From 92482fff0717b8761368772710fd091480c80dfd Mon Sep 17 00:00:00 2001
From: toravest <torave@stud.ntnu.no>
Date: Mon, 24 Mar 2025 11:25:58 +0100
Subject: [PATCH] add statistic historical data, rewrite functions for
 universal use

---
 notebooks/statistic_data_notebook.ipynb | 180 ++++++++++++++++++++++++
 src/my_package/fetch_data.py            |  46 +-----
 src/my_package/write_data.py            |   4 +-
 src/my_package/year_data.py             |  38 +++++
 4 files changed, 226 insertions(+), 42 deletions(-)
 create mode 100644 notebooks/statistic_data_notebook.ipynb
 create mode 100644 src/my_package/year_data.py

diff --git a/notebooks/statistic_data_notebook.ipynb b/notebooks/statistic_data_notebook.ipynb
new file mode 100644
index 0000000..e4b10d1
--- /dev/null
+++ b/notebooks/statistic_data_notebook.ipynb
@@ -0,0 +1,180 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Velg et sted i Norge å få statistisk data\n",
+    "\n",
+    "Denne API-en henter statistisk historisk data, herunder, statistisk data basert på de historiske dataene, ikke reele statistisk historisk. \n",
+    "\n",
+    "Statistikken er basert på de historiske datane total sett, ikke for hvert år."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import sys\n",
+    "import os\n",
+    "\n",
+    "# Gets the absolute path to the src folder\n",
+    "sys.path.append(os.path.abspath(\"../src\"))\n",
+    "\n",
+    "# Now we can import the fucntion from the module\n",
+    "from my_package.year_data import fetch_data\n",
+    "\n",
+    "# User input the city, for the weather\n",
+    "city_name = input(\"Enter a city in Norway: \")\n",
+    "\n",
+    "data, folder = fetch_data(city_name)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Lagre data i json-fil\n",
+    "\n",
+    "Skriv inn navn for til filen du vil lagre med dataen.\n",
+    "\n",
+    "Eks. test\n",
+    "Da vil filen lagres som data_**test**.json, i mappen \"../data/output_statistikk/data_{filnavn}.json\"\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Gets the absolute path to the src folder\n",
+    "sys.path.append(os.path.abspath(\"../src\"))\n",
+    "\n",
+    "from my_package.write_data import write_data\n",
+    "\n",
+    "filename = input(\"Write filename: \")\n",
+    "\n",
+    "write_data(data, folder, filename)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Lese fra fil\n",
+    "\n",
+    "Henter opp data lagret i filen, lagd over, og skriver ut lesbart ved hjelp av pandas"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "data = pd.read_json(f'../data/output_statistikk/data_{filename}.json')\n",
+    "\n",
+    "display(data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "if 'result' in data:\n",
+    "    df = pd.json_normalize(data['result'])\n",
+    "\n",
+    "    display(df)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Plotter data\n",
+    "Denne koden plotter data basert på gjennomsnitts temperatur gjennom året. For å sikre lagring av de ulike kjøringene, vil grafen bli lagret i mappen \"../data/output_fig/mean_temp_plot_{city_name}.json\"\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "import matplotlib.dates as mdates\n",
+    "import os\n",
+    "\n",
+    "output_folder = \"../data/output_fig\"\n",
+    "os.makedirs(output_folder, exist_ok=True)  # Create the folder if it doesn't exist\n",
+    "\n",
+    "# Converts to and make a new column with celsius temp, and not kelvin\n",
+    "df['temp.mean_celsius'] = df['temp.mean'] - 273.15\n",
+    "temp = df['temp.mean_celsius']\n",
+    "\n",
+    "# Convert from day and month, to datetime\n",
+    "# df['date'] = pd.to_datetime(df[['month', 'day']].assign(year=2024))\n",
+    "\n",
+    "# Create a new column that concatenates month and day (e.g., \"03-01\" for March 1)\n",
+    "df['month_day'] = df[['month', 'day']].apply(lambda x: f\"{x['month']:02d}-{x['day']:02d}\",axis=1)\n",
+    "\n",
+    "# Plot the graph of the mean temperature\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "plt.plot(df['month_day'], temp)\n",
+    "\n",
+    "# Label for easier reading and understanding of the plot\n",
+    "plt.title(f\"Mean temp - statistic historical {city_name}\")\n",
+    "plt.xlabel(\"Date\")\n",
+    "plt.ylabel(\"Temperature (°C)\")\n",
+    "\n",
+    "# Customize the x-axis to show ticks and labels only at the start of each month\n",
+    "plt.gca().xaxis.set_major_locator(mdates.MonthLocator())  \n",
+    "# Format ticks to show abbreviated month names (e.g., Jan, Feb)\n",
+    "plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%b')) \n",
+    "\n",
+    "plt.xticks(rotation=45)\n",
+    "plt.yticks(range(-20, 30, 2))\n",
+    "plt.tight_layout()\n",
+    "plt.grid()\n",
+    "\n",
+    "# Save the plot to the data/output_fig folder\n",
+    "plot_path = os.path.join(output_folder, f\"mean_temp_plot_{city_name}.png\")\n",
+    "plt.savefig(plot_path)  # Save the plot as a PNG file\n",
+    "\n",
+    "# Show the plot\n",
+    "plt.show()\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "venv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/src/my_package/fetch_data.py b/src/my_package/fetch_data.py
index a3097ef..f087d28 100644
--- a/src/my_package/fetch_data.py
+++ b/src/my_package/fetch_data.py
@@ -3,23 +3,17 @@
 import os
 from dotenv import load_dotenv
 
-
-# # only for change unix to date for plot
-# import matplotlib.pyplot as plt
-# import datetime
-# import time
-
 load_dotenv()
 
 # Gets the key, from my env file
 API_KEY = os.getenv("API_KEY")
 
 # city_name = "Trondheim"
-country_code = "NO"
+# country_code = "NO"
 
 # Temporarily standard times
-start_date = 1735686000
-end_date = 1740009323
+# start_date = 1735686000
+# end_date = 1740009323
 
 # Gets the data from the API - openweathermap.org
 def fetch_data(start_date, end_date, city_name):
@@ -36,39 +30,11 @@ def fetch_data(start_date, end_date, city_name):
         
         # Converts the data into json
         data = response.json()
+        folder = "../data/output_stedsnavn"
 
         print("Data fetch: ok")
-        return data
+        return data, folder
 
     else:
         # If html status code != 200, print the status code
-        print("Failed to fetch data from API. Status code:", response.status_code) 
-
-
-# myData = fetch_data(start_date, end_date, city_name)
-
-# if myData is not None:
-#     # Empty dict for temperature
-#     temperature = {}
-#     # Iterates through myData dict, and add the dt (unix_time) and temp to temperature dict
-#     if "list" in myData:
-#         for item in myData["list"]:
-#             if "main" in item and "temp" in item["main"]:
-#                 temperature[item["dt"]] = (item["main"]["temp"])
-
-#     # print(temperature)
-
-#     x_temp = [datetime.datetime.fromtimestamp(ts) for ts in temperature.keys()]
-#     # A test plot, to get a visual look of the data
-#     # x_temp = list(temperature.keys())
-#     y_temp = list(temperature.values())
-
-#     plt.plot(x_temp, y_temp)
-#     plt.xlabel('Unix Time')
-#     plt.ylabel('Temperature (°C)')
-#     plt.title('Temperature over Time')
-#     plt.grid()
-#     plt.show()
-
-# else:
-#     print("No data to process.")
\ No newline at end of file
+        print(f"Failed to fetch data for {city_name} from API. Status code:", response.status_code) 
diff --git a/src/my_package/write_data.py b/src/my_package/write_data.py
index 233bb91..e46eb2a 100644
--- a/src/my_package/write_data.py
+++ b/src/my_package/write_data.py
@@ -1,11 +1,11 @@
 import json
 import os
 
-def write_data(data, filename):
+def write_data(data, folder, filename):
     # Ensure the 'output_stedsdata' folder exists inside the 'data' folder at the root of the project
     script_dir = os.path.dirname(os.path.abspath(__file__))  # Get the directory of the script
     project_root = os.path.abspath(os.path.join(script_dir, os.pardir, os.pardir))  # Navigate to the root of the project
-    data_dir = os.path.join(project_root, 'data', 'output_stedsdata')
+    data_dir = os.path.join(project_root, 'data', folder)
     os.makedirs(data_dir, exist_ok=True)  # Creates 'data/output_stedsdata' folder if it doesn't exist
 
     # Write the JSON data to a file inside the 'output_stedsdata' folder
diff --git a/src/my_package/year_data.py b/src/my_package/year_data.py
new file mode 100644
index 0000000..3d6f780
--- /dev/null
+++ b/src/my_package/year_data.py
@@ -0,0 +1,38 @@
+# Import of needed libaries
+import requests
+import os
+from dotenv import load_dotenv
+
+
+load_dotenv()
+
+# Gets the key, from my env file
+API_KEY = os.getenv("API_KEY")
+
+# city_name = "Maura"
+
+# Gets the data from the API - openweathermap.org
+def fetch_data(city_name):
+   
+
+    # f-string url, to add the "custom" variables to the API-request
+    url = f"https://history.openweathermap.org/data/2.5/aggregated/year?q={city_name},NO&appid={API_KEY}&units=metric"
+    
+    # Saves the API-request for the url 
+    response = requests.get(url)
+
+    # Checks if the status code is OK
+    if response.status_code == 200:
+        
+        # Converts the data into json
+        data = response.json()
+        folder = "../data/output_statistikk"
+
+        print("Data fetch: ok")
+        return data, folder
+
+    else:
+        # If html status code != 200, print the status code
+        print(f"Failed to fetch data for city: {city_name} from API. Status code:", response.status_code) 
+
+# myData = fetch_data(city_name)
\ No newline at end of file