Add ability to work on links
I was told, that it would be valuable if mat2-web would be capable of cleaning files provided via a HTTP link, for example a .pdf made available on a website.
mat2-web should then probably download the file, clean it and serve it afterwards.
Without much thought put into this, I guess there are some dragons lurking here, as it might be easy to do some sort of denial of service, if the file is big, for example.
Content-Length should be checked and redirects should be ignored. Probably more stuff should be taken into consideration as well.