I found this design doc about Alphabetic Index, which is helpful: http://site.icu-project.org/design/alphabetic-index
I have a question about Han collation, specifically stroke count collation, and and getting the entries assigned to the correct index characters. The document above states:
"Rather than have some special syntax, the plan is to introduce special
private-use codes, one for each normal index boundary. These will be
added to the rules for each of the above collations."
How exactly would we go about using that to get an index to work correctly?
We sort by stroke count, and that works fine. We also have an index with digits (1 2 3 etc.). But it is unclear to us how we assign the characters to the buckets in the index?
AlphabeticIndex is very popular this week :-)
If you create an AlphabeticIndex for "zh_TW" or "zh_Hant" you should
automatically get it with stroke count labels/buckets. Call addRecord() and
let it bucket & sort for you. Or, call getBucketLabels() and
getBucketIndex() to get the bucketing and do your own in-bucket sorting.
Get latest updates about Open Source Projects, Conferences and News.