unify all handlers into one for clarity and proper text message handling

(first try)

delojza.py

@@ -22,8 +22,7 @@ import pytumblr
 import requests
 import telegram
 import youtube_dl
-from telegram import MessageEntity
+from telegram.ext import Updater, CommandHandler, MessageHandler
-from telegram.ext import Updater, CommandHandler, MessageHandler, Filters
 from youtube_dl.version import __version__ as YTDL_VERSION
 
 
@@ -58,13 +57,7 @@ class DelojzaBot:
 
         dp.add_handler(CommandHandler("start", self.tg_start))
         dp.add_error_handler(self.tg_error)
-        self.tg_url_handler = MessageHandler(Filters.entity(MessageEntity.URL), self.tg_handle_url)
+        dp.add_handler(MessageHandler(None, self.tg_handle))
-        dp.add_handler(self.tg_url_handler)
-        self.tg_rest_handler = MessageHandler(Filters.photo | Filters.video | Filters.video_note |
-                                              Filters.audio | Filters.voice | Filters.document, self.tg_handle_rest)
-        dp.add_handler(self.tg_rest_handler)
-        dp.add_handler(MessageHandler(Filters.entity(MessageEntity.HASHTAG), self.tg_handle_hashtag))
-        dp.add_handler(MessageHandler(Filters.text, self.tg_handle_text))
         dp.add_handler(CommandHandler("stats", self.tg_stats))
         dp.add_handler(CommandHandler("orphans", self.tg_orphan))
         dp.add_handler(CommandHandler("orphans_full", self.tg_orphan_full))
@@ -251,7 +244,7 @@ class DelojzaBot:
                     hashtags[i] = "PRAS"
         return hashtags
 
-    def get_hashtags(self, message):
+    def _get_hashtags(self, message):
         hashtags = self.extract_hashtags(message)
         if len(hashtags) == 0 and self.last_hashtags is not None:
             user, chat, ts, last_hashtags = self.last_hashtags
@@ -259,17 +252,6 @@ class DelojzaBot:
                 hashtags = last_hashtags
         return hashtags
 
-    def tg_handle_hashtag(self, bot, update):
-        hashtags = self.extract_hashtags(update.message)
-
-        self.handle_text(update.message.reply_to_message or update.message, hashtags)
-
-        if update.message.reply_to_message:
-            self.handle_tg_message(update.message.reply_to_message, bot, hashtags)
-            self.handle_urls(update.message.reply_to_message, hashtags)
-        else:
-            self.last_hashtags = update.message.from_user, update.message.chat, datetime.now(), hashtags
-
     def handle_text(self, message, hashtags):
         if len(hashtags) == 0 or hashtags[0] not in ('TEXT', 'TXT'):
             return
@@ -342,25 +324,6 @@ class DelojzaBot:
             if "Timed out" not in str(exc_value):
                 message.reply_text("Something is FUCKED: [{}] {}".format(exc_type, exc_value))
 
-    def tg_handle_url(self, _, update):
-        self.handle_urls(update.message, self.get_hashtags(update.message))
-
-    def handle_urls(self, message, hashtags):
-        urls = list(map(lambda e: message.parse_entity(e),
-                        filter(lambda e: e.type == 'url', message.entities)))
-        ytdl_urls = [url for url in urls if self.ytdl_can(url)]
-        normal_urls = [url for url in urls if not self.ytdl_can(url)]
-        if len(ytdl_urls) > 0:
-            self.handle(ytdl_urls, message, hashtags, self.download_ytdl)
-        if len(normal_urls) > 0:
-            file_urls = [url for url in normal_urls if
-                         "text" not in requests.head(url).headers.get("Content-Type", "text")]
-            if len(file_urls) > 0:
-                self.handle(file_urls, message, hashtags, self.download_raw)
-
-    def tg_handle_rest(self, bot, update):
-        self.handle_tg_message(update.message, bot, self.get_hashtags(update.message))
-
     def handle_tg_message(self, message, bot, hashtag):
         file, filetitle, tumblr = None, None, False
         if len(message.photo) > 0:
@@ -383,17 +346,43 @@ class DelojzaBot:
             url = bot.getFile(file).file_path
             self.handle([url], message, hashtag, self.download_raw, filetitle=filetitle)
 
-    def tg_handle_text(self, _, update):
+    def handle_urls(self, message, hashtags):
-        if self.markov:
+        urls = list(map(lambda e: message.parse_entity(e),
-            self.markov.add_to_corpus(update.message.text)
+                        filter(lambda e: e.type == 'url', message.entities)))
+        ytdl_urls = [url for url in urls if self.ytdl_can(url)]
+        normal_urls = [url for url in urls if not self.ytdl_can(url)]
+        if len(ytdl_urls) > 0:
+            self.handle(ytdl_urls, message, hashtags, self.download_ytdl)
+        if len(normal_urls) > 0:
+            file_urls = [url for url in normal_urls if
+                         "text" not in requests.head(url).headers.get("Content-Type", "text")]
+            if len(file_urls) > 0:
+                self.handle(file_urls, message, hashtags, self.download_raw)
 
-    def tag_dirs(self):
+    def tg_handle(self, bot, update):
+        self.handle_urls(update.message, self._get_hashtags(update.message))
+        self.handle_tg_message(update.message, bot, self._get_hashtags(update.message))
+
+        hashtags = self.extract_hashtags(update.message)
+        if len(hashtags) > 0:
+            self.handle_text(update.message.reply_to_message or update.message, hashtags)
+
+            if update.message.reply_to_message:
+                self.handle_tg_message(update.message.reply_to_message, bot, hashtags)
+                self.handle_urls(update.message.reply_to_message, hashtags)
+            else:
+                self.last_hashtags = update.message.from_user, update.message.chat, datetime.now(), hashtags
+        else:
+            if self.markov:
+                self.markov.add_to_corpus(update.message.text)
+
+    def _get_tag_dirs(self):
         return list(filter(lambda x: x.upper() == x,
                            filter(lambda directory: os.path.isdir(os.path.join(self.out_dir, directory)),
                                   os.listdir(self.out_dir))))
 
     def tg_stats(self, _, update):
-        tag_dirs = self.tag_dirs()
+        tag_dirs = self._get_tag_dirs()
         reply = "Total number of tags: {}\n\n".format(len(tag_dirs))
         counts = [(directory, os.listdir(os.path.join(self.out_dir, directory))) for directory in tag_dirs]
         counts.sort(key=itemgetter(0))
@@ -420,10 +409,9 @@ class DelojzaBot:
             reply += "\nFollowing tags are orphans: " + ", ".join(map(itemgetter(0), orphans))
         update.message.reply_text(reply, parse_mode=telegram.ParseMode.HTML)
 
-    def orphans(self):
+    def _get_orphan_tags(self):
         result = []
-        tag_dirs = self.tag_dirs()
+        for directory in self._get_tag_dirs():
-        for directory in tag_dirs:
             files = os.listdir(os.path.join(self.out_dir, directory))
             if len(files) == 1:
                 result.append((directory, files[0]))
@@ -432,7 +420,7 @@ class DelojzaBot:
         return sorted(result, key=itemgetter(0))
 
     def tg_orphan(self, _, update):
-        orphans = self.orphans()
+        orphans = self._get_orphan_tags()
         if len(orphans) == 0:
             update.message.reply_text("Good job, no orphan tags!")
         else:
@@ -440,7 +428,7 @@ class DelojzaBot:
                                       ", ".join(map(itemgetter(0), orphans)))
 
     def tg_orphan_full(self, _, update):
-        orphans = self.orphans()
+        orphans = self._get_orphan_tags()
         if len(orphans) == 0:
             update.message.reply_text("Good job, no orphan tags!")
         else:
@@ -497,7 +485,7 @@ class DelojzaBot:
             if update is not None:
                 default = "Mmmm, I like it..."
                 update.message.reply_text((self.markov.make_sentence(tries=100) if self.markov else default) or default)
-                self.tg_handle_rest(bot, update)
+                self.tg_handle(bot, update)
         else:
             if update is not None:
                 update.message.reply_text("Something is fucked: %s" % error)