try

2025-05-12 17:04:19 +10:00
parent 8a24a93878
commit e974bd1262
1 changed files with 67 additions and 88 deletions
@@ -8,18 +8,18 @@ import sys
 import fcntl
 import time
 from datetime import datetime, timedelta, timezone
-import tweepy
 from openai import OpenAI
-from foodie_utils import post_tweet, AUTHORS, SUMMARY_MODEL, load_json_file, check_author_rate_limit, update_system_activity
+from foodie_utils import AUTHORS, SUMMARY_MODEL, load_json_file, save_json_file, update_system_activity
 from foodie_config import X_API_CREDENTIALS, RECENT_POSTS_FILE
 from dotenv import load_dotenv
+import shutil

 load_dotenv()
-# Output file for weekly thread content
-WEEKLY_THREADS_FILE = "/home/shane/foodie_automator/weekly_threads.json"
+
 SCRIPT_NAME = "foodie_weekly_thread"
 LOCK_FILE = "/home/shane/foodie_automator/locks/foodie_weekly_thread.lock"
 LOG_FILE = "/home/shane/foodie_automator/logs/foodie_weekly_thread.log"
+WEEKLY_THREADS_FILE = "/home/shane/foodie_automator/weekly_threads.json"
 LOG_PRUNE_DAYS = 30
 MAX_RETRIES = 3
 RETRY_BACKOFF = 2
@@ -50,7 +50,7 @@ def setup_logging():
            with open(LOG_FILE, 'w') as f:
                f.writelines(pruned_lines)
        
-        logging.basicBasic(
+        logging.basicConfig(
            filename=LOG_FILE,
            level=logging.INFO,
            format='%(asctime)s - %(levelname)s - %(message)s',
@@ -59,7 +59,7 @@ def setup_logging():
        console_handler = logging.StreamHandler()
        console_handler.setFormatter(logging.Formatter('%(asctime)s - %(levelname)s - %(message)s'))
        logging.getLogger().addHandler(console_handler)
-        logging.getLogger("tweepy").setLevel(logging.WARNING)
+        logging.getLogger("openai").setLevel(logging.WARNING)
        logging.info("Logging initialized for foodie_weekly_thread.py")
    except Exception as e:
        print(f"Failed to setup logging: {e}")
@@ -81,7 +81,7 @@ def acquire_lock():
 def signal_handler(sig, frame):
    """Handle termination signals gracefully."""
    logging.info("Received termination signal, marking script as stopped...")
-    update_system_activity(SCRIPT_NAME, "stopped")  # Added to mark as stopped
+    update_system_activity(SCRIPT_NAME, "stopped")
    sys.exit(0)

 signal.signal(signal.SIGTERM, signal_handler)
@@ -97,82 +97,6 @@ except Exception as e:
    logging.error(f"Failed to initialize OpenAI client: {e}", exc_info=True)
    sys.exit(1)

-def validate_twitter_credentials():
-    """Validate Twitter API credentials for all authors."""
-    logging.info("Validating Twitter API credentials for all authors")
-    valid_credentials = []
-    for author in AUTHORS:
-        for attempt in range(MAX_RETRIES):
-            try:
-                remaining, reset = get_x_rate_limit_status(author)
-                if remaining is not None and reset is not None:
-                    logging.info(f"Credentials valid for {author['username']} (handle: {X_API_CREDENTIALS[author['username']]['x_username']})")
-                    valid_credentials.append(X_API_CREDENTIALS[author['username']])
-                    break
-                else:
-                    logging.error(f"Rate limit check failed for {author['username']} (attempt {attempt + 1})")
-                    if attempt < MAX_RETRIES - 1:
-                        time.sleep(RETRY_BACKOFF * (2 ** attempt))
-            except Exception as e:
-                logging.error(f"Failed to validate credentials for {author['username']} (attempt {attempt + 1}): {e}")
-                if attempt < MAX_RETRIES - 1:
-                    time.sleep(RETRY_BACKOFF * (2 ** attempt))
-        else:
-            logging.error(f"Credentials invalid for {author['username']} after {MAX_RETRIES} attempts")
-    if not valid_credentials:
-        logging.error("No valid Twitter credentials found for any author")
-        raise ValueError("No valid Twitter credentials found")
-    return valid_credentials
-
-def load_recent_posts():
-    """Load and deduplicate posts from recent_posts.json."""
-    logging.debug(f"Attempting to load posts from {RECENT_POSTS_FILE}")
-    posts = load_json_file(RECENT_POSTS_FILE)
-    
-    if not posts:
-        logging.warning(f"No valid posts loaded from {RECENT_POSTS_FILE}")
-        return []
-    
-    # Deduplicate posts
-    unique_posts = {}
-    for post in posts:
-        try:
-            required_fields = ["title", "url", "author_username", "timestamp"]
-            if not all(key in post for key in required_fields):
-                logging.warning(f"Skipping invalid post: missing fields {post}")
-                continue
-            datetime.fromisoformat(post["timestamp"].replace('Z', '+00:00'))
-            key = (post["title"], post["url"], post["author_username"])
-            if key not in unique_posts:
-                unique_posts[key] = post
-            else:
-                logging.debug(f"Skipping duplicate post: {post['title']}")
-        except (KeyError, ValueError) as e:
-            logging.warning(f"Skipping post due to invalid format: {e}")
-            continue
-    
-    deduped_posts = list(unique_posts.values())
-    logging.info(f"Loaded {len(deduped_posts)} unique posts from {RECENT_POSTS_FILE}")
-    return deduped_posts
-
-def filter_posts_for_week(posts, start_date, end_date):
-    """Filter posts within the given week range."""
-    filtered_posts = []
-    for post in posts:
-        try:
-            post_date = datetime.fromisoformat(post["timestamp"])
-            logging.debug(f"Checking post: title={post['title']}, timestamp={post_date}, in range {start_date} to {end_date}")
-            if start_date <= post_date <= end_date:
-                filtered_posts.append(post)
-                logging.debug(f"Included post: {post['title']}")
-            else:
-                logging.debug(f"Excluded post: {post['title']} (timestamp {post_date} outside range)")
-        except (KeyError, ValueError) as e:
-            logging.warning(f"Skipping post due to invalid format: {e}")
-            continue
-    logging.info(f"Filtered to {len(filtered_posts)} posts for the week")
-    return filtered_posts
-
 def generate_intro_tweet(author):
    """Generate an intro tweet for the weekly thread."""
    credentials = next((cred for cred in X_API_CREDENTIALS if cred["username"] == author["username"]), None)
@@ -267,7 +191,56 @@ def generate_final_cta(author):
                logging.info(f"Using fallback final CTA tweet: {fallback}")
                return fallback

-def post_weekly_thread():
+def load_recent_posts():
+    """Load and deduplicate posts from recent_posts.json."""
+    logging.debug(f"Attempting to load posts from {RECENT_POSTS_FILE}")
+    posts = load_json_file(RECENT_POSTS_FILE)
+    
+    if not posts:
+        logging.warning(f"No valid posts loaded from {RECENT_POSTS_FILE}")
+        return []
+    
+    # Deduplicate posts
+    unique_posts = {}
+    for post in posts:
+        try:
+            required_fields = ["title", "url", "author_username", "timestamp"]
+            if not all(key in post for key in required_fields):
+                logging.warning(f"Skipping invalid post: missing fields {post}")
+                continue
+            datetime.fromisoformat(post["timestamp"].replace('Z', '+00:00'))
+            key = (post["title"], post["url"], post["author_username"])
+            if key not in unique_posts:
+                unique_posts[key] = post
+            else:
+                logging.debug(f"Skipping duplicate post: {post['title']}")
+        except (KeyError, ValueError) as e:
+            logging.warning(f"Skipping post due to invalid format: {e}")
+            continue
+    
+    deduped_posts = list(unique_posts.values())
+    logging.info(f"Loaded {len(deduped_posts)} unique posts from {RECENT_POSTS_FILE}")
+    return deduped_posts
+
+def filter_posts_for_week(posts, start_date, end_date):
+    """Filter posts within the given week range."""
+    filtered_posts = []
+    for post in posts:
+        try:
+            post_date = datetime.fromisoformat(post["timestamp"])
+            logging.debug(f"Checking post: title={post['title']}, timestamp={post_date}, in range {start_date} to {end_date}")
+            if start_date <= post_date <= end_date:
+                filtered_posts.append(post)
+                logging.debug(f"Included post: {post['title']}")
+            else:
+                logging.debug(f"Excluded post: {post['title']} (timestamp {post_date} outside range)")
+        except (KeyError, ValueError) as e:
+            logging.warning(f"Skipping post due to invalid format: {e}")
+            continue
+    logging.info(f"Filtered to {len(filtered_posts)} posts for the week")
+    return filtered_posts
+
+def generate_weekly_thread():
    """Generate weekly thread content for each author and save to file on Mondays."""
    logging.info("Starting foodie_weekly_thread.py")
    
@@ -298,7 +271,7 @@ def post_weekly_thread():
    
    weekly_posts = filter_posts_for_week(deduped_posts, start_date, end_date)
    if not weekly_posts:
-        logging.warning(f"No posts found within the week range {start_date} to {end_date}, exiting post_weekly_thread")
+        logging.warning(f"No posts found within the week range {start_date} to {end_date}, exiting generate_weekly_thread")
        return
    
    # Group posts by author
@@ -369,6 +342,14 @@ def post_weekly_thread():
    # Save thread content to file, overwriting any existing content
    if thread_content:
        try:
+            # Backup existing file before overwriting
+            if os.path.exists(WEEKLY_THREADS_FILE):
+                backup_dir = "/home/shane/foodie_automator/backups"
+                os.makedirs(backup_dir, exist_ok=True)
+                backup_file = f"{backup_dir}/weekly_threads_{timestamp.replace(':', '-')}.json"
+                shutil.copy(WEEKLY_THREADS_FILE, backup_file)
+                logging.info(f"Backed up existing {WEEKLY_THREADS_FILE} to {backup_file}")
+                
            # Save new thread content, overwriting the file
            thread_data = {
                "week_start": start_date.isoformat(),
@@ -392,9 +373,7 @@ def main():
        lock_fd = acquire_lock()
        setup_logging()
        update_system_activity(SCRIPT_NAME, "running", os.getpid())  # Record start
-        # Skip Twitter credentials validation since we're not posting
-        # validate_twitter_credentials()
-        post_weekly_thread()
+        generate_weekly_thread()
        update_system_activity(SCRIPT_NAME, "stopped")  # Record stop
    except Exception as e:
        logging.error(f"Fatal error in main: {e}", exc_info=True)