[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: review of editor's notes for attributes of namespace text
hi Patrick,
i've done a review of the attributes in the text namespace in the ODF 1.2
committee draft.
chapter 18
text:active
according to Oliver Specht, for an inactive page variable,
a page-variable-get expands to nothing. also, an inactive page
variable is not incremented.
text:bibliography-type
i'm not sure i see a problem here. if a single bibliography contained
multiple templates for e.g. book, wouldn't that create a
inconsistently formatted bibliography?
and how would you choose which one you want for an entry?
text:caption-sequence-format
i would seriously doubt that whoever wrote this had any ISO
terminology in mind. probably better rewrite it to
"The attribute specifies...".
it seems to me that "the whole text" would be the concatentation of
the other two options.
text:copy-outline-levels
sorry, no idea...
text:current-value
ok, seems somewhat redundant.
text:date-value
note1: yes
note2: well, this is just an example (that would probably belong to
the description of text:fixed). if an author field were fixed,
it would display the author that marked it fixed.
text:default-style-name
i don't understand why there would be a conflict?
in case you insert a new note, and the default-style-name applies,
then the new note paragraph will have that style. if that style has
auto-update="true", then changing the properties of the paragraph
will change the style. i don't see a problem with that.
text:display
<text:table-formula>
i've no idea, maybe it's just badly formulated?
text:formula
ok.
text:index-name
ok.
text:list-tab-stop-position
if Oliver says so...
text:name
<text:alphabetical-index>
yes.
<text:bibliography>
yes.
<text:illustration-index>
yes.
<text:index-title>
yes.
<text:object-index>
yes.
<text:table-of-content>
yes.
<text:table-index>
yes.
<text:user-defined>
"The text:name attribute specifies a name defined by a user."
this is quite wrong.
this element displays the content of a <meta:user-defined> from
meta.xml, and refers to it by the text:name attribute.
<text:user-index>
yes.
<text:variable-input>
well, for the elements that introduce new entities in the document
like <bookmark>, <variable-decl> etc, the use must be unique,
while for the elements that reference other elements,
like <bookmark-end>, <variable-get>, etc. the use need not be unique;
in fact, it cannot be unique because the referenced entity must be
declared!
maybe it would make sense to use different attributes for these two
uses? and we already have <text:ref-name>, which is used for some
references... rather inconsistent.
text:note-class
this description is for the <text:note-ref> element.
you could generalize it by replacing "field references" with
"element specifies or references". but maybe separate descriptions
would be clearer.
text:outline-level
<text:user-index-entry-template>
i don't understand this note.
<text:user-index-mark-start>
why would <text:user-index-mark-end> have this attribute???
that seems like an error in the schema to me.
text:ref-name
sorry, i don't understand this note.
the description of this attribute only refers to the <sequence>
element, not to the other elements on which it may occur.
text:reference-format
yes, that makes sense.
text:restart-numbering
ouch...
text:select-page
hmmm.
text:separation-character
the first sentence is wrong: "The text:display-outline-level..."
should be "The text:separation-character..."
the other references seem correct
re: "sequence number": how about "the value of the sequence variable"?
text:sort-algorithm
hmmm... but do we say anything about possible values?
text:sort-ascending
yes.
text:start-value
<text:numbered-paragraph>
by "disconnected ranges of numbered paragraphs", you mean like:
<numbered-paragraph list-id="foo">...</numbered-paragraph>
<table>...</table>
<numbered-paragraph list-id="foo">...</numbered-paragraph>
i would interpret this case like a single list that is spread over
several <list> elements, with continue-list attributes.
re: second note: well, in ODF 1.0/1.1 the numbered-paragraph was
rather underspecified, so i would consider it an improvement
<text:outline-level-style>
hmm... it seems pretty clear to me: if the element is some kind of
style, then start-value specifies a number to be used for the first
element; if it is some content element, then start-value specifies
that numbering be restarted at the current element.
so you can interpret start-value at a style as an implicit start-value
at the first element where the style is applied.
text:string-value-phonetic
that note went way over my head
text:style-name
<text:a>
draft6 claims this is a text style:
<quote>If the link location of the hyperlink was not visited,
the text style specifies[sic] by the text:style-name attribute
is applied to the text of the hyperlink.</quote>
<text:alphabetical-index>
no, this should be a section style
<text:bibliography>
no, this should be a section style
<text:illustration-index>
no, this should be a section style
<text:index-title>
yes, this should be a section style
<text:object-index>
yes, this should be a section style
<text:section>
it would not entirely surprise me if only "section" styles were
allowed
<text:table-index>
yes, this should be a section style
<text:user-index>
no, this should be a section style
text:tab-ref
hmm, ok...
text:use-index-source-styles
i'm not Patrick :)
text:use-objects
i don't understand this note; it does not seem to belong here.
also, the description for this attribute is missing.
in draft6 it was: "Embedded Objects are included."
text:use-other-objects
"other objects" would be those _not_ covered by
text:use-chart-objects, text:use-draw-objects,
text:use-math-objects, and text:use-spreadsheet-objects.
xhtml:about xhtml:content xhtml:datatype xhtml:property
hmm, yes. this is still wrong: property may specify several
predicates, not just one. i've sent you a mail about this
some time ago.
chapter 19
text:anchor-page-number
this attribute is both in chapter 18 and 19 ???
it seems that the default anchor is set by _both_ attributes together
(but -page-number is only used for certain values of anchor-type).
but that should be clear from the text:anchor-type description,
so i don't mind if the sentence is gone.
text:condition
the latest schema seems to contain the deleted fragment again,
so this note is obsolete.
text:space-before
Oliver wanted to write about this one
regards,
michael
--
Michael Stahl mailto:michael.stahl@sun.com
http://www.sun.de OpenOffice.org/StarOffice Writer
Sun Microsystems GmbH Nagelsweg 55, 20097 Hamburg, Germany
-----------------------------------------------------------------------
Sitz der Gesellschaft:
Sun Microsystems GmbH, Sonnenallee 1, D-85551 Kirchheim-Heimstetten
Amtsgericht Muenchen: HRB 161028
Geschaeftsfuehrer: Thomas Schroeder, Wolfgang Engels, Dr. Roland Boemer
Vorsitzender des Aufsichtsrates: Martin Haering
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]