RE: [Fwd: XML Schema Part 2 should provide BNF for all primitive types.]

Alexander:
This is great!  Many thanks!
Ashok

-----Original Message-----
From: Alexander Falk [mailto:al@altova.com]
Sent: Tuesday, March 27, 2001 12:41 PM
To: 'Asir S Vedamuthu'; Noah_Mendelsohn@lotus.com; Ashok Malhotra
Cc: jjc@jclark.com; w3c-xml-schema-ig@w3.org;
www-xml-schema-comments@w3.org
Subject: RE: [Fwd: XML Schema Part 2 should provide BNF for all
primitive types.]



Hi, 

these are the regular expressions that we use within XML Spy, which were
created in accordance with the information in Part 2.

While they are still at CR level, I'd be perfectly willing to share them
as a starting point for maybe adding them to Part 2 for the REC version.
For CR you did pretty much the same thing with the EBNF that we had
written for the regular expressions that got added to the corresponding
chapter in Part 2.

        // primitive W3C Schema DataTypes 
        true|false
// DT_boolean,

        [\\-\\+]?(INF|NaN|(\\d*(\\.\\d*)?([eE]\\-?\\d+)?))
// DT_float,                    DT_r4 = DT_float,

        [\\-\\+]?(INF|NaN|(\\d*(\\.\\d*)?([eE]\\-?\\d+)?))
// DT_double,                   DT_r8 = DT_double,

        [\\-\\+]?\\d*(\\.\\d*)?
// DT_decimal,

        [\\-\\+]?P(\\d+Y)?(\\d+M)?(\\d+D)?(T(\\d+H)?(\\d+M)?(\\d+S)?)?
// DT_timeDuration, 
 
\\-?(\\d{2,4}|-)-(\\d{2}|-)-(\\d{2}|-)T(\\d{2}|-):(\\d{2}|-):(\\d{2}(\\.
\\d+)?|-)(Z|([\\-\\+]\\d{2}:\\d{2}))?                   //
DT_recurringDuration,

 
// DT_binary,

 
(([a-zA-Z][0-9a-zA-Z+\\-\\.]*:)?/{0,2}[0-9a-zA-Z;/?:@&=+$\\.\\-_!~*'()]+
)?(#[0-9a-zA-Z;/?:@&=+$\\.\\-_!~*'()]+)?        // DT_uri_reference,

        [\\p{L}_][\\p{L}\\d\\.\\-_]*
// DT_ID, 
        [\\p{L}_][\\p{L}\\d\\.\\-_]*
// DT_IDREF, 
        [\\p{L}_][\\p{L}\\d\\.\\-_]*
// DT_ENTITY, 
        ([\\p{L}_][\\p{L}\\d\\.\\-_]*:)?[\\p{L}_][\\p{L}\\d\\.\\-_]*
// DT_QName, 
    
        // derived W3C Schema DataTypes 
        [^\\n\\r\\t]*
// DT_CDATA,    // new: CR 10/24/00

        ([^\\n\\r\\t ]+)( [^\\n\\r\\t ]+)*
// DT_token,    // new: CR 10/24/00

        ([a-zA-Z]{2}|[iI]-[a-zA-Z]+|[xX]-[a-zA-Z]+)(-[a-zA-Z]+)*
// DT_language, 
        ([\\p{L}_][\\p{L}\\d\\.\\-_]*)( [\\p{L}_][\\p{L}\\d\\.\\-_]*)*
// DT_IDREFS, 
        ([\\p{L}_][\\p{L}\\d\\.\\-_]*)( [\\p{L}_][\\p{L}\\d\\.\\-_]*)*
// DT_ENTITIES, 
        [\\p{L}\\d\\.\\-_:]+
// DT_NMTOKEN,

        ([\\p{L}\\d\\.\\-_:]+)( [\\p{L}\\d\\.\\-_:]+)*
// DT_NMTOKENS, 
        [\\p{L}_:][\\p{L}\\d\\.\\-_:]*
// DT_Name, 
        [\\p{L}_][\\p{L}\\d\\.\\-_]*
// DT_NCName, 
        [\\p{L}_:][\\p{L}\\d\\.\\-_:]*
// DT_NOTATION, 
        [\\-\\+]?\\d*
// DT_integer,

        0+|-\\d+
// DT_non_positive_integer,

        -\\d+
// DT_negative_integer,

        [\\-\\+]?\\d*
// DT_long,

        [\\-\\+]?\\d*
// DT_int,                              DT_i4 = DT_int,

        [\\-\\+]?\\d*
// DT_short,                    DT_i2 = DT_short,

        [\\-\\+]?\\d*
// DT_byte,                             DT_i1 = DT_byte,

        \\+?\\d*
// DT_non_negative_integer,

        \\+?\\d*
// DT_unsigned_long,

        \\+?\\d*
// DT_unsigned_int,             DT_ui4 = DT_unsigned_int,

        \\+?\\d*
// DT_unsigned_short,   DT_ui2 = DT_unsigned_short,

        \\+?\\d*
// DT_unsigned_byte,    DT_ui1 = DT_unsigned_byte,

        \\+?\\d*
// DT_positive_integer,

 
\\-?\\d{4}-\\d{2}-\\d{2}T\\d{2}:\\d{2}:\\d{2}(\\.\\d+)?(Z|([\\-\\+]\\d{2
}:\\d{2}))?// DT_timePeriod, 
 
\\-?\\d{4}-\\d{2}-\\d{2}T\\d{2}:\\d{2}:\\d{2}(\\.\\d+)?(Z|([\\-\\+]\\d{2
}:\\d{2}))?// DT_timeInstant, 
        \\-?\\d{4}-\\d{2}
// DT_month,

        \\-?\\d{4}
// DT_year,

        \\-?\\d{2}
// DT_century,

        -?-?\\d{2}-\\d{2}
// DT_recurringDate,

        -?-?-?\\d{2}
// DT_recurringDay

        (\\d{2,4})?(-(\\d{2})?(-(\\d{2}))?)?
// DT_date, 
        \\d{2}:\\d{2}(:\\d{2}(\\.\\d+)?)?(Z|([\\-\\+]\\d{2}:\\d{2}))?
// DT_time, 

Please let me know, if you find any problems with these or think they
should be different. Also, again this is CR status and hasn't yet been
updated to PR.

Alexander 

... Alexander Falk 
... President & CEO of Altova, Inc. - The XML Spy Company 
... Member of the W3C Advisory Committee 
... Member of the W3C XML Schema Working Group 

========================================================================
= 
XML Spy 3.0  -  the first true Integrated Development Environment for
XML 
Visit http://www.xmlspy.com/ <http://www.xmlspy.com/>  to download a
free 30-day evaluation version 
========================================================================
= 



-----Original Message----- 
From: Asir S Vedamuthu [ mailto:asirv@webmethods.com
<mailto:asirv@webmethods.com> ] 
Sent: Tuesday, March 27, 2001 14:57 
To: Noah_Mendelsohn@lotus.com; Ashok Malhotra 
Cc: jjc@jclark.com; w3c-xml-schema-ig@w3.org; 
www-xml-schema-comments@w3.org 
Subject: Re: [Fwd: XML Schema Part 2 should provide BNF for all 
primitive types.] 


> grammar for the lexical forms, but formal mappings to the value space.
In 
> other words, show the polynomial that gives you the integer value, for

> example. 

I don't see a *reason* why we have to go this far. 

There are 19 primitive types. Of them, formal descriptions for some can
be 
found in related recommendations and standards, example [1]. I am sure
we 
can re-use these descriptions. For some, say anyURI, we do not have to 
provide a formal description. And, it is relatively easy to write a BNF
or 
RegEx for some datatypes, say 'boolean'. 

The big question is how long would it take to produce this? May be we
can 
give up Part 2 re-org. 

[1] http://www.w3.org/TR/1999/REC-xml-names-19990114/#ns-qualnames
<http://www.w3.org/TR/1999/REC-xml-names-19990114/#ns-qualnames>  

Regards, Asir 
----- Original Message ----- 
From: <Noah_Mendelsohn@lotus.com> 
To: "Ashok Malhotra" <ashokma@microsoft.com> 
Cc: <jjc@jclark.com>; <w3c-xml-schema-ig@w3.org>; 
<www-xml-schema-comments@w3.org> 
Sent: Tuesday, March 27, 2001 1:44 PM 
Subject: RE: [Fwd: XML Schema Part 2 should provide BNF for all
primitive 
types.] 


> Does it make any sense to do regex's or BNF as non-normative for 1.0, 
> normative for 1.1?  This might, editors' time permitting, let us get 
> something out, and still have the opportunigy to fix edge cases before
we 
> make it normative.  I've thought for a long time we need not only a
formal 
> grammar for the lexical forms, but formal mappings to the value space.
In 
> other words, show the polynomial that gives you the integer value, for

> example. 
> 
>
------------------------------------------------------------------------

> Noah Mendelsohn                                    Voice:
1-617-693-4036 
> Lotus Development Corp.                            Fax: 1-617-693-8676

> One Rogers Street 
> Cambridge, MA 02142 
>
------------------------------------------------------------------------

> 
>

Received on Tuesday, 27 March 2001 20:35:30 UTC