Ask HN: How does GoogleBot treat url encoding?

Suppose I have the following situation:

I used to have all my url's nicely url-encoded on my website in <a> tags.

Then someone made a mistake causing half of the url to be encoded and half of the url NOT to be encoded.

For normal webbrowsers, no problem, but what about GoogleBot?

Will GoogleBot understand that this is a single resource or will I end up with duplicate content & get punished by Google in the search rankings.