image is the go-to crate for image loading in the Rust gamedev world, which is a pretty thankless and tedious task a lot of the time. I got in a Mood and said "If I were to replace this relatively heavy dependency in ggez with a lighter-weight one that only did what I need it to, what would it look like?" and gave it a try. The result was only a couple hundred lines of code... mostly copy-pasted from image with minor modifications, very nicely readable, and using the tiff, jpeg-decoder etc crates which this project also maintains. It is simpler and compiles faster... but the differences aren't actually big enough to make it worth having a separate crate for this when you can just use image's feature flags to chop out a bunch of the things you don't need and get most of the same benefits. I've seldom been so happy to have wasted a couple hours on something that turned out unnecessary.
So, yeah, you could totally make a simpler image that was geared towards gamedev only, but it's not worth it. About all I could ask for is have a 1.0 release and split imageops into its own crate, and both of those seem to be on the way. So as a user, all I really want from this project is to keep doing what you're doing.
Thank you for making something so good I can't improve it!
image
is the go-to crate for image loading in the Rust gamedev world, which is a pretty thankless and tedious task a lot of the time. I got in a Mood and said "If I were to replace this relatively heavy dependency inggez
with a lighter-weight one that only did what I need it to, what would it look like?" and gave it a try. The result was only a couple hundred lines of code... mostly copy-pasted fromimage
with minor modifications, very nicely readable, and using thetiff
,jpeg-decoder
etc crates which this project also maintains. It is simpler and compiles faster... but the differences aren't actually big enough to make it worth having a separate crate for this when you can just useimage
's feature flags to chop out a bunch of the things you don't need and get most of the same benefits. I've seldom been so happy to have wasted a couple hours on something that turned out unnecessary.So, yeah, you could totally make a simpler
image
that was geared towards gamedev only, but it's not worth it. About all I could ask for is have a 1.0 release and splitimageops
into its own crate, and both of those seem to be on the way. So as a user, all I really want from this project is to keep doing what you're doing.Thank you for making something so good I can't improve it!