datadump.txt

================================================================================
DATA DUMP METHOD — SupleCost Product Import Pipeline
================================================================================
Date: 2026-06-30
Product: MuscleBlaze Biozyme Iso-Zero (Low Carb Ice Cream Chocolate, 500g)
Source: https://www.muscleblaze.com/sv/muscleblaze-biozyme-iso-zero/SP-45405
Target: Supabase (czdvxcuuylfgdltttnnt.supabase.co)
Website: https://varunsonline.com
================================================================================

OVERVIEW
--------
Extract product data from a product page using the built-in browser, then
insert/upsert all 50 fields directly into Supabase via the service role key.

NOTE: cost_per_serving is AUTO-CALCULATED as: best_price / total_servings
      This happens at 3 layers: DB trigger, API reshape, and import script.
      See: scripts/migration-cost-per-serving.sql

================================================================================
TOOLS USED
================================================================================

1. openwork-browser_navigate_page  — Navigate to product URL
2. openwork-browser_take_snapshot  — Read visible text (a11y tree)
3. openwork-browser_evaluate_script — Execute JS in page context
4. openwork-browser_click          — Click tabs to reveal hidden content
5. write                           — Create the import script
6. bash (npx tsx)                  — Run the import script

================================================================================
STEP-BY-STEP PROCESS
================================================================================

STEP 1: NAVIGATE
-----------------
Open the product page in the built-in browser.

STEP 2: EXTRACT PAGE TEXT
--------------------------
Take an a11y snapshot and extract visible text (product name, price, flavor,
description, rating, etc.).

STEP 3: EXTRACT STRUCTURED DATA (JavaScript)
----------------------------------------------
Inject JavaScript via evaluate_script to extract:

  a) window.__NEXT_DATA__ — Next.js serialized JSON containing:
     - Product name, brand, category, pricing (MRP, offer price, discount)
     - All image URLs with multiple resolutions
     - Variant/attribute details (flavors, weights, SKU IDs)
     - Manufacturer details, country of origin, FSSAI code
     - UPC/EAN codes (pv_upc, pr_Upc)
     - Ratings, review counts
     - Goals/concerns
     - Food type (vegetarian/vegan)
     - Shelf life
     - Vendor information
     - Available variants with per-variant pricing

  b) document.querySelectorAll() — DOM traversal for:
     - <meta> tags (OG description, meta description)
     - Tab content sections (Description, Key Benefits, Nutritional Info)

  c) Tab interaction — Click on each tab section (Key Benefits, Nutritional
     Information) to reveal hidden content and capture it via snapshot.

STEP 4: MAP TO 50-FIELD SCHEMA
--------------------------------
Map extracted data to database columns:

  #  Field                    DB Column            Table
  -- -----------------------  -------------------  ----------------
   1  Name                    name                 products
   2  Brand                   brand_id             products (FK→brands)
   3  Category                category_id          products (FK→categories)
   4  Slug                    slug                 products
   5  Description             description          products
   6  Short Description       short_description    products
   7  Image URLs              images[]             products
   8  Availability            is_active            products
   9  Net Weight              net_weight           products
  10  Serving Size            serving_size         products
  11  Total Servings          total_servings       products
  12  Cost Per Serving        cost_per_serving     products
  13  Protein Per Serving     protein_per_serving  products
  14  Creatine Per Serving    creatine_per_serving products
  15  Carbs Per Serving       carbs_per_serving    products
  16  Fats Per Serving        fats_per_serving     products
  17  Calories Per Serving    calories_per_serving products
  18  Flavor                  flavor[]             products
  19  Key Ingredients         key_ingredients[]    products
  20  Benefits                benefits[]           products
  21  Full Ingredients List   full_ingredients_list products
  22  Structured Ingredients  ingredients (JSONB)  products
  23  Nutrition Facts         nutrition (JSONB)    products
  24  Usage Instructions      usage_instructions   products
  25  Warnings                warnings             products
  26  Best Time To Use        best_time_to_use     products
  27  Manufacturing Standard  manufacturing_standard products
  28  Country Of Origin       country_of_origin    products
  29  UPC / EAN               upc_ean              products
  30  Goals                   goals[]              products
  31  Certifications          certifications[]     products
  32  Is Vegetarian           is_vegetarian        products
  33  Is Vegan                is_vegan             products
  34  Lab Tested              lab_tested           products
  35  Lab Test Details        lab_test_details     products
  36  Brand Logo URL          (via brands.logo_url) brands
  37  Value Score             value_score          products
  38  Meta Title              meta_title           products
  39  Meta Description        meta_description     products
  40  Best Price              best_price           products (denormalized)
  41  Original Price          original_price       products (denormalized)
  42  Discount Percentage     discount_percentage  products (denormalized)
  43  Retailer ID             retailer_id          product_prices
  44  Retailer Current Price  current_price        product_prices
  45  Retailer Original Price original_price       product_prices
  46  Retailer Discount %     discount_percentage  product_prices
  47  In Stock                in_stock             product_prices
  48  Buy URL                 url                  product_prices
  49  Is Main Link            is_main_link         product_prices
  50  Extended Nutrition      (stored in nutrition products
                              JSONB or as image URL)

STEP 5: WRITE IMPORT SCRIPT
----------------------------
Create a TypeScript file (scripts/import-<product-slug>.ts) that:

  1. Loads environment variables via dotenv
  2. Connects to Supabase using the service role key (supabaseAdmin)
  3. Ensures related records exist (brand, category, retailer)
  4. Upserts the product row
  5. Inserts product price per retailer
  6. Creates affiliate link record
  7. Links filter options (goals → product_filter_values)

STEP 6: EXECUTE SCRIPT
-----------------------
Run: npx tsx scripts/import-<product-slug>.ts

This inserts data directly into the same Supabase instance that powers
varunsonline.com (via the GET /api/supplements endpoint).

STEP 7: VERIFY
---------------
Re-query the database to confirm all fields were saved correctly.

================================================================================
DATABASE TABLES POPULATED
================================================================================

  Table               Purpose
  ------------------- -------------------------------------------
  brands              Product brand (MuscleBlaze)
  categories          Product category (Isolate Protein)
  retailers           Seller/store (MuscleBlaze Official)
  products            Core product data (all 50 fields)
  product_prices      Price per retailer
  affiliate_links     Trackable outbound URL
  product_filter_values  Link products to filter options (goals)

================================================================================
IMPORTANT NOTES
================================================================================

- The supabaseAdmin client uses SUPABASE_SERVICE_ROLE_KEY (bypasses RLS).
- The product must have is_active = true to appear on the public website.
- The products page queries /api/supplements which always filters
  is_active = true.
- The DEALS section on the homepage requires discount_percentage > 20%.
- The SUPPLECOST RECOMMENDED section requires brand_id = 'supplecost-products'.
- The product is immediately visible after upsert — no redeploy needed since
  the Vercel frontend reads from the same Supabase instance at runtime.

================================================================================
PRODUCT 2: MuscleBlaze Biozyme Whey PR
================================================================================

Source Page:
  https://www.muscleblaze.com/sv/muscleblaze-biozyme-whey-pr/SP-122029?navKey=VRNT-249465

Script:      scripts/import-muscleblaze-biozyme-whey-pr.ts
Product ID:  muscleblaze-biozyme-whey-pr
Slug:        muscleblaze-biozyme-whey-pr
Category:    whey-protein
Flavor:      Molten Chocolate Cake
Weight:      1 kg (2.2 lb)
Serving:     45 g
Servings:    22
Protein:     30 g
Calories:    171
Best Price:  ₹3,949
MRP:         ₹3,999
Discount:    1%
UPC:         8906165785375

Key Differentiator:
  Whey Concentrate with Creatine Monohydrate (CreAMP) & AstraGin®.
  Unlike Biozyme Performance Whey (25g protein, no creatine) and
  Biozyme Iso-Zero (27g isolate, no creatine), this PR variant
  delivers 30g protein + 3g creatine per serving.

Cost Per Serving:
  ₹3,949 / 22 servings = ₹179.50

Extraction:
  - Navigate to product URL
  - Extract __NEXT_DATA__ via evaluate_script
  - Key fields under pageProps.data.results:
    - nm → product name
    - offer_pr → best price
    - mrp → original price
    - discount → discount %
    - pr_img.o_link → main image
    - pv_upc → UPC/EAN
    - hasAttributes → serving size, servings count, protein %
    - urlFragment → product URL slug
    - nut_info_grp → nutritional highlights
  - Additional images from thumbnail img elements

Verification:
  https://varunsonline.com/products/muscleblaze-biozyme-whey-pr
  → All fields display correctly including COST/SERVING = ₹179.5

================================================================================