Level 4 Schema and Schematron Feedback

>>Return to testing instructions: Feedback_on_the_Best_Practices_for_TEI_in_Libraries_ODD_Documentation_and_Schemas

Tester: Michelle Dalmau

Comments/Feedback

• 1001 errors generated!

• Some are due to non-compliance for one reason or another (usually trivial oversights or value requirements). For example:
  • <tagUsage gi="div">No numbered divs used.</tagUsage> should have a value of either: "Numbered divs used." or "Unnumbered divs used."

This is your error per BP. (Kshawkin 12:02, 5 July 2011 (EDT))

• • element "editorialDecl" incomplete; missing required element "p" but I have a paragraph in there so I don't know what this error message means

Has been fixed. (Kshawkin 12:02, 5 July 2011 (EDT))

• • Description: element "back" incomplete; expected element "div" or "div1". Well I don't have back matter, but an empty <back/> tag caused the error.

Level 4 inherits from Level 3, which requires div or div1 in back if present. (Kshawkin 12:02, 5 July 2011 (EDT))

• Most complaining about an encoding practice either too stringently or not supported by the schema

• • Description: element "particDesc" not allowed anywhere; expected the element end-tag or element "langUsage" or "textClass". This error perplexes me. The particDesc appears where it should in the Header. It appears after langUsage and right before textClass. Even if prosopography can be considered a level 5 activity, should it choke on level 4?
    • Many errors generated as the result of the prosopography with <listPerson> <person> etc.
    • Many, many @corresp errors that reference the people in the prosopography.

particDesc is not ever allowed in BP headers. Will revise section "Level 4 Oral History" to remove references to use of this element in headers. Also removed vestigal name elements found in BP. (Kshawkin 12:02, 5 July 2011 (EDT))

• • element "pb" not allowed here; expected the element end-tag or element "div". In this case, my <pb> was inside the <front> tag. It shouldn't complain.

BP says that pb elements must be within divs. (Kshawkin 12:02, 5 July 2011 (EDT))

• • in the front matter is not supported ??? Description: attribute "type" (and "rend") not allowed here; expected attribute "n", "xml:id", "xml:lang" or "xml:space". Why not? And why can't we apply @rend to any chunk of markup?

@type and @rend now fixed. Per a recent change in ODD processing, to add <memberOf key="att.global"/> to every element definition in the ODD containing mode="replace" (Kshawkin 12:02, 5 July 2011 (EDT))

• • Can't have a <bibl> in my half_title div either: Description: element "bibl" not allowed here; expected the element end-tag or element "div", "figure", "gap", "head", "lb", "note", "opener" or "pb". Not sure why not.

bibl not allowed as child of div. However, we realized that listBibl (containing bibls) was not allowed in Level 4 for bibliographies, so we've added this. (Kshawkin 12:02, 5 July 2011 (EDT))

• Crazy errors
  • Error messages don't line up with the highlighting in the XML (or anywhere near it). Probably because of all the many errors so I'll try and isolate after I work through the rest of the messages.

Yes, this has to do with your XML editor and validator. (Kshawkin 12:02, 5 July 2011 (EDT))

• • Can't seem to type any of my divs. That's crazy:
    : Description: attribute "type" not allowed here; expected attribute "n", "xml:id", "xml:lang" or "xml:space". Those attributes seem to correspond with <pb> so I moved the surrounding <pb> inside and outside the div but the error message didn't change.

Now fixed. (Kshawkin 12:02, 5 July 2011 (EDT))

• • Description: element "fw" not allowed anywhere. FormeWorks occur after the page break tag. Why not?

We decided not to tag forme work. (Kshawkin 12:02, 5 July 2011 (EDT))

Tester: Kevin Hawkins

• Most <elementSpec>s have examples that are taken exactly from P5 but do not validate according to the ODD itself. Furthermore, definitions sometimes refer to sections of P5 with pointers such as <ptr target="#SACR"/> , which don't resolver. I recommend omitting all element definitions since creation and maintenance of these is too much for us to undertake.

Later, Syd will change XSLT so that it omits exemplum elements from output. Some day in the future, we will figure out a mechanism that allows us to supply our own examples not in P5 for use of particular elemens. The output will include only our examples and none of those in P5. (Kshawkin 10:05, 5 July 2011 (EDT))

Tester: Sebastian Rahtz

From an email from Sebastian:

[. . .]

b) surely &amp; inside <egXML> is not intended?

c) glancing over https://github.com/sydb/TEI-in-Libraries/blob/master/BestPractices/level4.odd,
I see 
        <elementSpec module="textstructure" ident="div1" mode="replace">
	    <classes mode="change">
	      <memberOf key="att.typed" mode="delete"/>
	      <memberOf key="model.div1Like"/>
	    </classes>

 ie you have removed the @type attribute. then later on it says

     <attList>
	      <attDef ident="type" mode="change">

  which seems contradictory of the above. how can this work?

Resolved (Kshawkin 17:44, 5 July 2011 (EDT))