gradioTimeSeries.py

import requests
import json
import os
import xmltodict
import gzip
import shutil

import matplotlib.pyplot as plt
import matplotlib.font_manager as fm

def crawl(url, filename):
    try:
        try:
            response = requests.get(url)
        except Exception as e:
            print("Error",e)
        with open(f'{filename}.xml.gz', 'wb') as f:
            f.write(response.content)
        
        # 解壓縮
        with gzip.open(f'{filename}.xml.gz', 'rb') as f_in:
            with open(f'{filename}.xml', 'wb') as f_out:
                shutil.copyfileobj(f_in, f_out)

        # xml to json
        with open(f'{filename}.xml', 'r') as f:
            data_dict = xmltodict.parse(f.read())
            data_json = json.dumps(data_dict, ensure_ascii=False)
            # dump to Etag.json with json
            json.dump(data_dict, open(f'{filename}.json', 'w'), ensure_ascii=False, indent=2)
    except Exception as e:
        print(e)
        print(f'{url} failed')
    print(f'{filename} done')

def timeSeries(startDate,endDate):
    print("startDate: ", startDate, type(startDate))
    print("endDate: ", endDate, type(endDate))

    EtagPairId = '01F0339S-01F0376S'
    EtagStart = '01F0339S'
    EtagEnd = '01F0376S'
    EtagStartRoad = ['五股','高公局','林口(文化一路)']
    VehicleTypeDict = {
        31: "Car",
        32: "Small Truck",
        41: "Bus", # (含計程車)
        42: "Heavy Truck",
        5: "Trailer"
    }

    time_series_31data = {}
    time_series_32data = {}
    time_series_41data = {}
    time_series_42data = {}
    time_series_5data = {}

    day_list = ['202405%02d'%(x) for x in range(startDate,endDate+1)]
    hour_list = ['%02d'%(i) for i in range(24)]
    five_minute_list = ['%02d'%(i) for i in range(0,60,5)]
    print("Lens = ", len(day_list) * len(hour_list) * len(five_minute_list))
    for day in day_list:
        for hour in hour_list:
            for minute in five_minute_list:
                i = hour + minute
                filename = f'./timeSeriesData/ETagPairLive_{day}_{str(i)}'
                # Check if the file exists
                if not os.path.exists(f'{filename}.json'):
                    url = f'https://tisvcloud.freeway.gov.tw/history/motc20/ETag/{day}/ETagPairLive_{str(i)}.xml.gz'
                    crawl(url, filename)
                try:
                    with open(f'{filename}.json', 'r') as f:
                        data = json.load(f)
                        EtagPairLive = data['ETagPairLiveList']['ETagPairLives']['ETagPairLive']
                        for ETagPair in EtagPairLive:
                            if ETagPair['ETagPairID'] == EtagPairId:
                                # print(ETagPair['ETagPairID'])
                                Flow = ETagPair['Flows']['Flow']
                                for vehicle in Flow:
                                    # print(f"VehicleType: {VehicleTypeDict[int(vehicle['VehicleType'])]}, VehicleNum: {vehicle['VehicleCount']}, Speed: {vehicle['SpaceMeanSpeed']}, Time: {vehicle['TravelTime']}")
                                    if int(vehicle['VehicleType']) == 31:
                                        time_series_31data[f'{day}_{i}'] = int(vehicle['VehicleCount'])
                                    elif int(vehicle['VehicleType']) == 32:
                                        time_series_32data[f'{day}_{i}'] = int(vehicle['VehicleCount'])
                                    elif int(vehicle['VehicleType']) == 41:
                                        time_series_41data[f'{day}_{i}'] = int(vehicle['VehicleCount'])
                                    elif int(vehicle['VehicleType']) == 42:
                                        time_series_42data[f'{day}_{i}'] = int(vehicle['VehicleCount'])
                                    elif int(vehicle['VehicleType']) == 5:
                                        time_series_5data[f'{day}_{i}'] = int(vehicle['VehicleCount'])
                except:
                    pass

    step = 6
    plt.figure(figsize=(30,10))
    plt.plot(list(time_series_31data.keys())[::step], list(time_series_31data.values())[::step], label=VehicleTypeDict[31])
    plt.plot(list(time_series_32data.keys())[::step], list(time_series_32data.values())[::step], label=VehicleTypeDict[32])
    plt.plot(list(time_series_41data.keys())[::step], list(time_series_41data.values())[::step], label=VehicleTypeDict[41])
    plt.plot(list(time_series_42data.keys())[::step], list(time_series_42data.values())[::step], label=VehicleTypeDict[42])
    plt.plot(list(time_series_5data.keys())[::step], list(time_series_5data.values())[::step], label=VehicleTypeDict[5])
    plt.xlabel('Time')
    plt.xticks(fontsize=14)
    plt.xticks(rotation=60)
    plt.ylabel('VehicleCount')
    plt.title(f'{EtagStart} to {EtagEnd}')
    plt.legend()
    # plt.show()

    return plt