From 623274b1a64295f302388539c3a7434b87027e09 Mon Sep 17 00:00:00 2001
From: Jean-Marie Favreau <j-marie.favreau@uca.fr>
Date: Wed, 26 Jun 2024 10:45:56 +0200
Subject: [PATCH] =?UTF-8?q?-=20Fix=20crash=20r=C3=A9cup=C3=A9ration=20?=
 =?UTF-8?q?=C3=A9v=C3=A9nement=20FB=20-=20Tentative=20d'acc=C3=A9l=C3=A9ra?=
 =?UTF-8?q?tion=20quand=20on=20utilise=20Selenium=20(objet=20cr=C3=A9?=
 =?UTF-8?q?=C3=A9=20au=20chargement=20de=20Celery)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 src/agenda_culturel/celery.py                  | 4 +++-
 src/agenda_culturel/import_tasks/downloader.py | 6 +++---
 src/agenda_culturel/import_tasks/extractor.py  | 1 +
 src/agenda_culturel/import_tasks/importer.py   | 5 ++++-
 4 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/src/agenda_culturel/celery.py b/src/agenda_culturel/celery.py
index 09c0b8e..efb5bd6 100644
--- a/src/agenda_culturel/celery.py
+++ b/src/agenda_culturel/celery.py
@@ -20,6 +20,8 @@ app = Celery("agenda_culturel")
 
 logger = get_task_logger(__name__)
 
+chromiumDownloader = ChromiumHeadlessDownloader()
+
 
 # Using a string here means the worker doesn't have to serialize
 # the configuration object to child processes.
@@ -96,7 +98,7 @@ def run_recurrent_import(self, pk):
     downloader = (
         SimpleDownloader()
         if rimport.downloader == RecurrentImport.DOWNLOADER.SIMPLE
-        else ChromiumHeadlessDownloader()
+        else chromiumDownloader
     )
     if rimport.processor == RecurrentImport.PROCESSOR.ICAL:
         extractor = ICALExtractor()
diff --git a/src/agenda_culturel/import_tasks/downloader.py b/src/agenda_culturel/import_tasks/downloader.py
index c6a9cdf..3a4776d 100644
--- a/src/agenda_culturel/import_tasks/downloader.py
+++ b/src/agenda_culturel/import_tasks/downloader.py
@@ -67,14 +67,14 @@ class ChromiumHeadlessDownloader(Downloader):
         self.options.add_argument("--disable-dev-shm-usage")
         self.options.add_argument("--no-sandbox")
         self.service = Service("/usr/bin/chromedriver")
+        self.driver = webdriver.Chrome(service=self.service, options=self.options)
 
-    def download(self, url, post=None):
+    def download(self, url, referer=None, post=None):
         if post:
             raise Exception("POST method with Chromium headless not yet implemented")
         print("Download {}".format(url))
-        self.driver = webdriver.Chrome(service=self.service, options=self.options)
 
         self.driver.get(url)
         doc = self.driver.page_source
-        self.driver.quit()
+        self.driver.close()
         return doc
diff --git a/src/agenda_culturel/import_tasks/extractor.py b/src/agenda_culturel/import_tasks/extractor.py
index 7870a7d..ccb4380 100644
--- a/src/agenda_culturel/import_tasks/extractor.py
+++ b/src/agenda_culturel/import_tasks/extractor.py
@@ -16,6 +16,7 @@ class Extractor(ABC):
         self.header = {}
         self.events = []
         self.downloader = None
+        self.referer = ""
 
     def guess_end_day(self, start_day, start_time, end_time):
         if end_time:
diff --git a/src/agenda_culturel/import_tasks/importer.py b/src/agenda_culturel/import_tasks/importer.py
index 960f93c..753381f 100644
--- a/src/agenda_culturel/import_tasks/importer.py
+++ b/src/agenda_culturel/import_tasks/importer.py
@@ -13,7 +13,10 @@ class URL2Events:
     def process(
         self, url, url_human=None, cache=None, default_values=None, published=False
     ):
-        content = self.downloader.get_content(url, cache, referer=self.extractor.url_referer)
+        referer = ""
+        if self.extractor:
+            referer = self.extractor.url_referer
+        content = self.downloader.get_content(url, cache, referer=referer)
 
         if content is None:
             return None