This adds support to CarrierWave for the ultrafast and resource efficient VIPS library. It is ready for production - but if you do encounter any problems please be sure to report them on Github so that we can fix them.
gem install carrierwave-vips
You will also need ruby-vips. For instructions on how to install that see the repo: https://github.com/jcupitt/ruby-vips
If you are using bundler, add this to your Gemfile:
See the instructions for CarrierWave to use this processor. You will need to include the Vips module in the upload class:
class AvatarUploader < CarrierWave::Uploader::Base include CarrierWave::Vips end
You can use the following methods to change your images. All methods keep aspect ratio:
resize_to_fill(x,y)Will increase/decrease the size of the image and match the specified dimensions exactly, chopping off any extraneous bits.
resize_to_fit(x,y)Will increase/decrease the size of the image to fit within the specified dimensions. One dimension may be less than specified.
resize_to_limit(x,y)Just like resize_to_fit except will not increase size of image.
convert("jpeg|png")Changes the format of the image
quality(0-100)Sets the quality of the image being saved if JPEG
stripRemoves any exif and ICC metadata contained in the image to reduce filesize.
Please note that GIF writing is not supported by libvips, and therefore cannot be supported by ruby-vips nor this library. GIF reading is still supported.
In order to use the strip method, a recent version of libvips is required. It is recommended to use at least 7.30.2. Tested on 7.30 on Debian + OS X 10.7.3/HomeBrew
To see how vips stands up to other image processing libraries, see this benchmark: https://github.com/stanislaw/carrierwave-vips-benchmarks
- The carrierwave-daltonize gem corrects images for color-blindness: https://github.com/gingerlime/carrierwave-daltonize
- John Cupitt (@jcupitt)
- Stanislaw Pankevich (@stanislaw)
- Mario Visic (@mariovisic)