Nationwide Pre-Opening Restaurant Scraper

I need an automated crawler that can visit every state’s DBPR (Department of Business and Professional Regulations) site, locate newly licensed restaurants that have not yet opened, and harvest their basic contact details. Each record must capture the restaurant’s name, full street address, email address, phone number, projected opening date, and the owner’s name. Because every state maintains its own DBPR portal with slightly different layouts, the robot has to recognise those variations, navigate pagination or search forms, and normalise the data before exporting it.

The finished script should run unattended, iterate through all fifty states on a schedule I can trigger, and store the collected information in a clean CSV. I want restaurant name, address and email in three distinct columns, followed by the extra fields—phone, opening date and owner—so six columns in total. Please include meaningful error-handling for captchas or downtime, and log any skipped entries so I can review them later.

Deliverables
• Fully commented source code for the scraper (Python with Scrapy, Selenium, or another robust framework).
• One sample CSV showing at least a few live entries from different states in the required column order.
• A brief README explaining setup, required libraries, and how to schedule future runs.

I will consider the project complete once the script reliably pulls fresh pre-opening data from every state’s DBPR portal and produces a validated CSV in the specified structure.