Current documented extension: v0.0.16
A code-aligned account of how the extension captures long pages, how the editor paginates images, which limits are enforced, and how product guides are reviewed.
The extension scrolls the active page in viewport-sized steps, records each PNG slice, and keeps fixed or sticky interface elements in the first slice only. It retries temporary Chrome capture throttling up to three times and restores the page after success or failure.
The editor processes images locally in Canvas. Near each proposed A4 / Letter boundary it measures horizontal pixel transitions and searches backward for a lower-detail line. The preview remains the final check because dense images may not contain a safe visual gap.
Each capture is bounded at 60,000 CSS pixels and 32 MiB of encoded screenshot data; a batch is bounded at 80 MiB. Ordinary pages are processed one at a time with an 8-million-pixel budget. Long single-page images and undo history adapt to device memory. A limit may stop capture earlier.
Screenshot pixels, annotations, watermarks, previews, and generated PDFs stay in the browser. Product analytics is separate and excludes document content; the Google tag loads with analytics storage denied until the user grants optional analytics consent.
Chrome internal pages and the Chrome Web Store cannot be captured. Virtualized lists, cross-origin frames, video, WebGL, infinite feeds, late-loading media, and very long pages can require smaller sectional captures or an official export from the source service.
Product-specific documentation is checked against the current implementation and dated when reviewed. General archiving guidance is kept separate from legal, tax, accessibility, security, or records-management advice.
Material corrections update the reviewed date. To report a reproducible problem, include the public test URL when possible, browser and extension versions, viewport size, capture mode, and the smallest non-sensitive example that demonstrates the issue.
support@screenshotpdf.com