rtm_scraper.py

from __future__ import unicode_literals
import requests
import json
from datetime import datetime
import re
from bs4 import BeautifulSoup

class RTMTraffic:
    def __init__(self):
        self.headers = {
            'User-Agent': 'RTM API Python Client',
            'From': 'https://github.com/your-username/your-repo'
        }
        
    def get_alerts(self, period='all', line_type=None):
        """
        Récupère les alertes de trafic RTM
        Args:
            period (str): 'today', 'coming' ou 'all'
            line_type (str): None, 'bus', 'metro' ou 'tram'
        Returns:
            dict: Dictionnaire contenant les alertes
        """
        if line_type is None:
            url = 'https://api.rtm.fr/front/getAlertes/FR/All'
            response = requests.get(url, headers=self.headers)
            data = response.json()['data']
            
            alerts_today = self._process_alerts(data.get('AlertesToday', []))
            alerts_coming = self._process_alerts(data.get('AlertesComing', []))
            
            if period == 'today':
                return alerts_today
            elif period == 'coming':
                return alerts_coming
            else:
                return {
                    'AlertesToday': alerts_today,
                    'AlertesComing': alerts_coming
                }
        else:
            url = f'https://api.rtm.fr/front/getAlertes/FR/{line_type}'
            response = requests.get(url, headers=self.headers)
            data = response.json()['data']
            return self._process_alerts(data.get('Alertes', []))
    
    def _extract_description(self, html_content):
        """
        Extrait la description depuis le contenu HTML
        Args:
            html_content (str): Contenu HTML de l'alerte
        Returns:
            str: Description nettoyée
        """
        if not html_content:
            return ""
            
        # Parser le HTML
        soup = BeautifulSoup(html_content, 'html.parser')
        
        # Supprimer les balises <style> et leur contenu
        for style in soup.find_all('style'):
            style.decompose()
            
        # Récupérer le texte et nettoyer
        description = soup.get_text(separator=' ', strip=True)
        
        # Nettoyer les espaces multiples et les sauts de ligne
        description = re.sub(r'\s+', ' ', description)
        description = description.replace('\\n', ' ').replace('\\r', '')
        
        return description.strip()
    
    def _process_alerts(self, alerts):
        """
        Traite les alertes pour un format plus lisible
        Args:
            alerts (list): Liste des alertes brutes
        Returns:
            list: Liste des alertes formatées
        """
        processed_alerts = []
        for alert in alerts:
            # Extraire la description depuis le HTML (MessageD)
            description = self._extract_description(alert.get('MessageD', ''))
            if not description:
                # Si pas de description dans MessageD, utiliser MessageB
                description = alert.get('MessageB', '').strip()
            
            processed_alert = {
                'title': alert.get('MessageA', ''),
                'description': description,
                'validity': alert.get('MessageC', ''),
                'type': alert.get('type', ''),
                'valid_until': alert.get('ValidUntilTime', ''),
                'affected_lines': []
            }
            
            # Traiter les lignes affectées
            for line in alert.get('AffectedLine', []):
                if 'PublicCode' in line:
                    processed_alert['affected_lines'].append(line['PublicCode'])
            
            processed_alerts.append(processed_alert)
        
        return processed_alerts

def get_rtm_traffic():
    """
    Récupère toutes les perturbations de trafic RTM
    Returns:
        list: Liste des perturbations
    """
    rtm = RTMTraffic()
    all_alerts = rtm.get_alerts()
    
    traffic_info = []
    
    # Traiter les alertes du jour
    if 'AlertesToday' in all_alerts:
        for alert in all_alerts['AlertesToday']:
            traffic_info.append({
                'title': alert['title'],
                'description': alert['description'],
                'affected_lines': alert['affected_lines'],
                'validity': alert['validity'],
                'type': alert['type']
            })
    
    # Traiter les alertes à venir
    if 'AlertesComing' in all_alerts:
        for alert in all_alerts['AlertesComing']:
            traffic_info.append({
                'title': f"[À venir] {alert['title']}",
                'description': alert['description'],
                'affected_lines': alert['affected_lines'],
                'validity': alert['validity'],
                'type': alert['type']
            })
    
    return traffic_info

if __name__ == "__main__":
    traffic_info = get_rtm_traffic()
    if traffic_info:
        for info in traffic_info:
            print("\nPerturbation :")
            print(f"Titre : {info['title']}")
            if info['description']:
                print(f"Description : {info['description']}")
            print(f"Type : {info['type']}")
            print(f"Validité : {info['validity']}")
            if info['affected_lines']:
                print(f"Lignes affectées : {', '.join(info['affected_lines'])}")
            print("-" * 50)
    else:
        print("\nAucune perturbation en cours")