From yartsev_mercdev at yahoo.com  Mon Sep  8 04:54:11 2008
From: yartsev_mercdev at yahoo.com (Vladimir Yartsev)
Date: Mon Sep  8 04:54:19 2008
Subject: [xsde-users] Support for iPhone platform
Message-ID: <324118.41000.qm@web57206.mail.re3.yahoo.com>

Are there statically-linked libraries of CodeSynthesis XSD/e for use with "iPhone SDK 2.0" available?

Thanks, Vladimir.




      ________________________________________________________
?? ??? ? Yahoo!? 
????????? ??????????? ? ??????????. Yahoo! ?????! http://ru.mail.yahoo.com

From boris at codesynthesis.com  Tue Sep  9 11:43:04 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Tue Sep  9 11:44:16 2008
Subject: [xsde-users] Support for iPhone platform
In-Reply-To: <324118.41000.qm@web57206.mail.re3.yahoo.com>
References: <324118.41000.qm@web57206.mail.re3.yahoo.com>
Message-ID: <20080909154304.GA4427@karelia>

Hi Vladimir,

Vladimir Yartsev <yartsev_mercdev@yahoo.com> writes:

> Are there statically-linked libraries of CodeSynthesis XSD/e for use
> with "iPhone SDK 2.0" available?

No, we normally don't provide pre-compiled XSD/e runtime library
mainly because there are many configuration parameters which would
require us to publish tens of variants just to cover the most common
cases.

I looked around and it should be fairly straightforward to build
the XSD/e runtime for iPhone using the SDK provided by Apple. I am
going to give it a try later today or tomorrow and will get back to
you with a sample config file and instructions.

Also, would you like the XSD/e compiler binary for Mac OS X?

Boris

From ntlkhuong at tma.com.vn  Wed Sep 10 04:48:31 2008
From: ntlkhuong at tma.com.vn (Khuong Nguyen Thi Lien)
Date: Wed Sep 10 04:50:47 2008
Subject: [xsde-users]Get the element name
In-Reply-To: <20080909154304.GA4427@karelia>
References: <324118.41000.qm@web57206.mail.re3.yahoo.com>
	<20080909154304.GA4427@karelia>
Message-ID: <006901c91322$0242fc50$828010ac@tma.com.vn>

Hi Boris, 

Could you show me how I can get the element-name in the schema file?


Thanks a lot.

Regards,
Khuong


From boris at codesynthesis.com  Wed Sep 10 05:08:31 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Wed Sep 10 05:09:44 2008
Subject: [xsde-users] Get the element name
Message-ID: <20080910090831.GC6428@karelia>

Hi Khuong,

Khuong Nguyen Thi Lien <ntlkhuong@tma.com.vn> writes:

> Could you show me how I can get the element-name in the schema file?

I am not sure what you mean by this. Can you explain what you are
trying to achieve in a bit more detail? Maybe also show a schema
and/or XML fragment.

Boris

From ntlkhuong at tma.com.vn  Wed Sep 10 05:33:08 2008
From: ntlkhuong at tma.com.vn (Khuong Nguyen Thi Lien)
Date: Wed Sep 10 05:35:19 2008
Subject: [xsde-users] Get the element name
In-Reply-To: <20080910090831.GC6428@karelia>
References: <20080910090831.GC6428@karelia>
Message-ID: <007001c91328$3c60db90$828010ac@tma.com.vn>

Hi Boris,

 

Sorry for the unclear question, I have the xml fragment below, when the
parsing engine reaches the <greeting> element in the greeting() callback
function, I would like to extract the name of the element right at this
function. In this case I would like the returned name should be "greeting". 

 

Thanks,

Khuong

struct hello_pimpl: hello_pskel

{

  virtual void

  greeting (const string& greeting)

  {

    greeting_ = greeting;

  }

 

  virtual void

  name (const string& name)

  {

            cout << greeting_ << ", " << name << "!" << endl;            

  }

 

private:

  string greeting_;

};

 

<hello xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

       xsi:noNamespaceSchemaLocation="hello.xsd">

 

  <greeting>Hello</greeting>

 

  <name>sun</name>

  <name>moon</name>

  <name>world</name>

 

</hello>

 

-----Original Message-----
From: Boris Kolpackov [mailto:boris@codesynthesis.com] 
Sent: Wednesday, September 10, 2008 4:09 PM
To: Khuong Nguyen Thi Lien
Cc: xsde-users@codesynthesis.com
Subject: Re: [xsde-users] Get the element name

 

Hi Khuong,

 

Khuong Nguyen Thi Lien <ntlkhuong@tma.com.vn> writes:

 

> Could you show me how I can get the element-name in the schema file?

 

I am not sure what you mean by this. Can you explain what you are

trying to achieve in a bit more detail? Maybe also show a schema

and/or XML fragment.

 

Boris

From RobertH at microsol.ie  Wed Sep 10 11:00:45 2008
From: RobertH at microsol.ie (Robert Hammond)
Date: Wed Sep 10 11:00:56 2008
Subject: [xsde-users] Using Type Mapping?
Message-ID: <974C3B6C9F0A4A43918BACCA33390AC237C6C32E07@mail.microsol.ie>

Hi, I am trying to work my way through the "people" example to understand how the Type maps work properly.

My assumption after reading the manual is that if I wanted to include my own type which was not in the schema file, that I could define it in a hxx file, include this and the parser would build the skeleton code for that type.
Have I got this right?
I notice in the example that the "gender" type is already specified in the Schema and then defined as an enum in the hxx file which is included from the map file.
I am very new to using xml and the Xsde Parser so I may have made a wrong assumption here.

Is there any way you could please clarify this?




>> from Manual

"The first line indicates that the generated code must include people.hxx in order to get the
definition for the gender type. The second line specifies that both argument and return types for
the gender XML Schema type should be the ::gender C++ enum (we use fully-qualified
C++ names to avoid name clashes)."

_______________________________________________________________________

This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to which they are addressed.

If you have received this email in error please notify the originator of the message. 
This footer also confirms that this email message has been scanned for the presence of computer viruses.

Any views expressed in this message are those of the individual sender, except where the sender specifies and with authority, states them to be the views of Microsol.

If you have received this mail in error you are requested to e-mail us (info@microsol.ie) or by telephoning Microsol on +353 14153700.

Please Visit our website at
http://www.microsol.com

_______________________________________________________________________

This email has been scanned for all viruses.
_______________________________________________________________________
From boris at codesynthesis.com  Thu Sep 11 04:51:21 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Thu Sep 11 04:52:52 2008
Subject: [xsde-users] Get the element name
In-Reply-To: <007001c91328$3c60db90$828010ac@tma.com.vn>
References: <20080910090831.GC6428@karelia>
	<007001c91328$3c60db90$828010ac@tma.com.vn>
Message-ID: <20080911085121.GB9675@karelia>

Hi Khuong,

Khuong Nguyen Thi Lien <ntlkhuong@tma.com.vn> writes:

> Sorry for the unclear question, I have the xml fragment below, when the
> parsing engine reaches the <greeting> element in the greeting() callback
> function, I would like to extract the name of the element right at this
> function. In this case I would like the returned name should be "greeting". 

My guess is that your schema is using substitution groups and element
names act as identifiers (i.e., different substituting elements use the
same type). If that's the case then I have already outlined the possible
ways to achieve this in the following post:

http://www.codesynthesis.com/pipermail/xsde-users/2008-July/000039.html

In your case the easiest way would be to override _start_element and
save the element name in a member variable. You can then access it
from the greeting function:

struct hello_pimpl: hello_pskel
{
  virtual void
  greeting (const string& greeting)
  {
    // element_name_ contains the element name. 
  }

  virtual void
  name (const string& name)
  {
  }

  virtual void
  _start_element (const xml_schema::ro_string& ns,
                  const xml_schema::ro_string& name
                  const char* type)
  {
     element_name_ = name;

     // Always call the base _start_element().
     //
     hello_pskel::_start_element (ns, name, type);
  }

private:
  string element_name_;
};


Boris

From ntlkhuong at tma.com.vn  Thu Sep 11 04:58:34 2008
From: ntlkhuong at tma.com.vn (Khuong Nguyen Thi Lien)
Date: Thu Sep 11 05:00:46 2008
Subject: [xsde-users] Get the element name
In-Reply-To: <20080911085121.GB9675@karelia>
References: <20080910090831.GC6428@karelia>
	<007001c91328$3c60db90$828010ac@tma.com.vn>
	<20080911085121.GB9675@karelia>
Message-ID: <005701c913ec$923caf70$828010ac@tma.com.vn>

Hi Boris , 

I got your idea.

Thanks you so much.

Khuong

-----Original Message-----
From: Boris Kolpackov [mailto:boris@codesynthesis.com] 
Sent: Thursday, September 11, 2008 3:51 PM
To: Khuong Nguyen Thi Lien
Cc: xsde-users@codesynthesis.com
Subject: Re: [xsde-users] Get the element name

Hi Khuong,

Khuong Nguyen Thi Lien <ntlkhuong@tma.com.vn> writes:

> Sorry for the unclear question, I have the xml fragment below, when the
> parsing engine reaches the <greeting> element in the greeting() callback
> function, I would like to extract the name of the element right at this
> function. In this case I would like the returned name should be
"greeting". 

My guess is that your schema is using substitution groups and element
names act as identifiers (i.e., different substituting elements use the
same type). If that's the case then I have already outlined the possible
ways to achieve this in the following post:

http://www.codesynthesis.com/pipermail/xsde-users/2008-July/000039.html

In your case the easiest way would be to override _start_element and
save the element name in a member variable. You can then access it
from the greeting function:

struct hello_pimpl: hello_pskel
{
  virtual void
  greeting (const string& greeting)
  {
    // element_name_ contains the element name. 
  }

  virtual void
  name (const string& name)
  {
  }

  virtual void
  _start_element (const xml_schema::ro_string& ns,
                  const xml_schema::ro_string& name
                  const char* type)
  {
     element_name_ = name;

     // Always call the base _start_element().
     //
     hello_pskel::_start_element (ns, name, type);
  }

private:
  string element_name_;
};


Boris

From boris at codesynthesis.com  Thu Sep 11 05:33:50 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Thu Sep 11 05:35:20 2008
Subject: [xsde-users] Using Type Mapping?
In-Reply-To: <974C3B6C9F0A4A43918BACCA33390AC237C6C32E07@mail.microsol.ie>
References: <974C3B6C9F0A4A43918BACCA33390AC237C6C32E07@mail.microsol.ie>
Message-ID: <20080911093350.GC9675@karelia>

Hi Robert,

Robert Hammond <RobertH@microsol.ie> writes:

> Hi, I am trying to work my way through the "people" example to understand
> how the Type maps work properly.
> 
> My assumption after reading the manual is that if I wanted to include my
> own type which was not in the schema file, that I could define it in a
> hxx file, include this and the parser would build the skeleton code for
> that type.
> Have I got this right?

Not exactly. For each type defined in XML Schema the XSD/e compiler
generates a parser skeleton. By default these parser skeletons don't
return anything from their post_*() functions.

You use type maps to tell the XSD/e compiler that a parser skeleton
corresponding to a particular schema type should return something.
Returning things from parser skeletons is useful when you need to
pass information from a "child" element/attribute parser to the
"parent" one. This is handy when, for example, you need to build
a custom in-memory representation of the data stored in XML.

Let's consider the gender type from the people example in the manual.
It is defined in XML Schema like so:

<xsd:simpleType name="gender">
  <xsd:restriction base="xsd:string">
    <xsd:enumeration value="male"/>
    <xsd:enumeration value="female"/>
  </xsd:restriction>
</xsd:simpleType>

If we compile this schema fragment without specifying any type
maps then we get the following parser skeleton:

class gender_pskel: public xml_schema::string_pskel
{
public:
  gender_pskel (xml_schema::string_pskel* base_impl);

  virtual void
  pre ();

  virtual void
  post_gender ();
};

As you can see, the post_gender() function returns void. This may be
perfectly sufficient if, for example, all you need to do is print the
gender value:

class gender_pimpl: public gender_pskel
{
public:
  gender_pimpl ()
    : gender_pskel (&base_impl_)
  {
  }

  virtual void
  post_gender ()
  {
    std::string s = post_string ();
    cout << "gender: " << s << endl;
  }

private:
  xml_schema::string_pimpl base_impl_;
};

But what if you cannot process the gender value in gender_pimpl and
instead needed to pass the gender value to the "parent" parser, the
one generated for the person type:

<xs:complexType name="person">
  <xs:sequence>
    <xs:element name="gender" type="gender"/>

    ...

  </xs:sequence>
</xs:complexType>

class person_pskel: public xml_schema::parser_complex_content
{
public:

  virtual void
  gender ();

  ...
};

One not very clean way would be to have a global variable that is set
in gender_pimpl and is examined in person_pimpl. Since passing data
from "child" parsers to "parent" parsers is a fairly common requirement,
XSD/e provides a mechanism for this. It allows you to specify, via
type maps, the return type of the "child" parser's post_*() function
(post_gender() in our case) and argument type of the "parent" parser's
callback (gender() in our case). The generated code is also responsible
for calling gender_pimpl::post_gender() and passing its return value
to person_pimpl::gender().

In our case we can create a C++ enum corresponding to the gender type
in XML Schema. We can place it, say, in gender.hxx:

enum gender
{
  male,
  female
};

We can then create a type map for our schema, people.map:

include "gender.hxx"
gender ::gender ::gender

In the second line the first identifier is of the schema type. The second
is the return type for gender_pimpl::post_gender() and the last one is the
argument type for the person_pimpl::gender() callback.

Once we compile our schema with this type map, the gender_pskel and
person_pskel skeletons change as follows:

#include "gender.hxx"

class gender_pskel: public xml_schema::string_pskel
{
public:
  gender_pskel (xml_schema::string_pskel* base_impl);

  virtual void
  pre ();

  virtual ::gender
  post_gender ();
};

class person_pskel: public xml_schema::parser_complex_content
{
public:

  virtual void
  gender (::gender);

  ...
};

And our implementation could look like this:

class gender_pimpl: public gender_pskel
{
public:
  gender_pimpl ()
    : gender_pskel (&base_impl_)
  {
  }

  virtual ::gender
  post_gender ()
  {
    std::string s = post_string ();
    return s == "male" : male ? female;
  }

private:
  xml_schema::string_pimpl base_impl_;
};

class person_pimpl: public person_pskel
{
public:

  virtual void
  gender (::gender g)
  {
    // g is the gender value returned by gender_pimpl.
  }
};

Boris

From boris at codesynthesis.com  Thu Sep 11 09:49:12 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Thu Sep 11 09:50:35 2008
Subject: [xsde-users] Support for iPhone platform
In-Reply-To: <20080909154304.GA4427@karelia>
References: <324118.41000.qm@web57206.mail.re3.yahoo.com>
	<20080909154304.GA4427@karelia>
Message-ID: <20080911134912.GA10501@karelia>

Hi Vladimir,

I've successfully compiled the XSD/e runtime and examples for
iPhoneOS 2.0 and the simulator. I also ran the examples built
for the simulator. I haven't had a chance to run the examples
on the actual device yet (there does not seem to be a console
on iPhone so we will need to create some kind of driver for
the examples).

I have prepared the XSD/e package for Mac OS X (Intel) which
you can download via this link:

http://www.codesynthesis.com/~boris/tmp/xsde-2.1.0-i686-macosx.tar.bz2

In this package you will find the XSD/e compiler binary as well
as the source code for the runtime and examples. It also contains
two sample configuration files in the etc/iphone/ directory: one
is for the device and one is for the simulator. You will need to
copy one of these files to the config/ directory and rename it
to config.make. If your iPhone SDK is installed in a location
other than the default (/Developer) then you will need to adjust
the paths at the beginning of the config file. You may also want
to check other configuration parameters (e.g., use of iostream,
STL, C++ exceptions, etc; they are all enabled by default).

Once you are happy with your config file you can build the runtime
library by changing to the libxsde/ directory and running make
from there. Once the build is complete you will find libxsde.a in
the libxsde/xsde/ directory.

I assume you are using the XCode IDE to build your application.
If that's the case then probably the easiest way to integrate
the generated code into your project would be to change the
generated source file suffix from .cxx to .mm as suggested
in the answer to the "How do I enable Objective-C++ compilation?"
question in the Apple FAQ:

https://developer.apple.com/iphone/library/codinghowtos/Tools/index.html

You can do it automatically with the '--cxx-suffix .mm' option. See
the XSD/e Compiler Manual for more information:

http://codesynthesis.com/projects/xsde/documentation/xsde.xhtml

Other than that you will need to add the libxsde/ directory to the
include directory list and link to libxsde.a found in libxsde/xsde/.

Let me know if there are any problems as well as if everything is
working fine.

Boris

From boris at codesynthesis.com  Fri Sep 12 07:27:12 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Fri Sep 12 07:28:41 2008
Subject: [xsde-users] XSD/e 2.1.0 package for Mac OS X/iPhone available
Message-ID: <20080912112712.GC13410@karelia>

Hi,

We have uploaded the XSD/e 2.1.0 package for Max OS X (Intel)
which includes sample configuration files for iPhone OS 2.x.
For more information on how to build the XSD/e runtime and
examples for the iPhone device or simulator refer to the
INSTALL file inside.

You can download this package from the XSD/e Download page:

http://www.codesynthesis.com/products/xsde/download.xhtml

The SHA1 checksum for the file:

9ed03d2b3d971d8b30f79712278216550f5a1745  xsde-2.1.0-i686-macosx.tar.bz2

Boris

From ntlkhuong at tma.com.vn  Wed Sep 17 04:26:07 2008
From: ntlkhuong at tma.com.vn (Khuong Nguyen Thi Lien)
Date: Wed Sep 17 04:28:29 2008
Subject: [xsde-users] Get the element name
In-Reply-To: <20080911085121.GB9675@karelia>
References: <20080910090831.GC6428@karelia>
	<007001c91328$3c60db90$828010ac@tma.com.vn>
	<20080911085121.GB9675@karelia>
Message-ID: <001601c9189f$08913310$828010ac@tma.com.vn>

Hi Boris,

 

Thanks for your response, instead of using a string member variable to store
the name, I would like to use the const char* to point to the name array as
the way to save space for my classes and we only use in the callback and
then no use anymore. So is there is there any risk by using this way?

 

Thanks,

Khuong

 

  class clearArpCacheEthernet_pimpl: public clearArpCacheEthernet_pskel

  {

    public:

    clearArpCacheEthernet_pimpl ();

 

    virtual void

    pre ();

 

    // Elements.

    //

  void clearArpCacheEthernet_pimpl::

  slot (ucg::AttributePtr& slot)

  {

    slot->setName(currentElementName);

    data->addAttribute(slot);

  }

 

  void clearArpCacheEthernet_pimpl::

  port (ucg::AttributePtr& port)

  {

    port->setName(currentElementName);

    data->addAttribute(port);

  }

 

    virtual ::ucg::NodePtr&

    post_clearArpCacheEthernet ();

 

    protected:

        bool _start_element_impl(const xsde::cxx::ro_string& ns, const
xsde::cxx::ro_string& n, const char* s)

        {

            currentElementName =  n.data();

            return clearArpCacheEthernet_pskel::_start_element_impl(ns, n,
s);

        }

        ::ucg::NodePtr data;

        const char* currentElementName;

  };

 

 

-----Original Message-----
From: Boris Kolpackov [mailto:boris@codesynthesis.com] 
Sent: Thursday, September 11, 2008 3:51 PM
To: Khuong Nguyen Thi Lien
Cc: xsde-users@codesynthesis.com
Subject: Re: [xsde-users] Get the element name

 

Hi Khuong,

 

Khuong Nguyen Thi Lien <ntlkhuong@tma.com.vn> writes:

 

> Sorry for the unclear question, I have the xml fragment below, when the

> parsing engine reaches the <greeting> element in the greeting() callback

> function, I would like to extract the name of the element right at this

> function. In this case I would like the returned name should be
"greeting". 

 

My guess is that your schema is using substitution groups and element

names act as identifiers (i.e., different substituting elements use the

same type). If that's the case then I have already outlined the possible

ways to achieve this in the following post:

 

http://www.codesynthesis.com/pipermail/xsde-users/2008-July/000039.html

 

In your case the easiest way would be to override _start_element and

save the element name in a member variable. You can then access it

from the greeting function:

 

struct hello_pimpl: hello_pskel

{

  virtual void

  greeting (const string& greeting)

  {

    // element_name_ contains the element name. 

  }

 

  virtual void

  name (const string& name)

  {

  }

 

  virtual void

  _start_element (const xml_schema::ro_string& ns,

                  const xml_schema::ro_string& name

                  const char* type)

  {

     element_name_ = name;

 

     // Always call the base _start_element().

     //

     hello_pskel::_start_element (ns, name, type);

  }

 

private:

  string element_name_;

};

 

 

Boris

From boris at codesynthesis.com  Wed Sep 17 06:29:45 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Wed Sep 17 06:32:06 2008
Subject: [xsde-users] Get the element name
In-Reply-To: <001601c9189f$08913310$828010ac@tma.com.vn>
References: <20080910090831.GC6428@karelia>
	<007001c91328$3c60db90$828010ac@tma.com.vn>
	<20080911085121.GB9675@karelia>
	<001601c9189f$08913310$828010ac@tma.com.vn>
Message-ID: <20080917102945.GB19700@karelia>

Hi Khuong,

Khuong Nguyen Thi Lien <ntlkhuong@tma.com.vn> writes:

> Thanks for your response, instead of using a string member variable to store
> the name, I would like to use the const char* to point to the name array as
> the way to save space for my classes and we only use in the callback and
> then no use anymore. So is there is there any risk by using this way?

This will work in the current implementation though the char array 
returned by n.data() may not be 0-terminated in the general case
(e.g., in the future we may add another XML parser to XSD/e which
may not pass element names as 0-terminated strings to avoid copying).

If you want your code to work in the general case, you will need
to also store the size:

currentElementName     =  n.data();
currentElementNameSize =  n.size();

And then:

slot->setName(currentElementName, currentElementNameSize);

Boris

From sneeraj2 at yahoo.com  Wed Sep 17 08:29:10 2008
From: sneeraj2 at yahoo.com (Neeraj Seshadri)
Date: Wed Sep 17 09:33:24 2008
Subject: [xsde-users] XSDE for WinCE (Microsoft Visual Studio 2005(V8.0)
	with Smart Devices support)
Message-ID: <595042.91502.qm@web31909.mail.mud.yahoo.com>

Hi,

Is it possible to build a binary of XSDE for WinCE
(Microsoft Visual Studio 2005(V8.0) with Smart Devices
support)?
I see that this is listed in the supported platforms
and compiler list.
However the same is not listed in the Download page.
Can the binary for Windows
x86(xsde-2.1.0-i686-windows.zip) be used for WinCE
also?

Thanks
Neeraj


      

From boris at codesynthesis.com  Wed Sep 17 10:01:33 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Wed Sep 17 10:03:46 2008
Subject: [xsde-users] XSDE for WinCE (Microsoft Visual Studio 2005(V8.0)
	with Smart Devices support)
In-Reply-To: <595042.91502.qm@web31909.mail.mud.yahoo.com>
References: <595042.91502.qm@web31909.mail.mud.yahoo.com>
Message-ID: <20080917140133.GG18097@karelia>

Hi Neeraj,

Neeraj Seshadri <sneeraj2@yahoo.com> writes:

> Is it possible to build a binary of XSDE for WinCE
> (Microsoft Visual Studio 2005(V8.0) with Smart Devices
> support)?

You normally don't need to build the compiler binary
for the target platform (WinCE). Usually, you run the
XSD/e compiler (to generate the source code) on your
host development platform (Windows in your case).
What you do need to build for the target is the XSD/e
runtime library. So XSD/e is distributed in the form
of a precompiled binary for a host development platform
(e.g., Windows, Linux, etc) as well as the source code
for the runtime library and examples. This way you
can use the binary to develop for any target device.

WinCE with VC++ 2005 is tested and fully supported.
For more information on how to build the XSD/e runtime
for this platform download xsde-2.1.0-i686-windows.zip
and refer to the INSTALL file inside.

Boris

From boris at codesynthesis.com  Fri Sep 19 11:59:38 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Fri Sep 19 12:02:10 2008
Subject: [xsde-users] Re: [xsd-users] XSDE support for element 'any'
In-Reply-To: <278881.46404.qm@web31911.mail.mud.yahoo.com>
References: <278881.46404.qm@web31911.mail.mud.yahoo.com>
Message-ID: <20080919155938.GC11775@karelia>

Hi Neeraj,

In the future please send questions about XSD/e to the xsde-users
mailing list instead of xsd-users. I have switched to this mailing
list in my reply.

Neeraj Seshadri <sneeraj2@yahoo.com> writes:

> Does the XSDE code generation support the element of
> form any as shown below
> 
> <xs:any namespace="##any" processContents="lax">
> </xs:any>

Yes, it does. Please see the 'wildcard' examples in the
examples/cxx/tree/ and examples/cxx/parser/ directories.


> When I tried to use this in my schema I am getting the
> following error.
> 
> :0:0: error: Complex type '__AnonC1' violates the
> Unique Particle Attribution rule in its components
> 'targetName' and '##any'

This error message indicates that your schema violates the
Unique Particle Attribution (UPA) rule. The specification
states that the XML Schema validator should be able to
uniquely attribute an element being validated to an XML
Schema declaration. Consider this example:

<complexType name="foo">
  <sequence>
    <element name="bar" type="string" minOccurs="0"/>
    <any namespace="##any" processContents="lax"/>
  </sequence>
</complexType>

This fragment violates the UPA rule because when a validator
sees the 'bar' element in an XML document, it can validate
it in two ways: it can attribute 'bar' to the element
declaration or it can assume the element declaration is
not present (minOccurs="0") and attribute 'bar' to the
any wildcard.

For more information refer to Section 3.8.6, "Constraints on
Model Group Schema Components" in the XML Schema specification:

http://www.w3.org/TR/xmlschema-1/#coss-modelGroup

Boris

From mark63 at azurebell.co.nz  Mon Sep 22 23:23:35 2008
From: mark63 at azurebell.co.nz (Mark Easton)
Date: Mon Sep 22 23:19:47 2008
Subject: [xsde-users] Newbie - How to traverse my data structures?
Message-ID: <01eb01c91d2b$c3906830$0601010a@markdevelope>

Hi,

I have started on using xsde today and am struggling a little. My problem is
quite basic I think, so I just need a stear in the right direction.

I have a site and that site has many areas.

What I cant understand is if I do an area_next() in my site_simpl class, how
my area_simple class knows what area record to refer to??

ie I have

xml_schema::int_simpl int_s;
xml_schema::string_simpl string_s;

area_simpl area_s;
site_simpl site_s;

site_s.area_serializer (area_s);
area_s.serializers(int_s, string_s);

xml_document_simpl doc_s(site_s, "site");
std::ostringstream os;

site_s.pre();
doc_s.serialize(os);
site_s.post();


So when in site_simpl I do an area_next(), how does area_simple know which
area record to refer to??


TIA
Mark


From mark63 at azurebell.co.nz  Mon Sep 22 23:44:29 2008
From: mark63 at azurebell.co.nz (Mark Easton)
Date: Mon Sep 22 23:40:36 2008
Subject: [xsde-users] Newbie - How to traverse my data structures?
In-Reply-To: <01eb01c91d2b$c3906830$0601010a@markdevelope>
References: <01eb01c91d2b$c3906830$0601010a@markdevelope>
Message-ID: <01f501c91d2e$af124100$0601010a@markdevelope>

Duhh, sorry. I just realised that from my site_serializer I can access my
area_serializer and therefore tell it which record to use. I guess that is
the right way to do it.

I found that I had to cast area_serializer to (area_simpl)area_serializer.

 

> -----Original Message-----
> From: xsde-users-bounces@codesynthesis.com 
> [mailto:xsde-users-bounces@codesynthesis.com] On Behalf Of Mark Easton
> Sent: Tuesday, September 23, 2008 3:24 PM
> To: xsde-users@codesynthesis.com
> Subject: [xsde-users] Newbie - How to traverse my data structures?
> 
> Hi,
> 
> I have started on using xsde today and am struggling a 
> little. My problem is quite basic I think, so I just need a 
> stear in the right direction.
> 
> I have a site and that site has many areas.
> 
> What I cant understand is if I do an area_next() in my 
> site_simpl class, how my area_simple class knows what area 
> record to refer to??
> 
> ie I have
> 
> xml_schema::int_simpl int_s;
> xml_schema::string_simpl string_s;
> 
> area_simpl area_s;
> site_simpl site_s;
> 
> site_s.area_serializer (area_s);
> area_s.serializers(int_s, string_s);
> 
> xml_document_simpl doc_s(site_s, "site"); std::ostringstream os;
> 
> site_s.pre();
> doc_s.serialize(os);
> site_s.post();
> 
> 
> So when in site_simpl I do an area_next(), how does 
> area_simple know which area record to refer to??
> 
> 
> TIA
> Mark
> 

From boris at codesynthesis.com  Tue Sep 23 03:43:29 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Tue Sep 23 03:46:32 2008
Subject: [xsde-users] Newbie - How to traverse my data structures?
In-Reply-To: <01f501c91d2e$af124100$0601010a@markdevelope>
References: <01eb01c91d2b$c3906830$0601010a@markdevelope>
	<01f501c91d2e$af124100$0601010a@markdevelope>
Message-ID: <20080923074329.GA15064@karelia>

Hi Mark,

Mark Easton <mark63@azurebell.co.nz> writes:

> Duhh, sorry. I just realised that from my site_serializer I can access my
> area_serializer and therefore tell it which record to use. I guess that is
> the right way to do it.
> 
> I found that I had to cast area_serializer to (area_simpl)area_serializer.

You can definitely do it this way, however, there is a mechanism that
allows you to pass custom data from "parent" serializers (site_serializer
in your case) to "child" serializers (area_serializer in your case).

This mechanism is explained in Chapter 4, "Type Maps" in the Embedded
C++/Serializer Mapping Getting Started Guide:

http://codesynthesis.com/projects/xsde/documentation/cxx/serializer/guide/#4

In particular, note how people_simpl passes individual person records
to person_simpl. You may also want to read the previous Chapter 3,
"Serializer Skeletons" since that's where the schema used in Chapter
4 is introduced.

Boris

From mark63 at azurebell.co.nz  Tue Sep 23 04:15:15 2008
From: mark63 at azurebell.co.nz (Mark Easton)
Date: Tue Sep 23 04:11:22 2008
Subject: [xsde-users] Newbie - How to traverse my data structures?
In-Reply-To: <20080923074329.GA15064@karelia>
References: <01eb01c91d2b$c3906830$0601010a@markdevelope>
	<01f501c91d2e$af124100$0601010a@markdevelope>
	<20080923074329.GA15064@karelia>
Message-ID: <025701c91d54$82a32c80$0601010a@markdevelope>

Oh, ok. Many thanks Boris! I had read chapter three quite quickly, but had
not got on to chapter four yet. Will study that tonight and change my code
tomorrow. Great tool!


 

> -----Original Message-----
> From: Boris Kolpackov [mailto:boris@codesynthesis.com] 
> Sent: Tuesday, September 23, 2008 7:43 PM
> To: Mark Easton
> Cc: xsde-users@codesynthesis.com
> Subject: Re: [xsde-users] Newbie - How to traverse my data structures?
> 
> Hi Mark,
> 
> Mark Easton <mark63@azurebell.co.nz> writes:
> 
> > Duhh, sorry. I just realised that from my site_serializer I 
> can access 
> > my area_serializer and therefore tell it which record to 
> use. I guess 
> > that is the right way to do it.
> > 
> > I found that I had to cast area_serializer to 
> (area_simpl)area_serializer.
> 
> You can definitely do it this way, however, there is a 
> mechanism that allows you to pass custom data from "parent" 
> serializers (site_serializer in your case) to "child" 
> serializers (area_serializer in your case).
> 
> This mechanism is explained in Chapter 4, "Type Maps" in the Embedded
> C++/Serializer Mapping Getting Started Guide:
> 
> http://codesynthesis.com/projects/xsde/documentation/cxx/seria
> lizer/guide/#4
> 
> In particular, note how people_simpl passes individual person 
> records to person_simpl. You may also want to read the 
> previous Chapter 3, "Serializer Skeletons" since that's where 
> the schema used in Chapter
> 4 is introduced.
> 
> Boris
> 

From mark63 at azurebell.co.nz  Thu Sep 25 22:36:48 2008
From: mark63 at azurebell.co.nz (Mark Easton)
Date: Fri Sep 26 04:00:36 2008
Subject: [xsde-users] Parsing - Getting Super Data
In-Reply-To: <20080923074329.GA15064@karelia>
References: <01eb01c91d2b$c3906830$0601010a@markdevelope>
	<01f501c91d2e$af124100$0601010a@markdevelope>
	<20080923074329.GA15064@karelia>
Message-ID: <03fd01c91f80$b996c2f0$0601010a@markdevelope>

Great tool. Got the serializers working great to produce a number of xml
documents from my data.

So now working on the parsers to load the data in to my server database. But
I cant work out how to access super data items.

I have a site which is made up of areas. I.e. SO have a site type and an
area type.

<site>
	<id>1</id>
	<name>Auckland</name>
	<area>
		<id>1</id>
		<name>Onehunga</name>
	</area>
</site>


So in my areapimpl class how do I access the id from sitepimpl (ie I need
the site_id for my area record)? I have used the type map, but it only seems
to be used to pass values from area up to site. Ie. class site-pskel has
virtual void area(int siteid). So that is the area-pskel class calling my
site-pskel class. I need it the other way around.

Help

Cheers
Mark

From mark63 at azurebell.co.nz  Thu Sep 25 23:13:06 2008
From: mark63 at azurebell.co.nz (Mark Easton)
Date: Fri Sep 26 04:00:59 2008
Subject: [xsde-users] RE: Parsing - Getting Super Data
References: <01eb01c91d2b$c3906830$0601010a@markdevelope>
	<01f501c91d2e$af124100$0601010a@markdevelope>
	<20080923074329.GA15064@karelia> 
Message-ID: <040d01c91f85$cb979470$0601010a@markdevelope>

The only solution I can think of is to change my serialised data to be as
follows:

 <site>
 	<id>1</id>
 	<name>Auckland</name>
 	<area>
 		<id>1</id>
		<siteid>1</siteid>
 		<name>Onehunga</name>
 	</area>
 </site>

So, add in the extra <siteid> tag. Isn't this duplicated data? With so much
data to transmit this will add to the size of the serialized xml document
substantially. Do I really have to do this?



> -----Original Message-----
> From: Mark Easton [mailto:mark63@azurebell.co.nz] 
> Sent: Friday, September 26, 2008 2:37 PM
> To: 'xsde-users@codesynthesis.com'
> Subject: Parsing - Getting Super Data
> 
> Great tool. Got the serializers working great to produce a 
> number of xml documents from my data.
> 
> So now working on the parsers to load the data in to my 
> server database. But I cant work out how to access super data items.
> 
> I have a site which is made up of areas. I.e. SO have a site 
> type and an area type.
> 
> <site>
> 	<id>1</id>
> 	<name>Auckland</name>
> 	<area>
> 		<id>1</id>
> 		<name>Onehunga</name>
> 	</area>
> </site>
> 
> 
> So in my areapimpl class how do I access the id from 
> sitepimpl (ie I need the site_id for my area record)? I have 
> used the type map, but it only seems to be used to pass 
> values from area up to site. Ie. class site-pskel has virtual 
> void area(int siteid). So that is the area-pskel class 
> calling my site-pskel class. I need it the other way around.
> 
> Help
> 
> Cheers
> Mark
> 

From boris at codesynthesis.com  Fri Sep 26 04:18:53 2008
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Fri Sep 26 04:22:21 2008
Subject: [xsde-users] Parsing - Getting Super Data
In-Reply-To: <03fd01c91f80$b996c2f0$0601010a@markdevelope>
References: <01eb01c91d2b$c3906830$0601010a@markdevelope>
	<01f501c91d2e$af124100$0601010a@markdevelope>
	<20080923074329.GA15064@karelia>
	<03fd01c91f80$b996c2f0$0601010a@markdevelope>
Message-ID: <20080926081853.GA24097@karelia>

Hi Mark,

Mark Easton <mark63@azurebell.co.nz> writes:

> I have a site which is made up of areas. I.e. SO have a site type 
> and an area type.
> 
> <site>
> 	<id>1</id>
> 	<name>Auckland</name>
> 	<area>
> 		<id>1</id>
> 		<name>Onehunga</name>
> 	</area>
> </site>
> 
> 
> So in my areapimpl class how do I access the id from sitepimpl (ie I
> need the site_id for my area record)? 

There is no built-in mechanism for this kind of access since the data 
normally flows from inner parsers to the outer ones. The recommended
way to handle this is to propagate the information from inner parsers
to the point where you have all the parts necessary to process it. In
your case, to handle area you need the area data itself as well as
the site id. The place where all this information can be made 
conveniently available is the site_pimpl::area() callback. Here is
an outline of the site_pimpl class:

// In the type map, map XML Schema type for area to this struct.
// 
//
struct area_data
{
  int id;
  string name;
};


struct site_pimpl: size_pskel
{

  virtual void id (int v)
  {
    site_id = v;
  }

  virtual void area (const area_data& ad)
  {
    // ad and site_id contain all the information we need.
  }

private:
  int site_id;
};

struct area_pimpl: area_pskel
{
  
  virtual void id (int v)
  {
    ad.id = v;
  }

  virtual void name (const string& s)
  {
    ad.name = s;
  }
  
  virtual area_data post_area ()
  {
    return ad;
  }

private:
  area_data ad;
};

Another alternative which is not as clean but can be useful if you
can't afford to copy, say, the name string to a temporary struct
is to arrange a connection between site_pimpl and area_pimpl:


struct site_pimpl: size_pskel
{
  virtual void id (int v)
  {
    site_id = v;
  }

  int site_id;
};

struct area_pimpl: area_pskel
{  
  area_pimpl (site_pimpl& s)
    : site_p (s)
  {
  }

  virtual void id (int v)
  {
    // site_p.site_id conatins site id
  }

  virtual void name (const string& s)
  {
    // site_p.site_id conatins site id
  }
 
private:
  site_pimpl& site_p;
};

Boris

From mark63 at azurebell.co.nz  Sat Sep 27 06:56:25 2008
From: mark63 at azurebell.co.nz (Mark Easton)
Date: Sat Sep 27 06:52:33 2008
Subject: [xsde-users] Parsing - Getting Super Data
In-Reply-To: <20080926081853.GA24097@karelia>
References: <01eb01c91d2b$c3906830$0601010a@markdevelope>
	<01f501c91d2e$af124100$0601010a@markdevelope>
	<20080923074329.GA15064@karelia>
	<03fd01c91f80$b996c2f0$0601010a@markdevelope>
	<20080926081853.GA24097@karelia>
Message-ID: <02c301c9208f$af932cc0$0601010a@markdevelope>

Many thanks Boris,

I think I will follow through on the first technique and put the data in to
data structures and then process these at the site level.

Cheers
Mark 

> -----Original Message-----
> From: Boris Kolpackov [mailto:boris@codesynthesis.com] 
> Sent: Friday, September 26, 2008 8:19 PM
> To: Mark Easton
> Cc: xsde-users@codesynthesis.com
> Subject: Re: [xsde-users] Parsing - Getting Super Data
> 
> Hi Mark,
> 
> Mark Easton <mark63@azurebell.co.nz> writes:
> 
> > I have a site which is made up of areas. I.e. SO have a 
> site type and 
> > an area type.
> > 
> > <site>
> > 	<id>1</id>
> > 	<name>Auckland</name>
> > 	<area>
> > 		<id>1</id>
> > 		<name>Onehunga</name>
> > 	</area>
> > </site>
> > 
> > 
> > So in my areapimpl class how do I access the id from 
> sitepimpl (ie I 
> > need the site_id for my area record)?
> 
> There is no built-in mechanism for this kind of access since 
> the data normally flows from inner parsers to the outer ones. 
> The recommended way to handle this is to propagate the 
> information from inner parsers to the point where you have 
> all the parts necessary to process it. In your case, to 
> handle area you need the area data itself as well as the site 
> id. The place where all this information can be made 
> conveniently available is the site_pimpl::area() callback. 
> Here is an outline of the site_pimpl class:
> 
> // In the type map, map XML Schema type for area to this struct.
> //
> //
> struct area_data
> {
>   int id;
>   string name;
> };
> 
> 
> struct site_pimpl: size_pskel
> {
> 
>   virtual void id (int v)
>   {
>     site_id = v;
>   }
> 
>   virtual void area (const area_data& ad)
>   {
>     // ad and site_id contain all the information we need.
>   }
> 
> private:
>   int site_id;
> };
> 
> struct area_pimpl: area_pskel
> {
>   
>   virtual void id (int v)
>   {
>     ad.id = v;
>   }
> 
>   virtual void name (const string& s)
>   {
>     ad.name = s;
>   }
>   
>   virtual area_data post_area ()
>   {
>     return ad;
>   }
> 
> private:
>   area_data ad;
> };
> 
> Another alternative which is not as clean but can be useful if you
> can't afford to copy, say, the name string to a temporary struct
> is to arrange a connection between site_pimpl and area_pimpl:
> 
> 
> struct site_pimpl: size_pskel
> {
>   virtual void id (int v)
>   {
>     site_id = v;
>   }
> 
>   int site_id;
> };
> 
> struct area_pimpl: area_pskel
> {  
>   area_pimpl (site_pimpl& s)
>     : site_p (s)
>   {
>   }
> 
>   virtual void id (int v)
>   {
>     // site_p.site_id conatins site id
>   }
> 
>   virtual void name (const string& s)
>   {
>     // site_p.site_id conatins site id
>   }
>  
> private:
>   site_pimpl& site_p;
> };
> 
> Boris
>