February 2016 - Tei-council

Testing Pure ODD
by James Cummings 09 Feb '16

09 Feb '16

Hi Lou and Council, I've been looking around for a TEI ODD file where I redefined content models. Mostly I just add/delete elements/attributes/classes which isn't really changing. Finally remembered I had changed some content models in my tei_corset ODD. This was a mass-digitisation ODD for the Stationers' Register project where the keying company was willing to use any XML schema but would charge per kilobyte of output. So I renamed everything to be shorter names, etc. in a tei_corset namespace which over a lot of text input saved us money and was easy to convert back. Where I used any relax ng in this it was fairly minimal and the conversion able to be done automatically with no problems: a) to rename <TEI> to be <file> and say that its content model was just <teiHeader> and <body> (both renamed elsewhere) Old ODD: https://github.com/jamescummings/conluvies/blob/master/tei_corset/tei_corse… Pure ODD: https://github.com/jamescummings/conluvies/blob/master/tei_corset/tei_corse… This just changes to two <elementRef>s. b) to rename <teiHeader> to be <header> and give it a content model of <title> followed by zero or more model.pLike Old ODD: https://github.com/jamescummings/conluvies/blob/master/tei_corset/tei_corse… Pure ODD: https://github.com/jamescummings/conluvies/blob/master/tei_corset/tei_corse… This just changes to an <elementRef> followed by a <classRef> c) for various reasons some empty elements were changed to allow text content. For example <gap> was renamed <gp> and had a content of <rng:text> Old ODD: https://github.com/jamescummings/conluvies/blob/master/tei_corset/tei_corse… Pure ODD:https://github.com/jamescummings/conluvies/blob/master/tei_corset/tei_c… <rng:text> just becomes <textNode/> d) exactly the same thing for <space> renamed <spc>: Old ODD: https://github.com/jamescummings/conluvies/blob/master/tei_corset/tei_corse… Pure ODD: https://github.com/jamescummings/conluvies/blob/master/tei_corset/tei_corse… The resulting schema validates against the files I had. I did have to change some other things in the ODD, but these had to do with the moving of rendition things into att.global.rendition from att.global. (Since this ODD was 3 or 4 years old.) I converted the Pure ODD to HTML and everything worked as expected. Clearly not an exhaustive test of all of the functionality of Pure ODD, but just another confirmation that migrating to Pure ODD is able to be semi-automated and for most people's low hanging fruit doesn't introduce any major problems. -James -- Dr James Cummings, James.Cummings(a)it.ox.ac.uk Academic IT Services, University of Oxford

2 1

Roma eXist-db down?
by Peter Stadler 08 Feb '16

08 Feb '16

Dear all, at present, I can’t access Roma (I may enter the first pages but Roma is unable to fetch module information etc.) and it seems the eXist-db is down. Looking in the logs at /var/www/vhosts/tei-c.org/opencms_tomcat/lv1/local/tomcat_6.0.14_teicms/webapps/exist/WEB-INF/logs/exist.log states: 2016-02-08 07:01:18,941 [main] INFO (BrokerPool.java [shutdown]:1495) - shutdown complete ! 2016-02-08 07:23:49,589 [http-8040-9] INFO (XQueryServlet.java [init]:135) - form-encoding = UTF-8 2016-02-08 07:23:49,590 [http-8040-9] INFO (XQueryServlet.java [init]:139) - container-encoding = UTF-8 2016-02-08 07:23:49,590 [http-8040-9] INFO (XQueryServlet.java [init]:143) - encoding = UTF-8 2016-02-08 07:23:49,631 [http-8040-9] WARN (Descriptor.java [<init>]:96) - Giving up unable to read descriptor file from /descriptor.xml 2016-02-08 07:23:49,631 [http-8040-9] WARN (Descriptor.java [<init>]:115) - Giving up unable to read descriptor.xml file from classloader in package org.exist.http So, it looks as if the db was shutdown and failed to restart. Does anyone know any details and how to proceed? (I’m unfamiliar with eXist within Tomcat …) Best Peter

4 5

proc mod queries
by Lou Burnard 06 Feb '16

06 Feb '16

I've made a bit more progress on the "processing model" extensions to ODD proposed by the TEI Simpletons a while ago, and now have a version of P5 based on the current dev branch which includes them AND builds correctly. But I havent yet finished hacking the prose into shape, and I have some questions. Here's the first. The @output attribute "the intended output method" (that's clearly wrong though; surely it should be "the intended output target"?) is defined with a closed list of values "web", "print", and "plaintext". a) why is this a closed list? I can think of lots of other possible values e.g. "mobile device", "PDF", "text-to-speech synthesizer" for starters... b) what exactly does "plaintext" mean? And here's the zero'th. To whom if not the Council should I be addressing this query?

3 8

processing model use of CSS
by Lou Burnard 05 Feb '16

05 Feb '16

Decision please. The current description of the processing model tacitly assumes that CSS is used to express output rendition, and it's hard to see how else the job would be done, but the text of the Guidelines is rather vague on the topic. Should we say a) outputRendition *must* be expressed using CSS b) it is *recommended* that CSS be used c) we don't care what you use, but our examples assume you will use CSS If we choose b or c, we will presumably need to add some complexity to the scheme (e.g. to say what has been used instead if it's not CSS)

5 17

weird example
by Lou Burnard 05 Feb '16

05 Feb '16

The spec for <model> currently has a whole slew of examples, some of which need to go elsewhere I think. And one or two of them are entirely mysterious. This one for example: <exemplum xml:lang="en"> <egXML xmlns="http://www.tei-c.org/ns/Examples"> <model predicate="parent::choice and count(parent::*/*) gt 1" behaviour="omit"/> </egXML> </exemplum> Can someone please explain to me what this is meant to achieve? It seems to be saying "if my parent is <choice> and any of my parents has more than one child node, then suppress me". In the immortal words of Manuel, che?

3 4

outputRendition spec
by Lou Burnard 05 Feb '16

05 Feb '16

The <remarks> for this element currently read: "The <gi>outputRendition</gi> element differs from the <gi>rendition</gi> element describes rendering information in one or more possible outputs, usually recorded in the TEI customization file, the latter must only be used to document the original rendition of the source text." - As far as I know <outputRendition> can only appear inside an ODD file, so why "usually" ? - If @useSourceRendition is true, then the <rendition> element *will* affect the output. I'm proposing the following revision: This component of an element specification describes the rendering or appearance intended for all occurrences of the element in a specified context for a specified type of output. The <rendition> element, by contrast, describes the actual rendering or appearance of all occurrences of the specified element in a source document. The <att>useSourceRendition</att> attribute may however be used to indicate that formatting information provided by a <gi>rendition</gi> element is to be combined with any provided by an <gi>outputRendition</gi> element. It is strongly recommended that the W3C Cascading Stylesheet language (CSS2 or later) be used to express the required formatting information. (The second paragraph modulo whatever we decide to do about the CSS or what debate)

3 2