From: Markus Scherer Date: Thu, 14 Aug 2014 20:36:37 +0000 (+0000) Subject: ICU-11010 improve CollationKey.merge() API doc X-Git-Tag: milestone-59-0-1~1734 X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=1e5c95645d395bf23a604900860cd791b56eb486;p=icu ICU-11010 improve CollationKey.merge() API doc X-SVN-Rev: 36167 --- diff --git a/icu4j/main/classes/collate/src/com/ibm/icu/text/CollationKey.java b/icu4j/main/classes/collate/src/com/ibm/icu/text/CollationKey.java index 2a983015bfa..af98ad942dd 100644 --- a/icu4j/main/classes/collate/src/com/ibm/icu/text/CollationKey.java +++ b/icu4j/main/classes/collate/src/com/ibm/icu/text/CollationKey.java @@ -475,8 +475,7 @@ public final class CollationKey implements Comparable } - - /** + /** * Merges this CollationKey with another. * The levels are merged with their corresponding counterparts * (primaries with primaries, secondaries with secondaries etc.). @@ -484,7 +483,18 @@ public final class CollationKey implements Comparable * *

This is useful, for example, for combining sort keys from first and last names * to sort such pairs. - * It is possible to merge multiple sort keys by consecutively merging + * See http://www.unicode.org/reports/tr10/#Merging_Sort_Keys + * + *

The recommended way to achieve "merged" sorting is by + * concatenating strings with U+FFFE between them. + * The concatenation has the same sort order as the merged sort keys, + * but merge(getSortKey(str1), getSortKey(str2)) may differ from getSortKey(str1 + '\uFFFE' + str2). + * Using strings with U+FFFE may yield shorter sort keys. + * + *

For details about Sort Key Features see + * http://userguide.icu-project.org/collation/api#TOC-Sort-Key-Features + * + *

It is possible to merge multiple sort keys by consecutively merging * another one with the intermediate result. * *

Only the sort key bytes of the CollationKeys are merged.