Update app.py

app.py

@@ -3,9 +3,9 @@ from sentence_transformers import SentenceTransformer, util
 import docx
 import os
 from PyPDF2 import PdfReader
+import re
 import requests
 import pandas as pd
-import re
 
 # Load pre-trained model for sentence embedding
 model = SentenceTransformer('paraphrase-MiniLM-L6-v2')
@@ -70,7 +70,7 @@ def system_prompt_to_extract_info(resume_text):
     """
     return prompt
 
-# Function to extract candidate information from resume text
+# Function to extract candidate information from resume text using Gemini API
 def extract_entities_via_gemini(resume_text):
     api_key = get_google_api_key()  # Fetch the API key from environment variables
     endpoint = "https://gemini.googleapis.com/v1/documents:analyzeEntities"  # Placeholder API endpoint (adjust as necessary)
@@ -86,42 +86,49 @@ def extract_entities_via_gemini(resume_text):
             "content": resume_text
         }
     }
-    
+
     # Send request to Gemini or another NLP API
     response = requests.post(endpoint, headers=headers, json=document)
 
+    # Debugging: Log raw response
+    print(f"Response Status Code: {response.status_code}")
+    print(f"Response Content: {response.text}")  # Log the raw response content
+
     if response.status_code != 200:
-        return {"error": "Failed to extract entities from resume", "status_code": response.status_code, "response": response.json()}
-    
-    # Process the response from the Gemini API (or similar NLP API)
-    entities = response.json().get('entities', [])
+        return {"error": "Failed to extract entities from resume", "status_code": response.status_code, "response": response.text}
+
+    # Use the raw text response (instead of parsing as JSON)
+    response_text = response.text
+
+    # You can now use `response_text` directly as you wish. For example, if you're extracting specific information:
     extracted_info = {"name": "Unknown Candidate", "email": "No Email", "contact": "No Contact"}
 
-    for entity in entities:
-        if entity['type'] == 'PERSON':
-            extracted_info['name'] = entity['name']
-        if entity['type'] == 'EMAIL':
-            extracted_info['email'] = entity['name']
-        if entity['type'] == 'PHONE_NUMBER':
-            extracted_info['contact'] = entity['name']
-    
+    # Check for keywords in the response text to extract candidate info
+    if "name" in response_text:
+        extracted_info['name'] = extract_info_from_text(response_text, 'name')
+    if "email" in response_text:
+        extracted_info['email'] = extract_info_from_text(response_text, 'email')
+    if "contact" in response_text:
+        extracted_info['contact'] = extract_info_from_text(response_text, 'contact')
+
     return extracted_info
 
-# Function to extract leadership experience (this function was missing previously)
-def extract_leadership_experience(resume_text):
-    # Define common leadership keywords/phrases
-    leadership_keywords = [
-        "leadership", "manager", "team lead", "supervisor", 
-        "director", "team leader", "managed", "led", 
-        "supervised", "coordinated", "coaching", "mentored"
-    ]
-    
-    # Search for the keywords in the resume text
-    for keyword in leadership_keywords:
-        if re.search(r"\b" + re.escape(keyword) + r"\b", resume_text, re.IGNORECASE):
-            return "Yes"  # Leadership experience found
-    
-    return "No"  # No leadership experience found
+# Helper function to extract specific information from raw response text
+def extract_info_from_text(response_text, info_type):
+    # Define simple patterns to match relevant information (you can improve this regex as needed)
+    if info_type == 'name':
+        match = re.search(r"Name: (\S+ \S+)", response_text)
+        if match:
+            return match.group(1)
+    elif info_type == 'email':
+        match = re.search(r"Email: (\S+@\S+)", response_text)
+        if match:
+            return match.group(1)
+    elif info_type == 'contact':
+        match = re.search(r"Contact: (\S+)", response_text)
+        if match:
+            return match.group(1)
+    return f"No {info_type}"
 
 # Function to check similarity between resumes and job description
 def check_similarity(job_description, resume_files):
@@ -141,7 +148,7 @@ def check_similarity(job_description, resume_files):
         # Convert similarity score to percentage
         similarity_percentage = similarity_score * 100
 
-        # Extract leadership experience
+        # Extract leadership experience (make sure this function is implemented)
         leadership_experience = extract_leadership_experience(resume_text)
 
         # Extract name, email, and contact info using Google Gemini API
@@ -197,11 +204,10 @@ def download_results(results):
 interface = gr.Interface(
     fn=check_similarity,
     inputs=[job_desc_input, resumes_input],
-    outputs=[results_output, gr.File(label="Download CSV")],
+    outputs=[results_output, gr.File(label="Download CSV", fn=download_results)],
     title="HR Assistant - Resume Screening & Leadership Experience",
     description="Upload job description and resumes to screen candidates for managerial and team leadership roles and extract candidate details.",
     allow_flagging="never"
 )
 
-# Launch the interface
 interface.launch()