'Why app Terminates due to memory waring when convert pdf page to high quality image in ios, swift in real device
I'm trying to get the image(high quality) of each pdf page. I'm using below code running through a for loop until page count and it works.
guard let document = CGPDFDocument(pdfurl as CFURL) else { return }
guard let page = document.page(at: i) else { return }
let dpi: CGFloat = 300.0/72.0
let pagerect = page.getBoxRect(.mediaBox)
print(pagebounds)
print(pagerect)
let render = UIGraphicsImageRenderer(size: CGSize(width: pagerect.size.width * dpi, height: pagerect.size.height * dpi))
let imagedata = render.jpegData(withCompressionQuality: 0.5, actions: { cnv in
UIColor.white.set()
cnv.fill(pagerect)
cnv.cgContext.translateBy(x: 0.0, y: pagerect.size.height * dpi)
cnv.cgContext.scaleBy(x: dpi, y: -dpi)
cnv.cgContext.drawPDFPage(page)
})
let image = UIImage(data: imagedata)
I'm getting following issues with this ...
- sometimes the
imageisnil. - When this runs, the usage of
memory is very high. - With the page count(number of pages), usage of memory is very very high, and sometimes it goes to
1.4 GBand suddenly it crashes the app with the warning :Terminate due to memory waring. then I tried to run above code insideautoreleasepool. it did work but when the memory usage is more high (when it near to RAM size), again app crashes with above warning.
How can I avoid this memory warning and get the quality image form pdf page. hope any help. have a nice day.
Solution 1:[1]
1) sometimes the image is nil.
- Is there a reason that you are generating a jpeg data then converting to
UIImageVS directly creating an uiimage (usingfunc image(actions: (UIGraphicsImageRendererContext) -> Void) -> UIImage)? - If you really need to use the jpeg method, then don't directly instantiate from
UIImage(data:), use CGImage'sinit?(jpegDataProviderSource source: CGDataProvider, decode: UnsafePointer<CGFloat>?, shouldInterpolate: Bool, intent: CGColorRenderingIntent)then useUIImage(cgImage:)to get yourUIImageinstance
2) When this runs, the usage of memory is very high
- Are you storing all images created per each page? If yes, then if you have a pdf of high number of pages then you will consume max memory at some point because of accumulated images. Why don't you store to disk each image created then releasing it afterwards so you don't accumulate the memory of storing all pages in memory.
- Sharing the loop(assuming there is a loop) outside of this snippet could help solve your issue more
Solution 2:[2]
If you are facing this issue then try this:
autoreleasepool {
guard let page = document.page(at: i) else { return }
// Fetch the page rect for the page we want to render.
let pageRect = page.getBoxRect(.mediaBox)
var dpi: CGFloat = 1.0
if pageRect.size.width > pageRect.size.height {
dpi = 3508.0 / pageRect.size.width
} else {
dpi = 3508.0 / pageRect.size.height
}
//dpi = 300
let format = UIGraphicsImageRendererFormat()
format.scale = 1
let renderer = UIGraphicsImageRenderer(size: CGSize(width: pageRect.size.width * dpi, height: pageRect.size.height * dpi), format: format)
let imagedata = renderer.jpegData(withCompressionQuality: 1.0, actions: { cnv in
UIColor.white.set()
cnv.fill(pageRect)
cnv.cgContext.translateBy(x: 0.0, y: pageRect.size.height * dpi)
cnv.cgContext.scaleBy(x: dpi, y: -dpi)
cnv.cgContext.drawPDFPage(page)
})
let image = UIImage(data: imagedata)
}
autoreleasepool - for permanent memory clearing
scale - so that images for different devices are not created, which increases their resolution by 2 or 3 times
changed the way to increase dpi as it can be initially more or less than 72
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | Joshua Francis Roman |
| Solution 2 |
