Zorba

"FULLY-NORMALIZED" normalization form not supported

Bug #1187813 reported by Sorin Marian Nasoi on 2013-06-05

This bug affects 1 person

Affects		Status	Importance	Assigned to	Milestone
	Zorba	Fix Committed	Undecided	Sorin Marian Nasoi

Bug Description

The following test-cases fail although we state that Zorba supports "FULLY-NORMALIZED" normalization form:

"cbcl-fn-normalize-unicode-001" and "cbcl-fn-normalize-unicode-006" from "fn-normalize-unicode" test-set: pelase see attached logs for details.

Tags:

Related branches

lp:~zorba-coders/zorba/fix_bug_1187813

Merged into lp:zorba at revision 11553

Chris Hillery: Approve on 2013-07-12

Sorin Marian Nasoi: Approve on 2013-07-12

Revision history for this message

Sorin Marian Nasoi (sorin.marian.nasoi) wrote on 2013-06-05:

detailed logs for the 2 test-case failures. Edit (932 bytes, application/x-tar)

Revision history for this message

Paul J. Lucas (paul-lucas) wrote on 2013-06-06:

Well, we erroneously made that statement then and it should be retracted. ICU (upon which we rely for Unicode normalization) doesn't support "FULLY-NORMALIZED."

Changed in zorba:
status:	New → Won't Fix

Revision history for this message

Sorin Marian Nasoi (sorin.marian.nasoi) wrote on 2013-07-10:

In light of comment from Paul we have to:

- update the FOTSZorbaManifest.xml to state that we do not support "FULLY-NORMALIZED" unicode normalization form
i.e. set
<dependency type="unicode-normalization-form" value="FULLY-NORMALIZED" satisfied="false"/>

- remove the impacted test-cases from the list of expected failures