W3C

XML Processing Model WG

Meeting 68, 17 May 2007

See also: IRC log

Attendees

Present
Alex, Richard, Paul, Mohamed, Rui, Andrew
Regrets
Norm, Henri, Alessandro
Chair
Alex
Scribe
Richard

Contents


Accept this agenda?

-> http://www.w3.org/XML/XProc/2007/05/10-agenda.html

Accepted.

Accept minutes from the previous meeting?

-> http://www.w3.org/XML/XProc/2007/05/03-minutes.html

Accepted.

Next meeting: telcon 24 May 2007

No regrets

Open action items

Alex to craft a proposal on serialization, complete

Henry to propose wording for grouping options for p:wrap, continued.

 

Step library review

http-request discussion deferred

load options

DTD validation, namespace support

Richard: should not allow turning off namespaces

Consensus: shouldn't have option to allow namespace-ill-formed docs

DTD validation: get a dynamic error if processor doesn't support it

If validated, dynamic error if not valid

Consensus: have this validate option on load

<alexmilowski> http://lists.w3.org/Archives/Public/public-xml-processing-model-wg/2007May/0008.html

<MoZ> http://lists.w3.org/Archives/Public/public-xml-processing-model-wg/2007May/thread.html#msg8

Action: Add 'validate' option where 'yes' means DTD validation will be performed. If the step doesn't support DTD validation, a dynamic error is thrown. If validation errors occur, a dynamic error is thrown.

Escaped markup

<MoZ> and match="//*" ?

<MoZ> match="//node()"

Discussion: do we need a match argument to specify what to escape, or should we jsut use a viewport?

Consensus: no match option, use a viewport

Boolean options

yes/no? true/false?

There was no active discussion of the issue.

Alex favours status quo (yes/no)

Moz favours status quo for now at least

Consensus: don't change it

Content-type for unescape-markup

Also question of what to do if you get text/html etc (it may be wf xml)

Alex: add HTML parsing to next week's agenda

(we don't have all the people this week)

wrapping a sequence

Should we have a component to take a sequence and wrap it into a single document?

Moz: there's no other way to do it

Alex: could allow "wrap" to do it

Moz: would be bad for wrap to be quite different for sequence and single doc

Consensus: have a separate component

Action: Add a wrap-sequence step to our library.

How to specify the wrap element name?

(can't use "name", because it's taken)

No strong preference although 'wrapper' and 'wrapper-name' were suggested--call it "wrapper" for now.

Action: Change the 'name' option on p:wrap to 'wrapper'

rename XSLT

(xslt->xslt1, we already have xslt2)

Moz: Norm's argument for xslt is good, are we going to put version numbers on all the other components?

<alexmilowski> http://lists.w3.org/Archives/Public/public-xml-processing-model-wg/2007May/0232.html

Consensus: stick the the status quo

Aggregate

Alex: is this the same as wrap-sequence?

Moz: no, it's more powerful. Wouldn't need wrap-sequence if we had aggregate
... or could having matching-document on wrap-sequence
... if we had recursive steps we could handle it that way, but we don't
... discussion of wrapping sequences of chapters etc

Alex:The main use case is now covered by a p:matching-documents step followed by a p:wrap-sequence. Mohamed will send additional issues to the list for other use cases.

Paul to send IRC log to construct minutes from

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.128 (CVS log)
$Date: 2007/05/24 16:47:02 $