content_quality_tool/title_creator_app/views.py

import os
import json
import time
import requests
import uuid
import threading
import pandas as pd
from bs4 import BeautifulSoup
from django.shortcuts import get_object_or_404, redirect, render
from django.core.files.storage import FileSystemStorage
from django.http import JsonResponse
from .models import TitleMapping, AttributeMaster,ProcessingTask # <--- THIS FIXES THE ERROR
from django.conf import settings
import cloudscraper
from django.contrib import messages
from django.contrib.auth import authenticate, login, logout
# from django.contrib.auth.decorators import login_required
from .decorators import login_required
from django.contrib.auth.hashers import make_password
import random
from rest_framework import status
from rest_framework.views import APIView
from django.utils import timezone
import logging
logger = logging.getLogger(__name__)

# To login
def login_view(request):
    if request.method == "POST":
        email = request.POST.get("username")
        password = request.POST.get("password")
        print("Email: ", email)
        print("Password: ", password)

        # Authenticate the user
        user = authenticate(request, username=email, password=password)
        print("user",user)
        
        if user is not None:
            print("User authenticated successfully.")
            login(request, user)
            request.session['user_email'] = user.email

            # request.session = user
            # request.session['full_name'] = f"{user.firstName} {user.lastName or ''}".strip()
            # # Store both human-readable role and code
            # request.session['role'] = user.get_role_display()   # 'Super Admin', 'Admin', 'RTA'
            # request.session['role_code'] = user.role            # '0', '1', '2'
            # request.session['joining_date'] = user.createdDate.strftime("%b, %Y")
            # request.session['userId'] = user.userId

            # 📌 Store client_id if user has a client associated
            # request.session['client_id'] = user.client.clientId if user.client else None

            return redirect('title_creator_home')
        else:
            print("Invalid credentials.")
            messages.error(request, "Invalid email or password.")
            return redirect('login')
    print("Rendering login page.")
    return render(request, 'login.html')


# To logout
@login_required
def logout_view(request):
    logout(request)
    messages.success(request, "You have been logged out successfully.")
    return redirect('login')

@login_required
def master_config_view(request):
    if request.method == 'POST':
        action = request.POST.get('action')
        
        # Part 1: Add New Attribute
        if action == 'add_attribute':
            name = request.POST.get('attr_name')
            is_m = request.POST.get('is_mandatory') == 'on'
            if name:
                AttributeMaster.objects.get_or_create(name=name.strip(), defaults={'is_mandatory': is_m})
        
        # Part 2: Add New Title Mapping (Product Type)
        # --- MAPPING ACTIONS (CREATE & UPDATE) ---
        elif action in ['add_mapping', 'update_mapping']:
            pt = request.POST.get('pt_name')
            seq = request.POST.get('sequence')
            edit_id = request.POST.get('edit_id')

            if action == 'update_mapping' and edit_id:
                # Update existing
                mapping = get_object_or_404(TitleMapping, id=edit_id)
                mapping.product_type = pt.strip()
                mapping.format_sequence = seq
                mapping.save()
            else:
                # Create new (using get_or_create to prevent exact duplicates)
                if pt:
                    TitleMapping.objects.get_or_create(
                        product_type=pt.strip(), 
                        defaults={'format_sequence': seq}
                    )

        # --- MAPPING DELETE ---
        elif action == 'delete_mapping':
            mapping_id = request.POST.get('id')
            TitleMapping.objects.filter(id=mapping_id).delete()
        
        # Part 3: Delete functionality
        elif action == 'delete_attribute':
            AttributeMaster.objects.filter(id=request.POST.get('id')).delete()
        
        return redirect('title_creator_master')

    # GET: Load all data
    context = {
        'attributes': AttributeMaster.objects.all().order_by('name'),
        'mappings': TitleMapping.objects.all().order_by('product_type'),
    }
    return render(request, 'title_creator_master.html', context)

def save_config_api(request):
    if request.method == 'POST':
        try:
            data = json.loads(request.body)
            
            # Update Mandatory Attributes
            # Expected data: { "mandatory_ids": [1, 3, 5] }
            AttributeMaster.objects.all().update(is_mandatory=False)
            AttributeMaster.objects.filter(id__in=data.get('mandatory_ids', [])).update(is_mandatory=True)
            
            # Update Title Sequences
            # Expected data: { "mappings": [{"id": 1, "sequence": "Brand,Color"}] }
            for m in data.get('mappings', []):
                TitleMapping.objects.filter(id=m['id']).update(format_sequence=m['sequence'])
                
            return JsonResponse({'success': True})
        except Exception as e:
            return JsonResponse({'success': False, 'error': str(e)})

# def extract_title_or_error(product,selected_pt):
#     # 1. Identify Product Type from JSON to fetch the correct Mapping
#     pt_name = selected_pt
#     try:
#         mapping = TitleMapping.objects.get(product_type=pt_name)
#         config_sequence = mapping.get_sequence_list()
#     except TitleMapping.DoesNotExist:
#         return f"No Title Configuration found for Product Type: {pt_name}"

#     # 2. Get Mandatory list from DB
#     mandatory_fields = list(AttributeMaster.objects.filter(is_mandatory=True).values_list('name', flat=True))

#     # 3. Data Extraction (Your logic)
#     extracted_data = {
#         "Brand": product.get("brand"),
#         "Product Type": pt_name
#     }
#     dimensions = {}

#     for group in product.get("attributeGroups", []):
#         for attr in group.get("attributes", []):
#             desc = attr.get("attributeDesc")
#             value = attr.get("attributeValue")
            
#             if desc == "Capacity":
#                 extracted_data[desc] = f"Capacity {value}"
#             if desc in ["Door Type", "Capacity", "Color"]:
#                 extracted_data[desc] = value
#             elif desc in ["Width", "Depth", "Height"]:
#                 dimensions[desc] = value

#     if {"Width", "Depth", "Height"}.issubset(dimensions):
#         # extracted_data["Dimensions"] = f'{dimensions["Width"]} x {dimensions["Depth"]} x {dimensions["Height"]}'
#         w, d, h = dimensions["Width"], dimensions["Depth"], dimensions["Height"]
#         extracted_data["Dimensions"] = f'{w}"w x {d}"d x {h}"h'

#     # 4. Build Title and Check Mandatory Rules from DB
#     final_title_parts = []
#     missing_mandatory = []

#     for attr_name in config_sequence:
#         val = extracted_data.get(attr_name)
        
#         if not val or str(val).strip() == "":
#             # If DB says it's mandatory, track the error
#             if attr_name in mandatory_fields:
#                 missing_mandatory.append(attr_name)
#             continue 
            
#         final_title_parts.append(str(val))

#     # 5. Result
#     if missing_mandatory:
#         return f"Could not found {', '.join(missing_mandatory)} on Product Details page"
    
#     return " ".join(final_title_parts)


def extract_title_or_error(product, selected_pt):
    # 1. Identify Product Type
    pt_name = selected_pt
    logger.info(f"IN extract_title_or_error")
    try:
        mapping = TitleMapping.objects.get(product_type=pt_name)
        config_sequence = mapping.get_sequence_list()
    except TitleMapping.DoesNotExist:
        return None,f"No Title Configuration found for Product Type: {pt_name}"

    mandatory_fields = list(AttributeMaster.objects.filter(is_mandatory=True).values_list('name', flat=True))
    # Loop through each group (e.g., Weights & Dimensions, Product Details)
    product_type = None
    # for product type
    for group in product.get("attributeGroups", []):
        # Loop through each attribute in that group
        for attr in group.get("attributes", []):
            if attr.get("attributeDesc") == "Type":
                product_type = attr.get("attributeValue")
                break  # Stop searching once found

    # 2. Data Extraction
    extracted_data = {
        "Brand": product.get("brand"),
        "Product Type": product_type
    }
    dimensions = {}

    for group in product.get("attributeGroups", []):
        for attr in group.get("attributes", []):
            desc = attr.get("attributeDesc")
            val = attr.get("attributeValue")
            
            if desc == "Capacity":
                extracted_data[desc] = f"{val} Capacity"
            elif desc in ["Door Type", "Color"]:
                extracted_data[desc] = val
            elif desc in ["Width", "Depth", "Height"]:
                dimensions[desc] = val

    if {"Width", "Depth", "Height"}.issubset(dimensions):
        w, d, h = dimensions["Width"], dimensions["Depth"], dimensions["Height"]
        # We use .replace(" in", "") to remove the existing unit before adding the " symbol
        w = dimensions["Width"].replace(" in", "").strip()
        d = dimensions["Depth"].replace(" in", "").strip()
        h = dimensions["Height"].replace(" in", "").strip()
        extracted_data["Dimensions"] = f'{w}"W x {d}"D x {h}"H'

    # 3. Build Title Parts
    final_title_parts = []
    missing_mandatory = []

    for attr_name in config_sequence:
        val = extracted_data.get(attr_name)
        if not val or str(val).strip() == "":
            if attr_name in mandatory_fields:
                missing_mandatory.append(attr_name)
            continue 
        final_title_parts.append(str(val))
    comment =  None
    if missing_mandatory:
        comment = f"Could not found {', '.join(missing_mandatory)} on Product Details page"
        # return f"Could not found {', '.join(missing_mandatory)} on Product Details page"

    # Helper function to join parts: Brand PT, Param1, Param2
    def construct_string(parts):
        if len(parts) <= 2:
            return " ".join(parts)
        return f"{parts[0]} {parts[1]}, {', '.join(parts[2:])}"

    current_title = construct_string(final_title_parts)

    # 4. Length Reduction Logic (Step-by-Step)
    print("Current Title 1 ######## ",current_title,len(current_title))
    logger.info(f"Current Title 1 Initial ########,{current_title},{len(current_title)}")
    # Step 1: Change "Capacity" -> "Cap."
    if len(current_title) > 100:
        for i, part in enumerate(final_title_parts):
            if "Capacity" in part:
                final_title_parts[i] = part.replace("Capacity", "Cap.")
        current_title = construct_string(final_title_parts)

    print("Current Title 2 ########",current_title,len(current_title))
    logger.info(f"Current Title 2 shorting capacity ########,{current_title},{len(current_title)}")    

    # Step 2: Shorten Product Type (e.g., Stainless Steel -> SS)
    # Step B: Dynamic Product Type Acronym
    # if len(current_title) > 100:
    #     pt_part = final_title_parts[1]
    #     words = pt_part.split()
    #     if len(words) > 1:
    #         # Takes first letter of every word in the Product Type
    #         final_title_parts[1] = "".join([w[0].upper() for w in words])
    #         current_title = construct_string(final_title_parts)
    # print("Current Title 3 ########",current_title,len(current_title))
    # logger.info(f"Current Title 3 change the title ########,{current_title},{len(current_title)}")
    # Step 3: Remove spaces from attributes starting from the back
    # Brand (0) and Product Type (1) are skipped
    if len(current_title) > 100:
        for i in range(len(final_title_parts) - 1, 1, -1):
            if len(current_title) <= 100:
                break
            # Remove white spaces from the current attribute part
            final_title_parts[i] = final_title_parts[i].replace(" ", "")
            current_title = construct_string(final_title_parts)
    print("Current Title 4 ########",current_title,len(current_title))
    logger.info(f"Current Title 4 Removing space ########,{current_title},{len(current_title)}")
    return current_title,comment

def construct_dynamic_title(raw_data,selected_pt):
    try:
        product = raw_data.get("props", {}).get("pageProps", {}).get("product", {})
        if not product: return "Product data not found"
        logger.info(f"IN construct_dynamic_title")
        return extract_title_or_error(product,selected_pt)
    except Exception:
        return None,"Could not found attribute name on product details page"

@login_required
def title_creator_view(request):
    if request.method == 'POST' and request.FILES.get('file'):
        fresh_token = get_fresh_token()
        logger.info(f"fresh_token Value: {fresh_token}")
        if not fresh_token:
            fresh_token = "Bearer eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzUxMiJ9.eyJzdWIiOiJERVYifQ.uOFB7h7_Aw6jbA1HSqVJ44tKMO7E1ljz1kV_JddeKL64YCOH57-l1ZX2Lly-Jnhdnxk3xMAeW5FawAgymEaMKA"
        excel_file = request.FILES['file']
        selected_pt = request.POST.get('product_type')
        fs = FileSystemStorage()
        filename = fs.save(excel_file.name, excel_file)
        file_path = fs.path(filename)

        try:
            # 1. Read Excel
            df = pd.read_excel(file_path)
            
            # 2. Add the NEW COLUMN if it doesn't exist
            if 'New_Generated_Title' not in df.columns:
                df['New_Generated_Title'] = ""
            if 'Comment' not in df.columns:
                df['Comment'] = ""    

            headers = {"User-Agent": "Mozilla/5.0"}
            results_for_ui = []
            # Specific Headers for the Item# API
            api_headers = {
                "accept": "application/json, text/plain, */*",
                "authorization": fresh_token,#"Bearer eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzUxMiJ9.eyJzdWIiOiJERVYifQ.uOFB7h7_Aw6jbA1HSqVJ44tKMO7E1ljz1kV_JddeKL64YCOH57-l1ZX2Lly-Jnhdnxk3xMAeW5FawAgymEaMKA",
                "client_id": "GEC",
                "referer": "https://www.globalindustrial.com/"
            }

            # 3. Process each row
            for index, row in df.iterrows():
                url = row.get('URL') # Assumes your excel has a 'URL' column
                item_number = row.get('Item#')
                new_title = ""
                final_url = None
                comment = ""

                # Step 1: Resolve the URL
                if pd.notna(url) and str(url).startswith('http'):
                    final_url = url
                elif pd.notna(item_number):
                    # Call API to get URL from Item#
                    api_url = f"https://www.globalindustrial.com/catalogApis/catalog/autosuggest?key={item_number}&features=true"
                    try:
                        api_resp = requests.get(api_url, headers=api_headers, timeout=10)
                        if api_resp.status_code == 200:
                            data = api_resp.json()
                            final_url = data.get('exactMatch', {}).get('canonicalLink')
                    except Exception as e:
                        new_title = f"API Error for Item# {item_number}"


                if pd.notna(final_url):
                    try:
                        resp = requests.get(final_url, headers=headers, timeout=10)
                        soup = BeautifulSoup(resp.content, 'html.parser')
                        script_tag = soup.find('script', id='__NEXT_DATA__')
                        
                        if script_tag:
                            raw_data = json.loads(script_tag.string)
                            new_title,comment = construct_dynamic_title(raw_data,selected_pt)
                        else:
                            new_title,comment = "Could not found attribute name on product details page",None
                    except:
                        new_title,comment = "Could not found attribute name on product details page",None
                else:
                    new_title,comment = "URL Missing",None

                # Update the DataFrame column for this row
                df.at[index, 'New_Generated_Title'] = new_title
                df.at[index, 'Comment'] = comment
                
                results_for_ui.append({
                    "id" : index + 1,
                    "url": final_url,
                    "new_title": new_title,
                    "comment": comment,
                    "status": True
                })
                # Generates a random float between 3.0 and 7.0
                time.sleep(random.uniform(3, 7))
                # time.sleep(1) # Safety delay

            # 4. Save the modified Excel to a new path
            output_filename = f"processed_{excel_file.name}"
            output_path = os.path.join(fs.location, output_filename)
            df.to_excel(output_path, index=False)

            return JsonResponse({
                'success': True,
                'results': results_for_ui,
                'download_url': fs.url(output_filename)
            })

        finally:
            if os.path.exists(file_path): os.remove(file_path)

    # GET request: Fetch all product types for the dropdown
    product_types = TitleMapping.objects.all().values_list('product_type', flat=True)
    return render(request, 'title_creator_index.html', {'product_types': product_types})
    # return render(request, 'title_creator_index.html')

def get_fresh_token():
    """Hits the homepage to extract the latest Bearer token."""
    base_url = "https://www.globalindustrial.com"
    try:
        # Use a session to persist cookies
        session = requests.Session()
        response = session.get(base_url, timeout=15, headers={"User-Agent": "Mozilla/5.0"})
        
        # 1. Try Cookies
        token = session.cookies.get('Authorization')
        if token:
            return token if "Bearer" in token else f"Bearer {token}"

        # 2. Try NEXT_DATA
        soup = BeautifulSoup(response.content, 'html.parser')
        script_tag = soup.find('script', id='__NEXT_DATA__')
        if script_tag:
            data = json.loads(script_tag.string)
            token = data.get('props', {}).get('pageProps', {}).get('token')
            if token:
                return f"Bearer {token}"
    except Exception as e:
        print(f"Token retrieval failed: {e}")
    return None

def process_excel_task(file_path, selected_pt, task_id):
    print("process excel task started.")
    # Retrieve the task record from the database
    # scraper = cloudscraper.create_scraper() # This replaces requests.get
    task = ProcessingTask.objects.get(task_id=task_id)
    fresh_token = get_fresh_token()
    logger.info(f"fresh_token Value: {fresh_token}")
    if not fresh_token:
            fresh_token = "Bearer eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzUxMiJ9.eyJzdWIiOiJERVYifQ.uOFB7h7_Aw6jbA1HSqVJ44tKMO7E1ljz1kV_JddeKL64YCOH57-l1ZX2Lly-Jnhdnxk3xMAeW5FawAgymEaMKA"
    
    try:
        # 1. Read Excel
        df = pd.read_excel(file_path)
        
        # 2. Add the NEW COLUMN if it doesn't exist
        if 'New_Generated_Title' not in df.columns:
            df['New_Generated_Title'] = ""
        if 'Comment' not in df.columns:
            df['Comment'] = ""        

        headers = {"User-Agent": "Mozilla/5.0"}
        # dynamic_token = await get_fresh_token(scraper)
        # Specific Headers for the Item# API
        api_headers = {
            "accept": "application/json, text/plain, */*",
            "authorization": fresh_token, #"Bearer eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzUxMiJ9.eyJzdWIiOiJERVYifQ.uOFB7h7_Aw6jbA1HSqVJ44tKMO7E1ljz1kV_JddeKL64YCOH57-l1ZX2Lly-Jnhdnxk3xMAeW5FawAgymEaMKA", #f"Bearer {dynamic_token}",
            "client_id": "GEC",
            "referer": "https://www.globalindustrial.com/"
        }

        # 3. Process each row
        for index, row in df.iterrows():
            logger.info(f"STARTED: {index}")
            url = row.get('URL')
            new_title = ""
            comment = ""
            item_number = row.get('Item#')
            final_url = None
            # Step 1: Resolve the URL
            if pd.notna(url) and str(url).startswith('http'):
                final_url = url
            elif pd.notna(item_number):
                # Call API to get URL from Item#
                api_url = f"https://www.globalindustrial.com/catalogApis/catalog/autosuggest?key={item_number}&features=true"
                try:
                    api_resp = requests.get(api_url, headers=api_headers, timeout=10)
                    if api_resp.status_code == 200:
                        data = api_resp.json()
                        final_url = data.get('exactMatch', {}).get('canonicalLink')
                except Exception as e:
                    new_title,comment = f"API Error for Item# {item_number}"

            if pd.notna(final_url):
                try:
                    # Scraping logic
                    # resp = scraper.get(url, timeout=15)
                    resp = requests.get(final_url, headers=headers, timeout=10)
                    if resp.status_code == 200:
                        soup = BeautifulSoup(resp.content, 'html.parser')
                        script_tag = soup.find('script', id='__NEXT_DATA__')
                        
                        if script_tag:
                            try:
                                raw_data = json.loads(script_tag.string)
                                # Calling your dynamic title helper
                                new_title,comment = construct_dynamic_title(raw_data, selected_pt)
                            except Exception:
                                new_title,comment = "Data Parsing Error",None
                        else:
                            new_title,comment = "Could not found attribute name on product details page",None
                    else:
                        new_title,comment = f"HTTP Error: {resp.status_code}",None
                except Exception:
                    new_title,comment = "Request Failed (Timeout/Connection)",None
            else:
                new_title,comment = "URL Missing",None

            # Update the DataFrame
            df.at[index, 'New_Generated_Title'] = new_title
            df.at[index, 'Comment'] = comment
            
            # Optional: Sleep to prevent getting blocked by the server
            # Generates a random float between 3.0 and 7.0
            time.sleep(random.uniform(3, 7))
            logger.info(f"ENDED: {index}")
            # time.sleep(1)

        # 4. Save the modified Excel to the MEDIA folder
        output_filename = f"completed_{task_id}_{task.original_filename}"
        
        # Ensure media directory exists
        if not os.path.exists(settings.MEDIA_ROOT):
            os.makedirs(settings.MEDIA_ROOT)
            
        output_path = os.path.join(settings.MEDIA_ROOT, output_filename)
        df.to_excel(output_path, index=False)

        # 5. Final Status Update
        task.status = 'COMPLETED'
        # Construct the URL for the frontend to download
        task.download_url = f"{settings.MEDIA_URL}{output_filename}"
        task.completed_at = timezone.now() # Sets the completion time to NOW (IST)
        task.save()
        print("process excel task ended.")

    except Exception as e:
        print(f"Critical Task Failure: {e}")
        task.status = 'FAILED'
        task.save()
        
    finally:
        # 6. Cleanup the temporary uploaded file
        if os.path.exists(file_path):
            os.remove(file_path)

@login_required
def title_creator_async_view(request):
    if request.method == 'POST' and request.FILES.get('file'):
        
        excel_file = request.FILES['file']
        selected_pt = request.POST.get('product_type')
        
        # 1. Save file temporarily
        fs = FileSystemStorage()
        filename = fs.save(f"temp_{uuid.uuid4().hex}_{excel_file.name}", excel_file)
        file_path = fs.path(filename)

        # 2. Create Task Record
        task_id = str(uuid.uuid4())
        ProcessingTask.objects.create(
            task_id=task_id,
            original_filename=excel_file.name,
            status='PENDING'
        )

        # 3. Start Background Thread
        thread = threading.Thread(
            target=process_excel_task, 
            args=(file_path, selected_pt, task_id)
        )
        thread.start()

        return JsonResponse({
            'status': 'started',
            'task_id': task_id,
            'message': 'File is processing in the background.'
        })
    
    return JsonResponse({'error': 'Invalid request'}, status=400)

# 2. This view is called repeatedly by pollStatus() in your JS
def check_status(request, task_id):
    # Look up the task in the database
    task = get_object_or_404(ProcessingTask, task_id=task_id)
    
    return JsonResponse({
        'status': task.status,               # 'PENDING', 'COMPLETED', or 'FAILED'
        'file_name': task.original_filename,
        'download_url': task.download_url    # This will be null until status is COMPLETED
    })

@login_required
def title_creator_history_page(request):
    # Renders the HTML page
    return render(request, 'title_creator_history.html')

@login_required
def get_title_creator_tasks_json(request):
    # Returns the list of tasks as JSON for the history table
    tasks = ProcessingTask.objects.all().order_by('-created_at')[:50]  # Latest 50 tasks
    data = []
    for t in tasks:
        data.append({
            'task_id': t.task_id,
            'filename': t.original_filename or "Unknown File",
            'status': t.status,
            'url': t.download_url,
            'date': t.created_at.strftime("%d %b %Y, %I:%M %p"),
            # Use a conditional (ternary) operator to handle the null case
            'completed_at': t.completed_at.strftime("%d %b %Y, %I:%M %p") if t.completed_at else ""
            # 'completed_at': t.completed_at.strftime("%d %b %Y, %I:%M %p")
        })
    return JsonResponse(data, safe=False)


class TokenFetcherAPI(APIView):
    def get(self, request):
        token = fetch_global_industrial_token()
        
        if token:
            return JsonResponse({
                "status": "success",
                "token": token
            }, status=status.HTTP_200_OK)
        
        return JsonResponse({
            "status": "error",
            "message": "Could not retrieve token"
        }, status=status.HTTP_500_INTERNAL_SERVER_ERROR)
    


def fetch_global_industrial_token():
    """Logic to scrape the token."""
    base_url = "https://www.globalindustrial.com"
    # Using cloudscraper to handle potential bot detection
    scraper = cloudscraper.create_scraper() 
    
    try:
        response = scraper.get(base_url, timeout=15)
        
        # 1. Check Cookies
        token = scraper.cookies.get('Authorization')
        if token:
            return token.replace('Bearer ', '').strip()

        # 2. Check __NEXT_DATA__
        soup = BeautifulSoup(response.content, 'html.parser')
        script_tag = soup.find('script', id='__NEXT_DATA__')
        if script_tag:
            data = json.loads(script_tag.string)
            token = data.get('props', {}).get('pageProps', {}).get('token')
            if token:
                return token
                
    except Exception as e:
        return None
    return None