follow navigation links in opds to retrieve entire catalog

konfluks/feed.py

@@ -155,8 +155,11 @@ def parse_enclosures(post_dir, entry):
         if "type" in e:
             print("found enclosed media", e.type)
             if "image/" in e.type:
+                if not os.path.exists(post_dir): #this might be redundant with create_post
+                    os.makedirs(post_dir)
                 featured_image = grab_media(post_dir, e.href)
-                entry["featured_image"] = featured_image
+                media_item = urlparse(e.href).path.split('/')[-1]
+                entry["featured_image"] = media_item
             else:
                 print("FIXME:ignoring enclosed", e.type)
     return entry
@@ -309,12 +312,15 @@ def create_opds_post(post_dir, entry):
         ft = item['type'].split('/')[-1]
         fn = item['rel'].split('/')[-1]
 
-        if fn == "acquisition":
+        # entry.links has image, thumbnail and publication/acquisition.
-            fn = "publication" #calling the publications acquisition is weird
+        # Only downloading image for now
+        #if fn == "acquisition": 
+        #fn = "publication" #calling the publications acquisition is weird 
 
+        if 'image' in fn:
             prefered_name = "{}-{}.{}".format(fn, slugify(entry['title']), ft)
-
             grab_media(post_dir, item['href'], prefered_name)
+            frontmatter['featured_image'] = prefered_name
 
         if "summary" in entry:
             summary = entry.summary
@@ -330,6 +336,18 @@ def create_opds_post(post_dir, entry):
         timestamp = arrow.get(entry['updated_parsed'])
         f.write(timestamp.format('X'))
 
+def opds_fetch_more(data):
+    """
+    Look for mode OPDS feeds to pull, untill we no longer hit the "next" navigation property. 
+    """
+    for link in data.feed.links:
+        for i in link:
+            if link[i] == 'next':
+                print(link['href'])
+                data = grab_feed(link['href'])
+                return data
+    return None
+
 
 def main():
     feed_urls = open("feeds_list.txt", "r").read().splitlines()
@@ -373,14 +391,25 @@ def main():
 
         data = grab_feed(feed_url)
 
-        if data:
+        if data: #whenever we get a 200
-
+            if data.feed: #only if it is an actual feed
                 opds_feed = False
+                opds_entries = []
+                if 'links' in data.feed:
                     for i in data.feed['links']:
                             if i['rel'] == 'self':
                                 if 'opds' in i['type']:
                                     opds_feed = True
                                     print("OPDS type feed!")
+                                    feed_data = data
+                                    while feed_data:
+                                        feed_data = opds_fetch_more(feed_data)
+                                        if feed_data:
+                                            for i in feed_data.entries:
+                                                opds_entries.append(i)
+                                    for i in opds_entries:
+                                        data['entries'].append(i)
+                                    
 
 
                 for entry in data.entries:
@@ -425,10 +454,14 @@ def main():
                             post_name
                         )  # create list of posts which have not been returned by the feed
 
+
                 for post in existing_posts:
                     # remove blog posts no longer returned by the RSS feed
-                print("deleted", post)
+                    post_dir = os.path.join(output_dir, feed_name, post)
-                shutil.rmtree(os.path.join(feed_dir, slugify(post)))
+                    shutil.rmtree(post_dir)
+                    print("deleted", post_dir)
+            else:
+                print(feed_url, "is not or no longer a feed!")
 
     end = time.time()