https://bz.apache.org/ooo/show_bug.cgi?id=97808
--- Comment #25 from Jim Jagielski <[email protected]> --- ICU uses the standard Unicode default for English, where lowercase sorts first and case only decides between otherwise identical strings. That gives io Io tu Tu, and Alan's abc aBc Abc ABC jump nice Nicely… The 4.1.17 output is called a "case-insensitive sort" - It isn't. A case-insensitive sort treats abc/aBc/Abc/ABC as equal, so their order would just follow the input. The output shows a consistent lowercase-first case-sensitive sort. We need to decide what the contract should be: 1. Keep the behaviour and fix the help. Lowercase-first is the Unicode default, so this is the lowest-risk option. 2. Honour the help in Calc only. Pass uppercase-first from Calc's sort path rather than changing the shared service. My vote would be for option 1 -- You are receiving this mail because: You are the assignee for the issue. You are on the CC list for the issue.
