Spaces:
Sleeping
Sleeping
Update src/streamlit_app.py
Browse files- src/streamlit_app.py +81 -21
src/streamlit_app.py
CHANGED
|
@@ -13,9 +13,7 @@ import feedparser
|
|
| 13 |
import requests
|
| 14 |
from bs4 import BeautifulSoup
|
| 15 |
from newspaper import Article # Make sure newspaper3k and lxml_html_clean are installed
|
| 16 |
-
# Removed torch, transformers imports for now to simplify
|
| 17 |
import time
|
| 18 |
-
# Removed geocoder import for now to simplify
|
| 19 |
import json
|
| 20 |
import datetime
|
| 21 |
|
|
@@ -33,22 +31,22 @@ TELUGU_CATEGORIES = {
|
|
| 33 |
"వైరల్": "Viral",
|
| 34 |
}
|
| 35 |
|
| 36 |
-
# --- RSS Feed URLs ---
|
| 37 |
# IMPORTANT: These URLs might be outdated or not provide full content.
|
| 38 |
# You might need to find current, active RSS feeds for Telugu news sources.
|
| 39 |
-
# Many news sites change their RSS feeds frequently or don't offer comprehensive ones.
|
| 40 |
RSS_FEEDS = {
|
| 41 |
-
"Sakshi": "https://www.sakshi.com/tags/rss",
|
| 42 |
-
"Eenadu": "https://www.eenadu.net/telugu-news/rss",
|
| 43 |
-
# Add more reliable Telugu news RSS feeds here if the above don't work well
|
| 44 |
-
# Example: "MyNewSource": "http://mynewsource.com/rss_feed_link",
|
| 45 |
}
|
| 46 |
|
| 47 |
# --- OpenWeatherMap API Configuration (for Realtime Weather) ---
|
| 48 |
-
# YOUR ACTUAL API KEY IS SET HERE:
|
| 49 |
OPENWEATHERMAP_API_KEY = "52f888dddede635f9b1d8a08141b5c49"
|
| 50 |
OPENWEATHERMAP_API_URL = "http://api.openweathermap.org/data/2.5/weather"
|
| 51 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
# --- Dummy Translation Function (Translation is temporarily disabled) ---
|
| 53 |
def translate_en_to_te(text):
|
| 54 |
return text # Just return the original text
|
|
@@ -66,20 +64,17 @@ def get_news_from_rss(url):
|
|
| 66 |
published = entry.published if hasattr(entry, 'published') else "N/A"
|
| 67 |
image_url = None
|
| 68 |
|
| 69 |
-
# Attempt to find image from media_content (common in some RSS feeds)
|
| 70 |
if hasattr(entry, 'media_content') and entry.media_content:
|
| 71 |
for media in entry.media_content:
|
| 72 |
if 'url' in media and media.get('type', '').startswith('image'):
|
| 73 |
image_url = media['url']
|
| 74 |
break
|
| 75 |
-
# Attempt to find image from description HTML (common in others)
|
| 76 |
elif hasattr(entry, 'description'):
|
| 77 |
soup = BeautifulSoup(entry.description, 'html.parser')
|
| 78 |
img_tag = soup.find('img')
|
| 79 |
if img_tag and 'src' in img_tag.attrs:
|
| 80 |
image_url = img_tag['src']
|
| 81 |
|
| 82 |
-
# Fallback to newspaper3k for better image/summary extraction from the article link
|
| 83 |
if link and "http" in link and link != "#" and (not image_url or not summary):
|
| 84 |
try:
|
| 85 |
article_parser = Article(link)
|
|
@@ -91,7 +86,7 @@ def get_news_from_rss(url):
|
|
| 91 |
summary = article_parser.text[:200] + "..." if len(article_parser.text) > 200 else article_parser.text
|
| 92 |
|
| 93 |
except Exception as e:
|
| 94 |
-
pass
|
| 95 |
|
| 96 |
if title.strip() and link.strip() != "#":
|
| 97 |
articles.append({
|
|
@@ -100,13 +95,55 @@ def get_news_from_rss(url):
|
|
| 100 |
"summary": summary,
|
| 101 |
"published": published,
|
| 102 |
"image_url": image_url,
|
| 103 |
-
"translated_title": None
|
| 104 |
})
|
| 105 |
return articles
|
| 106 |
except Exception as e:
|
| 107 |
st.error(f"Error fetching news from {url}: {e}")
|
| 108 |
return []
|
| 109 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 110 |
# --- Voice Search (Placeholder - requires a separate library/API) ---
|
| 111 |
def voice_search_widget():
|
| 112 |
st.write("### 🎙️ వాయిస్ సెర్చ్ (Voice Search)")
|
|
@@ -125,7 +162,7 @@ def get_user_location():
|
|
| 125 |
# --- Weather Integration ---
|
| 126 |
@st.cache_data(ttl=600)
|
| 127 |
def get_current_weather(city_name, api_key):
|
| 128 |
-
if not api_key:
|
| 129 |
return None
|
| 130 |
|
| 131 |
params = {
|
|
@@ -323,7 +360,6 @@ with main_tabs[0]:
|
|
| 323 |
articles = get_news_from_rss(rss_url)
|
| 324 |
|
| 325 |
for article in articles:
|
| 326 |
-
# Since translation is disabled, translated_title is just the original title
|
| 327 |
article['translated_title'] = article['title']
|
| 328 |
|
| 329 |
all_articles.extend(articles)
|
|
@@ -346,7 +382,7 @@ with main_tabs[0]:
|
|
| 346 |
if article['image_url']:
|
| 347 |
st.image(article['image_url'], use_column_width="always", caption="")
|
| 348 |
|
| 349 |
-
st.markdown(f"### {article['title']}")
|
| 350 |
|
| 351 |
st.markdown(f"<p>{article['summary']}</p>", unsafe_allow_html=True)
|
| 352 |
st.markdown(f"[పూర్తిగా చదవండి (Read More)]({article['link']})", unsafe_allow_html=True)
|
|
@@ -373,6 +409,7 @@ with main_tabs[1]:
|
|
| 373 |
st.info(f"'{user_city}' కోసం వాతావరణ డేటా అందుబాటులో లేదు. (Weather data not available for '{user_city}').")
|
| 374 |
|
| 375 |
st.subheader("స్థానిక వార్తలు (Local News)")
|
|
|
|
| 376 |
st.info(f"'{user_city}' కు సంబంధించిన స్థానిక వార్తలు ఇక్కడ ప్రదర్శించబడతాయి. (Local news for '{user_city}' will be displayed here.)")
|
| 377 |
|
| 378 |
st.subheader("అలర్ట్స్ (Alerts)")
|
|
@@ -399,9 +436,33 @@ with main_tabs[2]:
|
|
| 399 |
else:
|
| 400 |
st.info(f"'{search_city}' కోసం వాతావరణ డేటా అందుబాటులో లేదు. (Weather data not available for '{search_city}').")
|
| 401 |
|
| 402 |
-
st.subheader("స్థానిక వార్తలు (Local News)")
|
| 403 |
-
|
| 404 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 405 |
|
| 406 |
|
| 407 |
# --- Tab 4: Voice Search ---
|
|
@@ -483,7 +544,6 @@ st.sidebar.info("ఇక్కడ మీరు యాప్ సెట్టిం
|
|
| 483 |
|
| 484 |
if st.sidebar.button("రీఫ్రెష్ వార్తలు", key="refresh_news_button"):
|
| 485 |
st.cache_data.clear() # Clear all data caches (news, location, weather)
|
| 486 |
-
# st.cache_resource.clear() # Commented out as translation model is removed
|
| 487 |
st.rerun() # Rerun the app to load fresh data
|
| 488 |
st.sidebar.success("వార్తలు రీఫ్రెష్ చేయబడ్డాయి!")
|
| 489 |
|
|
|
|
| 13 |
import requests
|
| 14 |
from bs4 import BeautifulSoup
|
| 15 |
from newspaper import Article # Make sure newspaper3k and lxml_html_clean are installed
|
|
|
|
| 16 |
import time
|
|
|
|
| 17 |
import json
|
| 18 |
import datetime
|
| 19 |
|
|
|
|
| 31 |
"వైరల్": "Viral",
|
| 32 |
}
|
| 33 |
|
| 34 |
+
# --- RSS Feed URLs (for general headlines) ---
|
| 35 |
# IMPORTANT: These URLs might be outdated or not provide full content.
|
| 36 |
# You might need to find current, active RSS feeds for Telugu news sources.
|
|
|
|
| 37 |
RSS_FEEDS = {
|
| 38 |
+
"Sakshi": "https://www.sakshi.com/tags/rss",
|
| 39 |
+
"Eenadu": "https://www.eenadu.net/telugu-news/rss",
|
|
|
|
|
|
|
| 40 |
}
|
| 41 |
|
| 42 |
# --- OpenWeatherMap API Configuration (for Realtime Weather) ---
|
|
|
|
| 43 |
OPENWEATHERMAP_API_KEY = "52f888dddede635f9b1d8a08141b5c49"
|
| 44 |
OPENWEATHERMAP_API_URL = "http://api.openweathermap.org/data/2.5/weather"
|
| 45 |
|
| 46 |
+
# --- NewsAPI.org Configuration (for Local News Search) ---
|
| 47 |
+
NEWSAPI_API_KEY = "54035029265744dab172f28d58a93b56" # Your NewsAPI.org Key
|
| 48 |
+
NEWSAPI_EVERYTHING_URL = "https://newsapi.org/v2/everything"
|
| 49 |
+
|
| 50 |
# --- Dummy Translation Function (Translation is temporarily disabled) ---
|
| 51 |
def translate_en_to_te(text):
|
| 52 |
return text # Just return the original text
|
|
|
|
| 64 |
published = entry.published if hasattr(entry, 'published') else "N/A"
|
| 65 |
image_url = None
|
| 66 |
|
|
|
|
| 67 |
if hasattr(entry, 'media_content') and entry.media_content:
|
| 68 |
for media in entry.media_content:
|
| 69 |
if 'url' in media and media.get('type', '').startswith('image'):
|
| 70 |
image_url = media['url']
|
| 71 |
break
|
|
|
|
| 72 |
elif hasattr(entry, 'description'):
|
| 73 |
soup = BeautifulSoup(entry.description, 'html.parser')
|
| 74 |
img_tag = soup.find('img')
|
| 75 |
if img_tag and 'src' in img_tag.attrs:
|
| 76 |
image_url = img_tag['src']
|
| 77 |
|
|
|
|
| 78 |
if link and "http" in link and link != "#" and (not image_url or not summary):
|
| 79 |
try:
|
| 80 |
article_parser = Article(link)
|
|
|
|
| 86 |
summary = article_parser.text[:200] + "..." if len(article_parser.text) > 200 else article_parser.text
|
| 87 |
|
| 88 |
except Exception as e:
|
| 89 |
+
pass
|
| 90 |
|
| 91 |
if title.strip() and link.strip() != "#":
|
| 92 |
articles.append({
|
|
|
|
| 95 |
"summary": summary,
|
| 96 |
"published": published,
|
| 97 |
"image_url": image_url,
|
| 98 |
+
"translated_title": None
|
| 99 |
})
|
| 100 |
return articles
|
| 101 |
except Exception as e:
|
| 102 |
st.error(f"Error fetching news from {url}: {e}")
|
| 103 |
return []
|
| 104 |
|
| 105 |
+
# --- NewsAPI.org Fetching for Local News (Cached for performance) ---
|
| 106 |
+
@st.cache_data(ttl=600) # Cache for 10 minutes
|
| 107 |
+
def get_news_from_newsapi(query, api_key, language='en', page_size=10):
|
| 108 |
+
if not api_key:
|
| 109 |
+
return None
|
| 110 |
+
|
| 111 |
+
params = {
|
| 112 |
+
'q': query + " news", # Add " news" to the query for better results
|
| 113 |
+
'apiKey': api_key,
|
| 114 |
+
'language': language, # Can try 'te' but English coverage is better
|
| 115 |
+
'pageSize': page_size,
|
| 116 |
+
'sortBy': 'relevancy'
|
| 117 |
+
}
|
| 118 |
+
try:
|
| 119 |
+
response = requests.get(NEWSAPI_EVERYTHING_URL, params=params)
|
| 120 |
+
response.raise_for_status() # Raise an HTTPError for bad responses (4xx or 5xx)
|
| 121 |
+
news_data = response.json()
|
| 122 |
+
|
| 123 |
+
articles = []
|
| 124 |
+
if news_data['status'] == 'ok' and news_data['articles']:
|
| 125 |
+
for article in news_data['articles']:
|
| 126 |
+
if article['title'] and article['url']: # Basic validation
|
| 127 |
+
articles.append({
|
| 128 |
+
"title": article.get('title', 'No Title'),
|
| 129 |
+
"link": article.get('url', '#'),
|
| 130 |
+
"summary": article.get('description', article.get('content', 'No description available.')),
|
| 131 |
+
"published": article.get('publishedAt', 'N/A'),
|
| 132 |
+
"image_url": article.get('urlToImage'),
|
| 133 |
+
"source": article['source']['name'] if article.get('source') else 'Unknown Source',
|
| 134 |
+
"translated_title": None # Placeholder
|
| 135 |
+
})
|
| 136 |
+
return articles
|
| 137 |
+
except requests.exceptions.RequestException as e:
|
| 138 |
+
st.error(f"Error fetching news from NewsAPI for '{query}': {e}. Check API key or network.")
|
| 139 |
+
return None
|
| 140 |
+
except json.JSONDecodeError:
|
| 141 |
+
st.error(f"Error decoding NewsAPI response for '{query}'.")
|
| 142 |
+
return None
|
| 143 |
+
except Exception as e:
|
| 144 |
+
st.error(f"An unexpected error occurred with NewsAPI for '{query}': {e}")
|
| 145 |
+
return None
|
| 146 |
+
|
| 147 |
# --- Voice Search (Placeholder - requires a separate library/API) ---
|
| 148 |
def voice_search_widget():
|
| 149 |
st.write("### 🎙️ వాయిస్ సెర్చ్ (Voice Search)")
|
|
|
|
| 162 |
# --- Weather Integration ---
|
| 163 |
@st.cache_data(ttl=600)
|
| 164 |
def get_current_weather(city_name, api_key):
|
| 165 |
+
if not api_key:
|
| 166 |
return None
|
| 167 |
|
| 168 |
params = {
|
|
|
|
| 360 |
articles = get_news_from_rss(rss_url)
|
| 361 |
|
| 362 |
for article in articles:
|
|
|
|
| 363 |
article['translated_title'] = article['title']
|
| 364 |
|
| 365 |
all_articles.extend(articles)
|
|
|
|
| 382 |
if article['image_url']:
|
| 383 |
st.image(article['image_url'], use_column_width="always", caption="")
|
| 384 |
|
| 385 |
+
st.markdown(f"### {article['title']}")
|
| 386 |
|
| 387 |
st.markdown(f"<p>{article['summary']}</p>", unsafe_allow_html=True)
|
| 388 |
st.markdown(f"[పూర్తిగా చదవండి (Read More)]({article['link']})", unsafe_allow_html=True)
|
|
|
|
| 409 |
st.info(f"'{user_city}' కోసం వాతావరణ డేటా అందుబాటులో లేదు. (Weather data not available for '{user_city}').")
|
| 410 |
|
| 411 |
st.subheader("స్థానిక వార్తలు (Local News)")
|
| 412 |
+
# Local news for "My Location" still uses the RSS feeds, or you'd need another API for this
|
| 413 |
st.info(f"'{user_city}' కు సంబంధించిన స్థానిక వార్తలు ఇక్కడ ప్రదర్శించబడతాయి. (Local news for '{user_city}' will be displayed here.)")
|
| 414 |
|
| 415 |
st.subheader("అలర్ట్స్ (Alerts)")
|
|
|
|
| 436 |
else:
|
| 437 |
st.info(f"'{search_city}' కోసం వాతావరణ డేటా అందుబాటులో లేదు. (Weather data not available for '{search_city}').")
|
| 438 |
|
| 439 |
+
st.subheader(f"స్థానిక వార్తలు: {search_city} (Local News: {search_city})")
|
| 440 |
+
|
| 441 |
+
# --- NEW: Fetch and display local news using NewsAPI.org ---
|
| 442 |
+
local_news_articles = get_news_from_newsapi(search_city, NEWSAPI_API_KEY, language='en') # Use 'en' for broader coverage
|
| 443 |
+
|
| 444 |
+
if local_news_articles:
|
| 445 |
+
num_cols = 3
|
| 446 |
+
rows = []
|
| 447 |
+
for i in range(0, len(local_news_articles), num_cols):
|
| 448 |
+
rows.append(local_news_articles[i:i + num_cols])
|
| 449 |
+
|
| 450 |
+
for row_articles in rows:
|
| 451 |
+
cols = st.columns(num_cols)
|
| 452 |
+
for i, article in enumerate(row_articles):
|
| 453 |
+
with cols[i]:
|
| 454 |
+
with st.container():
|
| 455 |
+
st.markdown(f'<div class="news-card">', unsafe_allow_html=True)
|
| 456 |
+
if article['image_url']:
|
| 457 |
+
st.image(article['image_url'], use_column_width="always", caption="")
|
| 458 |
+
|
| 459 |
+
st.markdown(f"### {article['title']}")
|
| 460 |
+
st.markdown(f"**Source:** {article.get('source', 'Unknown')} | **Published:** {article['published']}")
|
| 461 |
+
st.markdown(f"<p>{article['summary']}</p>", unsafe_allow_html=True)
|
| 462 |
+
st.markdown(f"[పూర్తిగా చదవండి (Read More)]({article['link']})", unsafe_allow_html=True)
|
| 463 |
+
st.markdown('</div>', unsafe_allow_html=True)
|
| 464 |
+
else:
|
| 465 |
+
st.info(f"'{search_city}' కోసం స్థానిక వార్తలు అందుబాటులో లేవు. దయచేసి మరొక నగరాన్ని ప్రయత్నించండి లేదా మీ NewsAPI.org కీని తనిఖి చేయండి. (No local news available for '{search_city}'. Please try another city or check your NewsAPI.org key.)")
|
| 466 |
|
| 467 |
|
| 468 |
# --- Tab 4: Voice Search ---
|
|
|
|
| 544 |
|
| 545 |
if st.sidebar.button("రీఫ్రెష్ వార్తలు", key="refresh_news_button"):
|
| 546 |
st.cache_data.clear() # Clear all data caches (news, location, weather)
|
|
|
|
| 547 |
st.rerun() # Rerun the app to load fresh data
|
| 548 |
st.sidebar.success("వార్తలు రీఫ్రెష్ చేయబడ్డాయి!")
|
| 549 |
|