W3C home > Mailing lists > Public > public-ppl@w3.org > March 2012

Re: Workflow - on the wiki

From: Tony Graham <tgraham@mentea.net>
Date: Wed, 21 Mar 2012 09:10:41 -0000 (GMT)
Message-ID: <20782.>
To: "xsl-fo Community Group" <public-ppl@w3.org>
On Tue, March 20, 2012 5:28 pm, Dave Pawson wrote:
> On 20 March 2012 17:18, Tony Graham <tgraham@mentea.net> wrote:
>> I'm not sidestepping.  I was trying to show both that schemas can't
>> express enough detail and that users' expectations of an IDE are based
>> on
>> more than just schema-level checking.
> That has to be an assumption Tony?

No.  It's certainly what I expect.

>From http://www.oxygenxml.com/xslt_editor.html:

 - "Easy Identification of XSL Component References"

 - "XPath Aware Content Assistant"

 - "Content completion offers the list of components from
   included/imported XSLT stylesheets"

>From http://www.altova.com/xmlspy/xslt-editor.html:

 - "As you’re composing an XPath expression in Text View,
   Grid View, or in XMLSpy's XPath Analyzer window, XMLSpy
   provides you with valid XPath functions, as well as
   element and attribute names from the associated schema
   and XML instance(s)."

>From http://www.stylusstudio.com/xslt_editor.html:

 - "Stylus Studio« also supports code sensing and code
   completion for XPath functions, listing available
   functions, and their respective function prototypes.

> I guess you're making the same point as DC?

I think DC thought you wanted to recast the XSL-FO expressions that you
can't check with RELAX NG as sequences of multiple elements that you could
check with RELAX NG -- hence his reference co XQueryX -- but my impression
is that you want to lop off anything that you can't represent in RELAX NG.

>   My same answer, OK leave that out of the grammar, for later inclusion
> perhaps by Schematron / some other way.
>    Though I'd put money (a little) on these expressions being in the
> 20% or less?

Whose 20%?  I use expressions in property values.

>>> What about for that heap of ... that is the color definition?
> Quite simply, a relax NG grammar definition would be a thousand lines or
> more.
> I want to junk it and simplify.
> At least until some better validation tool comes along.

Have you tried RELAX NG + Schematron for your color issue?

I'd like to suggest that editing complete XSL-FO documents is in the 20%
to 0.01% of uses that we can safely ignore as inconvenient. [1]



[1] Actually, I wouldn't.
Received on Wednesday, 21 March 2012 09:11:08 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:57:24 UTC