I have several SVG images that I would like to use in an iOS application, and, in short, I would like to turn the SVG images into UIImages (or CGImages).
My goal is that I should be able to load the images from the .svg files at an arbitrary size (assuming correct W/H ratio) and store them as UIImages or CGImages without any loss of image quality. (Note this all has to happen at runtime, pre-converting the images to various sized .png files and putting them in the App bundle isn't a viable option.)
Is this possible, and if so, how could I go about doing this? I have a good working knowledge of Core Graphics but I have never worked with vector graphics before.
SVGKit could help you - this library renders SVGs onto CALayer instances; from that, you can easily composite the image to a bitmap and make a CGImage of it.
I would not recommend converting a vectored document to a bitmap in most instances, but if you must SVGgh has a class SVGRenderer which has a method -(UIImage*)asImageWithSize:(CGSize)maximumSize andScale:(CGFloat)scale which will create a UIImage.
This is assuming SVGgh renders your SVGs properly, it doesn't support image effects, for instance.
#import <SVGgh/SVGgh.h>
-(UIImage) imageFromSVGURL:(NSURL*)svgURL withSize:(CGSize)maximumSize andScale:(CGFloat)scale
{
SVGRenderer* renderer = [[SVGRenderer alloc] initWithContentsOfURL:svgURL];
return [renderer asImageWithSize:maximumSize andScale:scale];
}
Related
I have a series of images that I would look to loop through using iOS's [UIView startAnimating]. My trouble is that, when I exported the images, they all came standard in a 240x160 size, although only 50x50 contains the actual image, the rest being transparent parts that are just taking up space.
When I set the frame of the image automatically using image.size.width and image.size.height, iOS takes into images' original size of 240x160, so I am unable to get a frame that conforms to the actual parts of the image. I was wondering if there is a way using Illustrator or Photoshop, or any other graphics editing software for me to export the images based on their natural dimensions, and not a fixed dimension. Thanks!
I am a fan of vector graphics and thinks everything in the world should be vector ;-) so here is what you do in illustrator: file - document setup - edit artboards. Then click on the image, and the artboard should adjust to the exact size. You can of course have multiple artboards, or simply operate with one artboard and however-many images.
I have a big picture to show in UIScrollView.It's 28.1 MB.And it often crash the app.Is there some methods to deal with it so that app won't crash?
Here you go buddy,
Convert Large Image to tiles
Displaying tiled images in UIScrollview
Tiled Images in UIScrollView
If it's a static PNG image, run it through ImageAlpha and ImageOptim (Google these). You might be able to get it down by well over 50%. Bundle the resulting image with your app instead.
Note that ImageOptim is lossless whereas ImageAlpha is lossy. You can use them in conjunction.
Split up the image and load the pieces only when necessary.
My app uses flat graphics, which mostly consists of lines and flat surfaces...
Whats the best approach for building the UI?
use PNG images as much as possible (like for the attached picture, the whole background would be an image)
use 4 UIViews (in this example case) with background color to make the background and the 3 lines
use CoreGraphics to actually draw in drawRect (or somewhere else?)?
subclass UIView and draw PDF content
any other approach?
What are the performance impacts? The advantage of the first two, is that they can be done in IB, but is there a downside (like performance or quality or caching)? I also heard of a trend of using CoreGraphics for drawing all the time...
So basically the fastest way is to use PNG files (stretchable and normal), also if you are using PNG files it will be easy to change the design and fix eventual bugs.
When you are using PNG files, try to reduce the size of the PNGs (stretchable images for backgrounds) for a better performance and a small size of the app.
CoreGraphics is more complex and can add strange bugs or performance problems if you don't know what you are doing, but can be use also for simple view styling.
On every project that I've done I couldn't use only one of these two because of the project complexity or because I was to lazy to open photoshop and add extract some designs.
As H2CO3 said is not that simple. Each format suits particular requirements, there are no written laws, I just ca tell you what I do.
Small pieces of UI, such as buttons, icons etc: I normally use PNG or PDF. You can subclass a UIView to draw PDF vector content. The plus with PDF is that if the source is a vector image, you could not take care about physical-logical points.
Images or Hi-res images, such as image of a content: In this case I use JPG, for they small size respect of PNG. They take little bit more to decompress but usually the size is very smaller compared to PNG.
In general PNG can use alpha, JPG no
Image is a big topic, nowadays I always try a way to find symmetry in UI elements and use stretchable UIImages with PNG.
For you example I will use PNG witch stretchable left and right side. Instead of use 640 px image, you can use 10 px image, with stretchable right margin and stretchable left margin.
[update]
It should be said that if you use stretchable images you will get a lot of performance boost, since the image will be smaller, you will occupy less memory on the "disk", less memory on the heap thatnks to the GPU stretching and less decompressing times
.
I have been designing an app with oCanvas.js. It's a really nice canvas library that makes it much easier to create an app that can create and manipulate images, but I ran into a snag when I was trying to implement image filters:
I need transparent backgrounds so that I can have multiple layers, each of which is represented by its own display object, rendered separately (meaning one at a time) on a hidden "staging" canvas. Immediately after being rendered, a layer is then drawn on top of the previous layers on the visible canvas, so that different image filters can be applied to each layer independently during render.
The issue I am having is that, when attempting to extract the image from an oCanvas object's canvasElement, the resulting images never have a transparent background. For example: Imagine I have a 50x50 canvas that has been oCanvas.create() processed, but has display: none; (this is used as the rendering canvas) and another canvas (same dimensions) without an oCanvas instance. I am trying to do something like this (Pseudocode):
visibleCanvas.getContext("2d").drawImage(MyOcanvasCore.canvasElement, 0, 0);
I have also tried using URL = MyOcanvasCore.canvasElement.toDataURL() and then having my visibleCanvas do a drawImage with src=url.
The images always transfer, but they have a white background, even though I specify background: "transparent" during canvas.create(). As such, they completely overwrite all previous layers.
Do you have any tips for me? Am I doing it wrong? I tried transferring stuff from one canvas to another using classic drawRect, drawImage, etc methods, and transparency was retained. That's why I believe it is either the library or my code.
I guess you are using the image format other than png.
You should have your image format in PNG which keeps all the details of every pixel >including transparency of pixels
and not a compressed format.
So just try keep your image in png format after edting them in some image editor and save > result as .png .
Is there a way to use an icns file as the image source for I guess a UIImage? Or pick out one of the representations and resize?
NSImage does not seem to be available in cocoa-touch.
In particular I'd like to use them for drawing in the imageView part of a UITableViewCell.
There is no built-in way to do this and I see little benefit in doing so. It's best to just use the sizes that you need and ship them as individual images in your app. If you want to draw the images in a UITableViewCell, performance is an important aspect and resizing images won't be fast.
If you really need images in many different sizes and really don't want the best performance by including images for each size, you could use a PDF to make use of vector graphics. There is a nice article about this technique: http://mattgemmell.com/2012/02/10/using-pdf-images-in-ios-apps/
Honestly, don't try to use .icns files in your iOS app. It's just a container for images of different sizes and resizing is not a good idea, anyway. Either include individual images for any size you need or use vector graphics using a PDF file.