https://bugs.documentfoundation.org/show_bug.cgi?id=172646

--- Comment #6 from ady <[email protected]> ---
(In reply to Anders Lind from comment #5)
> I may be misunderstanding, but the characters I tested are standard Unicode
> characters rather than characters from the Private Use Area.
> 
> For example:
> 
> 🌴 = U+1F334
> 👑 = U+1F451
> 💶 = U+1F4B6
> 
> whereas the Unicode Private Use Areas are U+E000 - U+F8FF, U+F0000 - U+FFFFD
> and U+100000 - U+10FFFD. (ref.
> https://en.wikipedia.org/wiki/Private_Use_Areas )
> 
> Are you perhaps referring to some internal representation rather than the
> Unicode code points themselves?

You are correct. Here is what I meant...

In Calc, on a blank empty cell, go to menu Insert > Special Character.
Selecting the "Liberation Sans" font (used in attachment 207585 from comment
1), the set of available characters "ends" at U+FFFC. IOW, there is no way to
select / insert / search for U+1F334 or any of the other problematic characters
that you show in attachment 207585.

There is a similar limitation when using the Character Map on a Windows OS.

Since the characters are actually shown / displayed within Calc, I assume that
the actual font file(s) indeed include these characters.

Since this seems to be a regression, then at some point in time Calc was/is
limited in some of its built-in functions to handle these "out-of-range"
Unicode code points (?). Clearly the functions UNICODE() and UNICHAR() are not
limited, but the matching criteria in SUMIF() (and probably other
"text-criteria" dependent functions) are currently limited.

-- 
You are receiving this mail because:
You are the assignee for the bug.

Reply via email to