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 
 
-        prefered_name = "{}-{}.{}".format(fn, slugify(entry['title']), ft)
+        if 'image' in fn:
-
+            prefered_name = "{}-{}.{}".format(fn, slugify(entry['title']), ft)
-        grab_media(post_dir, item['href'], prefered_name)
+            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()
@@ -373,62 +391,77 @@ 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_feed = False
-            for i in data.feed['links']:
+                opds_entries = []
-                if i['rel'] == 'self':
+                if 'links' in data.feed:
-                    if 'opds' in i['type']:
+                    for i in data.feed['links']:
-                        opds_feed = True
+                            if i['rel'] == 'self':
-                        print("OPDS type feed!")
+                                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:
+                for entry in data.entries:
-                # if 'tags' in entry:
+                    # if 'tags' in entry:
-                #     for tag in entry.tags:
+                    #     for tag in entry.tags:
-                #        for x in ['lumbung.space', 'D15', 'lumbung']:
+                    #        for x in ['lumbung.space', 'D15', 'lumbung']:
-                #            if x in tag['term']:
+                    #            if x in tag['term']:
-                #                print(entry.title)
+                    #                print(entry.title)
-                entry["feed_name"] = feed_name
+                    entry["feed_name"] = feed_name
 
-                post_name = slugify(entry.title)
+                    post_name = slugify(entry.title)
 
-                # pixelfed returns the whole post text as the post name. max
+                    # pixelfed returns the whole post text as the post name. max
-                # filename length is 255 on many systems. here we're shortening
+                    # filename length is 255 on many systems. here we're shortening
-                # the name and adding a hash to it to avoid a conflict in a
+                    # the name and adding a hash to it to avoid a conflict in a
-                # situation where 2 posts start with exactly the same text.
+                    # situation where 2 posts start with exactly the same text.
-                if len(post_name) > 150:
+                    if len(post_name) > 150:
-                    post_hash = md5(bytes(post_name, "utf-8"))
+                        post_hash = md5(bytes(post_name, "utf-8"))
-                    post_name = post_name[:150] + "-" + post_hash.hexdigest()
+                        post_name = post_name[:150] + "-" + post_hash.hexdigest()
 
-                if opds_feed:
-                    entry['opds'] = True
-                    #format: Beyond-Debiasing-Report_Online-75535a4886e3
-                    post_name = slugify(entry['title'])+'-'+entry['id'].split('-')[-1]
-
-                post_dir = os.path.join(output_dir, feed_name, post_name)
-
-                if post_name not in existing_posts:
-                    # if there is a blog entry we dont already have, make it
                     if opds_feed:
-                        create_opds_post(post_dir, entry)
+                        entry['opds'] = True
-                    else:
+                        #format: Beyond-Debiasing-Report_Online-75535a4886e3
-                        create_post(post_dir, entry)
+                        post_name = slugify(entry['title'])+'-'+entry['id'].split('-')[-1]
 
-                elif post_name in existing_posts:
+                    post_dir = os.path.join(output_dir, feed_name, post_name)
-                    # if we already have it, update it
-                    if opds_feed:
-                        create_opds_post(post_dir, entry)
-                    else:
-                        create_post(post_dir, entry)
-                    existing_posts.remove(
-                        post_name
-                    )  # create list of posts which have not been returned by the feed
 
-            for post in existing_posts:
+                    if post_name not in existing_posts:
-                # remove blog posts no longer returned by the RSS feed
+                        # if there is a blog entry we dont already have, make it
-                print("deleted", post)
+                        if opds_feed:
-                shutil.rmtree(os.path.join(feed_dir, slugify(post)))
+                            create_opds_post(post_dir, entry)
+                        else:
+                            create_post(post_dir, entry)
+
+                    elif post_name in existing_posts:
+                        # if we already have it, update it
+                        if opds_feed:
+                            create_opds_post(post_dir, entry)
+                        else:
+                            create_post(post_dir, entry)
+                        existing_posts.remove(
+                            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
+                    post_dir = os.path.join(output_dir, feed_name, post)
+                    shutil.rmtree(post_dir)
+                    print("deleted", post_dir)
+            else:
+                print(feed_url, "is not or no longer a feed!")
 
     end = time.time()