W3C home > Mailing lists > Public > www-math@w3.org > November 2004

Re: Questions on accent attribute

From: Kobayashi <koba@antenna.co.jp>
Date: Tue, 09 Nov 2004 09:44:03 +0900
To: Robert Miner <RobertM@dessci.com>
Cc: www-math@w3.org
Message-Id: <20041109094403.5ea283%koba@antenna.co.jp>

Hello,

Thank you for your explanation. As I am not sure about my 
interpretation of English, your explanation is helpful.

As an explict value is more stronger than a defalut 
value, and operater dictionary is considered to set 
a default value, it is natural that an explicit value of 
mover overrides a default value of its child mo element.

Test suite says as follows: (moAccent10)
--------
This example tests the interaction of the 'accent' attribute 
for <mover> and <mo>. In particular, it tests that when 
the attribute is set on the enclosing <mover>, it overrides 
the value set on an operator in the second slot, whether 
computed or explicitly set
--------
This case says that explicit value of mover also overrides 
explict value of mo.

I understand an explict value of accent attribute of mover 
always overrides a same attribute value of mo.


Best regards,

Tokushige Kobayashi


> Reply to the original mail from:
>  Robert Miner <RobertM@dessci.com>
>


Hi.

This is confusing -- not the best designed part of the language perhaps...

The reason it works this way is so that in most cases, one will get
the right accent behavior automatically, without setting any attribute
values explicitly.  The point was to make accent behavior something
that was determined by a lookup in the operator dictionary.

The way I think of it is that is is the attribute on the <mover> that
actually controls the size and position of the overscript, while the
attribute on the <mo> is essentially a semantic hint (just like the
'fence' attribute) that doesn't really affect rendering except
indirectly.  

The 'accent' attribute on the <mo> in the script position can't override
an explicitly set value for the 'accent' attribute on the <mover>.  It
just supplies the default value in the case where there is no explicit
value for 'accent' set on the <mover>.

When there is an explicit 'accent' attribute set on the <mover>, then
it doesn't matter how the value is set on the <mo>.  The <mover> won't
look at it, and it doesn't itself change anything about the rendering
of the <mo>.

I'm not sure my explanation will really make anything any clearer, but
I tried!

--Robert

------------------------------------------------------------------
Dr. Robert Miner                                RobertM@dessci.com
W3C Math Interest Group Co-Chair                      651-223-2883
Design Science, Inc.   "How Science Communicates"   www.dessci.com
------------------------------------------------------------------

------- Original message -------
Date: Sun, 07 Nov 2004 17:51:20 +0900
From: Tokushige Kobayashi <koba@antenna.co.jp>
To: www-math@w3.org
Subject: Questions on accent attribute

Hello,

I am sorry for too many questions, but I can not understand how
the MathML renderer should behave for accent attribute of mover.

There is a description in 3.4.5.1 as follows:
- -------------
The default value of accent is false, unless overscript is an mo element .... 
If overscript is an mo element, the value of its accent attribute is used as the 
default value of accent for mover.
- ------------- 
This sentence says that the value of the child element (mo) influences
the default value of its parent element (mover).

There is also an example as bellow.

<mrow>
  <mover accent="true">
    <mi> x </mi>
    <mo> &Hat; </mo>
  </mover>
  <mtext>
  ..
  </mtext>
  <mover accent="false">
    <mi> x </mi>
    <mo> &Hat; </mo>
  </mover>
</mrow>

As a entry of operator dictionary for &Hat; specifies accent="true", 
the accent of both mo elements will be set as "true". But the rendering
image shows that accent attributes of mo are not effective for
rendering of &Hat.

The description in the spec seems to say that accent attribute
of mover overrides the value of accent attribute of its child (mo).

It seems to be strange that sometimes the attribute of parent
overrides the attribute of its child, but sometimes the attribute
value of child sets the value of its parent.

I hope someone will explain about the spec.

Best regards,


Tokushige Kobayashi
Antenna House, Inc.
E-mail koba@antenna.co.jp
WWW    http://www.antenna.co.jp/
WWW    http://www.antennahouse.com/ (English)
TEL    +81-3-3234-1361(direct call)
FAX    +81-3-3221-9975

Antenna House XSL School
http://www.antenna.co.jp/XML/school/xslday.htm
------- End of forwarded message -------



Tokushige Kobayashi
Antenna House, Inc.
E-mail koba@antenna.co.jp
WWW    http://www.antenna.co.jp/XML/ (Japanese)
WWW    http://www.antennahouse.com/ (English)
TEL    +81-3-3234-1361 (direct call)
FAX    +81-3-3221-9975

Let's migrate from a whiteboard to:
http://www.antenna.co.jp/ikisaki/
Received on Tuesday, 9 November 2004 01:03:45 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:12:56 GMT