un peu de refactoring pour les extracteurs

src/agenda_culturel/extractors.py

@@ -26,10 +26,20 @@ logger = get_task_logger(__name__)
 
 class Extractor:
 
+    name = None
+
+    @abstractmethod
+    def is_known_url(url):
+        pass
+
     @abstractmethod
     def extract(url):
         pass
 
+    @abstractmethod
+    def clean_url(url):
+        pass
+
     def download(url):
         try:
             options = Options()
@@ -69,6 +79,8 @@ class Extractor:
 
 class ExtractorFacebook(Extractor):
 
+    name = "Facebook"
+
     class SimpleFacebookEvent:
 
         def __init__(self, data):
@@ -194,8 +206,7 @@ class ExtractorFacebook(Extractor):
             image = self.get_element("image")
             local_image = None if image is None else Extractor.download_media(image)
 
-            u = urlparse(url)
-            unique_url = u.scheme + "://" + u.netloc + u.path
+
 
             return Event(title=self.get_element("name"), 
                 status=Event.STATUS.DRAFT,
@@ -208,7 +219,20 @@ class ExtractorFacebook(Extractor):
                 local_image=local_image,
                 image=self.get_element("image"),
                 image_alt=self.get_element("image_alt"),
-                reference_urls=[unique_url])
+                reference_urls=[url])
+
+
+    def clean_url(url):
+
+        if ExtractorFacebook.is_known_url(url):
+            u = urlparse(url)
+            return u.scheme + "://" + u.netloc + u.path
+        else:
+            return url
+
+    def is_known_url(url):
+        u = urlparse(url)
+        return u.netloc in ["facebook.com", "www.facebook.com", "m.facebook.com"]
 
 
     def process_page(txt, url):
@@ -230,6 +254,13 @@ class ExtractorFacebook(Extractor):
 
 class ExtractorAllURLs:
 
+    extractors = [ExtractorFacebook]
+
+    def clean_url(url):
+        result = url
+        for e in ExtractorAllURLs.extractors:
+            result = e.clean_url(result)
+        return result
 
     def extract(url):
         logger.info("Run extraction")
@@ -239,13 +270,12 @@ class ExtractorAllURLs:
             logger.info("Cannot download url")
             return None
 
-        result = ExtractorFacebook.process_page(txt, url)
+        for e in ExtractorAllURLs.extractors:
+            result = e.process_page(txt, url)
 
-        if result is not None:
-            return result
-        else:
-            logger.info("Not a Facebook link")
-
-        # TODO: add here other extrators
+            if result is not None:
+                return result
+            else:
+                logger.info("Not a " + e.name + " link")
 
         return None