Fix nouveau site puce à l'oreille

experimentations/get_lapucealoreille_events.py

@@ -29,8 +29,8 @@ from src.agenda_culturel.import_tasks.custom_extractors import *
 if __name__ == "__main__":
 
     u2e = URL2Events(SimpleDownloader(), lapucealoreille.CExtractor())
-    url = "https://www.lapucealoreille63.fr/programmation/"
-    url_human = "https://www.lapucealoreille63.fr/programmation/"
+    url = "https://www.lapucealoreille63.fr/agenda"
+    url_human = "https://www.lapucealoreille63.fr/agenda"
 
     try:
         events = u2e.process(url, url_human, cache = "cache-lapucealoreille.xml", default_values = {}, published = True)

src/agenda_culturel/import_tasks/custom_extractors/lapucealoreille.py

@@ -15,11 +15,6 @@ class CExtractor(TwoStepsExtractor):
             e_url = e.find("a")
             if e_url:
                 if self.add_event_url(e_url["href"]):
-                    title = e.select("div[data-testid=richTextElement] h1.font_0 span")
-                    if title:
-                        title = title[0].contents[0].get_text().replace("\n", " ")
-                        title = re.sub(" +", " ", title)
-                        self.add_event_title(e_url["href"], title)
 
     def add_event_from_content(
         self,
@@ -31,9 +26,12 @@ class CExtractor(TwoStepsExtractor):
     ):
         soup = BeautifulSoup(event_content, "html.parser")
 
+        title = soup.select("h2")[0].get_text()
+
         start_day = self.parse_french_date(
-            soup.find("h2").get_text()
+            soup.select("h2")[1].get_text()
         )  # pas parfait, mais bordel que ce site est mal construit
+        print(soup.select("h2")[1].get_text())
 
         spans = soup.select("div[data-testid=richTextElement] span")
         start_time = None
@@ -79,7 +77,7 @@ class CExtractor(TwoStepsExtractor):
 
         self.add_event_with_props(
             event_url,
-            None,
+            title,
             "Concert",
             start_day,
             location,