404-ed model still ended up in wllama_cache?
Closed this issue ยท 3 comments
Because I made a typo in the URL of a local model file I noticed something strange. It seems that invalid URL ended up in the wllama_cache
anyway.
I checked to see if my code was perhaps adding the URL to the cache manually, but didn't think so, and also didn't find that.
The result is that my projects thinks the model is loaded, but then Wllama crashes when loading it.
Perhaps it cached the 404 page?
(currently my code tries to recognise an event object, and see that as an indication that the (down)load process failed, and not the actual loading of the model itself. See also: #56
Oh yeah! I've faced this problem too!
FYI, for the downloading function, I'm planning to replace XMLHttpRequest
with modern fetch
API. This should fix the problem.