Automated Website Change Monitor with Email and Desktop Alerts (Python)

Websites change constantly — prices drop, admission deadlines move, job listings appear and disappear — but manually re-checking pages every day is tedious and easy to miss. This project solves that with an automated website change monitor written in Python: you register the URLs you care about, choose how often each one should be checked, and the system fetches the page on schedule, compares the fresh snapshot against the stored one, and alerts you the instant something relevant changes. Instead of diffing the entire page (which would trigger on every rotating ad or clock widget), it can watch just a CSS selector such as a price element or a notice board, and ignore rules filter out dynamic noise like visitor counters and timestamps.

The monitor sends two kinds of alerts: an email through SMTP with the change summary, and a desktop notification via the operating system's notification service. A web dashboard ties it all together — the tracked-site list with per-site status, a visual diff viewer showing exactly which lines changed, an alert history, and settings for check frequency, selectors and ignore rules. All snapshots are stored in SQLite, so any past change can be re-examined. It is a practical demonstration of web scraping, scheduling, text diffing and notification plumbing — skills that transfer directly to monitoring, QA automation and price-tracking work.

1. The user registers a URL with a check interval (default 30 minutes), an optional CSS selector to narrow the watched region, and optional ignore rules.
2. On schedule, the engine fetches the page with an HTTP timeout of 15 seconds, extracts the selected region, and normalizes whitespace.
3. The fresh content is hashed (SHA-256) and compared against the hash of the stored snapshot; ignore-rule elements are stripped before comparison.
4. If the hash differs, a line-level diff is computed and both snapshots are saved to the SQLite store with timestamps.
5. The engine fires the configured alerts — an SMTP email with the change summary and a desktop notification — and logs each alert with its delivery status.
6. If three consecutive fetches return HTTP 429, the site is auto-paused with exponential back-off and resumes automatically, respecting the server.
7. The dashboard reads the same SQLite store to show site status, the diff viewer, alert history and settings.

Tech stack:

• Python 3
• requests (page fetching)
• BeautifulSoup (HTML parsing & selector extraction)
• difflib (line-level diffing)
• smtplib + email (SMTP alerts)
• plyer (desktop notifications)
• SQLite (snapshot & alert store)
• HTML, CSS, JavaScript (dashboard)

• [Scheduled page checking] Each tracked URL is fetched on its own interval (default 30 minutes, configurable per site from 5 minutes to 6 hours), with automatic back-off after repeated HTTP 429 rate-limit responses.
• [CSS-selector targeting] Watch only part of a page — a .price-tag, a #notice-board — instead of the whole document, so unrelated layout edits never raise an alert.
• [Ignore rules] Selector-based exclusions (ads, live clocks, visitor counters, script/style blocks) are stripped before every comparison, eliminating the false positives that plague naive page monitors.
• [Visual diff viewer] Every detected change opens a line-by-line before/after diff with added and removed lines highlighted, exactly as the page source changed.
• [Email alerts] SMTP sends a change-summary email with the site name, what changed and a link to the diff; a daily digest mode can replace per-change emails.
• [Desktop notifications] OS-level popups appear on the machine running the monitor within seconds of a confirmed change, so alerts are seen even with the mailbox closed.
• [Snapshot history in SQLite] Every fetched snapshot is hashed and stored with its timestamp, making it possible to review what a page looked like at any point and when it changed.
• [Alert history & dashboard] A web dashboard lists all tracked sites with status, recent activity, the full alert log with delivery status, and per-site settings.

• Complete Python source code (fetch engine, scheduler, differ, alert modules)
• Web dashboard (tracked sites, diff viewer, alert history, settings UI)
• SQLite schema and snapshot/alert storage code
• Configuration guide (intervals, selectors, ignore rules, SMTP setup)
• Sample tracked-site configuration with example pages
• Project report PDF (background, methodology, architecture, results)
• PPT presentation for final review
• Viva Q&A preparation document (HTTP, scraping ethics, diffing, scheduling, notifications)