • Handles rate limiting by checking the HTTP response status. If status is 429 (Too Many Requests), it retries after the specified interval.

3. Currency Monitor

   • Files:
     • monitor.py
   • Description:
     • CurrencyMonitor Class:
       • Manages monitoring of multiple currency pairs.
       • Example: Monitors pairs like BTC-USD and ETH-USD concurrently.
       • Fetches current rates and compares them with the last recorded rates from DynamoDB. Logs an alert if the rate changes by the specified threshold (e.g., 0.01%).

4. DynamoDB Client

   • Files:
     • dynamodb_client.py
   • Description:
     • DynamoDBClient Class:
       • Interacts with DynamoDB to store and retrieve currency rates and configuration information.
       • Example:
         • get_last_alert
         • post_alert

5. Main Execution Script

   • Files:
     • main.py
   • Description:
     • Main Function:
       • Sets up logging and loads configuration.
       • Initializes CurrencyMonitor with parameters from currency_config.yaml.
       • Example: Starts monitoring process for pairs like BTC-USD, ETH-USD at intervals specified in the configuration.
       • Ensures DynamoDB tables are created using create_table() before starting the monitoring.

Asynchronous Programming

• Reason for Asynchronous Programming:
  • Efficiency: Allows handling multiple tasks simultaneously, crucial for IO-bound tasks like network requests.
  • Scalability: Manages concurrent HTTP requests for multiple currency pairs effectively.
  • Responsiveness: Keeps the bot responsive, handling retries and rate limits smoothly.
• Libraries Used:
  • aiohttp: For non-blocking HTTP requests to the Uphold API.
  • asyncio: For managing asynchronous tasks and concurrency.
  • aiobotocore: For asynchronous interactions with DynamoDB.

The following commands can be used to print the current DynamoDB table in different formats:

• Credentials:

export REGION_NAME='us-west-2'
export AWS_ACCESS_KEY_ID='fakeAccessKeyId'
export AWS_SECRET_ACCESS_KEY='fakeSecretAccessKey'

1. Print as JSON:

$ aws dynamodb scan --table-name CurrencyAlerts --endpoint-url http://localhost:8000 --output json

2. Print as formatted table:

$ aws dynamodb scan --table-name CurrencyAlerts --endpoint-url http://localhost:8000 --output json \
| jq -r '.Items[] | [.pair.S, .rate.N, (.timestamp.N | tonumber | strftime("%Y-%m-%d %H:%M:%S")), .change_threshold.N, .check_interval.N] | @tsv' \
| column -t

3. Print as CSV:

$ aws dynamodb scan --table-name CurrencyAlerts --endpoint-url http://localhost:8000 --output json \
| jq -r '["pair", "rate", "timestamp", "change_threshold", "check_interval"], (.Items[] | [.pair.S, .rate.N, (.timestamp.N | tonumber | strftime("%Y-%m-%d %H:%M:%S")), .change_threshold.N, .check_interval.N]) | @csv'