Exclusive C14N Last Call Comments
This page links to comments/issues raised on the list during Last Call (November 20 - December 11) of the Exclusive XML Canonicalization Version 1.0. It includes substantive issues, not necessarily the many welcome editorial suggestions and corrections.
A style sheet declaration is used to not render issues that are considered 'done'. Minority positions are also rendered as such. (Some issues which are considered done may not be rendered as such in order to draw attention to them for the Director review.) The source of an issue need not be its first instance, it also might reference a cogent description or WG poll. Also, this document may not capture editorial tweaks and errors that were easily and quickly remedied.
Resolutions are expressed as {email,minutes} » (resulting) draft
| Source | Isssue | Resolution |
| Donald Eastlake | Exclusive Canonicalization "with Comments"? | Reagle: Added a paremeter for including comments in the exc-c14n form. |
| David Orchard | the C14N specs do not deal with insignificant transformations permitted by various XML Schema built-in datatypes | Reagle: correct. |
| Paul Denning | Add a section 2.3 that uses actual SOAP messages as an example. | Reagle: I considered this but found it of little utility (see end of email for example). |
| Editorial Comments. | Reagle: done. | |
| Questions to the XP WG | Reagle: Interested in XP response but not in our path. | |
| Thomas Maslen | Many good questions on nuanced in the spec given implementation experience. | Reagle: done. |
| Joseph Reagle and John Boyer | When we moved from comma to whitespace delmited list of prefixes, we lost the capacity to specify default namespace nodes. | Reagle: fixed. |
| Philippe Le Hegaret | "According to the current draft, xml:base attributes are ommitted. This can lead to a wrong interpretation of the URIs inside the document subset." | Eastlake: "xml:base attributes are not omitted, they just aren't included from the nearest ancestor that has one in the apex of what is being canonicalized doesn't have one." » plh "I'm still not entirely convinced ... so I'm not opposed to move the WD forward. The CR phase would be the good step to test the relevance of the issue..." » Reagle: "It's definitely a tricky issue and we explicitly solicited feedback on these points in our CR, but haven't had any push-back, so I hope folks are largely satisfied and the text is clear." |
| Source | Isssue | Resolution |
| Joseph Reagle (UDDI) | When c14ning an attribute only, should its namespace declaration also be emitted? | Reagle: added warnings about "esoteric" nodesets. |
| Jeremy Carroll | Should a prefix that is utilized in an attribute value have its namespace declaration emitted? | Reagle: no, but we should give it some thought. Carroll: the result is understandable. |
| Joseph Reagle | Should xmlns:xml="http://www.w3.org/XML/1998/namespace" be emitted if utilized? (Or even accepted as legal?) Can another prefix be bound to the ".../1998/namespace" namespace? | Reagle: Tim Bray had commented that this was legal XML; though probably bad in practice. |
| Source | Isssue | Resolution |
| Merlin Hughes | A bug in the specification of the process and non-normative implementation | Reagle: Many improvements to specification. |
| John Boyer | Should xmlns="" be emitted at the root of fragments likely to be inserted elsewhere to isolate it from the parental target context? (This is similar to the problems of inserting XML into a document in which ancestors has the following elements, but the fragment doesn't (xml:base, xml:lang, xml:space)). | Hughes: This is a general problem of apps using XML, shouldn't include a fix in exc-c14n. Reagle: Following Merlin, includes text in xenc warning about this. |
| "If a document subset is to be canonicalized using 'Exclusive C14n', all namespace nodes in the original document are included in the document subset prior the serialization process; this inclusion is done regardless whether a namespace node is already in the subset or if it's excluded from the subset." This is done so as to ensure the result of the subsetting and exc-c14n is reparsable. | Hughes: it is possible for people to create problematic nodesets, but trying to prevent that via this menthod then creates problems we can not solve, and a different processing model than c14n. Reagle: this would be a substantive change to the processing. Boyer: satisfied with the wording. |
Last revised by Reagle $Date: 2002/07/11 19:51:54 $
=======