Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -21,22 +21,28 @@ search_tool = DuckDuckGoSearchTool()
|
|
| 21 |
@tool
|
| 22 |
def get_latest_news() -> Dict[str, List[Dict]]:
|
| 23 |
"""
|
| 24 |
-
Tool returns latest news from major news outlets using RSS feeds.
|
| 25 |
-
Uses verified RSS feed URLs and detailed error reporting.
|
| 26 |
Returns:
|
| 27 |
Dict[str, List[Dict]]: A dictionary where keys are news sources and values are lists of news items.
|
| 28 |
"""
|
| 29 |
-
# Verified working RSS feeds
|
| 30 |
rss_feeds = {
|
| 31 |
-
"Reuters": {
|
| 32 |
-
"Top News": "https://feeds.reuters.com/reuters/topNews",
|
| 33 |
-
"Business": "https://feeds.reuters.com/reuters/businessNews",
|
| 34 |
-
"World": "https://feeds.reuters.com/reuters/worldNews"
|
| 35 |
-
},
|
| 36 |
"NPR": {
|
| 37 |
"News": "https://feeds.npr.org/1001/rss.xml",
|
| 38 |
"World": "https://feeds.npr.org/1004/rss.xml",
|
|
|
|
| 39 |
"Business": "https://feeds.npr.org/1006/rss.xml"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
}
|
| 41 |
}
|
| 42 |
|
|
@@ -44,53 +50,54 @@ def get_latest_news() -> Dict[str, List[Dict]]:
|
|
| 44 |
|
| 45 |
for source, feeds in rss_feeds.items():
|
| 46 |
news_items[source] = []
|
|
|
|
| 47 |
|
| 48 |
for feed_name, feed_url in feeds.items():
|
| 49 |
try:
|
| 50 |
-
# Add modern browser headers
|
| 51 |
headers = {
|
| 52 |
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
|
| 53 |
'Accept': 'application/rss+xml,application/xml;q=0.9,*/*;q=0.8'
|
| 54 |
}
|
| 55 |
|
| 56 |
-
print(f"Fetching {
|
| 57 |
|
| 58 |
req = urllib.request.Request(feed_url, headers=headers)
|
| 59 |
response = urllib.request.urlopen(req, timeout=10)
|
| 60 |
xml_data = response.read().decode('utf-8')
|
| 61 |
|
| 62 |
-
# Parse XML
|
| 63 |
root = ET.fromstring(xml_data)
|
| 64 |
|
| 65 |
-
#
|
| 66 |
items = root.findall('.//item')
|
| 67 |
if not items:
|
| 68 |
-
items = root.findall('./channel/item')
|
| 69 |
|
| 70 |
-
|
|
|
|
| 71 |
title = item.find('title')
|
| 72 |
link = item.find('link')
|
| 73 |
pub_date = item.find('pubDate')
|
| 74 |
description = item.find('description')
|
| 75 |
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
|
|
|
|
|
|
| 85 |
|
| 86 |
-
print(f"
|
| 87 |
|
| 88 |
except Exception as e:
|
| 89 |
-
|
| 90 |
-
print(error_message) # Debug info
|
| 91 |
news_items[source].append({
|
| 92 |
'category': feed_name,
|
| 93 |
-
'title':
|
| 94 |
'link': '',
|
| 95 |
'published': '',
|
| 96 |
'summary': ''
|
|
|
|
| 21 |
@tool
|
| 22 |
def get_latest_news() -> Dict[str, List[Dict]]:
|
| 23 |
"""
|
| 24 |
+
Tool returns latest news from major news outlets using reliable RSS feeds.
|
|
|
|
| 25 |
Returns:
|
| 26 |
Dict[str, List[Dict]]: A dictionary where keys are news sources and values are lists of news items.
|
| 27 |
"""
|
|
|
|
| 28 |
rss_feeds = {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
"NPR": {
|
| 30 |
"News": "https://feeds.npr.org/1001/rss.xml",
|
| 31 |
"World": "https://feeds.npr.org/1004/rss.xml",
|
| 32 |
+
"Politics": "https://feeds.npr.org/1014/rss.xml",
|
| 33 |
"Business": "https://feeds.npr.org/1006/rss.xml"
|
| 34 |
+
},
|
| 35 |
+
"BBC": {
|
| 36 |
+
"Top Stories": "http://feeds.bbci.co.uk/news/rss.xml",
|
| 37 |
+
"World": "http://feeds.bbci.co.uk/news/world/rss.xml",
|
| 38 |
+
"Politics": "http://feeds.bbci.co.uk/news/politics/rss.xml",
|
| 39 |
+
"Business": "http://feeds.bbci.co.uk/news/business/rss.xml"
|
| 40 |
+
},
|
| 41 |
+
"ABC News": {
|
| 42 |
+
"Top Stories": "https://abcnews.go.com/abcnews/topstories",
|
| 43 |
+
"World News": "https://abcnews.go.com/abcnews/worldnewsheadlines",
|
| 44 |
+
"Politics": "https://abcnews.go.com/abcnews/politicsheadlines",
|
| 45 |
+
"Business": "https://abcnews.go.com/abcnews/moneyheadlines"
|
| 46 |
}
|
| 47 |
}
|
| 48 |
|
|
|
|
| 50 |
|
| 51 |
for source, feeds in rss_feeds.items():
|
| 52 |
news_items[source] = []
|
| 53 |
+
print(f"\nTrying source: {source}")
|
| 54 |
|
| 55 |
for feed_name, feed_url in feeds.items():
|
| 56 |
try:
|
|
|
|
| 57 |
headers = {
|
| 58 |
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
|
| 59 |
'Accept': 'application/rss+xml,application/xml;q=0.9,*/*;q=0.8'
|
| 60 |
}
|
| 61 |
|
| 62 |
+
print(f" Fetching {feed_name}...", end=" ")
|
| 63 |
|
| 64 |
req = urllib.request.Request(feed_url, headers=headers)
|
| 65 |
response = urllib.request.urlopen(req, timeout=10)
|
| 66 |
xml_data = response.read().decode('utf-8')
|
| 67 |
|
|
|
|
| 68 |
root = ET.fromstring(xml_data)
|
| 69 |
|
| 70 |
+
# Try different possible XML paths for items
|
| 71 |
items = root.findall('.//item')
|
| 72 |
if not items:
|
| 73 |
+
items = root.findall('./channel/item')
|
| 74 |
|
| 75 |
+
successful_items = 0
|
| 76 |
+
for item in items[:5]:
|
| 77 |
title = item.find('title')
|
| 78 |
link = item.find('link')
|
| 79 |
pub_date = item.find('pubDate')
|
| 80 |
description = item.find('description')
|
| 81 |
|
| 82 |
+
# Only add items that have at least a title
|
| 83 |
+
if title is not None and title.text:
|
| 84 |
+
news_item = {
|
| 85 |
+
'category': feed_name,
|
| 86 |
+
'title': title.text.strip(),
|
| 87 |
+
'link': link.text.strip() if link is not None and link.text else '',
|
| 88 |
+
'published': pub_date.text if pub_date is not None else '',
|
| 89 |
+
'summary': description.text[:200] + '...' if description is not None and description.text else ''
|
| 90 |
+
}
|
| 91 |
+
news_items[source].append(news_item)
|
| 92 |
+
successful_items += 1
|
| 93 |
|
| 94 |
+
print(f"Success! Found {successful_items} articles")
|
| 95 |
|
| 96 |
except Exception as e:
|
| 97 |
+
print(f"Failed: {str(e)}")
|
|
|
|
| 98 |
news_items[source].append({
|
| 99 |
'category': feed_name,
|
| 100 |
+
'title': f"Error fetching {feed_name} feed: {str(e)}",
|
| 101 |
'link': '',
|
| 102 |
'published': '',
|
| 103 |
'summary': ''
|