From 850b30e353e634cb2f7eb7113414197966770a9d Mon Sep 17 00:00:00 2001
From: Tanner Collin <git@tannercollin.com>
Date: Sat, 4 Jul 2020 00:25:41 +0000
Subject: [PATCH] Add requests timeouts and temporary logging

---
 apiserver/feed.py         | 25 ++++++++++++++++++++-----
 apiserver/feeds/tildes.py |  8 +++++++-
 apiserver/server.py       |  2 +-
 3 files changed, 28 insertions(+), 7 deletions(-)

diff --git a/apiserver/feed.py b/apiserver/feed.py
index 33d15b0..8682309 100644
--- a/apiserver/feed.py
+++ b/apiserver/feed.py
@@ -63,7 +63,7 @@ def get_first_image(text):
         first_img = soup.find('img')
         url = first_img['src']
         headers = {'User-Agent': 'Twitterbot/1.0'}
-        length = requests.get(url, headers=headers).headers['content-length']
+        length = requests.get(url, headers=headers, timeout=4).headers['content-length']
         if int(length) > 1000000: raise
         return url
     except:
@@ -72,9 +72,15 @@ def get_first_image(text):
 def get_content_type(url):
     try:
         headers = {'User-Agent': 'Twitterbot/1.0'}
-        return requests.get(url, headers=headers).headers['content-type']
+        return requests.get(url, headers=headers, timeout=2).headers['content-type']
     except:
-        return ''
+        pass
+
+    try:
+        headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:77.0) Gecko/20100101 Firefox/77.0'}
+        return requests.get(url, headers=headers, timeout=2).headers['content-type']
+    except:
+        return 'text/'
 
 def update_story(story, is_manual=False):
     res = {}
@@ -90,26 +96,35 @@ def update_story(story, is_manual=False):
     elif story['source'] == 'manual':
         res = manual.story(story['ref'])
 
+    logging.info('Got story')
+
     if res:
         story.update(res) # join dicts
     else:
-        logging.info('Article not ready yet')
+        logging.info('Story not ready yet')
         return False
 
+    logging.info('story joined')
+
     if story['date'] and not is_manual and story['date'] + TWO_DAYS < time.time():
-        logging.info('Article too old, removing')
+        logging.info('Story too old, removing')
         return False
 
+    logging.info('story age good')
+
     if story.get('url', '') and not story.get('text', ''):
+        logging.info('inside if')
         if not get_content_type(story['url']).startswith('text/'):
             logging.info('URL invalid file type / content type:')
             logging.info(story['url'])
             return False
+        logging.info('content type good')
 
         if any([domain in story['url'] for domain in INVALID_DOMAINS]):
             logging.info('URL invalid domain:')
             logging.info(story['url'])
             return False
+        logging.info('domain good')
 
         logging.info('Getting article ' + story['url'])
         story['text'] = get_article(story['url'])
diff --git a/apiserver/feeds/tildes.py b/apiserver/feeds/tildes.py
index 83a2e86..920c751 100644
--- a/apiserver/feeds/tildes.py
+++ b/apiserver/feeds/tildes.py
@@ -73,10 +73,14 @@ def story(ref):
         html = api(API_ITEM(ref))
     if not html: return False
 
+    logging.info('Got Tildes html')
+
     soup = BeautifulSoup(html, features='html.parser')
     a = soup.find('article', class_='topic-full')
     if a is None: return False
 
+    logging.info('Got article tag')
+
     h = a.find('header')
     lu = h.find('a', class_='link-user')
 
@@ -102,6 +106,8 @@ def story(ref):
     ch = a.find('header', class_='topic-comments-header')
     s['num_comments'] = int(ch.h2.string.split(' ')[0]) if ch else 0
 
+    logging.info('Parsed html data')
+
     if s['score'] < 8 and s['num_comments'] < 6:
         return False
 
@@ -122,7 +128,7 @@ if __name__ == '__main__':
     #print(self_post)
     #li_comment = story('gqx')
     #print(li_comment)
-    broken = story('n03')
+    broken = story('q4y')
     print(broken)
 
     # make sure there's no self-reference
diff --git a/apiserver/server.py b/apiserver/server.py
index 659ced9..f7716d4 100644
--- a/apiserver/server.py
+++ b/apiserver/server.py
@@ -184,7 +184,7 @@ def feed_thread():
                     database.del_ref(item['ref'])
                     logging.info('Removed ref {}'.format(item['ref']))
 
-            gevent.sleep(6)
+            gevent.sleep(60)
 
             news_index += 1
             if news_index == FEED_LENGTH: news_index = 0