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
There are two major stepping stones that cause troubles for us on Python 3 and we will have to decide how to deal with that. As it stands currently, URLs on Python 3 are a major pain since you can no longer have them encoded and pass them around user code without the user becoming incredibly frustrated.
Werkzeug currently sees bytestrings as representing URIs and unicode strings representing IRIs. This will no longer work on Python 3. Now the question is how we deal with that. We could change URIs to be unicode as well and forcing users to use IRIs everywhere internal. We once did this per accident for a single header and double encoding was the result. :-/
The other one is that we have some streams that accept both unicode and bytes currently, something that is impossible to do on Python 3. What do we do there?
The text was updated successfully, but these errors were encountered:
There are two major stepping stones that cause troubles for us on Python 3 and we will have to decide how to deal with that. As it stands currently, URLs on Python 3 are a major pain since you can no longer have them encoded and pass them around user code without the user becoming incredibly frustrated.
Werkzeug currently sees bytestrings as representing URIs and unicode strings representing IRIs. This will no longer work on Python 3. Now the question is how we deal with that. We could change URIs to be unicode as well and forcing users to use IRIs everywhere internal. We once did this per accident for a single header and double encoding was the result. :-/
The other one is that we have some streams that accept both unicode and bytes currently, something that is impossible to do on Python 3. What do we do there?
The text was updated successfully, but these errors were encountered: