In addition, the default setting in Table 14, UTS #10, 6.0.0 are strength: tertiary alternative: shifted
But the setting won't generate the conformant behavior specified by CollationTest_SHIFTED.txt I think when alternative is set to shifted, strength should be set to quaternary (as default) unless it is explicitly set. Thanks, Matt On Tue, Nov 29, 2011 at 12:55 PM, Matt Ma <matt.ma.um...@gmail.com> wrote: > Thanks for clarification. But to pass UCA conformance test on Shifted, > does the strength have to be set to quaternary? Howeve, it is stated > in UCA, C2, "A conformant implementation shall support at least three > levels of collation". > > Does this mean a UCA conformant implementation only need pass UCA > conformance test on Non-Ignorable? > > Regards, > Matt > > On Tue, Nov 29, 2011 at 12:49 PM, Mark Davis ☕ <m...@macchiato.com> wrote: >> Yes, if the strength is tertiary, then Blanked and Shifted give the same >> results. >> http://www.unicode.org/reports/tr10/proposed.html#Variable_Weighting >> >> Mark >> — Il meglio è l’inimico del bene — >> [https://plus.google.com/114199149796022210033] >> >> >> On Tue, Nov 29, 2011 at 19:11, Matt Ma <matt.ma.um...@gmail.com> wrote: >>> >>> Hi, >>> >>> Does Shifted implies strength being quaternary? If strength stays as >>> tertiary (default or explicitly set), it seems the collation behavior >>> is Blanked. Please clarify. >>> >>> Thanks, >>> Matt >>> >> >> >