OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

docbook-apps message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [docbook-apps] Custom ISO Schematron validation for modular Docbook XML with extensive profiling


We currently use single-pass profiling with a custom version of fo/profile-docbook.xsl (Includes our custom docbook.xsl rather than the stock stylesheet. No customizations to the actual profiling). Our only profiling attribute is profile.audience, and it is usually 1:1 mapped to a specific book element and thus a specific PDF. 

I guess that we need to switch to two-pass profiling to use profiling/profile.xsl ? 

Will you or someone else with Schematron knowledge be at tcworld2016 next week? It would be great to follow up this in more detail.

Bergfrid Skaara Dias

On Wed, Oct 26, 2016 at 3:07 PM, Radu Coravu <radu_coravu@sync.ro> wrote:
Dear Bergfrid,

Schematron is in the end translated to XSLT.
And the XSLT can be applied directly over the XML content and it will generate messages for each assert.
As you will do it outside of Oxygen you will not have fancy backmapping, you might not have a clear indication where the particular error was originally located.
There is also a Docbook XSLT stylesheet:


which can be applied over the XML along with parameters which specify profiling in order to filter the XML and remove from it the excluded content.

So if you first apply the profile.xsl and then the XSLT generated from Schematron over the original XML, you should also get reports about missing IDs.
This cascade of stylesheets should probably be applied on the XML for each combination of profiling attribute values that you are using.


Radu Coravu
<oXygen/>  XML Editor, Schema Editor and XSLT Editor/Debugger

On 10/26/2016 3:17 PM, Bergfrid Skaara wrote:
What are our options for adding custom ISO Schematron validation to our
DocBook publishing pipeline ?

The code base is large and extremely modular. We also use extensive
profiling to leverage single source reuse as much as possible. The
output is currently PDF only, but we might add HTML to the mix next
year. Our technical editors use oXygen, and we validate and transform
single documents there. In addition, we have a continuous integration
pipeline that builds all PDFs when something is committed to Git.

On-the-fly Schematron validation seems doable in oXygen. However, I find
it tricky to combine profiling with validating after a commit to check
that changes did not break other documents and that business rules are
adhered to.

Bergfrid Skaara Dias

To unsubscribe, e-mail: docbook-apps-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: docbook-apps-help@lists.oasis-open.org

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]