Closed daidaiworm closed 1 year ago
same problem, any soluton around ? Somehow instead of original text inside txt file, we get url from where it was downloaded
When passing just 1 arguments, the path gets validated to sniff between urls and arbitrary text content that should be downloaded with a generated file name. Leading dots in a url will break this detection because it looks for the given path appearing in a fully qualified path, and the fully qualified path doesn't contain the leading ".". This would be hard to patch without breaking existing workflows. I would suggest you normalize the url before passing to download(), here's a helper function to do that:
function normalize(path){
var a = document.createElement("a");
a.href=path;
return a.href;
}
tried both download.js version4.2 and version4.21.
My intention was to download a linked txt file. For some reason I can't specific absolute address, such as http://foobar.com/path/txtfile.txt What I can do is to give a relative address, like "./path1/path2/txtfile.txt"
It's ok to run
download(txtfile.txt)
orhttp://foobar.com/path/txtfile.txt
But when run
download(./txtfile.txt)
or../path/txtfile.txt
The txtfile.txt didn't get downloaded. What we get is just a file with contains the string only, for this case you get "../path/txtfile.txt"