W3C home > Mailing lists > Public > www-math@w3.org > March 2013

Re: Definition of numbers in the MathML RelaxNG

From: Neil Soiffer <NeilS@dessci.com>
Date: Sat, 9 Mar 2013 10:00:55 -0800
Message-ID: <CAESRWkANn=p6uCzRYOt-+Gnh+cdDNn-VaizOTwy4CMC6H2ZJrQ@mail.gmail.com>
To: David Carlisle <davidc@nag.co.uk>
Cc: Frédéric WANG <fred.wang@free.fr>, "www-math@w3.org" <www-math@w3.org>
Those shouldn't be legal, so I like the change.

    Neil



On Sat, Mar 9, 2013 at 8:47 AM, David Carlisle <davidc@nag.co.uk> wrote:

> On 19/12/2012 12:27, Frédéric WANG wrote:
>
>> Hi Math WG,
>>
>> I notice that in the regular expression for "length" and
>> "mpadded-length", you use the following definition of numbers:
>>
>> [0-9]*(\.[0-9]*)?
>>
>> which allows empty string as numbers and whitespace-only "length".
>> I'm wondering if you could replace this regular expression by:
>>
>> [0-9]*(\.[0-9]*|[0-9])
>>
>> That will make the schema a bit stricter and help me for a tool I'm
>> developing.
>>
>>
> Stephen pointed out off list that this would still allow a single "."
> with no digit and suggested
>
> [0-9]*([0-9]\.?|\.[0-9])[0-9]*
>
> I have finally got round to updating this so the draft spec at
>
> http://www.w3.org/Math/draft-**spec/appendixa-d.html#parsing_**length<http://www.w3.org/Math/draft-spec/appendixa-d.html#parsing_length>
>
> and
>
> http://www.w3.org/Math/draft-**spec/appendixa-d.html#parsing_**
> mpadded-length<http://www.w3.org/Math/draft-spec/appendixa-d.html#parsing_mpadded-length>
>
> shows this change.
>
> The RelaxNG and XSD schemas are also updated at
>
> http://www.w3.org/Math/**RelaxNG/ <http://www.w3.org/Math/RelaxNG/>
>
> and
>
> http://www.w3.org/Math/**XMLSchema/ <http://www.w3.org/Math/XMLSchema/>
>
>
> These changes mean that the file
>
> <math xmlns="http://www.w3.org/1998/**Math/MathML<http://www.w3.org/1998/Math/MathML>"
> >
> <mo lspace=".pt"/>
> <mo lspace="pt"/>
> <mpadded width=".pt"><mo>+</mo></**mpadded>
> <mpadded width="pt"><mo>+</mo></**mpadded>
> </math>
>
> which was previously reported as valid is now reported as invalid, which
> matches the prose description of the syntax of these attributes given in
> the spec.
>
> David
>
>
>
>
Received on Saturday, 9 March 2013 18:01:27 GMT

This archive was generated by hypermail 2.3.1 : Saturday, 9 March 2013 18:01:27 GMT