You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
... crashes. Expected result is obviously no crashing, but to _correctly_ deal with UTF-8 errors as specified in The Unicode 5.2 standard, Chapter 3, section 3.9 Unicode Encoding Forms.
---------------------------------------
The following can be turned back in to a file by selecting all the pre lines and then using
This will create a file UTF-8-test.json. The reason for all this is that the file contains JSON with _very_ specific UTF-8 encoding errors that are likely to be stripped or mutilated. It requires bit perfect binary replication. Even editing with something like TextEdit.app is likely to destroy the UTF-8 errors contained within.
FYI, I'm the author of JSONKit. This is one of my "torture tests", but it's really nothing more than a JSON'ified version of the UTF-8 torture test (the link to the original is given in the .json file). But it does make for a good test!
Crap, how did this get closed? Did I click on the "save and close" on my comment? Can you re-open this? Is there seriously no way on github to "undo" this? :(
I have actually fixed this issue. (The fix to issue 42 fixed it, though only the crash; it's still rejecting the utf8 as malformed.) But I'm not closing it yet because I haven't actually added a test that shows this yet.
Using the following code (and the very special uuencoded file embedded in this ticket):
... crashes. Expected result is obviously no crashing, but to _correctly_ deal with UTF-8 errors as specified in The Unicode 5.2 standard, Chapter 3, section 3.9 Unicode Encoding Forms.
---------------------------------------
The following can be turned back in to a file by selecting all the
pre
lines and then usingThis will create a file
UTF-8-test.json
. The reason for all this is that the file contains JSON with _very_ specific UTF-8 encoding errors that are likely to be stripped or mutilated. It requires bit perfect binary replication. Even editing with something like TextEdit.app is likely to destroy the UTF-8 errors contained within.The text was updated successfully, but these errors were encountered: