In this repository, you'll find two Python scripts that fetch weather data from JSON and CSV files, storing it in an SQLite3 database. The scripts generate a view combining data from IMIS and SMN (VQHA80 & VQHA98) stations and another view with a summary of the data, allowing for calculations like total snowfall in the last hours or days.
-
Install the required packages using the following command:
pip3 install -r requirements.txt
-
Initialize the SQLite database:
sqlite3 meteo.db
-
Create the "VQHA80" table with the following SQL command:
```sql CREATE TABLE IF NOT EXISTS VQHA80 ( id INTEGER PRIMARY KEY AUTOINCREMENT, location TEXT, rre150z0 FLOAT, date DATETIME, CONSTRAINT location_date UNIQUE (location, date) );
-
Create the "VQHA98" table with the following SQL command:
CREATE TABLE IF NOT EXISTS VQHA98 ( id INTEGER PRIMARY KEY AUTOINCREMENT, location TEXT, date DATETIME, rre150z0 FLOAT, CONSTRAINT location_date UNIQUE (location, date) );
-
Create the "IMIS" table with the following SQL command:
CREATE TABLE IF NOT EXISTS IMIS ( id INTEGER PRIMARY KEY AUTOINCREMENT, station_code TEXT, measure_date DATETIME, RR_10MIN_SUM FLOAT );
- Data Retrieval: Extracting CSV and JSON data from URLs.
- Database Integration: Efficient storage in an SQLite3 database.
- Combine Data: Merge SMN and IMIS stations' weather data and generate another view with a time summary of the weather database.
JSON:
- IMIS Weather Station:
IMIS Measurements Precipitation
CSV:
- VQHA80 Weather Station:
VQHA80 Measurements
- VQHA98 Weather Station:
VQHA98 Measurements
SQLite3
Python
pandas