Ticket #3884 (closed defect: worksforme)

SVN Diffs for #3884

 

Opened 4 years ago

Last modified 1 month ago

Collation contractions

Reported by: mark.davis(at)us.ibm.com Assigned to: eric
Priority: major Milestone: 4.1.1
Component: collation Version:
Keywords: collation Cc:
Load: Xref:
Java Version: Operating System: all
Project (C/J): ICU4C,ICU4J and ICU4JNI Weeks: 0.1
Review:

Description (Last modified by srl)

1. The function getContractions() that returns the contractions in the UCA is missing in Java; needs to be ported from C. There also needs to be a way to find out if a tailoring overrides a UCA contraction, so that it is not a contraction any more.

2. All rules that use prefixes need to be treated as contractions. Thus, because there is a rule:

= \u30A1|\u30FC

the string "\u30A1\u30FC" must show up in the tailored characters in getTailoredSet (UErrorCode &status) [both C and Java!]

Attachments

Change History

12/31/69 17:29:00 changed by notes2

part of this is done with 4165. Need to look into the tailoring set and NFC issue.

12/31/69 17:29:01 changed by auditor

  • 07/06/04 13:44:16 guest sent reply 1
  • 07/07/04 14:35:28 dougfelt changed notes2
  • 07/07/04 14:35:28 dougfelt moved from incoming to collation
  • 07/12/04 01:36:41 weiv changed notes2
  • Wed Sep 28 10:42:21 2005 weiv changed notes2: target: "3.2" to "3.6", comments: "
  • " to "part of this is done with 4165. Need to look into the tailoring set and NFC issue.",
  • Fri Oct 13 18:08:15 2006 andy changed notes2: target: "3.6" to "UNSCH",

07/06/04 13:44:16 changed by mark.davis(at)us.ibm.com

(Guest Reply)

Also, make sure that the tailoring set is NFC.

07/07/08 12:38:03 changed by srl

  • load changed.
  • xref changed.
  • java changed.
  • description changed.
  • priority changed from major to assess.
  • owner changed from weiv to eric.
  • revw changed.

07/21/08 08:59:09 changed by hchapman

  • priority changed from assess to major.
  • weeks changed from 0.5 to 0.1.
  • milestone changed from UNSCH to 4.1.1.

08/05/08 17:46:47 changed by eric

  • status changed from new to closed.
  • resolution set to worksforme.

Here's an email conversation with Mark Davis:

Hi Mark,

I'm investigating the subject bug, which you submitted four years ago. It seems that most of this has been done; I know that the APIs exist in Java, because your minLengthInChars program uses them.

Can you give me some idea about what's left to do and how important it is?

Regards, Eric

A quick test would be if <ka><length mark> in Japanese gets included in the contractions for Japanese. If that's in there, then the bug is all done.

Mark

I see <ka><length mark> in the *tailorings* set but not in the contractions set. Does this mean that the bug is not done?

If not, can you give me an idea of how important this bug is?

Regards, Eric

... closing is fine.

Mark

Add/Change #3884 (Collation contractions)




Anti spam check: