slight restructuring around tracking our posts

src/ttweetqueue.py

@@ -15,10 +15,13 @@ class TalentTweetQueue:
         TalentTweetQueue.instance = self
         self.queue_path = util.get_queue_path()
         self.queue_backup_path = util.get_queue_backup_path()
+        self.current_ttweet_path = f'{util.get_project_dir()}/_current_ttweet.txt'
+        self.finished_ttweets_path = f'{util.get_project_dir()}/finished_ttweets.txt'
+        self.is_good = True
+        self.__sorted = False
         self.finished_user_timestamps = dict()
         self.ttweets_dict = dict()
-        self.good = False # if true, overwrite queue.txt on destruction
-        self.__sorted = False
+        self.finished_ttweets = list()
 
         ## file check, backup copy
         if os.path.exists(self.queue_backup_path):
@@ -30,59 +33,104 @@ class TalentTweetQueue:
 
         ## initialize structures
         # user timestamps
-        with open(self.queue_path, 'r') as f:
-            for line in f:
-                tokens = line.split()
-                if len(tokens) == 0: continue
+        try:
+            with open(self.queue_path, 'r') as f:
+                for line in f:
+                    tokens = line.split()
+                    if len(tokens) == 0: continue
 
-                if tokens[0][0] != '#':
-                    print(f'Stopped finding user timestamps at {line}')
-                    # reached end of accounts list
-                    break
-                if tokens[2] != '-1':
-                    self.finished_user_timestamps[int(tokens[1])] = float(tokens[2])
-        
-        # tweets
-        with open(self.queue_path, 'r') as f: # reset seek head
-            # Get existing queued TalentTweets
-            for line in f:
-                tokens = line.split()
-                if len(tokens) == 0 or tokens[0][0] == '#':
-                    continue
-                ttweet = tt.TalentTweet.deserialize(line)
-                self.ttweets_dict[ttweet.tweet_id] = ttweet
-            print(f'Found {len(self.finished_user_timestamps)} scraped accounts and {len(self.ttweets_dict)} tweets in queue.')
+                    if tokens[0][0] != '#':
+                        print(f'Stopped finding user timestamps at {line}')
+                        # reached end of accounts list
+                        break
+                    if tokens[2] != '-1':
+                        self.finished_user_timestamps[int(tokens[1])] = float(tokens[2])
+        except: pass
+        # ttweets
+        try:
+            with open(self.queue_path, 'r') as f: # reset seek head
+                # Get existing queued TalentTweets
+                for line in f:
+                    tokens = line.split()
+                    if len(tokens) == 0 or tokens[0][0] == '#':
+                        continue
+                    ttweet = tt.TalentTweet.deserialize(line)
+                    self.ttweets_dict[ttweet.tweet_id] = ttweet
+                print(f'Found {len(self.finished_user_timestamps)} scraped accounts and {len(self.ttweets_dict)} tweets in queue.')
+        except: pass
+        # finished ttweets
+        try:
+            with open(self.finished_ttweets_path, 'r') as f:
+                for line in f:
+                    self.finished_ttweets.append(int(line))
+        except: pass
+
+
+    def is_empty(self):
+        return self.get_count() <= 0
 
     def add_ttweet(self, ttweet):
         self.__sorted = False
         self.ttweets_dict[ttweet.tweet_id] = ttweet
 
     def get_ttweet(self, id):
-         return self.ttweets_dict[id]
+        return self.ttweets_dict[id]
+
+    def get_next_ttweet(self):
+        if os.path.exists(self.current_ttweet_path):
+            with open(self.current_ttweet_path, 'r') as f:
+                return tt.TalentTweet.deserialize(f.readline())
+
+        self.__sort_ttweets_dict()
+        key = list(self.ttweets_dict.keys())[0]
+        ttweet = self.ttweets_dict.pop(key)
+        with open(self.current_ttweet_path, 'w') as f:
+            f.write(ttweet.serialize())
+        self.is_good = False
+        return ttweet
     
-    def get_ttweets_dict(self):
-        self.__sort_ttweets_dict() if not self.__sorted else None
-        return self.ttweets_dict
+    def get_count(self):
+        return len(self.ttweets_dict)
+    
+    ## Call when the TalentTweet retrieved from get_next_ttweet() was
+    #  posted successfully.
+    def good(self):
+        with open(self.current_ttweet_path, 'r') as f:
+            ttweet = tt.TalentTweet.deserialize(f.readline())
+
+        self.add_finished_tweet(ttweet.tweet_id)
+        os.remove(self.current_ttweet_path)
+        self.save_file()
+        self.is_good = True
     
     # overwrite queue.txt
     def save_file(self):
+        shutil.copyfile(self.queue_path, self.queue_backup_path)
         self.__sort_ttweets_dict()
         with open(self.queue_path, 'w') as f:
             # write timestamps
             for (id, timestamp) in self.finished_user_timestamps.items():
                 f.write(f'# {id} {timestamp}\n')
+
             f.write('\n')
+
             # write sorted ttweets
             for ttweet in self.ttweets_dict.values():
                 f.write(ttweet.serialize() + '\n')
+
+    def add_finished_tweet(self, id):
+        self.finished_ttweets.append(id)
+        with open(self.finished_ttweets_path, 'a') as f:
+            f.write(f'{id}\n')
     
     def __sort_ttweets_dict(self):
-        self.ttweets_dict = dict(sorted(self.ttweets_dict.items()))
+        if not self.__sorted:
+            self.ttweets_dict = dict(sorted(self.ttweets_dict.items()))
         self.__sorted = True
     
     # destructor
     def __del__(self):
-        if self.good:
+        if self.is_good:
             print('Ended in good state, deleting backup queue...')
             os.remove(self.queue_backup_path)
         else: