Data Faker

😫 The Pain Point

You are testing a new CRM or Website. You need 1,000 users to test performance. You can’t use real customer data (GDPR/Privacy). Typing “User1”, “User2” manually is slow and doesn’t reveal real-world bugs (like long names breaking the layout).

🚀 Agentic Solution

Faker Library: Generates realistic data for any country (Vietnam, US, Japan…).

Key Features:

• Localization: Generates “Nguyen Van A” for VN locale, “John Smith” for US locale.
• Diversity: Can generate Addresses, Job Titles, Credit Card Numbers, etc.

⚔️ Phase 1: Commander (Quick Fix)

For generating a quick test file.

Prompt:

“Generate a CSV file named fake_users.csv with 100 rows. Columns: Name, Phone, Email, Address. Use the faker library with Vietnamese locale (vi_VN) to make it look realistic.”

Result: A CSV file full of fake data.

🏗️ Phase 2: Architect (Permanent Tool)

For QA Testers/Developers.

Engineering Prompt:

**Role:** Python Tool Developer
**Task:** Create a "Test Data Generator".
**Requirements:**
1.  **GUI:**
    *   Input: Number of rows (e.g., 1000).
    *   Checkboxes for columns: Name, Email, Phone, Address, Company, Job.
    *   Dropdown: Locale (en_US, vi_VN, ja_JP).
    *   "Generate" button.
2.  **Logic:**
    *   Initialize `Faker(locale)`.
    *   Loop N times.
    *   Create DataFrame.
    *   Export to Excel/CSV.
3.  **Deliverables:** `data_faker.py`, `run.bat` (Windows), `run.sh` (Mac).

🧠 Prompt Decoding

• Locale Support: The Faker library is magical because it knows the context. If you choose ‘vi_VN’, it won’t just give Vietnamese names, it will give addresses like “Quan 1, TP HCM” instead of “New York”.

🛠️ Instructions

1. Copy Prompt -> Paste -> Run.
2. Select Rows & Type -> Generate.