This commit is contained in:
2025-11-10 19:21:58 +01:00
parent ac5738c29d
commit bcd0a10576

View File

@@ -38,6 +38,7 @@ class Config:
# Newsletter
MAX_ARTICLES = int(os.getenv('NEWSLETTER_MAX_ARTICLES', '10'))
HOURS_LOOKBACK = int(os.getenv('NEWSLETTER_HOURS_LOOKBACK', '24'))
WEBSITE_URL = os.getenv('WEBSITE_URL', 'http://localhost:3000')
@@ -48,19 +49,51 @@ articles_collection = db['articles']
subscribers_collection = db['subscribers']
def get_latest_articles(max_articles=10):
def get_latest_articles(max_articles=10, hours=24):
"""
Get latest articles with AI summaries from database
Get latest articles with AI summaries from database (from today only)
Args:
max_articles: Maximum number of articles to return
hours: Number of hours to look back (default 24)
Returns:
list: Articles with summaries
list: Articles with summaries published today
"""
cursor = articles_collection.find(
{'summary': {'$exists': True, '$ne': None}}
).sort('created_at', -1).limit(max_articles)
from datetime import timedelta
# Calculate cutoff time (e.g., 24 hours ago)
cutoff_time = datetime.utcnow() - timedelta(hours=hours)
# Get start of today (00:00:00 UTC)
today_start = datetime.utcnow().replace(hour=0, minute=0, second=0, microsecond=0)
# Query for articles with summaries published today OR created today
# This ensures we only get fresh articles from today
cursor = articles_collection.find({
'summary': {'$exists': True, '$ne': None},
'$or': [
# Articles published today (if published_at is available)
{'published_at': {'$gte': today_start}},
# Articles created today (fallback if published_at is missing)
{'created_at': {'$gte': today_start}}
]
}).sort('created_at', -1).limit(max_articles)
articles = []
for doc in cursor:
# Double-check the date to ensure it's from today
published_at = doc.get('published_at')
created_at = doc.get('created_at')
# Skip if both dates are old (extra safety check)
if published_at and isinstance(published_at, datetime):
if published_at < today_start:
continue
elif created_at and isinstance(created_at, datetime):
if created_at < today_start:
continue
articles.append({
'title': doc.get('title', ''),
'author': doc.get('author'),
@@ -179,20 +212,23 @@ def send_newsletter(max_articles=None, test_email=None):
'error': 'Email credentials not configured'
}
# Get articles
# Get articles from today only
max_articles = max_articles or Config.MAX_ARTICLES
print(f"\nFetching latest {max_articles} articles with AI summaries...")
articles = get_latest_articles(max_articles)
today_date = datetime.now().strftime('%B %d, %Y')
print(f"\nFetching articles published TODAY ({today_date})...")
print(f" Max articles: {max_articles}")
articles = get_latest_articles(max_articles, hours=Config.HOURS_LOOKBACK)
if not articles:
print("❌ No articles with summaries found")
print(" Run the crawler with Ollama enabled first")
print("❌ No articles from today with summaries found")
print(f" No articles published today ({today_date})")
print(" Run the crawler with Ollama enabled to get fresh content")
return {
'success': False,
'error': 'No articles with summaries'
'error': f'No articles published today'
}
print(f"✓ Found {len(articles)} articles")
print(f"✓ Found {len(articles)} recent article(s)")
# Get subscribers
if test_email:
@@ -255,21 +291,24 @@ def send_newsletter(max_articles=None, test_email=None):
}
def preview_newsletter(max_articles=None):
def preview_newsletter(max_articles=None, hours=None):
"""
Generate newsletter HTML for preview (doesn't send)
Args:
max_articles: Maximum number of articles to include
hours: Hours to look back (default from config)
Returns:
str: HTML content
"""
max_articles = max_articles or Config.MAX_ARTICLES
articles = get_latest_articles(max_articles)
hours = hours or Config.HOURS_LOOKBACK
articles = get_latest_articles(max_articles, hours=hours)
if not articles:
return "<h1>No articles with summaries found</h1><p>Run the crawler with Ollama enabled first.</p>"
today_date = datetime.now().strftime('%B %d, %Y')
return f"<h1>No articles from today found</h1><p>No articles published today ({today_date}). Run the crawler with Ollama enabled to get fresh content.</p>"
return render_newsletter_html(articles)