Most particularly at a glance, I suspect ES3 allowed a percent sequence starting with an encoded octet with 5 or 6 upper bits set because Unicode allowed this at the time, but modern Unicode allows at most 4 upper bits to be set, and ES5 makes >4 an error (assuming it isn't already -- if it is, SpiderMonkey's current algorithm obscures it more than ES5's algorithm does).
Fixed at some point, jsstr.cpp's Decode now has |goto report_bad_uri;| if |n > 4|.
Status: NEW → RESOLVED
Last Resolved: 6 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.