On Fri, 4 Jan 2013 14:19:15 -0800
Markus Scherer <markus.icu_at_gmail.com> wrote:
> It is sometimes useful to design very low-level data structures or
> algorithms in alignment with some "block" boundaries, but they are
> really just artifacts of previous character allocations, and for
> nearly all text processing the blocks are not useful. Please use the
> Script and other properties for text processing.
For example, if unassigned characters were treated as belonging to the
script suggested by their block location, you could end up with
strings being correctly rendered even though they used characters
introduced in a later version of Unicode than the application's
components were designed for, and we wouldn't want that, would we?
Richard.
Received on Sat Jan 19 2013 - 10:15:02 CST
This archive was generated by hypermail 2.2.0 : Sat Jan 19 2013 - 10:15:04 CST