The exceptions you added a hotfix for in 0443f54 seem to have been caused by the browser trying to add the response to the browser state (when downloading files). Using open_novisit() leaves the state unchanged and seems more appropriate for file downloading.
Also, it ignores urls with https://learn.uwaterloo.ca/d2l/lor/viewer/view.d2l since they are not files in the same sense as the others. It may be worth eventually adding support for this custom content type.
The exceptions you added a hotfix for in 0443f54 seem to have been caused by the browser trying to add the response to the browser state (when downloading files). Using open_novisit() leaves the state unchanged and seems more appropriate for file downloading.
Also, it ignores urls with
https://learn.uwaterloo.ca/d2l/lor/viewer/view.d2l
since they are not files in the same sense as the others. It may be worth eventually adding support for this custom content type.