FileRepositoryDownloader returns 406 error when file name contains # character -even when URL encode

Hi everyone,

I’m running into an issue with ThingWorx FileRepositoryDownloader when trying to download files that contain a # character in their name.

For example, I have a file in my repository called:

When I build the URL dynamically using encodeURIComponent, it looks like this:

However, the request always fails with:

and the response shows Content-Length: 0 from the ThingWorx Nginx server.

If I rename the file (for example to test_2.pdf), the same request works perfectly.

It seems that even though the filename is correctly encoded (%23), the FileRepositoryDownloader servlet doesn’t handle # correctly inside the download-path query string.

Has anyone else encountered this before?

i am using webgl widget to render pdf in browser..

 output = "/Thingworx/FileRepositoryDownloader?download-repository=" + fileRepository + "&directRender=true&download-path=" + encodeURIComponent(path);

Is there a known workaround or a configuration change to make ThingWorx accept such filenames (without renaming them)?

Note All German characters in pdf name were causing an issue but after adding this encodeURIComponent(path) in generating pdf path all problems were resolved. special characters mean ü ö but this only file which has # hash sign is causing an issue. 

Any help or confirmation would be greatly appreciated