Re: Editor's draft for review by 28 June 2007

Despite what I said on the call, I think I failed to fix these.

/ ht@inf.ed.ac.uk (Henry S. Thompson) was heard to say:
| 3.6 -- 3.8 are in an odd order, and 3.7 seems to have lost the
| definition of 'extension element' which used to be there (parallel to
| the definition of 'extension attribute' in 3.6).

Oops.

| 3.7 and 3.8 appear to me to contradict each other (extensions are
| allowed in subpipes vs. extensions only allowed as children of
| p:pipeline(-library), and I can't tell by looking at your relaxng
| grammar which you intend.  The difference between being ignored and
| being an extension is not at all clear.  Is the following a correct
| summary of what's intended:
|
|  1) ignore-prefixes may appear on p:pipeline or p:pipeline-library;

Right.

|  2) elements in namespaces declared as ignorable (_ignorable
|     elements_) may appear only as the children of containers (and
|     p:pipeline-library?);

Right.

|  3) ignorable elements are either recognised by implementations as
|     extensions, and interpreted as per the rules in 3.7, or ignored;

Right.

|  4) prefixed attributes are _all_ ignorable;

Right.

|  5) ignorable attributes are either recognised by implementations as
|     extensions, and interpreted as per the rules in 3.6, or ignored.

Right.

| If that's what you mean, I don't thing the current prose says so very
| clearly at all, sorry. . .

Well. Ok. I'll take another pass for next time.

| It also might help to clarify that the consequence of this, and of
| steps being named by QNames, is that the decision tree for elements in
| containers is:
|
|  1) in XProc namespace?
|     Check against grammar, interpret per spec.
|  2) in ignorable namespace?
|     2a) Known extension?
|         Process as appropriate.
|     2b) Otherwise
|         Ignore.
|  3) names a declared step type?
|     Check against grammar, interpret per spec.
|  4) otherwise
|     Error.
|
| In particular, is the ordering fo (2) and (3) above correct?  Is this
| stated in the spec. anywhere?

I think that's right. I'll work to clarify it.

| The glossary is not in alphabetical order????

That did get fixed.

| Your relaxng grammar still treats stuff in the XHTML ns specially, but
| I think we abandoned that. . .

Right you are. Drat.

Oh, well, there's always the next draft.

                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com> | Everything should be made as simple as
http://nwalsh.com/            | possible, but no simpler.