malikolivier / fitrs Goto Github PK

Hi, this code panicked:

fn main(){
    let data = vec![0; 20*20];
    let mut primary_hdu = fitrs::Hdu::new(&[20, 20], data);
    primary_hdu.insert("KEYFLOAT", 299.999);
    fitrs::Fits::create("new_file.fits", primary_hdu).expect("Failed to create");
}
// thread 'main' panicked at 'attempt to subtract with overflow', fitrs-0.5.0\src\fits.rs:1011:25

Hi!

First, thanks for your work on this library; it has been useful for me.

Second, I apologise if this request comes as rude or inconsiderate - it is not meant to be as such.

My understanding is that fitrs, as it is GPLv3 licensed, will require any software that uses fitrs to also have a GPLv3 license. Personally, I do not have an issue with this. However, recently, I have learned that some of my collaborators at a university (I shall not mention which) are not allowed to use any GPL libraries (perhaps out of fear that the copyleft nature of the license would prevent any profits on any profitable software). In my industry (academia/astrophysics), we're unlikely to make any software of monetary value, but all the same, it seems that some people would not be able to access your code simply because of the license. For this reason, personally, I will be implementing the Mozilla Public License (https://www.mozilla.org/en-US/MPL/2.0/FAQ/) in all of my future works, as well as existing ones (when I get around to them). I wonder if this would be a suitable replacement for fitrs, such that everyone could benefit from the code. A good example of how the GPL license can generate a lot of extra work is here: https://astropy-healpix.readthedocs.io/en/latest/about.html#about

Again, personally, I don't have an issue with the GPL, and I strongly endorse the spirit of it. However, I feel that making my code as available as possible is more important, and the MPL seems to be a good fit for that. Having said all that, feel free to disregard everything I've said - it is your code!