Update app.py

app.py

@@ -1,8 +1,31 @@
+# 🚀 0. BRUTE FORCE INSTALLER
+import subprocess
+import sys
+import os
+
+# This block forces the installation of the browser Playwright needs.
+# It's a workaround for stubborn server environments.
+try:
+    print("--- Ensuring Playwright browsers are installed ---")
+    # We use 'playwright install chromium' to be specific and save space.
+    # Using sys.executable ensures we use the python env's playwright.
+    process = subprocess.run(
+        [sys.executable, "-m", "playwright", "install", "chromium"],
+        capture_output=True,
+        text=True,
+        check=True
+    )
+    print(process.stdout)
+    print("--- Playwright browsers installation check complete ---")
+except Exception as e:
+    print(f"--- An error occurred during browser installation: {e} ---")
+    # Even if it fails, we try to continue.
+    pass
+
 # 🚀 1. IMPORTS & SETUP
 import asyncio
 import datetime
 import logging
-import os
 import random
 import gradio as gr
 from cachetools import TTLCache
@@ -44,7 +67,13 @@ async def _extract_user_request(question: str) -> dict:
         ("human", "{format_instructions}\n{request}")
     ]).partial(format_instructions=parser.get_format_instructions())
     logging.info(f"🔍 Extracting details from: {question}")
-    return await (prompt | llm | parser).ainvoke({"request": question})
+    # Default to a one-way trip if end_date isn't found
+    extracted = await (prompt | llm | parser).ainvoke({"request": question})
+    if not extracted.get("end_date"):
+        start_dt = datetime.datetime.strptime(extracted["start_date"], "%Y-%m-%d")
+        end_dt = start_dt + datetime.timedelta(days=7) # Default to 7 days later
+        extracted["end_date"] = end_dt.strftime("%Y-%m-%d")
+    return extracted
 
 
 # 🛠 4. CACHE WRAPPER
@@ -61,7 +90,7 @@ async def _cached_search(func, details):
 
 # 🌐 5. PLAYWRIGHT SCRAPER FUNCTIONS
 
-async def _search_google_flights(details):
+async def _search_Google Flights(details):
     results = []
     try:
         async with async_playwright() as pw:
@@ -69,20 +98,15 @@ async def _search_google_flights(details):
             context = await browser.new_context(user_agent=random.choice(USER_AGENTS))
             page = await context.new_page()
             
-            # Add random delay
-            await asyncio.sleep(random.uniform(1, 3))
-            
             o = details["origin_city"]
             d = details["destination_city"]
             sd = details["start_date"]
             ed = details["end_date"]
             
-            # More reliable Google Flights URL
             url = f"https://www.google.com/travel/flights/search?tfs=CBwQAhokEgoyMDI1LTA4LTAxagwIAhIIL20vMDVxdGwyDAg"
             
             await page.goto(url, timeout=60000)
             
-            # Wait for any flight results to load
             try:
                 await page.wait_for_selector('[data-testid="flight-offer"]', timeout=20000)
                 cards = await page.query_selector_all('[data-testid="flight-offer"]')
@@ -90,9 +114,6 @@ async def _search_google_flights(details):
                 for card in cards[:3]:
                     try:
                         price_el = await card.query_selector('[data-testid="price-text"]')
-                        if not price_el:
-                            price_el = await card.query_selector('span[aria-label*="dollars"]')
-                        
                         if price_el:
                             price_text = await price_el.inner_text()
                             price_clean = ''.join(filter(str.isdigit, price_text))
@@ -108,32 +129,14 @@ async def _search_google_flights(details):
                     except Exception as e:
                         logging.warning(f"Error parsing Google Flights card: {e}")
                         continue
-                        
             except Exception as e:
                 logging.warning(f"Google Flights selector timeout: {e}")
-                # Fallback: try to find any price-like elements
-                price_elements = await page.query_selector_all('span[aria-label*="dollar"], span[aria-label*="price"]')
-                for elem in price_elements[:3]:
-                    try:
-                        text = await elem.inner_text()
-                        if '$' in text:
-                            results.append({
-                                "source": "Google Flights", 
-                                "type": "flight", 
-                                "details": text.strip(),
-                                "price": 0.0,
-                                "link": url
-                            })
-                    except:
-                        continue
-                        
             await browser.close()
-            
     except Exception as e:
         logging.error(f"Google Flights scraping failed: {e}")
-    
     return results
 
+# ... (rest of the scraper functions remain the same)
 async def _search_kayak(details):
     results = []
     try:
@@ -153,13 +156,7 @@ async def _search_kayak(details):
             
             await page.goto(url, timeout=60000)
             
-            # Multiple selector attempts
-            selectors_to_try = [
-                ".resultWrapper",
-                "[data-testid='result-item']",
-                ".Common-Booking-MultiBookProvider",
-                ".result-column"
-            ]
+            selectors_to_try = [".resultWrapper", "[data-testid='result-item']", ".Common-Booking-MultiBookProvider"]
             
             cards = []
             for selector in selectors_to_try:
@@ -173,14 +170,7 @@ async def _search_kayak(details):
             
             for card in cards[:3]:
                 try:
-                    # Try multiple price selectors
-                    price_selectors = [
-                        ".price-text",
-                        "[data-testid='price']",
-                        ".Common-Booking-MultiBookProvider-price",
-                        "span[aria-label*='price']"
-                    ]
-                    
+                    price_selectors = [".price-text", "[data-testid='price']", ".Common-Booking-MultiBookProvider-price"]
                     price_text = None
                     for price_sel in price_selectors:
                         try:
@@ -194,24 +184,16 @@ async def _search_kayak(details):
                     if price_text:
                         price_clean = ''.join(filter(str.isdigit, price_text))
                         price = float(price_clean) if price_clean else 0.0
-                        
                         results.append({
-                            "source": "Kayak", 
-                            "type": "flight", 
-                            "details": price_text.strip(),
-                            "price": price,
-                            "link": url
+                            "source": "Kayak", "type": "flight", "details": price_text.strip(),
+                            "price": price, "link": url
                         })
-                        
                 except Exception as e:
                     logging.warning(f"Error parsing Kayak card: {e}")
                     continue
-                    
             await browser.close()
-            
     except Exception as e:
         logging.error(f"Kayak scraping failed: {e}")
-    
     return results
 
 async def _search_trabber(details):
@@ -221,22 +203,16 @@ async def _search_trabber(details):
             browser = await pw.chromium.launch(headless=True)
             context = await browser.new_context(user_agent=random.choice(USER_AGENTS))
             page = await context.new_page()
-            
             await asyncio.sleep(random.uniform(1, 3))
-            
             o = details["origin_city"].upper()
             d = details["destination_city"].upper()
             sd_parts = details["start_date"].split('-')
-            sd = f"{sd_parts[2]}{sd_parts[1]}{sd_parts[0][2:]}"  # DDMMYY format
-            
+            sd = f"{sd_parts[2]}{sd_parts[1]}{sd_parts[0][2:]}"
             url = f"https://www.trabber.ca/flights-from-{o}-to-{d}-on-{sd}"
-            
             await page.goto(url, timeout=60000)
-            
             try:
                 await page.wait_for_selector("#results_list_det tr", timeout=20000)
                 rows = await page.query_selector_all("#results_list_det tr")
-                
                 for row in rows[:3]:
                     try:
                         price_el = await row.query_selector("td.results_price a, .results_price")
@@ -244,26 +220,18 @@ async def _search_trabber(details):
                             price_text = await price_el.inner_text()
                             price_clean = ''.join(filter(str.isdigit, price_text))
                             price = float(price_clean) if price_clean else 0.0
-                            
                             results.append({
-                                "source": "Trabber", 
-                                "type": "flight", 
-                                "details": price_text.strip(),
-                                "price": price,
-                                "link": url
+                                "source": "Trabber", "type": "flight", "details": price_text.strip(),
+                                "price": price, "link": url
                             })
                     except Exception as e:
                         logging.warning(f"Error parsing Trabber row: {e}")
                         continue
-                        
             except Exception as e:
                 logging.warning(f"Trabber selector timeout: {e}")
-                
             await browser.close()
-            
     except Exception as e:
         logging.error(f"Trabber scraping failed: {e}")
-    
     return results
 
 async def _search_travala(details):
@@ -273,21 +241,11 @@ async def _search_travala(details):
             browser = await pw.chromium.launch(headless=True)
             context = await browser.new_context(user_agent=random.choice(USER_AGENTS))
             page = await context.new_page()
-            
             await asyncio.sleep(random.uniform(1, 3))
-            
             url = "https://www.travala.com/deals"
-            
             await page.goto(url, timeout=60000)
-            
             try:
-                # Try multiple selectors for deals
-                selectors = [
-                    "div[class*='DealCard_container__']",
-                    ".deal-card",
-                    "[data-testid='deal-card']"
-                ]
-                
+                selectors = ["div[class*='DealCard_container__']", ".deal-card", "[data-testid='deal-card']"]
                 cards = []
                 for selector in selectors:
                     try:
@@ -297,15 +255,9 @@ async def _search_travala(details):
                             break
                     except:
                         continue
-                
                 for card in cards[:5]:
                     try:
-                        title_selectors = [
-                            "p[class*='DealCard_title__']",
-                            ".deal-title",
-                            "h3, h4"
-                        ]
-                        
+                        title_selectors = ["p[class*='DealCard_title__']", ".deal-title", "h3, h4"]
                         title = "Travala Deal"
                         for title_sel in title_selectors:
                             try:
@@ -315,27 +267,17 @@ async def _search_travala(details):
                                     break
                             except:
                                 continue
-                        
                         results.append({
-                            "source": "Travala", 
-                            "type": "deal", 
-                            "details": title[:100], 
-                            "price": 0, 
-                            "link": url
+                            "source": "Travala", "type": "deal", "details": title[:100], "price": 0, "link": url
                         })
-                        
                     except Exception as e:
                         logging.warning(f"Error parsing Travala card: {e}")
                         continue
-                        
             except Exception as e:
                 logging.warning(f"Travala selector timeout: {e}")
-                
             await browser.close()
-            
     except Exception as e:
         logging.error(f"Travala scraping failed: {e}")
-    
     return results
 
 async def _search_secretflying(details):
@@ -345,21 +287,11 @@ async def _search_secretflying(details):
             browser = await pw.chromium.launch(headless=True)
             context = await browser.new_context(user_agent=random.choice(USER_AGENTS))
             page = await context.new_page()
-            
             await asyncio.sleep(random.uniform(1, 3))
-            
             url = "https://www.secretflying.com/canada-deals/"
-            
             await page.goto(url, timeout=60000)
-            
             try:
-                # Try multiple selectors for posts
-                selectors = [
-                    ".post-item",
-                    "article",
-                    ".entry"
-                ]
-                
+                selectors = [".post-item", "article", ".entry"]
                 posts = []
                 for selector in selectors:
                     try:
@@ -369,19 +301,11 @@ async def _search_secretflying(details):
                             break
                     except:
                         continue
-                
                 for post in posts[:5]:
                     try:
-                        title_selectors = [
-                            "h2.entry-title a",
-                            ".entry-title a",
-                            "h3 a",
-                            "h2 a"
-                        ]
-                        
+                        title_selectors = ["h2.entry-title a", ".entry-title a", "h3 a", "h2 a"]
                         title = "Deal"
                         link = url
-                        
                         for title_sel in title_selectors:
                             try:
                                 title_el = await post.query_selector(title_sel)
@@ -393,34 +317,23 @@ async def _search_secretflying(details):
                                     break
                             except:
                                 continue
-                        
                         results.append({
-                            "source": "SecretFlying", 
-                            "type": "deal", 
-                            "details": title[:100], 
-                            "price": 0, 
-                            "link": link
+                            "source": "SecretFlying", "type": "deal", "details": title[:100], "price": 0, "link": link
                         })
-                        
                     except Exception as e:
                         logging.warning(f"Error parsing SecretFlying post: {e}")
                         continue
-                        
             except Exception as e:
                 logging.warning(f"SecretFlying selector timeout: {e}")
-                
             await browser.close()
-            
     except Exception as e:
         logging.error(f"SecretFlying scraping failed: {e}")
-    
     return results
 
-
 # 🔄 6. GATHER ALL DATA
 async def _gather_travel_data(req):
     tasks = [
-        _cached_search(_search_google_flights, req),
+        _cached_search(_search_Google Flights, req),
         _cached_search(_search_kayak, req),
         _cached_search(_search_trabber, req),
         _cached_search(_search_travala, req),
@@ -435,12 +348,10 @@ async def _gather_travel_data(req):
             logging.error(f"A scraper task failed: {r}")
     return combined
 
-
 # ✍️ 7. FORMAT RESULTS
 def _format_response(req, data):
     dest = req["destination_city"]
     md = f"## 🌏 Travel Plan to **{dest}**\n\n"
-    
     flights = sorted([i for i in data if i["type"]=="flight" and i.get("price", 0) > 0], key=lambda x: x["price"])
     if flights:
         md += "### ✈️ Flights\n"