Repository navigation
RecordScrape v1.0.0
·
70 commits
to 710003d578030c553dc7af9940614cc1486cd827
since this release
The first release of the Playwright based RecordScrape.
- Record a session in a real browser: clicks, typing and scrolling are kept across navigations and replayed on every run.
- Pick single elements, or a repeating list as a row table with two example clicks, then follow it across pages with a Next button or infinite scroll.
- Download any run as CSV, JSON or JSONL. CSV cells that a spreadsheet would run as formulas are written as text.
- Export a session as a flow file, edit it, and import it back.
- Record and run each session on Chromium, Patchright or CloakBrowser (optional extra), through an HTTP, HTTPS or SOCKS5 proxy whose credentials can come from environment variables.
- Run on a server through Waitress or Docker, with a bearer token required off loopback.
- Schedule a session to run every N minutes.