Update app.py

app.py

@@ -6,34 +6,32 @@ import gradio as gr
 
 
 # ---------------------------------
-# Extract closest name near email
+# Extract name from same block
 # ---------------------------------
-def find_name_near_email(tag):
-    # Look upward in same block
-    parent = tag.find_parent()
-
-    if not parent:
-        return "Not found"
+def extract_name(tag):
+    # Step 1: check headings inside same block
+    for h in tag.find_all(["h1", "h2", "h3", "h4", "strong", "b"]):
+        name = h.get_text(strip=True)
+        if 2 <= len(name.split()) <= 5:
+            return name
+
+    # Step 2: fallback to nearby text
+    text = tag.get_text("\n", strip=True)
+    lines = text.split("\n")
 
-    # Check inside same block
-    candidates = parent.find_all(["h1", "h2", "h3", "h4", "strong", "b"])
+    for line in lines:
+        line = line.strip()
 
-    for c in candidates:
-        text = c.get_text(strip=True)
+        if "@" in line:
+            continue
 
-        if 2 <= len(text.split()) <= 5:
-            return text
+        if any(word in line.lower() for word in [
+            "department", "school", "iit", "university"
+        ]):
+            continue
 
-    # Fallback: search nearby text
-    text = parent.get_text("\n", strip=True)
-    lines = text.split("\n")
-
-    for line in lines:
-        if 2 <= len(line.split()) <= 5 and "@" not in line:
-            if not any(word in line.lower() for word in [
-                "department", "school", "university", "iit"
-            ]):
-                return line
+        if 2 <= len(line.split()) <= 5:
+            return line
 
     return "Not found"
 
@@ -65,17 +63,20 @@ def get_professor_details(college_url):
                 res = requests.get(page, headers=headers, timeout=10)
                 page_soup = BeautifulSoup(res.text, "html.parser")
 
-                # BEST: mailto-based extraction
-                for a in page_soup.find_all("a", href=True):
-                    if "mailto:" in a["href"]:
-                        email = a["href"].replace("mailto:", "").strip()
+                # 🔥 KEY: scan structured blocks
+                for tag in page_soup.find_all(["div", "li", "td"]):
+                    text = tag.get_text(" ", strip=True)
+
+                    emails = re.findall(r"[\w\.-]+@[\w\.-]+", text)
 
-                        name = find_name_near_email(a)
+                    if emails:
+                        name = extract_name(tag)
 
-                        results.append({
-                            "name": name,
-                            "email": email
-                        })
+                        for email in emails:
+                            results.append({
+                                "name": name,
+                                "email": email
+                            })
 
             except:
                 continue
@@ -110,8 +111,8 @@ iface = gr.Interface(
     fn=get_professor_details,
     inputs=gr.Textbox(label="Enter College Website URL"),
     outputs=gr.Textbox(label="Professor Details"),
-    title="🎓 Professor Scraper (Accurate Version)",
-    description="Extracts correct professor names and emails"
+    title="🎓 Professor Scraper (Final Working)",
+    description="Extracts professor names and emails from college websites"
 )
 
 iface.launch()