Abstract api server feeds

apiserver/.gitignore

@@ -104,3 +104,6 @@ ENV/
 
 # DB
 db.sqlite3
+
+praw.ini
+data.db

apiserver/feed.py

@@ -0,0 +1,41 @@
+import logging
+logging.basicConfig(
+        format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
+        level=logging.INFO)
+
+import requests
+
+from feeds import hackernews
+
+READ_API = 'http://127.0.0.1:33843'
+
+def list():
+    feed = []
+    feed += [(x, 'hackernews') for x in hackernews.feed()]
+    return feed
+
+def get_article(url):
+    try:
+        r = requests.post(READ_API, data=dict(url=url), timeout=10)
+        if r.status_code != 200:
+            raise
+        return r.text
+    except BaseException as e:
+        logging.error('Problem getting article: {}'.format(str(e)))
+        return ''
+
+def update_story(story):
+    res = {}
+
+    if story['source'] == 'hackernews':
+        res = hackernews.story(story['ref'])
+    else:
+        return
+
+    if res:
+        story.update(res)
+    if story.get('url', '') and not story.get('text', ''):
+        if not story['url'].endswith('.pdf'):
+            story['text'] = get_article(story['url'])
+        else:
+            story['text'] = '<p>Unsupported article type.</p>'

apiserver/feeds/hackernews.py

@@ -54,7 +54,6 @@ def story(ref):
     s['date'] = r.get('created_at_i', 0)
     s['title'] = r.get('title', '')
     s['link'] = SITE_LINK(ref)
-    s['source'] = 'hackernews'
     s['url'] = r.get('url', '')
     s['comments'] = [comment(i) for i in r['children']]
     s['num_comments'] = comment_count(s) - 1

apiserver/server.py

@@ -7,17 +7,16 @@ import copy
 import threading
 import time
 import random
-import requests
 import shelve
 import string
 
-from feeds import hackernews
+import feed
+
 from flask import abort, Flask, request
 from flask_cors import CORS
 
 CACHE_LENGTH = 300
 DATA_FILE = 'data/data'
-READ_API = 'http://127.0.0.1:33843'
 
 news_index = 0
 
@@ -65,36 +64,25 @@ def new_id():
             nid = gen_rand_id()
     return nid
 
-def get_article(url):
-    try:
-        r = requests.post(READ_API, data=dict(url=url), timeout=10)
-        if r.status_code != 200:
-            raise
-        return r.text
-    except BaseException as e:
-        logging.error('Problem getting article: {}'.format(str(e)))
-        return ''
-
 try:
     while True:
         if news_index == 0:
-            feed = hackernews.feed()
+            feed_list = feed.list()
-            new_refs = [ref for ref in feed if ref not in news_list]
+            new_items = [(ref, source) for ref, source in feed_list if ref not in news_list]
-            for ref in new_refs:
+            for ref, source in new_items:
                 news_list.insert(0, ref)
                 nid = new_id()
                 news_ref_to_id[ref] = nid
-                news_cache[nid] = dict(id=nid, ref=ref)
+                news_cache[nid] = dict(id=nid, ref=ref, source=source)
 
-            if len(new_refs):
+            if len(new_items):
-                logging.info('Added {} new refs.'.format(len(new_refs)))
+                logging.info('Added {} new refs.'.format(len(new_items)))
 
             while len(news_list) > CACHE_LENGTH:
                 old_ref = news_list.pop()
                 old_story = news_cache.pop(news_ref_to_id[old_ref])
                 old_id = news_ref_to_id.pop(old_ref)
                 logging.info('Removed ref {} id {}.'.format(old_ref, old_id))
-                if old_story and old_id:
                 with shelve.open(DATA_FILE) as db:
                     db[old_id] = old_story
 
@@ -102,14 +90,7 @@ try:
             update_ref = news_list[news_index]
             update_id = news_ref_to_id[update_ref]
             news_story = news_cache[update_id]
-            story = hackernews.story(update_ref)
+            feed.update_story(news_story)
-            if story:
-                news_story.update(story)
-            if news_story.get('url', '') and not news_story.get('text', ''):
-                if not news_story['url'].endswith('.pdf'):
-                    news_story['text'] = get_article(news_story['url'])
-                else:
-                    news_story['text'] = '<p>Unsupported article type.</p>'
 
         time.sleep(1)