V-Mart Retail Store Data Extraction

Solution & Approach

• Dynamic Pagination Discovery: Built intelligent scraper that automatically detects total page count by parsing the 'Last' page link from V-Mart's store directory (stores.vmartretail.com). Implemented sequential page iteration to systematically crawl all store listings without hardcoding page limits, ensuring scalability as store count grows
• Multi-Attribute Store Parsing: Used BeautifulSoup with CSS class selectors to extract 5 key attributes per store. Retrieved store name from H2 tags, parsed address from span.store-address-info, extracted operating hours from span.store-time-info, and captured contact information from phone number spans. Implemented robust error handling for missing or malformed data fields
• Embedded Coordinate Extraction: Developed URL parsing logic to extract latitude/longitude coordinates directly from Google Maps direction links embedded in each store listing. Split href attributes on '/' and '=' delimiters to isolate coordinate strings in decimal-degree format, avoiding external geocoding API costs while maintaining accuracy
• Phone Number Processing: Implemented string manipulation to clean and standardize phone numbers from various formats. Removed special characters, normalized spacing, and validated digit counts to ensure consistent contact data across all store records
• Pandas DataFrame Structuring: Aggregated extracted data into Python lists during scraping phase, then batch-converted to Pandas DataFrame for efficient data manipulation. Structured dataset with indexed rows and labeled columns (Store Name, Address, Hours, Coordinates, Phone) for standardized output format
• CSV Export with UTF-8 Encoding: Generated output.csv with proper UTF-8 encoding to handle special characters in Indian store names and addresses. Exported 465 complete store records with sequential indexing, ready for integration with BI tools, GIS software, or database import