I want to allow the user to draw on an iOS 11 PDFKit document viewed in a PDFView. The drawing should ultimately be embedded inside the PDF.
The latter I have solved by
In the end I solved the problem by creating a PDFViewController class extending UIViewController and UIGestureRecognizerDelegate. I added a PDFView as a subview, and a UIBarButtonItem to the navigationItem, that serves to toggle annotation mode.
I record the touches in a UIBezierPath called signingPath, and have the current annotation in currentAnnotation of type PDFAnnotation using the following code:
override func touchesBegan(_ touches: Set, with event: UIEvent?) {
if let touch = touches.first {
let position = touch.location(in: pdfView)
signingPath = UIBezierPath()
signingPath.move(to: pdfView.convert(position, to: pdfView.page(for: position, nearest: true)!))
annotationAdded = false
UIGraphicsBeginImageContext(CGSize(width: 800, height: 600))
lastPoint = pdfView.convert(position, to: pdfView.page(for: position, nearest: true)!)
}
}
override func touchesMoved(_ touches: Set, with event: UIEvent?) {
if let touch = touches.first {
let position = touch.location(in: pdfView)
let convertedPoint = pdfView.convert(position, to: pdfView.page(for: position, nearest: true)!)
let page = pdfView.page(for: position, nearest: true)!
signingPath.addLine(to: convertedPoint)
let rect = signingPath.bounds
if( annotationAdded ) {
pdfView.document?.page(at: 0)?.removeAnnotation(currentAnnotation)
currentAnnotation = PDFAnnotation(bounds: rect, forType: .ink, withProperties: nil)
var signingPathCentered = UIBezierPath()
signingPathCentered.cgPath = signingPath.cgPath
signingPathCentered.moveCenter(to: rect.center)
currentAnnotation.add(signingPathCentered)
pdfView.document?.page(at: 0)?.addAnnotation(currentAnnotation)
} else {
lastPoint = pdfView.convert(position, to: pdfView.page(for: position, nearest: true)!)
annotationAdded = true
currentAnnotation = PDFAnnotation(bounds: rect, forType: .ink, withProperties: nil)
currentAnnotation.add(signingPath)
pdfView.document?.page(at: 0)?.addAnnotation(currentAnnotation)
}
}
}
override func touchesEnded(_ touches: Set, with event: UIEvent?) {
if let touch = touches.first {
let position = touch.location(in: pdfView)
signingPath.addLine(to: pdfView.convert(position, to: pdfView.page(for: position, nearest: true)!))
pdfView.document?.page(at: 0)?.removeAnnotation(currentAnnotation)
let rect = signingPath.bounds
let annotation = PDFAnnotation(bounds: rect, forType: .ink, withProperties: nil)
annotation.color = UIColor(hex: 0x284283)
signingPath.moveCenter(to: rect.center)
annotation.add(signingPath)
pdfView.document?.page(at: 0)?.addAnnotation(annotation)
}
}
The annotation toggle button just runs:
pdfView.isUserInteractionEnabled = !pdfView.isUserInteractionEnabled
This was really the key to it, as this disables scrolling on the PDF and enables me to receive the touch events.
The way the touch events are recorded and converted into PDFAnnotation immediately means that the annotation is visible while writing on the PDF, and that it is finally recorded into the correct position in the PDF - no matter the scroll position.
Making sure it ends up on the right page is just a matter of similarly changing the hardcoded 0 for page number to the pdfView.page(for: position, nearest:true) value.