2639 – Hex and octal string values not completely specified

The spec implies that a string literal containing hex and octal values is assembled in UTF-8, but doesn't say so explicitly. The text says that hex and octal values can be used to build binary data. If you have something like:
dstring ds = "\U00101234\x7f"d
what are the actual bytes in the string ds? Without defining explicitly, it allows for
00 10 12 34 7f 00 00 00 
on a big-endian machine as opposed to the probably intended
00 10 12 34 00 00 00 7f

Also, if you define arbitrary binary data that is not valid utf-8, should an error reported when the 'd' suffix is processed by the compiler?
If the spec more clearly says that the stuff between quotes is treated as if a UTF-8 string is created before suffixes are applied, then I think it formalizes the intent.

*** Issue 3784 has been marked as a duplicate of this issue. ***

Commit pushed to https://github.com/D-Programming-Language/d-programming-language.org
https://github.com/D-Programming-Language/d-programming-language.org/commit/15151c41c4b35deaa690e7e052f74fb8f9d56010
fix Issue 2639 - Hex and octal string values not completely specified