From boris at codesynthesis.com  Tue Mar  1 08:28:39 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Tue Mar  1 08:27:21 2011
Subject: [xsd-users] enumerations not exported in dll when
	"generate-inline" enabled
In-Reply-To: <C141EE9A-11B7-4A3C-9B7E-23B59E5FDB70@gblsys.com>
References: <C141EE9A-11B7-4A3C-9B7E-23B59E5FDB70@gblsys.com>
Message-ID: <boris.20110301152501@codesynthesis.com>

Hi Matt,

Matt Burnham <mburnham@gblsys.com> writes:

> The resulting class is put in a DLL and when I try and link  an application 
> with the DLL, I get the following link error:
> unresolved external symbol "public: static char const * const * const tmp::action::_xsd_action_literals_" (?_xsd_action_literals_@action@tmp@@2QBQBDB)

Your DLL export file has the following definitions:

#  ifdef MY_DLL_EXPORT
#    define MY_DLL_DECL __declspec(dllexport)
#  elif MY_DLL_IMPORT
#    define MY_DLL_DECL __declspec(dllimport)
#  else
#    define MY_DLL_DECL // for static lib
#  endif

In particular, it does not import things by default; one has to explicitly
define MY_DLL_IMPORT for that to happen. Naturally, your test does not define
this macro. If add it to the preprocessor definitions in its project file,
everything links fine.

Boris

From alessandro.carpi at gmail.com  Tue Mar  1 15:05:30 2011
From: alessandro.carpi at gmail.com (Alessandro Carpi)
Date: Tue Mar  1 15:05:57 2011
Subject: [xsd-users] Report line number when error is encountered
Message-ID: <AANLkTikWuwvvRi981DHods_bBX3qdB9KLwS2gph0NLSS@mail.gmail.com>

Hi,
I am very happy to use XSD C++/Parser for a personal project in which
actually I'm involved.

I have a simple question: is there a way to raise an XML exception when a
particular condition was encountered during the parsing of a file?

Specifically, I have a language that internally uses bison to parse an
expression, something like this:

<if expr="a>b">
 <if-true>
   [stuff]
 </if-true>
 <if-false>
   [other-stuff]
 </if-false>
</if>
Obviously expressions should be written according to a specific grammar
defined by the external bison parser. Actually I can recognize the event
into the code which parses the attribute "expr" (the bison parser returns a
null pointer).

It would nice if, when this kind of errors are encountered (aka. expression
parser returns a null pointer because is written wrongly) , the XML parser
prints the XML line of the wrong expression.

Thank you very much.

--
Alessandro
From mjklaim at gmail.com  Wed Mar  2 03:40:12 2011
From: mjklaim at gmail.com (=?UTF-8?Q?Klaim_=2D_Jo=C3=ABl_Lamotte?=)
Date: Wed Mar  2 03:40:19 2011
Subject: [xsd-users] Hierarchies and code generation.
Message-ID: <AANLkTi=R+R3V7NSeoW-L1H7V-8fyuDJWaDHJDO44e+nV@mail.gmail.com>

Hi,

I'm having troubles working on an xsd definition to generate an XML-format
language parser in C++.

Here is what I want to have in the end in C++ (in a specific namespace) :

 1. an abstract class Object. It's a common interface.
 2. child classes of Object : Object_Unknown, Object_Video, Object_Sprite,
Object_Audio, Object_Sprite
 3. an Object should have a "list" of children Object objects. As Element is
abstract, the children should be of Object child class types (2.).
 4. I still want to use the specific xml interface of the child types in the
"list" element :

<object id="my_object" >
    <children>
        <object ... />
        <sprite > <position x="1" y="2" /> </sprite> <--- specific interface
of a child type
    </children>
</object>



To achieve all this I tried several variant of the following :

<xs:complexType name="object" abstract="true" ... >

    ...blahblah
    <xs:sequence>
         <xs:element name="children" type="list_object" />
    </xs:sequence>


</xs:complexType>

child element types...
<xs:complexType name="object_unknown" ... />
<xs:complexType name="object_sprite" ... />
<xs:complexType name="object_audio" ... />

etc.

<xs:complexType name="list_object">

   <xs:sequence maxOccur="unbouded" minOccur="0" >
       <xs:element name="object" type="object_unknown" />
       <xs:element name="sprite" type="object_sprite" />
       <xs:element name="audio" type="object_audio" />
        etc.
   </xs:sequence>


But the generated C++ code of list_object will never expose a sequence of
"object" types, using
the common interface to list all the object instances without knowing the
child type.
It currently expose sequences of child object types, and that follows the
xsd correctly.


So, is there a way to make my list_object provide a sequence of "object"?

Thanks for your time.


Jo?l Lamotte
From ivan.lelann at free.fr  Wed Mar  2 04:03:32 2011
From: ivan.lelann at free.fr (Ivan Le Lann)
Date: Wed Mar  2 04:03:42 2011
Subject: [xsd-users] Hierarchies and code generation.
In-Reply-To: <975839662.2447481299056598643.JavaMail.root@zimbra36-e6.priv.proxad.net>
Message-ID: <1384720355.2447521299056612237.JavaMail.root@zimbra36-e6.priv.proxad.net>


----- "Klaim - Jo?l Lamotte" <mjklaim@gmail.com> a ?crit :

> Hi,
> 
> I'm having troubles working on an xsd definition to generate an
> XML-format
> language parser in C++.
> 
> Here is what I want to have in the end in C++ (in a specific
> namespace) :
> 
>  1. an abstract class Object. It's a common interface.
>  2. child classes of Object : Object_Unknown, Object_Video,
> Object_Sprite,
> Object_Audio, Object_Sprite
>  3. an Object should have a "list" of children Object objects. As
> Element is
> abstract, the children should be of Object child class types (2.).
>  4. I still want to use the specific xml interface of the child types
> in the
> "list" element :
> 
> <object id="my_object" >
>     <children>
>         <object ... />
>         <sprite > <position x="1" y="2" /> </sprite> <--- specific
> interface
> of a child type
>     </children>
> </object>
> 
> 
> 
> To achieve all this I tried several variant of the following :
> 
> <xs:complexType name="object" abstract="true" ... >
> 
>     ...blahblah
>     <xs:sequence>
>          <xs:element name="children" type="list_object" />
>     </xs:sequence>
> 
> 
> </xs:complexType>
> 
> child element types...
> <xs:complexType name="object_unknown" ... />
> <xs:complexType name="object_sprite" ... />
> <xs:complexType name="object_audio" ... />
> 
> etc.
> 
> <xs:complexType name="list_object">
> 
>    <xs:sequence maxOccur="unbouded" minOccur="0" >
>        <xs:element name="object" type="object_unknown" />
>        <xs:element name="sprite" type="object_sprite" />
>        <xs:element name="audio" type="object_audio" />
>         etc.
>    </xs:sequence>
> 
> 
> But the generated C++ code of list_object will never expose a sequence
> of
> "object" types, using
> the common interface to list all the object instances without knowing
> the
> child type.

Indeed. I fail to see how it could when that information is not in the schema.

> So, is there a way to make my list_object provide a sequence of
> "object"?
> 

See the schema taken from XSD/e Hybrid  "polymorphic" example below.
I have no XSD distribution now on my computer, but I guess a similar example 
is in there too.

Regards,
Ivan

<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema">

  <xsd:complexType name="person">
    <xsd:sequence>
      <xsd:element name="name" type="xsd:string"/>
    </xsd:sequence>
  </xsd:complexType>

  <!-- substitution group root -->
  <xsd:element name="person" type="person"/>


  <xsd:complexType name="superman">
    <xsd:complexContent>
      <xsd:extension base="person">
        <xsd:attribute name="can-fly" type="xsd:boolean" use="required"/>
      </xsd:extension>
    </xsd:complexContent>
  </xsd:complexType>

  <xsd:element name="superman" type="superman" substitutionGroup="person"/>

  <xsd:complexType name="batman">
    <xsd:complexContent>
      <xsd:extension base="superman">
        <xsd:attribute name="wing-span" type="xsd:unsignedInt" use="required"/>
      </xsd:extension>
    </xsd:complexContent>
  </xsd:complexType>

  <xsd:complexType name="supermen">
    <xsd:sequence>
      <xsd:element ref="person" maxOccurs="unbounded"/>
    </xsd:sequence>
  </xsd:complexType>

  <xsd:element name="supermen" type="supermen"/>

</xsd:schema>

From mjklaim at gmail.com  Wed Mar  2 04:15:36 2011
From: mjklaim at gmail.com (=?UTF-8?Q?Klaim_=2D_Jo=C3=ABl_Lamotte?=)
Date: Wed Mar  2 04:15:44 2011
Subject: [xsd-users] Hierarchies and code generation.
In-Reply-To: <1384720355.2447521299056612237.JavaMail.root@zimbra36-e6.priv.proxad.net>
References: <975839662.2447481299056598643.JavaMail.root@zimbra36-e6.priv.proxad.net>
	<1384720355.2447521299056612237.JavaMail.root@zimbra36-e6.priv.proxad.net>
Message-ID: <AANLkTi=S-jO9wjnv+8KGQwHB1WhC4MxKJ-Z=5tQgc9Me@mail.gmail.com>

2011/3/2 Ivan Le Lann <ivan.lelann@free.fr>

> <xsd:element name="superman" type="superman" *substitutionGroup*
> ="person"/>


O__O

I think that's the part I didn't know about.

I'm not sure I understand it fully though, because of the use of elements
outside of complexType definition (I have only the root element in my xsd)

Why is an element required for the base (root) type?
Should I do this for my abstract element types?


Thank you very much!

Jo?l
From victor.manoliu at yahoo.com  Wed Mar  2 05:03:47 2011
From: victor.manoliu at yahoo.com (Adrian Victor Manoliu)
Date: Wed Mar  2 05:03:54 2011
Subject: [xsd-users] CityGML deserialization - CityModel generation
In-Reply-To: <boris.20110228163211@codesynthesis.com>
References: <995783.93915.qm@web45504.mail.sp1.yahoo.com>
	<boris.20110222174426@codesynthesis.com>
	<399571.66645.qm@web45511.mail.sp1.yahoo.com>
	<boris.20110223152512@codesynthesis.com>
	<392645.14182.qm@web45503.mail.sp1.yahoo.com>
	<boris.20110228163211@codesynthesis.com>
Message-ID: <645117.22384.qm@web45512.mail.sp1.yahoo.com>


Hello,

Thanks for the casting method advice, however I still don't have anything to 
iterate upon, after the parsing is done. From what I've seen, the only way to go 
through the created elements is via the 
_GenericApplicationPropertyOfCityModel_(const_)iterator. In this case, this 
iterator does not work because the begin() and end() methods look in the v_ 
field, which is empty. The field where the items exist perfectly well is 
container_, but I don't have access to that through any public method.

Adrian




----- Original Message ----
From: Boris Kolpackov <boris@codesynthesis.com>
To: Adrian Victor Manoliu <victor.manoliu@yahoo.com>
Cc: xsd-users@codesynthesis.com
Sent: Mon, February 28, 2011 3:48:35 PM
Subject: Re: [xsd-users] CityGML deserialization - CityModel generation

Hi Adrian,

Adrian Victor Manoliu <victor.manoliu@yahoo.com> writes:

> const std::string fName = "waldbruecke_v1.0.0.gml";
> std::auto_ptr<CityModelType> iCity(CityModel(fName, flags.dont_validate)); 
> 
> [...]
> 
> Now, once I have iCity, I simply want to iterate over its elements, that 
> is the Buildings, BuildingParts etc. by doing something like this (I have
> been doing this from the start):
> 
> CityModelType::_GenericApplicationPropertyOfCityModel_const_iterator ci;
> 
> for (ci  = iCity->_GenericApplicationPropertyOfCityModel().begin();
>      ci != iCity->_GenericApplicationPropertyOfCityModel().end();
>      ++ci)
> {
>     [...]
> }  
> 
> however in this case I don't see how I could know what type to cast these 
> objects to

You need to study the GML/CityGML schemas to see which concrete elements
(and types) substitute which abstract elements.

Here is an example body of the above for-loop. Hopefully, it will get you
started:

for (ci  = iCity->_GenericApplicationPropertyOfCityModel().begin();
     ci != iCity->_GenericApplicationPropertyOfCityModel().end();
     ++ci)
{
  xml_schema::type& x (*ci);

  if (gml::FeaturePropertyType* cityObjectMember = 
        dynamic_cast<gml::FeaturePropertyType*> (&x))
  {
    gml::FeaturePropertyType::_Feature_const_iterator fi;

    for (fi = cityObjectMember->_Feature ().begin ();
         fi != cityObjectMember->_Feature ().end ();
         ++fi)
    {
      gml::AbstractFeatureType& f (*fi);

      if (citygml::building::BuildingType* b = 
        dynamic_cast<citygml::building::BuildingType*> (&f);
      {
        // We've got a building.
      }
      else if (...)
      {
        ...
      }
    }
  }
}

Boris



      

From victor.manoliu at yahoo.com  Wed Mar  2 05:08:07 2011
From: victor.manoliu at yahoo.com (Adrian Victor Manoliu)
Date: Wed Mar  2 05:08:15 2011
Subject: [xsd-users] CityGML deserialization - CityModel generation
In-Reply-To: <boris.20110228163211@codesynthesis.com>
References: <995783.93915.qm@web45504.mail.sp1.yahoo.com>
	<boris.20110222174426@codesynthesis.com>
	<399571.66645.qm@web45511.mail.sp1.yahoo.com>
	<boris.20110223152512@codesynthesis.com>
	<392645.14182.qm@web45503.mail.sp1.yahoo.com>
	<boris.20110228163211@codesynthesis.com>
Message-ID: <784323.88556.qm@web45505.mail.sp1.yahoo.com>


Hello,

Thanks for the casting method advice, however I still don't have anything to 
iterate upon, after the parsing is done. From what I've seen, the only way to go 

through the created elements is via the 
_GenericApplicationPropertyOfCityModel_(const_)iterator. In this case, this 
iterator does not work because the begin() and end() methods look in the v_ 
field, which is empty. The field where the items exist perfectly well is 
container_, but I don't have access to that through any public method.

Adrian



----- Original Message ----
From: Boris Kolpackov <boris@codesynthesis.com>
To: Adrian Victor Manoliu <victor.manoliu@yahoo.com>
Cc: xsd-users@codesynthesis.com
Sent: Mon, February 28, 2011 3:48:35 PM
Subject: Re: [xsd-users] CityGML deserialization - CityModel generation

Hi Adrian,

Adrian Victor Manoliu <victor.manoliu@yahoo.com> writes:

> const std::string fName = "waldbruecke_v1.0.0.gml";
> std::auto_ptr<CityModelType> iCity(CityModel(fName, flags.dont_validate)); 
> 
> [...]
> 
> Now, once I have iCity, I simply want to iterate over its elements, that 
> is the Buildings, BuildingParts etc. by doing something like this (I have
> been doing this from the start):
> 
> CityModelType::_GenericApplicationPropertyOfCityModel_const_iterator ci;
> 
> for (ci  = iCity->_GenericApplicationPropertyOfCityModel().begin();
>      ci != iCity->_GenericApplicationPropertyOfCityModel().end();
>      ++ci)
> {
>     [...]
> }  
> 
> however in this case I don't see how I could know what type to cast these 
> objects to

You need to study the GML/CityGML schemas to see which concrete elements
(and types) substitute which abstract elements.

Here is an example body of the above for-loop. Hopefully, it will get you
started:

for (ci  = iCity->_GenericApplicationPropertyOfCityModel().begin();
     ci != iCity->_GenericApplicationPropertyOfCityModel().end();
     ++ci)
{
  xml_schema::type& x (*ci);

  if (gml::FeaturePropertyType* cityObjectMember = 
        dynamic_cast<gml::FeaturePropertyType*> (&x))
  {
    gml::FeaturePropertyType::_Feature_const_iterator fi;

    for (fi = cityObjectMember->_Feature ().begin ();
         fi != cityObjectMember->_Feature ().end ();
         ++fi)
    {
      gml::AbstractFeatureType& f (*fi);

      if (citygml::building::BuildingType* b = 
        dynamic_cast<citygml::building::BuildingType*> (&f);
      {
        // We've got a building.
      }
      else if (...)
      {
        ...
      }
    }
  }
}

Boris



      

From boris at codesynthesis.com  Wed Mar  2 08:55:19 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Wed Mar  2 08:54:06 2011
Subject: [xsd-users] Hierarchies and code generation.
In-Reply-To: <AANLkTi=S-jO9wjnv+8KGQwHB1WhC4MxKJ-Z=5tQgc9Me@mail.gmail.com>
References: <975839662.2447481299056598643.JavaMail.root@zimbra36-e6.priv.proxad.net>
	<1384720355.2447521299056612237.JavaMail.root@zimbra36-e6.priv.proxad.net>
	<AANLkTi=S-jO9wjnv+8KGQwHB1WhC4MxKJ-Z=5tQgc9Me@mail.gmail.com>
Message-ID: <boris.20110302155254@codesynthesis.com>

Hi Jo?l,

Klaim - Jo?l Lamotte <mjklaim@gmail.com> writes:

> I'm not sure I understand it fully though, because of the use of elements
> outside of complexType definition (I have only the root element in my xsd)
>
> Why is an element required for the base (root) type?
> Should I do this for my abstract element types?

That's how the substitution groups mechanism works in XML Schema.

For XSD-specific details on this feature, see Section 2.11, "Mapping for
xsi:type and Substitution Groups" in the C++/Tree Mapping User Manual:

http://www.codesynthesis.com/projects/xsd/documentation/cxx/tree/manual/#2.11

Boris

From boris at codesynthesis.com  Wed Mar  2 09:11:57 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Wed Mar  2 09:10:44 2011
Subject: [xsd-users] Report line number when error is encountered
In-Reply-To: <AANLkTikWuwvvRi981DHods_bBX3qdB9KLwS2gph0NLSS@mail.gmail.com>
References: <AANLkTikWuwvvRi981DHods_bBX3qdB9KLwS2gph0NLSS@mail.gmail.com>
Message-ID: <boris.20110302160438@codesynthesis.com>

Hi Alessandro,

Alessandro Carpi <alessandro.carpi@gmail.com> writes:

> It would nice if, when this kind of errors are encountered (aka. expression
> parser returns a null pointer because is written wrongly) , the XML parser
> prints the XML line of the wrong expression.

In other words you want to get the line/column information of the current
position so that you can print it together with your application-specific
error.

The way to get to this information depends on the underlying XML parser
you are using (Expat or Xerces-C++). But the general approach would be
to create the parser yourself, then use this parser to parse the document
and query it for the current line/column information in case of an error.
This is fairly easy to implement with Expat and slightly more difficult
with Xerces-C++.

Boris

From mjklaim at gmail.com  Wed Mar  2 09:32:31 2011
From: mjklaim at gmail.com (=?UTF-8?Q?Klaim_=2D_Jo=C3=ABl_Lamotte?=)
Date: Wed Mar  2 09:32:40 2011
Subject: [xsd-users] Hierarchies and code generation.
In-Reply-To: <boris.20110302155254@codesynthesis.com>
References: <975839662.2447481299056598643.JavaMail.root@zimbra36-e6.priv.proxad.net>
	<1384720355.2447521299056612237.JavaMail.root@zimbra36-e6.priv.proxad.net>
	<AANLkTi=S-jO9wjnv+8KGQwHB1WhC4MxKJ-Z=5tQgc9Me@mail.gmail.com>
	<boris.20110302155254@codesynthesis.com>
Message-ID: <AANLkTi=s4HPTPtiE=21L83prG9xKYd+__1Cp+ffFTEOC@mail.gmail.com>

>
>
> > Why is an element required for the base (root) type?
> > Should I do this for my abstract element types?
>
> That's how the substitution groups mechanism works in XML Schema.
>
> For XSD-specific details on this feature, see Section 2.11, "Mapping for
> xsi:type and Substitution Groups" in the C++/Tree Mapping User Manual:
>
>
> http://www.codesynthesis.com/projects/xsd/documentation/cxx/tree/manual/#2.11
>
> Boris
>

Thank your very very very much, that's the answer to all my questions. :D
From alessandro.carpi at gmail.com  Wed Mar  2 09:35:49 2011
From: alessandro.carpi at gmail.com (Alessandro Carpi)
Date: Wed Mar  2 09:36:16 2011
Subject: [xsd-users] Report line number when error is encountered
In-Reply-To: <boris.20110302160438@codesynthesis.com>
References: <AANLkTikWuwvvRi981DHods_bBX3qdB9KLwS2gph0NLSS@mail.gmail.com>
	<boris.20110302160438@codesynthesis.com>
Message-ID: <AANLkTikwUiKLMhcDGt=G1Q9pjkwWAwUUnXheHhx-S_4=@mail.gmail.com>

2011/3/2 Boris Kolpackov <boris@codesynthesis.com>

> Hi Alessandro,
>
> Alessandro Carpi <alessandro.carpi@gmail.com> writes:
>
> > It would nice if, when this kind of errors are encountered (aka.
> expression
> > parser returns a null pointer because is written wrongly) , the XML
> parser
> > prints the XML line of the wrong expression.
>
> In other words you want to get the line/column information of the current
> position so that you can print it together with your application-specific
> error.
>
Actually yes. I was thinking to throw a sort of "not valid xml  exception"
 in order to obtain the printing of line and column for free :-)

>
> The way to get to this information depends on the underlying XML parser
> you are using (Expat or Xerces-C++). But the general approach would be
> to create the parser yourself, then use this parser to parse the document
> and query it for the current line/column information in case of an error.
> This is fairly easy to implement with Expat and slightly more difficult
> with Xerces-C++.
>
> I'm using Xercesc-C++ because it is already present to the target system.
If the above solution is not praticable, how can I obtain such informations
from the underlying parser ?

 Thank you very much :-)

-- 
--
Alessandro Carpi
From boris at codesynthesis.com  Wed Mar  2 09:49:10 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Wed Mar  2 09:47:58 2011
Subject: [xsd-users] Report line number when error is encountered
In-Reply-To: <AANLkTikwUiKLMhcDGt=G1Q9pjkwWAwUUnXheHhx-S_4=@mail.gmail.com>
References: <AANLkTikWuwvvRi981DHods_bBX3qdB9KLwS2gph0NLSS@mail.gmail.com>
	<boris.20110302160438@codesynthesis.com>
	<AANLkTikwUiKLMhcDGt=G1Q9pjkwWAwUUnXheHhx-S_4=@mail.gmail.com>
Message-ID: <boris.20110302164445@codesynthesis.com>

Hi Alessandro,

Alessandro Carpi <alessandro.carpi@gmail.com> writes:

> Actually yes. I was thinking to throw a sort of "not valid xml  exception"
> in order to obtain the printing of line and column for free :-)

Yes, this approach is probably the easiest. Simply derive your own exception
class from schema_exception like this:

#include <xsd/cxx/parser/schema-exceptions.hxx>

struct my_exception: xsd::cxx::parser::schema_exception<char>
{
  virtual std::string
  message () const
  {
    return "...";
  }
};

And throw it from one of the callbacks. You should then see the message along
with the file/line/column information as if it was an XML validation error.

Boris

From alessandro.carpi at gmail.com  Wed Mar  2 11:40:11 2011
From: alessandro.carpi at gmail.com (Alessandro Carpi)
Date: Wed Mar  2 11:40:42 2011
Subject: [xsd-users] Report line number when error is encountered
In-Reply-To: <boris.20110302164445@codesynthesis.com>
References: <AANLkTikWuwvvRi981DHods_bBX3qdB9KLwS2gph0NLSS@mail.gmail.com>
	<boris.20110302160438@codesynthesis.com>
	<AANLkTikwUiKLMhcDGt=G1Q9pjkwWAwUUnXheHhx-S_4=@mail.gmail.com>
	<boris.20110302164445@codesynthesis.com>
Message-ID: <AANLkTintQCxvecF605nsdQRYjxWQOj5=xvPJf+=2VAv2@mail.gmail.com>

2011/3/2 Boris Kolpackov <boris@codesynthesis.com>

> Hi Alessandro,
>
> Alessandro Carpi <alessandro.carpi@gmail.com> writes:
>
> > Actually yes. I was thinking to throw a sort of "not valid xml
>  exception"
> > in order to obtain the printing of line and column for free :-)
>
> Yes, this approach is probably the easiest. Simply derive your own
> exception
> class from schema_exception like this:
>
> #include <xsd/cxx/parser/schema-exceptions.hxx>
>
> struct my_exception: xsd::cxx::parser::schema_exception<char>
> {
>  virtual std::string
>  message () const
>  {
>    return "...";
>  }
> };
>
> And throw it from one of the callbacks. You should then see the message
> along
> with the file/line/column information as if it was an XML validation error.
>
> Boris
>

Thank you! It is exactly what i need :-)


-- 
--
Alessandro Carpi
From stargate625 at gmail.com  Thu Mar  3 12:49:15 2011
From: stargate625 at gmail.com (William Canton)
Date: Thu Mar  3 14:44:07 2011
Subject: [xsd-users] Error in building the generated header due to XSD
	runtime version mismatch
Message-ID: <AANLkTim-JHKKyhj4Yg+-YKdAfcz2eFGGe7RunYjhVdCf@mail.gmail.com>

I am not sure how I got the following fatal error during the build on the
VC++ 9.0:

fatal error C1189: #error :  XSD runtime version mismatch

This error was reported in one of my generated *.hxx file at the line:

#if (XSD_INT_VERSION != 3030000L)
#error XSD runtime version mismatch
#endif

But hovering over the macro XSD_INT_VERSION showed a correct version number
of 3030000L.  How could this happen?

Wiilliam
From boris at codesynthesis.com  Thu Mar  3 15:00:02 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Thu Mar  3 14:59:04 2011
Subject: [xsd-users] Error in building the generated header due to XSD
	runtime version mismatch
In-Reply-To: <AANLkTim-JHKKyhj4Yg+-YKdAfcz2eFGGe7RunYjhVdCf@mail.gmail.com>
References: <AANLkTim-JHKKyhj4Yg+-YKdAfcz2eFGGe7RunYjhVdCf@mail.gmail.com>
Message-ID: <boris.20110303215434@codesynthesis.com>

Hi William,

William Canton <stargate625@gmail.com> writes:

> This error was reported in one of my generated *.hxx file at the line:
> 
> #if (XSD_INT_VERSION != 3030000L)
> #error XSD runtime version mismatch
> #endif
> 
> But hovering over the macro XSD_INT_VERSION showed a correct version number
> of 3030000L.  How could this happen?

Hm, this is strange. Does it happen with all the generated files or only this
one?

Also, can you open (by right-clicking on it in the source code and selecting
the "Open Document" menu item) the <xsd/cxx/config.hxx> file that is included
at the beginning of the generated header file. Then do the same for
<xsd/cxx/version.hxx> and see if the version defined there is correct.

You may also want to double check if there is a previous version of XSD
installed somewhere on your machine.

Boris

From stargate625 at gmail.com  Thu Mar  3 17:23:57 2011
From: stargate625 at gmail.com (William Canton)
Date: Fri Mar  4 10:21:43 2011
Subject: [xsd-users] Error in building the generated header due to XSD
	runtime version mismatch
In-Reply-To: <boris.20110303215434@codesynthesis.com>
References: <AANLkTim-JHKKyhj4Yg+-YKdAfcz2eFGGe7RunYjhVdCf@mail.gmail.com>
	<boris.20110303215434@codesynthesis.com>
Message-ID: <AANLkTi=fXunvma+98-gb=j=qy5p+xTiUZKXLa+kcWtwM@mail.gmail.com>

Hi Boris,

I eventually found out that my workspace in Visual Studio C++ has multiple
projects.  As it happened, some of the projects still referenced the old 3.2
version.  After I fixed all the project properties, the CodeSynthesis could
generate the version number embedded in the corresponding *.hxx files
correctly.

However, there is another problem though.  The CodeSynthesis doesn't seem to
be able to parse a valid XML document with optional field elements.  For
example, if my XML file contains an empty optional <EISSDescriptor/>, the
xerces parser would throw exception at runtime.  Again, it took me a while
to discover this problem by removing the empty EISSDescriptor elements from
the XML document.

Is this a bug in the CodeSynthesis?  Attached is the standard XML schema
file that I used to generate the classes.  Please try it to see if you will
encounter the same anomaly that I stumbled upon.

William


On Thu, Mar 3, 2011 at 3:00 PM, Boris Kolpackov <boris@codesynthesis.com>wrote:

> Hi William,
>
> William Canton <stargate625@gmail.com> writes:
>
> > This error was reported in one of my generated *.hxx file at the line:
> >
> > #if (XSD_INT_VERSION != 3030000L)
> > #error XSD runtime version mismatch
> > #endif
> >
> > But hovering over the macro XSD_INT_VERSION showed a correct version
> number
> > of 3030000L.  How could this happen?
>
> Hm, this is strange. Does it happen with all the generated files or only
> this
> one?
>
> Also, can you open (by right-clicking on it in the source code and
> selecting
> the "Open Document" menu item) the <xsd/cxx/config.hxx> file that is
> included
> at the beginning of the generated header file. Then do the same for
> <xsd/cxx/version.hxx> and see if the version defined there is correct.
>
> You may also want to double check if there is a previous version of XSD
> installed somewhere on your machine.
>
> Boris
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: BigHunt_example.xml
Type: text/xml
Size: 11255 bytes
Desc: not available
Url : http://codesynthesis.com/pipermail/xsd-users/attachments/20110303/9b2f21b9/BigHunt_example-0001.bin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: standard_cdf.xsd
Type: application/octet-stream
Size: 75353 bytes
Desc: not available
Url : http://codesynthesis.com/pipermail/xsd-users/attachments/20110303/9b2f21b9/standard_cdf-0001.obj
From rlischner at proteuseng.com  Fri Mar  4 09:49:54 2011
From: rlischner at proteuseng.com (Ray Lischner)
Date: Fri Mar  4 10:21:44 2011
Subject: [xsd-users] shared objects that load compiled schemas from a shared
	object
Message-ID: <E7AA20464DFA774A9ACC4B8EC01CA08105BB586852@SMD-EX01.proteus-technologies.com>

We are using RedHat Linux 5.x, g++ 4.1.2, xsd 3.3.0.
We compile a set of schemas that use substitutionGroups into a shared object X.so.
We build two more shared objects, A.so and B.so, which make use of the schemas in X.so. We build A.so and B.so with -lX.

When we load A.so and B.so into an application, the substitutionGroups don't work. That is, a dynamic_cast<Type>() of an object of type Type returns a null pointer. When the same code executes with only A.so loaded, or only B.so loaded, the code works. It fails only when we load A.so and B.so into the same process space.

If we set LD_PRELOAD=X.so, the program works with A.so and B.so loaded.

Is there a way we can make this work without the LD_PRELOAD? I thought the loader would be able to detect that it needs to load only one copy of X.so, but apparently I am mistaken.

Ray Lischner,
Distinguished Member of Technical Staff
133 National Business Pkwy, Ste 150     t. 443.539.3448
Annapolis Junction, MD 20701                c. 410.854.5170
rlischner@proteuseng.com                     f. 443.539.3370

This electronic message and any files transmitted with it contain information
which may be privileged and/or proprietary. The information is intended for use
solely by the intended recipient(s). If you are not the intended recipient, be
aware that any disclosure, copying, distribution or use of this information is
prohibited. If you have received this electronic message in error, please advise
the sender by reply email or by telephone (443.539.3400) and delete the message.
From boris at codesynthesis.com  Fri Mar  4 10:31:08 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Fri Mar  4 10:30:10 2011
Subject: [xsd-users] Error in building the generated header due to XSD
	runtime version mismatch
In-Reply-To: <AANLkTi=fXunvma+98-gb=j=qy5p+xTiUZKXLa+kcWtwM@mail.gmail.com>
References: <AANLkTim-JHKKyhj4Yg+-YKdAfcz2eFGGe7RunYjhVdCf@mail.gmail.com>
	<boris.20110303215434@codesynthesis.com>
	<AANLkTi=fXunvma+98-gb=j=qy5p+xTiUZKXLa+kcWtwM@mail.gmail.com>
Message-ID: <boris.20110304172423@codesynthesis.com>

Hi William,

William Canton <stargate625@gmail.com> writes:

> However, there is another problem though.  The CodeSynthesis doesn't seem to
> be able to parse a valid XML document with optional field elements.

Your XML is invalid per your schema. EISSDescriptorType has the following
definition:

<xs:complexType name="EISSDescriptorType">
  <xs:sequence>
    <xs:element name="Data" type="DataType"/>
    <xs:element name="BIFPlatform" type="BIFPlatformType" minOccurs="0" maxOccurs="unbounded"/>
  </xs:sequence>
  ...
</xs:complexType>

Which means that the Data element must be present in every EISSDescriptor
element.

I think you are confusing minOccurs="0" with empty content.

When you see something like this:

<EISS>
  <!-- No mentioning of EISSDescriptor -->
  ...
</EISS>

Then the minOccurs="0" is at play here. However, if you see something
like this:

<EISS>
  <EISSDescriptor/>
  ...
</EISS>

Then the element is present and the empty content must be allowed by the
element's type (EISSDescriptorType in this case). EISSDescriptorType clearly
does not allow empty content.

Boris

From stargate625 at gmail.com  Fri Mar  4 12:39:19 2011
From: stargate625 at gmail.com (William Canton)
Date: Sat Mar  5 04:52:10 2011
Subject: [xsd-users] Error in building the generated header due to XSD
	runtime version mismatch
In-Reply-To: <boris.20110304172423@codesynthesis.com>
References: <AANLkTim-JHKKyhj4Yg+-YKdAfcz2eFGGe7RunYjhVdCf@mail.gmail.com>
	<boris.20110303215434@codesynthesis.com>
	<AANLkTi=fXunvma+98-gb=j=qy5p+xTiUZKXLa+kcWtwM@mail.gmail.com>
	<boris.20110304172423@codesynthesis.com>
Message-ID: <AANLkTi=ZC=6SCfg6uz2-rB7WcxxjAUBZ5F74oSomKAuP@mail.gmail.com>

Hi Boris,

Thank you for pointing out the required Data element within the
EISSDescriptorType element.  Indeed, the Data element is required to be
non-empty once presented.

The XML document was generated from a tool and I just assumed that it was
created properly.  I tested the same document on Java platform and didn't
see any problem.  Java XML parser seems more lenient than the CodeSynthesis
parser in this regard.

William


On Fri, Mar 4, 2011 at 10:31 AM, Boris Kolpackov <boris@codesynthesis.com>wrote:

> Hi William,
>
> William Canton <stargate625@gmail.com> writes:
>
> > However, there is another problem though.  The CodeSynthesis doesn't seem
> to
> > be able to parse a valid XML document with optional field elements.
>
> Your XML is invalid per your schema. EISSDescriptorType has the following
> definition:
>
> <xs:complexType name="EISSDescriptorType">
>  <xs:sequence>
>    <xs:element name="Data" type="DataType"/>
>    <xs:element name="BIFPlatform" type="BIFPlatformType" minOccurs="0"
> maxOccurs="unbounded"/>
>  </xs:sequence>
>  ...
> </xs:complexType>
>
> Which means that the Data element must be present in every EISSDescriptor
> element.
>
> I think you are confusing minOccurs="0" with empty content.
>
> When you see something like this:
>
> <EISS>
>  <!-- No mentioning of EISSDescriptor -->
>  ...
> </EISS>
>
> Then the minOccurs="0" is at play here. However, if you see something
> like this:
>
> <EISS>
>  <EISSDescriptor/>
>  ...
> </EISS>
>
> Then the element is present and the empty content must be allowed by the
> element's type (EISSDescriptorType in this case). EISSDescriptorType
> clearly
> does not allow empty content.
>
> Boris
>
From jclements at comcast.net  Sat Mar  5 22:14:03 2011
From: jclements at comcast.net (Jeff Clements)
Date: Sun Mar  6 08:02:19 2011
Subject: [xsd-users] C++/Tree parsing: expected element exception query
Message-ID: <000001cbdbac$8ce496c0$a6adc440$@net>

 

When attempting to parse an XML document using a C++/Tree generated parser,
I get an Expected Element exception that looks like this:

 

Exception: expected element my.name.space#minimized

 

The "my.name.space"  namespace, I believe, is properly formatted (as I get
an unexpected exception if I modify it in any way), so I'm wondering what
other checks are being performed that could cause the exception to be
thrown?  Also, what is the significance of the "minimized" message?  

 

Thanks in advance for any advice.

Jeff

 

From alessandro.carpi at gmail.com  Sun Mar  6 10:18:21 2011
From: alessandro.carpi at gmail.com (Alessandro Carpi)
Date: Sun Mar  6 10:18:48 2011
Subject: [xsd-users] [C++/Parser] Passing arguments to the document parser
Message-ID: <AANLkTikvUngkziMvrOvR6RV7HhHwOgm4hibNFVFf_-ZB@mail.gmail.com>

Hello people,

I am using C++/Parser and I have a simple question: is there a way to pass a
specific argument to the document parser, it will be visible from my
callbacks ?

I have a data structure that is filled outside the code implementing my
parser, but I need to use such structure in order to verify if informations
read from the XML are consistent with what I should expect.

For instance, such data structure contains a list of possibile names. When
an attribute "name" was parsed I must check if the value is present into
this names' list.

Thank you very much,
-- 
--
Alessandro Carpi
From boris at codesynthesis.com  Mon Mar  7 07:56:13 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Mon Mar  7 07:55:31 2011
Subject: [xsd-users] Error in building the generated header due to XSD
	runtime version mismatch
In-Reply-To: <AANLkTi=ZC=6SCfg6uz2-rB7WcxxjAUBZ5F74oSomKAuP@mail.gmail.com>
References: <AANLkTim-JHKKyhj4Yg+-YKdAfcz2eFGGe7RunYjhVdCf@mail.gmail.com>
	<boris.20110303215434@codesynthesis.com>
	<AANLkTi=fXunvma+98-gb=j=qy5p+xTiUZKXLa+kcWtwM@mail.gmail.com>
	<boris.20110304172423@codesynthesis.com>
	<AANLkTi=ZC=6SCfg6uz2-rB7WcxxjAUBZ5F74oSomKAuP@mail.gmail.com>
Message-ID: <boris.20110307145307@codesynthesis.com>

Hi William,

William Canton <stargate625@gmail.com> writes:

> The XML document was generated from a tool and I just assumed that it was
> created properly. I tested the same document on Java platform and didn't
> see any problem. Java XML parser seems more lenient than the CodeSynthesis
> parser in this regard.

This is a very basic validation rule and I would be very surprised if
any well-known Java parser ignored it. Something else most likely is
going on (like the Java parser cannot find the schema and disables
validation altogether because of that; this is the default behavior
in Xerces-J, for example).

Boris

From boris at codesynthesis.com  Mon Mar  7 08:23:33 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Mon Mar  7 08:22:52 2011
Subject: [xsd-users] shared objects that load compiled schemas from a
	shared object
In-Reply-To: <E7AA20464DFA774A9ACC4B8EC01CA08105BB586852@SMD-EX01.proteus-technologies.com>
References: <E7AA20464DFA774A9ACC4B8EC01CA08105BB586852@SMD-EX01.proteus-technologies.com>
Message-ID: <boris.20110307151950@codesynthesis.com>

Hi Ray,

Ray Lischner <rlischner@proteuseng.com> writes:

> When we load A.so and B.so into an application, the substitutionGroups 
> don't work. That is, a dynamic_cast<Type>() of an object of type Type
> returns a null pointer.

Found this seemingly relevant discussion:

http://stackoverflow.com/questions/2351786/dynamic-cast-fails-when-used-with-dlopen-dlsym

Boris

From boris at codesynthesis.com  Mon Mar  7 08:32:55 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Mon Mar  7 08:32:13 2011
Subject: [xsd-users] C++/Tree parsing: expected element exception query
In-Reply-To: <000001cbdbac$8ce496c0$a6adc440$@net>
References: <000001cbdbac$8ce496c0$a6adc440$@net>
Message-ID: <boris.20110307152829@codesynthesis.com>

Hi Jeff,

Jeff Clements <jclements@comcast.net> writes:

> Exception: expected element my.name.space#minimized

The "my.name.space" is the namespace and "minimized" is the name of the
expected root document element. When you call one of the parsing functions,
it expects to find the corresponding element as the root document element.
If the element namespace/name does not match, you get this exception.

If you want to handle documents with varying root elements, check the
'multiroot' example in the XSD distribution.

Boris

From boris at codesynthesis.com  Mon Mar  7 08:37:15 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Mon Mar  7 08:36:33 2011
Subject: [xsd-users] [C++/Parser] Passing arguments to the document parser
In-Reply-To: <AANLkTikvUngkziMvrOvR6RV7HhHwOgm4hibNFVFf_-ZB@mail.gmail.com>
References: <AANLkTikvUngkziMvrOvR6RV7HhHwOgm4hibNFVFf_-ZB@mail.gmail.com>
Message-ID: <boris.20110307153303@codesynthesis.com>

Hi Alessandro,

Alessandro Carpi <alessandro.carpi@gmail.com> writes:

> I am using C++/Parser and I have a simple question: is there a way to pass a
> specific argument to the document parser, it will be visible from my
> callbacks ?

No, but you can pass anything you want to the parser implementation's
constructors.


> I have a data structure that is filled outside the code implementing my
> parser, but I need to use such structure in order to verify if informations
> read from the XML are consistent with what I should expect.
> 
> For instance, such data structure contains a list of possibile names. When
> an attribute "name" was parsed I must check if the value is present into
> this names' list.

You could pass/store a reference to this data structure in the parser
implementation that handles the "name" attribute. This way you can change
the list of names between the parse sessions if you would like to reuse
the same set of parsers to parse multiple documents.

Boris

From js.homer at yahoo.com  Thu Mar 17 19:55:02 2011
From: js.homer at yahoo.com (Homer J S)
Date: Thu Mar 17 19:55:11 2011
Subject: [xsd-users] How to get around invalid characters in UTF-8 string
Message-ID: <399415.65185.qm@web45604.mail.sp1.yahoo.com>


Hello everyone,
When parsing an UTF-8 encode xml file I got the following error from the parser:

"An exception occurred! Type:UTFDataFormatException, Message:Exceede bytes 
limits , 6-byte sequence"

I believe the reason is due to a string that contain invalid byte sequence for 
UTF-8 encoding. I am on the receiver side of this xml message and cannot control 
what's in it.

Is there a way to to get the parser to bypass those characters, strip them out, 
or replace them with something else?

Many thanks,

JS


      
From fschmidt at techfak.uni-bielefeld.de  Fri Mar 18 05:46:42 2011
From: fschmidt at techfak.uni-bielefeld.de (Florian Paul Schmidt)
Date: Fri Mar 18 05:47:03 2011
Subject: [xsd-users] How to get around invalid characters in UTF-8 string
In-Reply-To: <399415.65185.qm@web45604.mail.sp1.yahoo.com>
References: <399415.65185.qm@web45604.mail.sp1.yahoo.com>
Message-ID: <4D832A02.8060009@techfak.uni-bielefeld.de>

On 03/18/2011 12:55 AM, Homer J S wrote:
> Hello everyone,
> When parsing an UTF-8 encode xml file I got the following error from the parser:
>
> "An exception occurred! Type:UTFDataFormatException, Message:Exceede bytes
> limits , 6-byte sequence"
>
> I believe the reason is due to a string that contain invalid byte sequence for
> UTF-8 encoding. I am on the receiver side of this xml message and cannot control
> what's in it.
>
> Is there a way to to get the parser to bypass those characters, strip them out,
> or replace them with something else?
>
> Many thanks,
>
> JS

This doesn't sound like a job for the parser really. Aren't there nice 
usable unicode libraries that you can use to filter the input before 
passing it to the parser?

Flo

>
>

From mikhail.evstiounin at gmail.com  Thu Mar 17 19:28:57 2011
From: mikhail.evstiounin at gmail.com (Mikhail Evstiounin)
Date: Fri Mar 18 09:33:21 2011
Subject: [xsd-users] XSD C++ code generation error
Message-ID: <AANLkTin-bRq-e+GtF6=z1TJ_BTw5XDVP5Pdh-WF25mV0@mail.gmail.com>

Hi,

XSD generated the following code:

hxx file:

namespace xml_schema

{

?

?

  typedef ::xsd::cxx::tree::type type;

?

namespace up

{

?

  class results: public ::xml_schema::type

  {

?

    static const success_type success_default_value_;

?

  };

?

}

?

}







cxx file:

namespace xml_schema

{

?

namespace up

{

?

    typedef ::xml_schema::type success_type;

?

  const results::success_type results::success_default_value_( );

?

}

?

}



It produces the following error (gcc 4.1.1):





document-envelope.cxx:2882: error: no ?const xml_schema::type
up::results::success_default_value_()? member function declared in class
'up::results'

document-envelope.cxx:2882: error: declaration of ?const xml_schema::type
up::results::success_default_value_()? outside of class is not definition



If you change it  a little bit:

namespace xml_schema

{

?

namespace up

{

?

  const results::success_type results::success_default_value_;// ( );

?

}

?

}

It compiles normally.


Since there is no default value this type of declaration (of a class static
member) is considered as a function definition. Per many C++ name resolution
discussions, rationale is made to lean to a "function definition", not a
"variable declaration" -- too many code rely on it.

Proposal -- to fix XSD in part of static member declaration not to use "()"
if default c-tor is used.

Truly yours
Mikhail Evstiounin
From boris at codesynthesis.com  Fri Mar 18 09:42:51 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Fri Mar 18 09:43:20 2011
Subject: [xsd-users] How to get around invalid characters in UTF-8 string
In-Reply-To: <399415.65185.qm@web45604.mail.sp1.yahoo.com>
References: <399415.65185.qm@web45604.mail.sp1.yahoo.com>
Message-ID: <boris.20110318153329@codesynthesis.com>

Hi,

Homer J S <js.homer@yahoo.com> writes:

> Is there a way to to get the parser to bypass those characters, strip them 
> out, or replace them with something else?

There is no out of the box support for this. And I agree with Florian
that this is something that is better to handle before XML parsing since
"bypassing", "stripping", and "replacing" can be very application-
specific. Also note that such stripping can render the resulting XML
malformed (e.g., by removing '<' from a closing tag).

The best way to do this would be to filter the input by providing a
custom input stream (e.g., an implementation of std::istream or
xercesc::InputSource; the latter is probably easier). In this
implementation you can either use some existing library or validate
and "correct" UTF-8 yourself. You can base this on the 'compression'
example from the XSD distribution which uses this technique to inflate
compressed XML on the fly.

Boris

From boris at codesynthesis.com  Fri Mar 18 10:20:10 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Fri Mar 18 10:20:39 2011
Subject: [xsd-users] XSD C++ code generation error
In-Reply-To: <AANLkTin-bRq-e+GtF6=z1TJ_BTw5XDVP5Pdh-WF25mV0@mail.gmail.com>
References: <AANLkTin-bRq-e+GtF6=z1TJ_BTw5XDVP5Pdh-WF25mV0@mail.gmail.com>
Message-ID: <boris.20110318161835@codesynthesis.com>

Hi Mikhail,

Mikhail Evstiounin <mikhail.evstiounin@gmail.com> writes:

> document-envelope.cxx:2882: error: no ?const xml_schema::type
> up::results::success_default_value_()? member function declared in class
> 'up::results'

Fixed. Thanks for reporting this! If you would like, I can build you
a pre-release binary (just let me know which platform you need).

Boris

From mikhail.evstiounin at gmail.com  Fri Mar 18 10:31:14 2011
From: mikhail.evstiounin at gmail.com (Mikhail Evstiounin)
Date: Fri Mar 18 11:04:53 2011
Subject: [xsd-users] XSD C++ code generation error
In-Reply-To: <boris.20110318161835@codesynthesis.com>
References: <AANLkTin-bRq-e+GtF6=z1TJ_BTw5XDVP5Pdh-WF25mV0@mail.gmail.com>
	<boris.20110318161835@codesynthesis.com>
Message-ID: <AANLkTik-pJn2vtKpB42G1MHiMWPVT_3wUhSS1QqXV2Mn@mail.gmail.com>

Hi Boris,

Thanks!

If you can build pre-realese binary tehn I would appreciate it.

/login/evstiomv/tests/xsd-3.3.0-i686-linux-gnu/ -- I think, it should be
enough information. :)

On Fri, Mar 18, 2011 at 10:20 AM, Boris Kolpackov
<boris@codesynthesis.com>wrote:

> Hi Mikhail,
>
> Mikhail Evstiounin <mikhail.evstiounin@gmail.com> writes:
>
> > document-envelope.cxx:2882: error: no ?const xml_schema::type
> > up::results::success_default_value_()? member function declared in class
> > 'up::results'
>
> Fixed. Thanks for reporting this! If you would like, I can build you
> a pre-release binary (just let me know which platform you need).
>
> Boris
>
From boris at codesynthesis.com  Fri Mar 18 11:05:24 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Fri Mar 18 11:05:55 2011
Subject: [xsd-users] XSD C++ code generation error
In-Reply-To: <AANLkTik-pJn2vtKpB42G1MHiMWPVT_3wUhSS1QqXV2Mn@mail.gmail.com>
References: <AANLkTin-bRq-e+GtF6=z1TJ_BTw5XDVP5Pdh-WF25mV0@mail.gmail.com>
	<boris.20110318161835@codesynthesis.com>
	<AANLkTik-pJn2vtKpB42G1MHiMWPVT_3wUhSS1QqXV2Mn@mail.gmail.com>
Message-ID: <boris.20110318170443@codesynthesis.com>

Mikhail Evstiounin <mikhail.evstiounin@gmail.com> writes:

> If you can build pre-realese binary tehn I would appreciate it.

Here you go:

http://www.codesynthesis.com/~boris/tmp/xsd-4.0.0.a4-i686-linux-gnu.tar.bz2

Boris

From florin.hillebrand at streamunlimited.com  Mon Mar 21 11:19:00 2011
From: florin.hillebrand at streamunlimited.com (Florin Hillebrand)
Date: Mon Mar 21 11:19:07 2011
Subject: [xsd-users] Warning while using maxUnsignedInt with c++98 standard
	with xsd3.3
Message-ID: <1300720740.14135.57.camel@flozdeskub>

Hi all,

first my environment:

- ubuntu 10.04 on x86 32bit
- xsd3.3
- g++-4.4 with -stdc=c++98 -Wall compiler options.

I have an xsd which generates me a element with an attribute of type
xsd:unsignedInt with a default value. The default value has to be the
maximal unsigned int value (4294967295).

----------
<xsd:element name="Element">
  <xsd:complexType>
      <xsd:attribute name="attr" use="optional" type="xsd:unsignedInt" default="4294967295" />
  </xsd:complexType>
</xsd:element>
----------

When xsd 3.2 generated the cxx-tree i got back a constant and a method:

----------
const Element::attr_type Element::attr_default_value_ (
        ::xsd::cxx::tree::traits< ::xml_schema::unsigned_int, char >::create (
          ::std::string ("4294967295"), 0, 0, 0));


const Element::attr_type& Element::attr_default_value () {
        return attr_default_value_;
      }
---------

This works well without any compiler warnings, but when I updated xsd to
3.3 it generates me:

---------
const Element::attr_type& Element::attr_default_value () {
        return 4294967295;
      }
---------

This generates a g++ warning (Warnings treaded as errors):

------------
error: this decimal constant is unsigned only in ISO C90
------------

c++98 standard says:

---cite---- ISO/IEC 14882:1998(E)  page 42 point 2 ------
If it is decimal and has no suffix, it has
the first of these types in which its value can be represented: int,
long int;
-----------

So this won't work, but:

---cite---- ISO/IEC 14882:1998(E)  page 42 point 2 ------
If it is suffixed by u or U, its type is the first of these types in
which its value can be represented: unsigned int, unsigned long int;
-----------

This would work but i can't append an u or U to the default value,
because regex wouldn't pass.

I would appreciate any suggestions.

With kind regards,

Florin

-- 
The information contained in this message may be confidential and
legally protected under applicable law. The message is intended solely
for the addressee(s). If you are not the intended recipient, you are
hereby notified that any use, forwarding, dissemination, or reproduction
of this message is strictly prohibited and may be unlawful. If you are
not the intended recipient, please contact the sender by return e-mail
and destroy all copies of the original message.

From boris at codesynthesis.com  Mon Mar 21 11:55:04 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Mon Mar 21 11:55:55 2011
Subject: [xsd-users] Warning while using maxUnsignedInt with c++98
	standard with xsd3.3
In-Reply-To: <1300720740.14135.57.camel@flozdeskub>
References: <1300720740.14135.57.camel@flozdeskub>
Message-ID: <boris.20110321175208@codesynthesis.com>

Hi Florin,

Florin Hillebrand <florin.hillebrand@streamunlimited.com> writes:

> return 4294967295;
>
> This generates a g++ warning (Warnings treaded as errors):
>
> error: this decimal constant is unsigned only in ISO C90

I believe we have already fixed this issue for the upcoming release:

http://scm.codesynthesis.com/?p=xsd/xsd.git;a=commit;h=54c3d3ecf7abaac4c2239f44340f7f9dbdadfe3a

You can try this pre-release binary to see if it helps your case:

http://www.codesynthesis.com/~boris/tmp/xsd-4.0.0.a4-i686-linux-gnu.tar.bz2

Boris

From dv7777 at gmail.com  Tue Mar 22 10:44:37 2011
From: dv7777 at gmail.com (dv)
Date: Tue Mar 22 10:44:45 2011
Subject: [xsd-users] custom xml parser?
Message-ID: <4D88B5D5.8070602@gmail.com>

Hello,

I've been looking at the XSD code, and am I right to assume that adding 
support for an XML parser is a non-trivial task? I see code for Expat 
and Xerces, and it is scattered all over the place. I am trying to use a 
custom parser, so far I don't see how. Would this be easier with XSD/e ?

From boris at codesynthesis.com  Tue Mar 22 13:31:36 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Tue Mar 22 13:32:31 2011
Subject: [xsd-users] custom xml parser?
In-Reply-To: <4D88B5D5.8070602@gmail.com>
References: <4D88B5D5.8070602@gmail.com>
Message-ID: <boris.20110322193038@codesynthesis.com>

Hi,

dv <dv7777@gmail.com> writes:

> I've been looking at the XSD code, and am I right to assume that adding  
> support for an XML parser is a non-trivial task? I see code for Expat  
> and Xerces, and it is scattered all over the place. I am trying to use a  
> custom parser, so far I don't see how. Would this be easier with XSD/e ?

Which mapping would you like to use, C++/Parser or C++/Tree?

Boris

From dv7777 at gmail.com  Wed Mar 23 06:45:15 2011
From: dv7777 at gmail.com (dv)
Date: Wed Mar 23 06:45:28 2011
Subject: [xsd-users] custom xml parser?
In-Reply-To: <boris.20110322193038@codesynthesis.com>
References: <4D88B5D5.8070602@gmail.com>
	<boris.20110322193038@codesynthesis.com>
Message-ID: <4D89CF3B.7090100@gmail.com>

On 03/22/2011 06:31 PM, Boris Kolpackov wrote:
> Hi,
>
> dv<dv7777@gmail.com>  writes:
>
>> I've been looking at the XSD code, and am I right to assume that adding
>> support for an XML parser is a non-trivial task? I see code for Expat
>> and Xerces, and it is scattered all over the place. I am trying to use a
>> custom parser, so far I don't see how. Would this be easier with XSD/e ?
>
> Which mapping would you like to use, C++/Parser or C++/Tree?
>
> Boris
>

I would like to use the C++/Tree mapping.

From boris at codesynthesis.com  Wed Mar 23 09:25:59 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Wed Mar 23 09:26:59 2011
Subject: [xsd-users] custom xml parser?
In-Reply-To: <4D89CF3B.7090100@gmail.com>
References: <4D88B5D5.8070602@gmail.com>
	<boris.20110322193038@codesynthesis.com>
	<4D89CF3B.7090100@gmail.com>
Message-ID: <boris.20110323151456@codesynthesis.com>

Hi,

dv <dv7777@gmail.com> writes:

> I would like to use the C++/Tree mapping.

C++/Tree is integrated pretty tightly with Xerces-C++ (there are a
number of benefits to such tight integration, including representation
of wildcard/mixed content as DOM fragments, XPath support, etc). Plus,
C++/Tree relies on the underlying XML parser for XML Schema validation
so if your parser does not support it, then you will only be able to
parse in non-validating mode.

On the other hand, the C++/Hybrid mapping in XSD/e, which is similar
to C++/Tree, doesn't know anything about the underlying XML parser.
It is all handled by C++/Parser which in turn uses customized Expat
parser and performs XML Schema validation in the generated code. It
should be quite easy to customize the XSD/e runtime to use a parser
other than Expat. Pretty much all teh parser-specific logic is in
the libxsde/xsde/cxx/parser/expat/ directory with just a few
references from the rest of the C++/Parser runtime. This would be
very easy to change. If you would like to give it try and need
help, feel free to ask on the xsde-users mailing list.

Boris

From stargate625 at gmail.com  Mon Mar 28 11:43:19 2011
From: stargate625 at gmail.com (William Canton)
Date: Mon Mar 28 11:56:34 2011
Subject: [xsd-users] Compiler errors occurred in the generated *.hxx files
 from CodeSynthesis 3.2
Message-ID: <AANLkTik9oKVeP4d5pM=7DPEfn8xLuGNm7MocvLfDjR_x@mail.gmail.com>

I am not sure what was causing the following compiler errors in the
generated header file:

1>c:\test\generated\codf.hxx(391) : error C2146: syntax error : missing ';'
before identifier 'FileSystem_type'
1>c:\test\generated\codf.hxx(391) : error C2838: 'FileSystemType' : illegal
qualified name in member declaration
1>c:\test\generated\codf.hxx(391) : error C4430: missing type specifier -
int assumed. Note: C++ does not support default-int
1>c:\test\generated\codf.hxx(392) : error C2065: 'FileSystem_type' :
undeclared identifier
1>c:\test\generated\codf.hxx(393) : error C2065: 'FileSystem_type' :
undeclared identifier
1>c:\test\codf.hxx(402) : error C4430: missing type specifier - int assumed.
Note: C++ does not support default-int
1>c:\test\codf.hxx(402) : error C2143: syntax error : missing ',' before '&'
1>c:\test\generated\codf.hxx(408) : error C2065: 'FileSystem_type' :
undeclared identifier
1>c:\test\generated\codf.hxx(532) : error C2079: 'CarouselType::FileSystem_'
uses undefined class 'xsd::cxx::tree::optional'
1>c:\test\generated\codf.hxx(6546) : error C4430: missing type specifier -
int assumed. Note: C++ does not support default-int
1>c:\test\generated\codf.hxx(6546) : error C2143: syntax error : missing ','
before '&'
1>MyApp.cpp

The codf.hxx was generated by the CodeSynthesis after compiling the XML
Schema file codf.xsd.  But when I included this header file in my MyApp.cpp
file, the build failed with various mysterious syntax errors while
processing the codf.hxx header.  When I opened the codf.hxx file, it has a
snippet of content as follows surrounding the location of the reported
errors:

386  void
387  OCConfiguration (const OCConfiguration_sequence& s);
388
389  // FileSystem
390  //
391  typedef ::FileSystemType FileSystem_type;
392  typedef ::xsd::cxx::tree::optional< FileSystem_type >
FileSystem_optional;
393  typedef ::xsd::cxx::tree::traits< FileSystem_type, char >
FileSystem_traits;

It's unclear to me why my Visual Studio C++ 2008 reported an error on line
number 391.  The was the first build error detected.  Before this line,
there is a function on line 387 returning a "void" type.  This could have
caused the error.  It seemed like the compiler somehow could not even
recognize the keyword "void."

Any idea what could have caused these strange compiler's behavior?

William
From boris at codesynthesis.com  Mon Mar 28 12:02:11 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Mon Mar 28 12:03:44 2011
Subject: [xsd-users] Compiler errors occurred in the generated *.hxx
	files from CodeSynthesis 3.2
In-Reply-To: <AANLkTik9oKVeP4d5pM=7DPEfn8xLuGNm7MocvLfDjR_x@mail.gmail.com>
References: <AANLkTik9oKVeP4d5pM=7DPEfn8xLuGNm7MocvLfDjR_x@mail.gmail.com>
Message-ID: <boris.20110328175611@codesynthesis.com>

Hi William,

William Canton <stargate625@gmail.com> writes:

> 1>c:\test\generated\codf.hxx(391) : error C2146: syntax error : missing ';'
> before identifier 'FileSystem_type'
>
> [...]
>
> 391  typedef ::FileSystemType FileSystem_type;

Can you check that the FileSystemType class is actually defined in the
generated codf.hxx header?

One thing that might be causing this is the name clash with some existing
function called FileSystemType (possibly defined by one of the system
headers). You can try to tell the XSD compiler that it is a reserved
name with the --reserved-name option:

--reserved-name FileSystemType

Better yet, you may want to place the generated code into a namespace
instead of the global scope. This can be achieved with the --namespace-map
options:

--namespace-map =codf

Boris

From stargate625 at gmail.com  Mon Mar 28 12:10:07 2011
From: stargate625 at gmail.com (William Canton)
Date: Mon Mar 28 12:14:55 2011
Subject: [xsd-users] Compiler errors occurred in the generated *.hxx files
	from CodeSynthesis 3.2
In-Reply-To: <boris.20110328175611@codesynthesis.com>
References: <AANLkTik9oKVeP4d5pM=7DPEfn8xLuGNm7MocvLfDjR_x@mail.gmail.com>
	<boris.20110328175611@codesynthesis.com>
Message-ID: <AANLkTi=C3pgVB1=vYAo243V3CzGeZ9-t-5XC3beoNB8G@mail.gmail.com>

The FileSystemType is defined by CodeSynthesis in the same generated header
codf.hxx on line 5028.  There is also a forward declaration of this type on
line 285 in the same header.


On Mon, Mar 28, 2011 at 12:02 PM, Boris Kolpackov
<boris@codesynthesis.com>wrote:

> Hi William,
>
> William Canton <stargate625@gmail.com> writes:
>
> > 1>c:\test\generated\codf.hxx(391) : error C2146: syntax error : missing
> ';'
> > before identifier 'FileSystem_type'
> >
> > [...]
> >
> > 391  typedef ::FileSystemType FileSystem_type;
>
> Can you check that the FileSystemType class is actually defined in the
> generated codf.hxx header?
>
> One thing that might be causing this is the name clash with some existing
> function called FileSystemType (possibly defined by one of the system
> headers). You can try to tell the XSD compiler that it is a reserved
> name with the --reserved-name option:
>
> --reserved-name FileSystemType
>
> Better yet, you may want to place the generated code into a namespace
> instead of the global scope. This can be achieved with the --namespace-map
> options:
>
> --namespace-map =codf
>
> Boris
>
From stargate625 at gmail.com  Mon Mar 28 12:17:01 2011
From: stargate625 at gmail.com (William Canton)
Date: Mon Mar 28 12:21:55 2011
Subject: [xsd-users] Compiler errors occurred in the generated *.hxx files
	from CodeSynthesis 3.2
In-Reply-To: <AANLkTi=C3pgVB1=vYAo243V3CzGeZ9-t-5XC3beoNB8G@mail.gmail.com>
References: <AANLkTik9oKVeP4d5pM=7DPEfn8xLuGNm7MocvLfDjR_x@mail.gmail.com>
	<boris.20110328175611@codesynthesis.com>
	<AANLkTi=C3pgVB1=vYAo243V3CzGeZ9-t-5XC3beoNB8G@mail.gmail.com>
Message-ID: <AANLkTik23K4NxVJp02fpzSbMkMahp0G0Tz8G=-3uVZ=4@mail.gmail.com>

Hi Boris,

After adding the codf as namespace as you suggested, it created another
problem:  CodeSynthesis never generated any headers or cpp files from the
schema.

William

On Mon, Mar 28, 2011 at 12:10 PM, William Canton <stargate625@gmail.com>wrote:

> The FileSystemType is defined by CodeSynthesis in the same generated header
> codf.hxx on line 5028.  There is also a forward declaration of this type on
> line 285 in the same header.
>
>
> On Mon, Mar 28, 2011 at 12:02 PM, Boris Kolpackov <boris@codesynthesis.com
> > wrote:
>
>> Hi William,
>>
>> William Canton <stargate625@gmail.com> writes:
>>
>> > 1>c:\test\generated\codf.hxx(391) : error C2146: syntax error : missing
>> ';'
>> > before identifier 'FileSystem_type'
>> >
>> > [...]
>> >
>> > 391  typedef ::FileSystemType FileSystem_type;
>>
>> Can you check that the FileSystemType class is actually defined in the
>> generated codf.hxx header?
>>
>> One thing that might be causing this is the name clash with some existing
>> function called FileSystemType (possibly defined by one of the system
>> headers). You can try to tell the XSD compiler that it is a reserved
>> name with the --reserved-name option:
>>
>> --reserved-name FileSystemType
>>
>> Better yet, you may want to place the generated code into a namespace
>> instead of the global scope. This can be achieved with the --namespace-map
>> options:
>>
>> --namespace-map =codf
>>
>> Boris
>>
>
>
From stargate625 at gmail.com  Mon Mar 28 12:27:30 2011
From: stargate625 at gmail.com (William Canton)
Date: Mon Mar 28 12:32:59 2011
Subject: [xsd-users] Compiler errors occurred in the generated *.hxx files
	from CodeSynthesis 3.2
In-Reply-To: <AANLkTik23K4NxVJp02fpzSbMkMahp0G0Tz8G=-3uVZ=4@mail.gmail.com>
References: <AANLkTik9oKVeP4d5pM=7DPEfn8xLuGNm7MocvLfDjR_x@mail.gmail.com>
	<boris.20110328175611@codesynthesis.com>
	<AANLkTi=C3pgVB1=vYAo243V3CzGeZ9-t-5XC3beoNB8G@mail.gmail.com>
	<AANLkTik23K4NxVJp02fpzSbMkMahp0G0Tz8G=-3uVZ=4@mail.gmail.com>
Message-ID: <AANLkTim7ZSMeSjC50a5a4NhLAd09eivAnMAQUvg6rLRM@mail.gmail.com>

Oops! I didn't use the correct syntax last time in the --namespace-Map
option.  I should have used:

--namespace-map xml_schema=codf

This compiled.  However in the end, the same compiler errors kept popping
up.


On Mon, Mar 28, 2011 at 12:17 PM, William Canton <stargate625@gmail.com>wrote:

>
> Hi Boris,
>
> After adding the codf as namespace as you suggested, it created another
> problem:  CodeSynthesis never generated any headers or cpp files from the
> schema.
>
> William
>
>
> On Mon, Mar 28, 2011 at 12:10 PM, William Canton <stargate625@gmail.com>wrote:
>
>> The FileSystemType is defined by CodeSynthesis in the same generated
>> header codf.hxx on line 5028.  There is also a forward declaration of this
>> type on line 285 in the same header.
>>
>>
>> On Mon, Mar 28, 2011 at 12:02 PM, Boris Kolpackov <
>> boris@codesynthesis.com> wrote:
>>
>>> Hi William,
>>>
>>> William Canton <stargate625@gmail.com> writes:
>>>
>>> > 1>c:\test\generated\codf.hxx(391) : error C2146: syntax error : missing
>>> ';'
>>> > before identifier 'FileSystem_type'
>>> >
>>> > [...]
>>> >
>>> > 391  typedef ::FileSystemType FileSystem_type;
>>>
>>> Can you check that the FileSystemType class is actually defined in the
>>> generated codf.hxx header?
>>>
>>> One thing that might be causing this is the name clash with some existing
>>> function called FileSystemType (possibly defined by one of the system
>>> headers). You can try to tell the XSD compiler that it is a reserved
>>> name with the --reserved-name option:
>>>
>>> --reserved-name FileSystemType
>>>
>>> Better yet, you may want to place the generated code into a namespace
>>> instead of the global scope. This can be achieved with the
>>> --namespace-map
>>> options:
>>>
>>> --namespace-map =codf
>>>
>>> Boris
>>>
>>
>>
>
From boris at codesynthesis.com  Mon Mar 28 12:35:28 2011
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Mon Mar 28 12:37:00 2011
Subject: [xsd-users] Compiler errors occurred in the generated *.hxx
	files from CodeSynthesis 3.2
In-Reply-To: <AANLkTim7ZSMeSjC50a5a4NhLAd09eivAnMAQUvg6rLRM@mail.gmail.com>
References: <AANLkTik9oKVeP4d5pM=7DPEfn8xLuGNm7MocvLfDjR_x@mail.gmail.com>
	<boris.20110328175611@codesynthesis.com>
	<AANLkTi=C3pgVB1=vYAo243V3CzGeZ9-t-5XC3beoNB8G@mail.gmail.com>
	<AANLkTik23K4NxVJp02fpzSbMkMahp0G0Tz8G=-3uVZ=4@mail.gmail.com>
	<AANLkTim7ZSMeSjC50a5a4NhLAd09eivAnMAQUvg6rLRM@mail.gmail.com>
Message-ID: <boris.20110328183138@codesynthesis.com>

Hi William,

William Canton <stargate625@gmail.com> writes:

> Oops! I didn't use the correct syntax last time in the --namespace-Map
> option.  I should have used:
> 
> --namespace-map xml_schema=codf

I am not so sure. This would be correct if your schema had this in the
root <xs:schema> element:

targetNamespace="xml_schema"

Looking at your original code fragment, I don't think you have this.
If your schema doesn't have the targetNamespace attribute, then the
correct option is:

--namespace-map =codf

That is, it maps empty namespace (global scope) to codf.

If you still cannot figure this out, please send me your schema (off
list) and I will take a look.

Boris