How OnlineDocumentViewer Renders Documents in the Browser with Doconut SDK
OnlineDocumentViewer shows PDFs, Word files, CAD drawings, and dozens of other formats directly inside any web page. It does that by delegating the heavy‑lifting to Doconut SDK, a server‑side rendering engine that converts the original file into a pixel‑perfect image stream, then streams those images to the browser through a lightweight JavaScript wrapper.
Doconut SDK: The Rendering Core Behind Every Preview
Doconut SDK understands more than 50 native file types – PDF, DOCX, XLSX, PPTX, DWG, PSD, SVG, and many others. When a request arrives, Doconut parses the source document, resolves fonts and layers, and produces a raster representation that matches the original layout pixel for pixel. Because the conversion happens on the server, the client never needs Microsoft Office, AutoCAD, or any other desktop application.
From Token to Pixels: The End‑to‑End Request Flow
- Generate a token‑signed URL – Your back‑end calls the OnlineDocumentViewer REST API, passing the file identifier and optional flags such as
disableDownload: true. The API returns a short‑lived URL that embeds a cryptographic token. - Server invokes Doconut – The OnlineDocumentViewer service forwards the request to Doconut SDK, which performs the file conversion in a sandboxed environment.
- Pixel stream is sent back – Doconut streams the rendered pages as image tiles. The SDK stitches the tiles into a scrollable canvas, preserving zoom levels and scroll positions.
- Viewer displays the canvas – The front‑end sees a seamless, plugin‑free preview that behaves like a native PDF viewer.
This flow works the same way for a .NET back‑end, a Node.js service, or any platform that can issue an HTTP request, satisfying the “cross‑platform” requirement.
Pixel‑Perfect Fidelity: What High‑Fidelity Rendering Means for Your UI
High‑fidelity rendering guarantees that every glyph, vector shape, and layer appears exactly as the source author intended. For a PDF that contains embedded OpenType fonts, Doconut extracts the font data and rasterizes each glyph at the requested zoom level, so no fallback fonts appear. For a CAD drawing (DWG), line weights, hatch patterns, and layer visibility are reproduced without distortion. The result is a preview that looks identical to opening the file in its native desktop application, eliminating the visual gaps that often force developers to fall back to screenshots.
Security, Privacy, and GDPR Compliance Built In
- Token‑Based Access Control – Every preview URL contains a signed token that the server validates before invoking Doconut. If the token is missing or expired, the request is rejected.
- Automatic File Deletion – After the rendering pipeline finishes, the original file is removed from temporary storage. No persistent copy remains on the server.
- End‑to‑End Encryption – All traffic between your front‑end, the OnlineDocumentViewer service, and Doconut runs over TLS.
- Domain Restrictions – You can whitelist the domains that are allowed to load the viewer, preventing click‑jacking attacks.
These safeguards let you embed document previews in SaaS products that must meet strict data‑protection regulations.
Sub‑Second Load Times Even at Scale
Because Doconut performs the conversion on powerful cloud instances, most documents render in under one second, regardless of whether the client is on Windows, macOS, Linux, iOS, or Android. The viewer streams only the tiles that are currently visible, so bandwidth usage stays low even on mobile networks. The architecture is cloud‑native: multiple instances of the rendering service autoscale behind a global load balancer, delivering consistent latency worldwide.
What to Try Next
- Sign up at Online Document Viewer (https://onlinedocumentviewer.com) and upload a sample PDF.
Explore more about the rendering engine on the Doconut website: https://doconut.com.
For additional integration guides, visit https://doconut.app.
