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
Though the paper was not polled, discussion suggested an emerging consensus for a different approach; to specify that passing one or more code unit sequences encoded in encoding A to an interface that expects the sequence to be encoded in encoding B results in the code unit sequence(s) being interpreted in encoding B. In other words, each code unit of the sequence retains its numeric value with the result that decoding it in encoding B potentially results in decoding errors and/or a different sequence of characters than would be produced using encoding A. Further effects are then proscribed by the receiving interface; if the code unit sequence (interpreted as encoding B) violates preconditions of the interface, then the behavior is undefined; otherwise, the behavior is well-defined, but erroneous if any of the code unit sequences would be interpreted differently using encoding A as opposed to encoding B.
I'm retaining the SG16 label and adding the needs-revision label. SG16 will look forward to reviewing an updated paper.
Activity
tahonermann commentedon Jun 5, 2025
SG16 reviewed P3671R0 during the 2025-05-28 SG16 meeting. No polls were taken; feedback was provided to the author.
Though the paper was not polled, discussion suggested an emerging consensus for a different approach; to specify that passing one or more code unit sequences encoded in encoding A to an interface that expects the sequence to be encoded in encoding B results in the code unit sequence(s) being interpreted in encoding B. In other words, each code unit of the sequence retains its numeric value with the result that decoding it in encoding B potentially results in decoding errors and/or a different sequence of characters than would be produced using encoding A. Further effects are then proscribed by the receiving interface; if the code unit sequence (interpreted as encoding B) violates preconditions of the interface, then the behavior is undefined; otherwise, the behavior is well-defined, but erroneous if any of the code unit sequences would be interpreted differently using encoding A as opposed to encoding B.
I'm retaining the SG16 label and adding the needs-revision label. SG16 will look forward to reviewing an updated paper.