Comment utiliser correctement les métriques d’autorité SEO ?

Le code

Tout d’abord, vous aurez besoin des packages suivants (et donc de lancer les commandes suivantes) :

pip install git+https://github.com/BabbarTech/BabbarPy.git
pip install configparser, pandas, numpy, plotly

Et enfin, le code :

import csv
from BabbarPy.host import h_history
import os
import configparser
import pandas as pd
import numpy as np
import plotly.express as px
import plotly.graph_objects as go

def create_hosts_list():
    hosts = []
    while True:
        host = input("Veuillez entrer le host à vérifier (ou appuyez sur entrée pour terminer) : ")
        if host == '':
            break
        hosts.append(host)
    with open('hosts.txt', 'w') as f:
        for host in hosts:
            f.write(f"{host}\n")

def get_api_key():
    config = configparser.ConfigParser()
    if not os.path.exists('config.ini') or not config.read('config.ini') or not 'API' in config or not 'api_key' in config['API']:
        api_key = input("Entrez votre clé API: ")
        config['API'] = {'api_key': api_key}
        with open('config.ini', 'w') as configfile:
            config.write(configfile)
        return api_key
    else:
        return config['API']['api_key']

def convert_scholar_score(score):
    score_map = {
        'A+': 13, 'A': 12, 'A-': 11,
        'B+': 10, 'B': 9, 'B-': 8,
        'C+': 7, 'C': 6, 'C-': 5,
        'D+': 4, 'D': 3, 'D-': 2,
        'F': 1
    }
    return score_map.get(score, np.nan)

def plot_metrics(csv_file):
    # Read the CSV file into a DataFrame
    df = pd.read_csv(csv_file, parse_dates=['Date'])
    df.set_index('Date', inplace=True)

    # Separate the data into two DataFrames for each metric
    bas_df = df[[col for col in df.columns if 'bas_' in col]]
    scholar_df = df[[col for col in df.columns if 'scholar_' in col]]

    # Ensure all data is numeric and handle missing values
    bas_df = bas_df.apply(pd.to_numeric, errors='coerce')
    scholar_df = scholar_df.apply(pd.to_numeric, errors='coerce')

    # Plot the Babbar Authority Scores using Plotly
    fig_bas = go.Figure()
    for column in bas_df.columns:
        fig_bas.add_trace(go.Scatter(x=bas_df.index, y=bas_df[column], mode='lines+markers', name=column))
    fig_bas.update_layout(
        title='Comparison of Babbar Authority Scores Over Time',
        xaxis_title='Date',
        yaxis_title='Babbar Authority Score',
        hovermode='x'
    )
    fig_bas.write_html("babbar_authority_scores.html")

    # Plot the Scholar Scores using Plotly
    fig_scholar = go.Figure()
    for column in scholar_df.columns:
        fig_scholar.add_trace(go.Scatter(x=scholar_df.index, y=scholar_df[column], mode='lines+markers', name=column))
    fig_scholar.update_layout(
        title='Comparison of Scholar Scores Over Time',
        xaxis_title='Date',
        yaxis_title='Scholar Score',
        hovermode='x'
    )
    fig_scholar.write_html("scholar_scores.html")

def plot_ranked_metrics(csv_file):
    # Read the CSV file into a DataFrame
    df = pd.read_csv(csv_file, parse_dates=['Date'])
    df.set_index('Date', inplace=True)

    # Separate the BAS data
    bas_df = df[[col for col in df.columns if 'bas_' in col]]

    # Ensure all data is numeric and handle missing values
    bas_df = bas_df.apply(pd.to_numeric, errors='coerce')

    # Rank the sites for each month based on BAS
    ranked_df = bas_df.rank(axis=1, method='min', ascending=False)

    # Plot the ranked Babbar Authority Scores using Plotly
    fig_ranked_bas = go.Figure()
    for column in ranked_df.columns:
        fig_ranked_bas.add_trace(go.Scatter(x=ranked_df.index, y=ranked_df[column], mode='lines+markers', name=column))
    fig_ranked_bas.update_layout(
        title='Monthly Rank of Babbar Authority Scores',
        xaxis_title='Date',
        yaxis_title='Rank',
        yaxis=dict(autorange='reversed'),
        hovermode='x'
    )
    fig_ranked_bas.write_html("ranked_babbar_authority_scores.html")

def main():
    confupdate = "N"
    api_key = get_api_key()
    if os.path.exists("hosts.txt"):
        confupdate = input("hosts.txt up to date ? Y/N: ")
    if confupdate != "Y":
        create_hosts_list()
    hosts = []
    metrics = {}
    with open('hosts.txt', 'r') as f:
        for line in f:
            hosts.append(line.strip())
    for host in hosts:
        history = h_history(host, api_key)
        if isinstance(history, list):
            bas_values = {entry['date']: (entry['babbarAuthorityScore'], convert_scholar_score(entry['scholarScore'])) for entry in history}
        else:
            bas_values = {date: (data['babbarAuthorityScore'], convert_scholar_score(data['scholarScore'])) for date, data in history.items()}
        metrics[host] = bas_values

    # Get all unique dates from all hosts
    all_dates = sorted({date for values in metrics.values() for date in values.keys()})

    # Write to CSV
    with open('metrics.csv', 'w', newline='') as csvfile:
        fieldnames = ['Date'] + [f'bas_{host}' for host in hosts] + [f'scholar_{host}' for host in hosts]
        writer = csv.DictWriter(csvfile, fieldnames=fieldnames)

        writer.writeheader()
        for date in all_dates:
            row = {'Date': date}
            for host in hosts:
                row[f'bas_{host}'] = metrics.get(host, {}).get(date, (np.nan, np.nan))[0]
                row[f'scholar_{host}'] = metrics.get(host, {}).get(date, (np.nan, np.nan))[1]
            writer.writerow(row)
    
    plot_metrics('metrics.csv')
    plot_ranked_metrics('metrics.csv')

if __name__ == "__main__":
    main()