W3C home > Mailing lists > Public > public-xml-processing-model-comments@w3.org > August 2008

Re: a few comments on latest draft

From: Norman Walsh <ndw@nwalsh.com>
Date: Thu, 21 Aug 2008 07:49:55 -0400
To: public-xml-processing-model-comments@w3.org
Message-ID: <m2y72qwpss.fsf@nwalsh.com>
/ James Fuller <james.fuller.2007@gmail.com> was heard to say:
| Example 3. A validate and transform pipeline
| propose highlighting the use of an Optional step (p:validate-with-xml-schema)

I suppose we could, but does it add value?

| should we explicitly define primary outputs in Standard and Optional steps ?

I don't understand the question.

| in section 1 Introduction
| 'The pipeline document determines how the steps are connected together
| inside the pipeline. How inputs are connected to XML documents outside
| the pipeline is implementation-defined. How pipeline outputs are
| connected to XML documents outside the pipeline is
| implementation-defined.'
| do we want to delineate between the 'outside world' e.g. the top level
| pipeline versus a pipeline that is executing in the context of a
| nested pipeline ?

Isn't that what that paragraph does?

| in section 2 Pipeline Concepts
| 'A pipeline must behave as if it evaluated each step each time it occurs.'
| is it more valid to say  memoisation is not allowed ... or is this too
| constraining ?

For the benefit of readers unfamiliar with the technical term, I'm not
inclined to make this change.

| what is the primary scenario where this applies ?

It's just a way of saying that implementations can't cache results and
only evaluate a step once.

| in section 1 it says
| 'There are two kinds of steps: atomic steps and compound steps.'
| in section 2.1 Steps
| 'There are three kinds of steps: atomic, compound, and multi-container.'
| I propose syncing these

Yes. Done.

| in section 2.1.1 Step names
| I am a unsure about the need to specifically prescribe the manufactured format
| I propose to keep existing text but frame it as an 'example'.

I think we decided that we'd like to define it, though I suppose
that's not strictly necessary.

| I must have lost the thread on this discussion but what happens when
| <p:declare-step
|   xpath-version? = string>
| and a pipelines xpath-version
| have different values ... is this a static error ? just need a pointer here

That text has been clarified, I think. If you don't find it clear now,
please let us know.

| probably need to expand the definition of pfx:user-pipeline ... somewhere
| also, shouldn't p:standard-step be pfx:atomic-step ?

I think we worked on this too. Let us know if you still have questions.

                                        Be seeing you,

Norman Walsh <ndw@nwalsh.com> | To what excesses will men not go for
http://nwalsh.com/            | the sake of a religion in which they
                              | believe so little and which they
                              | practice so imperfectly!--La Bruyère

Received on Thursday, 21 August 2008 11:51:01 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:41:07 UTC