chrw error 5

[Elroy]

Option Explicit

Private Sub Form_Load()

  ' Of course this works because VB6 thinks it's assigning TWO characters, not one!
  Dim sUTF16 As String: sUTF16 = ChrW2(&H1F601)

  Debug.Print Len(sUTF16)       ' As proof, Len doesn't work.  It reports 2.

  ' And yeah, of course this works because it's just adding to the TWO characters (which should be ONE character in UTF-16.
  Dim sConct As String: sConct = " " & sUTF16 & " " 'String-Concatentation-Operators work

  '
  ' Now, let's show that InStr, InstrRev, Mid, DON'T work.


  sConct = ChrW2(&H1F601) & ChrW2(&H1F602) & ChrW2(&H1F603)
  sUTF16 = ChrW2(&H1F602)


  Debug.Print InStr(sConct, sUTF16)     ' Prints out 3, when the correct answer is 2.

  Debug.Print InStrRev(sConct, sUTF16)  ' Prints out 3, when the correct answer is 2.


  ' Knowing our sConct character is the second character, we could do the following:
  Dim sTest As String: sTest = Mid$(sConct, 2)
  ' But this isn't going to work at all, instead returning garbage because it cut one of our double-word characters in-two.


  ' I'll give you that Trim$ and StrComp probably work.
  ' Trim$ works because it's just looking for leading and trailing spaces.
  ' StrComp works because it's just comparing each of those double-word characters as two-characters-each.


  'And TextControl-Prop-assignments work too (incl. visualization)
  ' And I've got no idea what TextBoxW1 is, obviously another work-around.
  'TextBoxW1.Text = Trim$(sConct) '(as do Trim, LTrim and RTrim)
End Sub


'
' A work-around to make VB6 work with UTF-16.
' If we're willing to engage in work-arounds,
' we can make virtually any language do anything.
'
Public Function ChrW2(ByVal CharCode As Long) As String
  Const POW10 As Long = 2 ^ 10
  If CharCode <= &HFFFF& Then ChrW2 = ChrW$(CharCode) Else _
                              ChrW2 = ChrW$(&HD800& + (CharCode And &HFFFF&) \ POW10) & _
                                      ChrW$(&HDC00& + (CharCode And (POW10 - 1)))
End Function

Regarding the rest of post #12, as stated above, it's just more hyperbole, and not worthy of a response from me. I'll let others decide how to interpret it.

Hopefully, the above code sets the record straight about what works and what doesn't.

EDIT1: And just one last thought. To call UTF-16's double-word characters "corner-cases" just isn't right. That's just another way of saying that VB6 is really UCS-2, because the difference between UCS-2 and UTF-16 is precisely these "corner-cases". That's analogous to saying that all non-one-byte characters in UTF-8 are "corner-cases", which just isn't right at all. There are many languages that deal perfectly well with UTF-8. And those that don't, clearly state that they're ASCII/ANSI, and never try to claim that they're UTF-8 compatible, but just have trouble with the "corner-cases".