- From: David A. Lee <dlee@calldei.com>
- Date: Mon, 29 Dec 2008 09:18:00 -0500
- To: "XProc Dev" <xproc-dev@w3.org>
- Message-ID: <D50550153D834F1795B42F7E5A1890C6@calldei.com>
Looking at the test case add-xml-base-003.xml I believe the intent is clear, but I cant see how the specs supports the interpretation. The relevent specs: " if the value of the relative option is true, a string which, when resolved against the parent's [base URI], will give the element's [base URI], otherwise the element's [base URI]." Given the parent's base URI: http://tests.xproc.org/tests/doc/xml-base-test.xml And the element's base URI: http://tests.xproc.org/tests/doc/xml-base-chap.xml And using the definition of "resolve" from both the Java SDK for URI and the RFC 2396 The string which resolved against the parent's [base URI] to give the element's baseURI is the full URI http://tests.xproc.org/tests/doc/xml-base-chap.xml But the test expectations (and what I think is the "intent") is that the base URI for purposes of resolve (but NOT for purposes of the xml:base attribute) is the parent's base URI minus the final file component. e.g. http://tests.xproc.org/tests/doc Hence the resultant string is "xml-base-chap.xml" which when resolved against http://tests.xproc.org/tests/doc produces http://tests.xproc.org/tests/doc/xml-base-chap.xml Hence I think the spec needs to be changed to specify that resolution is NOT against the parent's base URI but rather against the parent's base URI with the final path component removed. ----------------------------------------------------------- David A. Lee dlee@calldei.com http://www.calldei.com http://www.xmlsh.org
Received on Monday, 29 December 2008 14:18:36 UTC