You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-dev@axis.apache.org by Russell Butek <bu...@us.ibm.com> on 2002/03/05 16:23:21 UTC
Re: date vs dateTime (was: cvs commit: xml-axis/java/src/org/apache/axis/encoding
DefaultTypeMappingImpl.java)
The only way to fix this is to have a different Java class for date vs
dateTime. Do we really want to roll our own DateTime class?
I don't know whether we want to fix it. I just want us all to be aware
that we WILL have users run into this issue and that we WILL have to admit
that it's broken, and unlikely to be fixed, unless someone wants to create
a special class.
This problem will exist all over the place, not just here. We don't
support xsd:Name, xsd:ID, xsd:NMTOKEN, etc. All of these would probably
map to java.lang.String since they're all derived from xsd:string. But
that would break roundtripping unless we create a new class for each of
these XML types. Ditto for xsd:integer derived types like
nonPositiveInteger, nonNegativeInteger, etc.
I'd prefer that this were fixed in one place (like the WSDL and/or JAX-RPC
specs) rather than haphazardly in our implementation. The specs should not
allow many-to-one mappings.
Russell Butek
butek@us.ibm.com
Davanum Srinivas <di...@yahoo.com> on 03/05/2002 09:12:01 AM
Please respond to axis-dev@xml.apache.org
To: axis-dev@xml.apache.org
cc:
Subject: Re: date vs dateTime (was: cvs commit:
xml-axis/java/src/org/apache/axis/encoding
DefaultTypeMappingImpl.java)
Russell,
How do i fix it? Any pointers? I ran "all-tests" before checking the code
in.
Thanks,
dims
--- Russell Butek <bu...@us.ibm.com> wrote:
> WARNING!! WARNING!!
>
> You've just broken WSDL->Java->WSDL roundtripping!
>
> WSDL date -> Java Date -> WSDL dateTime.
>
> Last week at the interop meetings we've discovered that we really need
> WSDL->Java->WSDL roundtripping for the round III group D tests. You've
> introduced a case where a group D scenario won't work.
>
> I'm not saying we shouldn't support date (breaking round tripping isn't
as
> bad as not working at all), but we MUST be VERY aware of the hole we're
> digging for ourselves.
>
> Russell Butek
> butek@us.ibm.com
> ---------------------- Forwarded by Russell Butek/Austin/IBM on
03/05/2002
> 08:26 AM ---------------------------
>
>
>
>
>
> dims@apache.org on 03/05/2002 08:19:57 AM
>
> Please respond to axis-dev@xml.apache.org
>
> To: xml-axis-cvs@apache.org
> cc:
>
> Subject: cvs commit: xml-axis/java/src/org/apache/axis/encoding
> DefaultTypeMappingImpl.java
>
>
>
>
> dims 02/03/05 06:19:57
>
> Modified: java/src/org/apache/axis Constants.java
> java/src/org/apache/axis/encoding
> DefaultTypeMappingImpl.java
> Log:
> adding support for "http://www.w3.org/2001/XMLSchema:date". "date" is a
> perfectly valid XML Schema data type and .NET and other WS frameworks
> handle it. Also see
> http://www.w3.org/2001/XMLSchema.xsd.
>
> Revision Changes Path
> 1.55 +1 -0 xml-axis/java/src/org/apache/axis/Constants.java
>
> Index: Constants.java
> ===================================================================
> RCS file: /home/cvs/xml-axis/java/src/org/apache/axis/Constants.java,v
> retrieving revision 1.54
> retrieving revision 1.55
> diff -u -r1.54 -r1.55
> --- Constants.java 28 Feb 2002 18:47:21 -0000 1.54
> +++ Constants.java 5 Mar 2002 14:19:57 -0000 1.55
> @@ -442,6 +442,7 @@
> public static final QName SOAP_VECTOR = new QName
> ("http://xml.apache.org/xml-soap", "Vector");
>
> public static QName XSD_DATE = new
> QName(Constants.URI_CURRENT_SCHEMA_XSD, "dateTime");
> + public static QName XSD_DATE1 = new
> QName(Constants.URI_CURRENT_SCHEMA_XSD, "date");
> public static QName XSD_DATE2= new
> QName(Constants.URI_1999_SCHEMA_XSD, "timeInstant");
> public static QName XSD_DATE3= new
> QName(Constants.URI_2000_SCHEMA_XSD, "timeInstant");
>
>
>
>
> 1.11 +7 -1
> xml-axis/java/src/org/apache/axis/encoding/DefaultTypeMappingImpl.java
>
> Index: DefaultTypeMappingImpl.java
> ===================================================================
> RCS file:
>
/home/cvs/xml-axis/java/src/org/apache/axis/encoding/DefaultTypeMappingImpl.java,v
>
> retrieving revision 1.10
> retrieving revision 1.11
> diff -u -r1.10 -r1.11
> --- DefaultTypeMappingImpl.java 12 Feb 2002 16:50:20 -0000 1.10
> +++ DefaultTypeMappingImpl.java 5 Mar 2002 14:19:57 -0000 1.11
> @@ -264,7 +264,13 @@
> new DateDeserializerFactory(java.util.Date.class,
> Constants.XSD_DATE3),
> true);
> - myRegister(Constants.XSD_DATE, java.util.Date.class,
> + myRegister(Constants.XSD_DATE1, java.util.Date.class,
> + new DateSerializerFactory(java.util.Date.class,
> + Constants.XSD_DATE1),
> + new DateDeserializerFactory(java.util.Date.class,
> + Constants.XSD_DATE1),
> + true);
> + myRegister(Constants.XSD_DATE, java.util.Date.class,
> new DateSerializerFactory(java.util.Date.class,
> Constants.XSD_DATE),
> new DateDeserializerFactory(java.util.Date.class,
>
>
>
>
>
=====
Davanum Srinivas - http://xml.apache.org/~dims/
__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/
Re: date vs dateTime (was: cvs commit: xml-axis/java/src/org/apache/axis/encoding DefaultTypeMappingImpl.java)
Posted by Davanum Srinivas <di...@yahoo.com>.
Russell,
Quick Question. I see byte[].class used with XSD_HEXBIN, SOAP_BASE64 and XSD_BASE64...How is
dateTime/date different?
Thanks,
dims
--- Russell Butek <bu...@us.ibm.com> wrote:
> The only way to fix this is to have a different Java class for date vs
> dateTime. Do we really want to roll our own DateTime class?
>
> I don't know whether we want to fix it. I just want us all to be aware
> that we WILL have users run into this issue and that we WILL have to admit
> that it's broken, and unlikely to be fixed, unless someone wants to create
> a special class.
>
> This problem will exist all over the place, not just here. We don't
> support xsd:Name, xsd:ID, xsd:NMTOKEN, etc. All of these would probably
> map to java.lang.String since they're all derived from xsd:string. But
> that would break roundtripping unless we create a new class for each of
> these XML types. Ditto for xsd:integer derived types like
> nonPositiveInteger, nonNegativeInteger, etc.
>
> I'd prefer that this were fixed in one place (like the WSDL and/or JAX-RPC
> specs) rather than haphazardly in our implementation. The specs should not
> allow many-to-one mappings.
>
> Russell Butek
> butek@us.ibm.com
>
>
> Davanum Srinivas <di...@yahoo.com> on 03/05/2002 09:12:01 AM
>
> Please respond to axis-dev@xml.apache.org
>
> To: axis-dev@xml.apache.org
> cc:
> Subject: Re: date vs dateTime (was: cvs commit:
> xml-axis/java/src/org/apache/axis/encoding
> DefaultTypeMappingImpl.java)
>
>
>
> Russell,
>
> How do i fix it? Any pointers? I ran "all-tests" before checking the code
> in.
>
> Thanks,
> dims
>
> --- Russell Butek <bu...@us.ibm.com> wrote:
> > WARNING!! WARNING!!
> >
> > You've just broken WSDL->Java->WSDL roundtripping!
> >
> > WSDL date -> Java Date -> WSDL dateTime.
> >
> > Last week at the interop meetings we've discovered that we really need
> > WSDL->Java->WSDL roundtripping for the round III group D tests. You've
> > introduced a case where a group D scenario won't work.
> >
> > I'm not saying we shouldn't support date (breaking round tripping isn't
> as
> > bad as not working at all), but we MUST be VERY aware of the hole we're
> > digging for ourselves.
> >
> > Russell Butek
> > butek@us.ibm.com
> > ---------------------- Forwarded by Russell Butek/Austin/IBM on
> 03/05/2002
> > 08:26 AM ---------------------------
> >
> >
> >
> >
> >
> > dims@apache.org on 03/05/2002 08:19:57 AM
> >
> > Please respond to axis-dev@xml.apache.org
> >
> > To: xml-axis-cvs@apache.org
> > cc:
> >
> > Subject: cvs commit: xml-axis/java/src/org/apache/axis/encoding
> > DefaultTypeMappingImpl.java
> >
> >
> >
> >
> > dims 02/03/05 06:19:57
> >
> > Modified: java/src/org/apache/axis Constants.java
> > java/src/org/apache/axis/encoding
> > DefaultTypeMappingImpl.java
> > Log:
> > adding support for "http://www.w3.org/2001/XMLSchema:date". "date" is a
> > perfectly valid XML Schema data type and .NET and other WS frameworks
> > handle it. Also see
> > http://www.w3.org/2001/XMLSchema.xsd.
> >
> > Revision Changes Path
> > 1.55 +1 -0 xml-axis/java/src/org/apache/axis/Constants.java
> >
> > Index: Constants.java
> > ===================================================================
> > RCS file: /home/cvs/xml-axis/java/src/org/apache/axis/Constants.java,v
> > retrieving revision 1.54
> > retrieving revision 1.55
> > diff -u -r1.54 -r1.55
> > --- Constants.java 28 Feb 2002 18:47:21 -0000 1.54
> > +++ Constants.java 5 Mar 2002 14:19:57 -0000 1.55
> > @@ -442,6 +442,7 @@
> > public static final QName SOAP_VECTOR = new QName
> > ("http://xml.apache.org/xml-soap", "Vector");
> >
> > public static QName XSD_DATE = new
> > QName(Constants.URI_CURRENT_SCHEMA_XSD, "dateTime");
> > + public static QName XSD_DATE1 = new
> > QName(Constants.URI_CURRENT_SCHEMA_XSD, "date");
> > public static QName XSD_DATE2= new
> > QName(Constants.URI_1999_SCHEMA_XSD, "timeInstant");
> > public static QName XSD_DATE3= new
> > QName(Constants.URI_2000_SCHEMA_XSD, "timeInstant");
> >
> >
> >
> >
> > 1.11 +7 -1
> > xml-axis/java/src/org/apache/axis/encoding/DefaultTypeMappingImpl.java
> >
> > Index: DefaultTypeMappingImpl.java
> > ===================================================================
> > RCS file:
> >
> /home/cvs/xml-axis/java/src/org/apache/axis/encoding/DefaultTypeMappingImpl.java,v
>
> >
> > retrieving revision 1.10
> > retrieving revision 1.11
> > diff -u -r1.10 -r1.11
> > --- DefaultTypeMappingImpl.java 12 Feb 2002 16:50:20 -0000 1.10
> > +++ DefaultTypeMappingImpl.java 5 Mar 2002 14:19:57 -0000 1.11
> > @@ -264,7 +264,13 @@
> > new DateDeserializerFactory(java.util.Date.class,
> > Constants.XSD_DATE3),
> > true);
> > - myRegister(Constants.XSD_DATE, java.util.Date.class,
> > + myRegister(Constants.XSD_DATE1, java.util.Date.class,
> > + new DateSerializerFactory(java.util.Date.class,
> > + Constants.XSD_DATE1),
> > + new DateDeserializerFactory(java.util.Date.class,
> > + Constants.XSD_DATE1),
> > + true);
> > + myRegister(Constants.XSD_DATE, java.util.Date.class,
> > new DateSerializerFactory(java.util.Date.class,
> > Constants.XSD_DATE),
> > new DateDeserializerFactory(java.util.Date.class,
> >
> >
> >
> >
> >
>
>
> =====
> Davanum Srinivas - http://xml.apache.org/~dims/
>
> __________________________________________________
> Do You Yahoo!?
> Try FREE Yahoo! Mail - the world's greatest free email!
> http://mail.yahoo.com/
>
>
=====
Davanum Srinivas - http://xml.apache.org/~dims/
__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/