Pedro Guardini developer challenge

.gitignore

@@ -0,0 +1,26 @@
+# Environment & Secrets
+.env
+.venv/
+env/
+venv/
+*.env
+
+# Python artifacts
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.pytest_cache/
+.coverage
+htmlcov/
+
+# Database / PostgreSQL local artifacts
+# (If using local pgAdmin or SQLite for testing)
+*.db
+*.sqlite3
+.pgpass
+.pgdata
+
+# OS specific
+.DS_Store
+Thumbs.db

signal-processing-api/.python-version

@@ -0,0 +1 @@
+3.13

signal-processing-api/Dockerfile

@@ -0,0 +1,15 @@
+FROM python:3.13-slim
+
+WORKDIR /code
+
+COPY ./pyproject.toml /code/pyproject.toml
+
+COPY ./app /code/app
+
+COPY ./tests /code/tests
+
+RUN pip install --no-cache-dir --upgrade /code/.
+
+EXPOSE 80
+
+CMD ["fastapi", "run", "app/main.py", "--port", "80"]

signal-processing-api/README.md

@@ -0,0 +1,91 @@
+<h1>Signal Processor API</h1>
+
+
+- [Setting up the API](#setting-up-the-api)
+  - [Using Docker](#using-docker)
+- [API Routes](#api-routes)
+  - [Simple Data Retrieve Routes](#simple-data-retrieve-routes)
+  - [Operations Routes](#operations-routes)
+  - [Metrics and metadata routes](#metrics-and-metadata-routes)
+- [Running tests](#running-tests)
+
+___
+
+
+## Setting up the API
+
+### Using Docker
+
+You can run the API using the compose file in root folder as it follows:
+
+```bash
+docker compose up -d --build
+
+```
+
+Docker will check if the database service is healthy and so start the API at 127.0.0.1/80
+
+The database is already populated with some air humidity time-series data and this data can be acessed by the following routes.
+
+___
+
+## API Routes
+
+In general, we have 7 routes, correspondig all to time-series related data
+
+### Simple Data Retrieve Routes
+- GET: */series/*
+
+  This route retrieves all time-series registered in database, exposing only id, sensor name and register date of time-series. It's not necessary passing any parameter to get this data
+
+___
+
+- GET: */series/full_series/*
+
+  In another hand, this route retrieves all time-series registered, but with all corresponding measurements. No parameter is necessary
+
+___
+
+- GET: */series/time_series_id*
+
+  In this route we can get the data about a specific time-series passing as parameter its id. The response includes id, sensor, creation date and all related measurements.
+
+___
+
+### Operations Routes
+- POST:  */series/* 
+
+  In this route it's possible to create a new time-series record, just passing a json with: sensor and a list of measurements (with its timestamp and air humidity data)
+
+___
+
+- DELETE: */series/time_series_id* 
+
+  Here it's possible delete an existent time-series by passing its id. Following this route all related time-series data, like measurements points, are automatically deleted too
+
+___
+
+### Metrics and metadata routes
+
+- GET: */series/metrics/time_series_id*
+
+  This route is a more sofisticated one, here you can pass as parameter the selected time-series id and receive some metrics about the registered measurements as:
+  - Min/Max air humidity: showing which was the maximum and minimum air humidity registered by the sensor
+  - Avg air humidity: the average of all air humidity measurements
+  - Stddev air humidity: which was the standard deviation of all measurements in relation to the average air humidity
+
+- GET: */series/count/*
+
+  If you want to see how many time-series are stored, you can just follow this route without parameters and it returns the total number of registered time-series (ignoring related data)
+
+___
+
+## Running tests
+
+All the project tests are located at tests/ folder.
+
+We made a total of 14 tests, considering success cases and failures cases.
+
+We try to include as many user stories as possible to check the API response.
+
+For more instructions about running tests see [Readme Tests](/signal-processing-api/tests/README.md)

signal-processing-api/app/config/database.py

@@ -0,0 +1,36 @@
+import os
+from pathlib import Path
+from sqlmodel import SQLModel, create_engine, Session
+from sqlalchemy import text
+from fastapi import Depends
+
+from models.timeseries import TimeSeries
+from models.measurement import Measurement
+
+
+DATABASE_URL = os.getenv('DATABASE_URL', 
+                         'postgresql://signal_api:dynamox@database:5432/signal')
+
+engine = create_engine(DATABASE_URL)
+
+def create_db_and_tables():
+    """Function that create database and all registered tables"""
+    SQLModel.metadata.create_all(engine)
+
+def get_session():
+    with Session(engine) as session:
+        yield session
+
+def seed_db():
+    if 'postgresql' in DATABASE_URL:
+        current_dir = Path(__file__).parent
+
+        seed_path = f'{current_dir}/seed.sql'
+        with Session(engine) as session:
+            with open(seed_path, 'r', encoding='UTF-8') as f:
+                seeding_commands = f.read()
+
+            session.execute(text(seeding_commands))
+            session.commit()
+    else: 
+        return