20672 2013-01-15 10:35:20 +0000 [XT3TS] type-functions-0304 2015-10-16 11:20:46 +0000 1 1 1 Unclassified XPath / XQuery / XSLT XSLT 3.0 Test Suite Working drafts PC Windows NT CLOSED FIXED P2 normal --- 20602 1 tim abel.online abel.braaksma mike public-qt-comments oldest_to_newest 81387 0 tim 2013-01-15 10:35:20 +0000 This test assumes that URIs end up being percent encoded. e.g. <assert>/out/e3 = "true , http://base.example.org/ns/two%20organizations"</assert> Regarding resolve-uri, F&O 3.0 says "No percent-encoding takes place." 97300 1 abel.braaksma 2013-12-09 01:12:27 +0000 It looks like this, and related issues with resolve-uri do not fully adhere to this special case. I propose to go over the resolve-uri tests and improve and adjust where needed. 97303 2 tim 2013-12-09 09:39:11 +0000 Thanks. 119547 3 abel.braaksma 2015-04-15 18:56:07 +0000 Fixed for the mentioned test-case, by s/%20/ /. There were no other escapes in this test-set that I could find and a trivial search elsewhere didn't come up with other tests that have the same issue. Marked resolved/fixed, changes pushed. 122339 4 mike 2015-07-31 10:01:40 +0000 Reopening this, as the decision on this bug appears to contradict the decision on bug #20602, which I have reopened as a spec bug. 123724 5 abel.braaksma 2015-10-15 17:20:05 +0000 As discussed during the XSLWG telcon of 2015-10-15 this bug was re-assessed in light of bug 20602. Conclusion: the issue can be closed, there are no ambiguities with the QT30 test suite anymore.