Showing documents

The signer reads the document in your application, page by page, with the values already filled into its form drawn as they are. Every page drawn is remembered with a hash of its pixels, so a reviewer can later tell which pages were on screen before the signature was given.

See it running: Show a PDF.

Setup

1. Add SignArgs → PDF Worker to any object in the scene. One per scene is enough; every SignArgs component shares it.
2. Under a canvas, create a RawImage for the page and add an AspectRatioFitter set to Fit In Parent, so pages keep their proportions whatever their size.
3. Add SignArgs → PDF View and set:

Runtime

await view.OpenAsync(pdfBytes);            // opens and shows page 1; closes what was open
await view.OpenAsync(pdfBytes, password);  // for an encrypted PDF

view.NextPage();                           // for buttons; clamps at the ends, logs failures
view.PreviousPage();
await view.ShowPageAsync(2);               // zero-based; completes when the page is on screen

view.PageShown += index => label.text = $"Page {index + 1} of {view.PageCount}";

view.Close();                              // frees the document once a render in flight ends

IsOpen, PageCount and PageIndex (-1 before the first page shows) describe the current state. Page requests are served in order, so pressing Next three times quickly shows three pages in turn.

Opening throws PdfLoadException when the bytes are not a PDF, the password is wrong, or the file is encrypted in a way the engine does not support; its Error says which.

What gets recorded

view.Rendering returns the resolution and render flags plus every page shown since the document was opened, each with its pixel size and SHA-256 over its pixels. LocalFinalizer.ConfirmAsync reads it for you and puts it into the signature record. Each distinct rendering of a page is recorded once, in the order it was first shown; showing the same page again with the same pixels adds nothing.

Form fields

PdfView draws the values of the document’s AcroForm fields as part of the page. To read which fields a page declares, for example to find where its signature boxes are:

IReadOnlyList<PdfFormField> fields = await view.GetFormFieldsAsync(pageIndex);
foreach (var field in fields)
    Debug.Log($"{field.Name} {field.Kind} {field.Area.X},{field.Area.Y} {field.IsReadOnly}");

Areas are in PDF points, origin at the bottom-left of the page. Field values are drawn but not editable: PdfView is a renderer, not a form editor. See Templates and form fields for turning signature boxes into signing fields.

Performance

On the reference phone (a OnePlus Nord on Android 12, IL2CPP, stripping “High”), rendering the first page of a 100-page document at 150 dpi took 42 to 101 ms depending on the build, a page turn including texture upload about 12 ms, and the Sign a document sample shows its first page about 300 ms after Play, opening and hashing included. The pixel buffer and texture are reused between pages of the same size. See Platforms for the dated measurements.

Limits

PdfView shows one page at a time. It has no text selection, search, zoom or pan gesture, thumbnails, links, outlines or annotation editing. Put the RawImage inside a ScrollRect or your own zoom control if you need them; a higher dpi keeps zoomed text sharp at the cost of memory.