[osis-core] empty tag / milestone proposal

Patrick Durusau osis-core@bibletechnologieswg.org
Wed, 19 Jun 2002 15:12:14 -0400


Troy,

Let's start with the text as you typed it:

Troy A. Griffitts wrote:

> As I understand it, <head>text</head> and <head/>text<head/> are NOT 
> identical.

<snip>

> <verse verseID="jn.1.1">In the beginning...</verse>
> <verse verseID="jn.1.1" mStart="jn.1.1"/>In the beginning...<verse 
> mEnd="jn.1.1"/>
>
Now, other than the attributes that you inserted (which I now delete for 
illustration only, you are correct, I did not add them properly), how is:

<head>text</head> and <head/>text<head/> different from:

<verse verseID="jn.1.1">In the beginning...</verse> and <verse />In the 
beginning...<verse />

It seems to me that other than the attributes you added (which I agree 
need to be added so that is not an issue) the two examples are the same.

>
> mStart and mEnd should NOT be global attributes.  <div>, for example, 
> should NOT allow this type of behaviour.

Agreed.

>
> Please tell me why this sucks.  I'm perfectly ok with someone 
> explaining why this sucks and finally implementing my problem passages 
> some other way, but I really have been frustrated, and frozen from 
> completing ABS work because this issue has not been resolved.  Can we 
> get more than 2 of us on the phone or in person to deal with these 
> issues that have lagged for more than a month?

Sorry that you feel they have lagged. I feel that I have answered them 
much as I have above before. I am really trying to get the example to 
the point that we can either agree or find where it is that we disagree. 
I really don't see the difference (other than the attributes and yes, 
they should be properly added).

Patrick


>
>     -Troy.
>
>
>
>
>
>
>
> Patrick Durusau wrote:
>
>> (Got a delayed start this morning getting house guests on the road. 
>> Expect the next version just after NOON East Coast time.)
>>
>> Troy,
>>
>> XML treats <head></head> and <head/> as syntactically equivalent. 
>> Unless an element has required content (in other words, elements that 
>> must occur)  I think you already have milestones. (Am I missing 
>> something here?)
>>
>> Since <div> is mixed content, you could write all of the elements 
>> that may occur in <div> as empty elements, or so I am reading the XML 
>> spec.
>>
>> I may be missing the purpose of the patch. Is it to add the mstart, 
>> mstop attributes? Is so, can I add them to global?
>>
>> I would like to leave the generic milestone for cases where people 
>> want to use standard containers and yet also do some milestone markup.
>>
>> Patrick
>>
>> Troy A. Griffitts wrote:
>>
>>> Attached is a proposal to allow some elements to be used as empty 
>>> milestones optionally.  Please consider applying this patch.
>>>
>>> I would also suggest removing the generic milestone element, but 
>>> this patch does not do this.
>>>
>>>
>>> ------------------------------------------------------------------------ 
>>>
>>>
>>> --- osisCore_1test12.xsd.orig    Tue Jun 18 19:45:00 2002
>>> +++ osisCore_1test12.xsd    Tue Jun 18 20:13:05 2002
>>> @@ -308,6 +308,7 @@
>>>                            <xs:element ref="w"/>
>>>         </xs:choice>
>>>         <xs:attributeGroup ref="globalAttributes"/>
>>> +        <xs:attributeGroup ref="milestoneAttributes"/>
>>>        </xs:complexType>
>>> </xs:element>
>>>
>>> @@ -765,6 +766,7 @@
>>>                <xs:element ref="w"/>
>>>         </xs:choice>
>>>         <xs:attributeGroup ref="globalAttributes"/>
>>> +        <xs:attributeGroup ref="milestoneAttributes"/>
>>>        </xs:complexType>
>>> </xs:element>
>>>
>>> @@ -798,6 +800,7 @@
>>>         <xs:attribute name="next" type="xs:string" use="optional"/>
>>>         <xs:attribute name="prev" type="xs:string" use="optional"/>
>>>         <xs:attributeGroup ref="globalAttributes"/>
>>> +        <xs:attributeGroup ref="milestoneAttributes"/>
>>>        </xs:complexType>
>>> </xs:element>
>>>
>>> @@ -895,6 +898,7 @@
>>>         <xs:attribute name="next" type="xs:string" use="optional"/>
>>>         <xs:attribute name="prev" type="xs:string" use="optional"/>
>>>         <xs:attributeGroup ref="globalAttributes"/>
>>> +        <xs:attributeGroup ref="milestoneAttributes"/>
>>>        </xs:complexType>
>>> </xs:element>
>>>
>>> @@ -976,6 +980,7 @@
>>>         <xs:attribute name="prev" type="xs:string" use="optional"/>
>>>             <xs:attribute name="who" type="xs:string" use="optional"/>
>>>         <xs:attributeGroup ref="globalAttributes"/>
>>> +        <xs:attributeGroup ref="milestoneAttributes"/>
>>>        </xs:complexType>
>>> </xs:element>
>>>
>>> @@ -1027,6 +1032,7 @@
>>>         </xs:choice>
>>>         <xs:attribute name="changeType" type="changeType" 
>>> use="optional"/>
>>>         <xs:attributeGroup ref="globalAttributes"/>
>>> +        <xs:attributeGroup ref="milestoneAttributes"/>
>>>        </xs:complexType>
>>> </xs:element>
>>>
>>> @@ -1059,6 +1065,7 @@
>>>         <xs:attribute name="next" type="xs:string" use="optional"/>
>>>         <xs:attribute name="prev" type="xs:string" use="optional"/>
>>>         <xs:attributeGroup ref="globalAttributes"/>
>>> +        <xs:attributeGroup ref="milestoneAttributes"/>
>>>        </xs:complexType>
>>> </xs:element>
>>>
>>> @@ -1117,6 +1124,11 @@
>>>     <xs:attribute name="outCite" type="osisRef" use="optional"/>
>>> </xs:attributeGroup>
>>>
>>> +<!-- allow some container elements to be left empty and used as 
>>> milestones. eg. either <verse verseID="jn.1.1">In the 
>>> beginning...</verse> OR <verse verseID="jn.1.1" mStart="jn.1.1"/>In 
>>> the beginning...<verse mEnd="jn.1.1"/> -->
>>> +<xs:attributeGroup name="milestoneAttributes">
>>> +    <xs:attribute name="mStart" type="xs:string" use="optional"/>
>>> +    <xs:attribute name="mEnd" type="xs:string" use="optional"/>
>>> +</xs:attributeGroup>
>>>
>>> <!-- simple types -->
>>>
>>> @@ -1369,4 +1381,4 @@
>>>     <xs:union memberTypes="roleOSIS attributeExtension"/>
>>> </xs:simpleType>
>>>
>>> -</xs:schema>
>>> \ No newline at end of file
>>> +</xs:schema>
>>>
>>
>
>

-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
pdurusau@emory.edu