simple-support — General user and support issues

[Simple-support] add forward references

[Nuno F. <nfa...@is...>]

Hi,

Would it be possible for Simple to support forward references, such as:

<model>
   <object1 id="obj1">
       <x>4</x>
       <y>3</y>
   </object1>
   <object2 reference="obj1" />
   <object3 reference="obj4" />
   <object4 id="obj4">
       <x>2</x>
       <y>1</y>
   </object4>
</model>

Using CyclicStrategy Simple allows that object2 references object1; 
however it's not currently possible for object3 to reference object4.
Could this feature be implemented? Instead of firing the exception when 
the reference is not found, Simple could parse the all of the XML and 
then resolve references; then, if an invalid reference is found, the 
exception would be thrown.

Thank you,

Nuno Fachada
ISR - IST - Portugal

Re: [Simple-support] Syntax highlighting in tutorial page, II

[Timo R. <cr...@ol...>]

Hello Niall,

>> Niall, please have a look at the end of the tutorial page. I wanted
>> to read about "Version tolerant serialization" but it's a bit hard
>> to read it that way ;)

> If you click on the view source button it will show you the source
> in a pop up window. It's the icon that looks like this <>.

oh, I think that was a misunderstanding. I meant a real display-
/formatting-problem:

I checked it via Internet Explorer and everything is ok. But with
Opera (using version 9.60), the whole display of the source code is
mixed with the normal paragraphs; or in other words, from the section
"Version tolerant serialization" down to the end, *everthing* is
formatted as source code :) That is what I meant with "hard to read".

If you have a chance looking at it with the Opera browser, you can
see the problem.


Thanks


Best regards,
Timo Rumland

Re: [Simple-support] Syntax highlighting in tutorial page, II

[<Nia...@ub...>]

Hi,

If you click on the view source button it will show you the source in a
pop up window. It's the icon that looks like this <>. 

Niall

-----Original Message-----
From: Timo Rumland [mailto:cr...@ol...] 
Sent: 26 March 2009 22:43
To: sim...@li...
Subject: [Simple-support] Syntax highlighting in tutorial page, II

Hello,

well, perhaps I was a bit quick with my commendation ;)

Niall, please have a look at the end of the tutorial page. I wanted to
read about "Version tolerant serialization" but it's a bit hard to read
it that way ;)


Best regards,
Timo Rumland


------------------------------------------------------------------------
------
_______________________________________________
Simple-support mailing list
Sim...@li...
https://lists.sourceforge.net/lists/listinfo/simple-support
Visit our website at http://www.ubs.com

This message contains confidential information and is intended only 
for the individual named.  If you are not the named addressee you 
should not disseminate, distribute or copy this e-mail.  Please 
notify the sender immediately by e-mail if you have received this 
e-mail by mistake and delete this e-mail from your system.
	
E-mails are not encrypted and cannot be guaranteed to be secure or 
error-free as information could be intercepted, corrupted, lost, 
destroyed, arrive late or incomplete, or contain viruses.  The sender 
therefore does not accept liability for any errors or omissions in the 
contents of this message which arise as a result of e-mail transmission.  
If verification is required please request a hard-copy version.  This 
message is provided for informational purposes and should not be 
construed as a solicitation or offer to buy or sell any securities 
or related financial instruments.

UBS Limited is a company registered in England & Wales under company
number 2035362, whose registered office is at 1 Finsbury Avenue,
London, EC2M 2PP, United Kingdom.

UBS AG (London Branch) is registered as a branch of a foreign company
under number BR004507, whose registered office is at
1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

UBS Clearing and Execution Services Limited is a company registered
in England & Wales under company number 03123037, whose registered
office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

[Simple-support] Syntax highlighting in tutorial page, II

[Timo R. <cr...@ol...>]

Hello,

well, perhaps I was a bit quick with my commendation ;)

Niall, please have a look at the end of the tutorial page. I wanted
to read about "Version tolerant serialization" but it's a bit hard to
read it that way ;)


Best regards,
Timo Rumland

[Simple-support] Syntax highlighting in tutorial page

[Timo R. <cr...@ol...>]

Hello Niall,

I just noticed the new syntax highlighting on your tutorial page.
It's a really nice improvement, clear and easy to read.

Thanks for that :)


Best regards,
Timo Rumland

Re: [Simple-support] Deserialize a list of objects from one XML

[Niall G. <gal...@ya...>]

Hi,

I am afraid this is a StAX problem. So it all depends on the StAX implementation you use. Some StAX parsers will read the whole file and find that it is not valid, others may not. Simple does not deal with the stream directly.

You can write to a single file. But reading multiple XML serializations from a single file will depend on how stax does it.

Niall


--- On Sun, 3/15/09, Timo Rumland <cr...@ol...> wrote:

> From: Timo Rumland <cr...@ol...>
> Subject: [Simple-support] Deserialize a list of objects from one XML
> To: sim...@li...
> Date: Sunday, March 15, 2009, 3:36 PM
> Hello,
> 
> > In answer to your question yes, you can stream many
> objects to one
> > file. Also, simple should be much more memory
> efficient than XStream
> > at just about everything.
> >
> > Niall
> 
> Niall, you wrote these lines on June 3rd (2008), but I
> could not find
> another question/answer related to this topic.
> 
> Let's say, several objects, serialized to XML, are in
> one file (so
> there is more than one xml root in one file). Is it
> possible to
> deserialize it to a list of objects? I'm missing a
> method with a
> signature like
> 
>   public < T > List< T >
> serializer.read(Class< ? extends T > type, String xml
> );
> 
> (assuming the types of the objects in the file are all the
> same.)
> 
> Is this possible out-of-the-box with simpleXML? Or do I
> have to
> preprocess and parse the XML code and then call the
> serializer.read
> method subsequently?
> 
> 
> Thanks a lot
> 
> 
> Best regards,
> Timo
> 
> 
> ------------------------------------------------------------------------------
> Apps built with the Adobe(R) Flex(R) framework and Flex
> Builder(TM) are
> powering Web 2.0 with engaging, cross-platform
> capabilities. Quickly and
> easily build your RIAs with Flex Builder, the
> Eclipse(TM)based development
> software that enables intelligent coding and step-through
> debugging.
> Download the free 60 day trial.
> http://p.sf.net/sfu/www-adobe-com
> _______________________________________________
> Simple-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simple-support


      

[Simple-support] Deserialize a list of objects from one XML

[Timo R. <cr...@ol...>]

Hello,

> In answer to your question yes, you can stream many objects to one
> file. Also, simple should be much more memory efficient than XStream
> at just about everything.
>
> Niall

Niall, you wrote these lines on June 3rd (2008), but I could not find
another question/answer related to this topic.

Let's say, several objects, serialized to XML, are in one file (so
there is more than one xml root in one file). Is it possible to
deserialize it to a list of objects? I'm missing a method with a
signature like

  public < T > List< T > serializer.read(Class< ? extends T > type, String xml );

(assuming the types of the objects in the file are all the same.)

Is this possible out-of-the-box with simpleXML? Or do I have to
preprocess and parse the XML code and then call the serializer.read
method subsequently?


Thanks a lot


Best regards,
Timo

Re: [Simple-support] Simple @Replace (how to fit an "Element" in an Attribute)

[Fabian K. <fab...@we...>]

Niall Gallagher schrieb:
> Hi,
> 
> Make the field java.lang.Object then it should work. Simple is very strict on types and does not allow a Data to be written as a double. But should allow a java.lang.Object to be represented as a java.lang.Double.
> 
> Niall
> 

Hi,

actualliy, it does not. :(

@Root
public class Data {
	@Attribute
	public double value;

	@Transient
	int stuff;
	
	@Replace
	Object replace() {
		System.out.println("replace called");
		return Double.valueOf(this.value);
	}
}

public class RootClass {
	@Attribute
	Object data;

	// ...
}

But maybe this is not the right way... I think I understand now, that 
replace is not meant for this.

However, using Transform, i can achieve the desired xml.
There is just one little problem:
I need to define a custom Matcher for the Persister (to map 
DataTransform to Data), I think it would be better if I could annotate 
Data with the information about the Transformer.

Fabian

Re: [Simple-support] Simple @Replace (how to fit an "Element" in an Attribute)

[Niall G. <gal...@ya...>]

Hi,

Make the field java.lang.Object then it should work. Simple is very strict on types and does not allow a Data to be written as a double. But should allow a java.lang.Object to be represented as a java.lang.Double.

Niall


--- On Sun, 3/15/09, Fabian Kürten <fab...@we...> wrote:

> From: Fabian Kürten <fab...@we...>
> Subject: Re: [Simple-support] Simple @Replace (how to fit an "Element" in an Attribute)
> To: gal...@ya..., sim...@li...
> Date: Sunday, March 15, 2009, 5:06 AM
> Niall Gallagher schrieb:
> > Hi,
> > 
> > Yes, this will give you an error. How is Data to be
> represented as an attribute. It is a composite of several
> types. Only primitives can be attributes. That is only types
> that can be represented as strings can be attributes.
> > 
> > If you like you can create a custom Transform object
> to represent Data. However I think it best that the object
> model reflects the data. Perhaps Data should be an @Element
> instead of an @Attribute.
> > 
> > Thanks,
> > Niall
> > 
> 
> Hi,
> 
> thanks for your advice, I agree Data might be better an
> Element. However I still want to understand why I can't
> use Data as an Attribute after I use Replace to
> "make" it an double?
> 
> Thanks,
> Fabian


      

Re: [Simple-support] Simple @Replace (how to fit an "Element" in an Attribute)

[Niall G. <gal...@ya...>]

Hi,

You get this because the @Root annotation is not on the Data object. Try this.


@Root
public class Data {
 [blah blah blah....]
}

Let me know how it goes!
Niall

--- On Sat, 3/14/09, Fabian Kürten <fab...@we...> wrote:

> From: Fabian Kürten <fab...@we...>
> Subject: Re: [Simple-support] Simple @Replace (how to fit an "Element" in an Attribute)
> To: gal...@ya..., sim...@li...
> Date: Saturday, March 14, 2009, 2:39 PM
> Niall Gallagher wrote:
> > Hi,
> > 
> > How do you mean it does not work. What happens, do you
> get an exception?
> > 
> > Niall
> 
> Hi,
> 
> yes, I get an exception.
> Some more details:
> 
> Exception in thread "main"
> org.simpleframework.xml.transform.TransformException:
> Transform of class Data not supported
> 
> Fabian
> 
> PS: How should I attach sourcecode?


      

Re: [Simple-support] Simple @Replace (how to fit an "Element" in an Attribute)

[Fabian K. <fab...@we...>]

Niall Gallagher wrote:
> Hi,
> 
> How do you mean it does not work. What happens, do you get an exception?
> 
> Niall

Hi,

yes, I get an exception.
Some more details:

Exception in thread "main" 
org.simpleframework.xml.transform.TransformException: Transform of class 
Data not supported

Fabian

PS: How should I attach sourcecode?

Re: [Simple-support] Simple @Replace (how to fit an "Element" in an Attribute)

[Niall G. <gal...@ya...>]

Hi,

How do you mean it does not work. What happens, do you get an exception?

Niall


--- On Sat, 3/14/09, Fabian Kürten <fab...@we...> wrote:

> From: Fabian Kürten <fab...@we...>
> Subject: [Simple-support] Simple @Replace (how to fit an "Element" in an Attribute)
> To: sim...@li...
> Date: Saturday, March 14, 2009, 1:30 PM
> Hi List,
> 
> I am currently trying to serialize (and later deserialize)
> a simple 
> class containing only one relevant double in another class
> as Attribute.
> 
> public class Data {
> 	@Attribute
> 	public double data;
> 
> 	@Transient
> 	[...]
> }
> 
> public class Root {
> 	@Attribute
> 	Data data
> }
> 
> The serialization does not work.
> 
> I think the Problem could be solved with the @Replace
> Annotation, but I 
> could not find any information how to use it.
> 
> I am currently trying
> (in Data)
> @Replace
> public double replace() {
> 	return this.data;
> }
> but this doesn't change anything.
> 
> Has someone any clues?
> 
> Thanks, Fabian
> 
> ------------------------------------------------------------------------------
> Apps built with the Adobe(R) Flex(R) framework and Flex
> Builder(TM) are
> powering Web 2.0 with engaging, cross-platform
> capabilities. Quickly and
> easily build your RIAs with Flex Builder, the
> Eclipse(TM)based development
> software that enables intelligent coding and step-through
> debugging.
> Download the free 60 day trial.
> http://p.sf.net/sfu/www-adobe-com
> _______________________________________________
> Simple-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simple-support


      

[Simple-support] Simple @Replace (how to fit an "Element" in an Attribute)

[Fabian K. <fab...@we...>]

Hi List,

I am currently trying to serialize (and later deserialize) a simple 
class containing only one relevant double in another class as Attribute.

public class Data {
	@Attribute
	public double data;

	@Transient
	[...]
}

public class Root {
	@Attribute
	Data data
}

The serialization does not work.

I think the Problem could be solved with the @Replace Annotation, but I 
could not find any information how to use it.

I am currently trying
(in Data)
@Replace
public double replace() {
	return this.data;
}
but this doesn't change anything.

Has someone any clues?

Thanks, Fabian

Re: [Simple-support] Little error in simple documentation?

[Niall G. <gal...@ya...>]

Hi,

Thanks, did not notice that. Ill make sure to update it.

Niall


--- On Sat, 3/14/09, Timo Rumland <cr...@ol...> wrote:

> From: Timo Rumland <cr...@ol...>
> Subject: [Simple-support] Little error in simple documentation?
> To: sim...@li...
> Date: Saturday, March 14, 2009, 2:40 AM
> Hello,
> 
> I just found a little error in the Simple documentation on
> the
> website ("tutorial"). Please see this sample XML
> code:
> 
> 
>   <dateList created="2007-01-03 18:05:11.234
> GMT">
>       <list>
>           <object>2007-01-03 18:05:11.234
> GMT</date>
>           <object>2007-01-03 18:05:11.234
> GMT</date>
>       </list>
>   </dateList>
> 
> section "Serializing Java language types")
> 
> The element starts with "<object>" and ends
> with "</date>", not quite
> right, isn't is? :)
> 
> 
> Best Regards,
> Timo
> 
> 
> ------------------------------------------------------------------------------
> Apps built with the Adobe(R) Flex(R) framework and Flex
> Builder(TM) are
> powering Web 2.0 with engaging, cross-platform
> capabilities. Quickly and
> easily build your RIAs with Flex Builder, the
> Eclipse(TM)based development
> software that enables intelligent coding and step-through
> debugging.
> Download the free 60 day trial.
> http://p.sf.net/sfu/www-adobe-com
> _______________________________________________
> Simple-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simple-support


      

[Simple-support] Little error in simple documentation?

[Timo R. <cr...@ol...>]

Hello,

I just found a little error in the Simple documentation on the
website ("tutorial"). Please see this sample XML code:


  <dateList created="2007-01-03 18:05:11.234 GMT">
      <list>
          <object>2007-01-03 18:05:11.234 GMT</date>
          <object>2007-01-03 18:05:11.234 GMT</date>
      </list>
  </dateList>

section "Serializing Java language types")

The element starts with "<object>" and ends with "</date>", not quite
right, isn't is? :)


Best Regards,
Timo

Re: [Simple-support] any performance tricks?

[Kevin D. <ke...@tr...>]

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content=text/html;charset=ISO-8859-1>
<META content="MSHTML 6.00.6000.16788" name=GENERATOR></HEAD>
<BODY text=#000000 bgColor=#ffffff leftMargin=1 topMargin=1 rightMargin=1><FONT face=Arial size=2>
<DIV>At the risk of starting a war here (not my intent!):</DIV>
<DIV>&nbsp;</DIV>
<DIV>This is a case where text based, xml files are always going to be many orders of magnitude less efficient than binary representations.&nbsp; Protocol Buffers were designed for very high speed (and cross language compatability).&nbsp; XML is really just not going to be fast (even formats like JSON which were developed to address some of XML's verboseness for wire-protocol use will be considerably slower than binary representations).</DIV>
<DIV>&nbsp;</DIV>
<DIV>The cost of generating and parsing the text in XML is just always going to be several orders of magnitude higher.&nbsp; If you took the protocol buffer output and converted it to XML, you'd probably find that the overall performance is equivalent to what you are seeing from Simple.&nbsp; Protobuffers may still be&nbsp;a bit faster because they don't use reflection, but I'm pretty sure that the bottleneck here is the text processing...</DIV>
<DIV>&nbsp;</DIV>
<DIV>&nbsp;</DIV>
<DIV>XML is a *heck* of a lot easier to read, perform transformations on and troubleshoot&nbsp;than binary data, though.&nbsp; So the substantial inneficienies inherent to XML (and any text based data transfer format) are overcome by ease of development and troubleshooting.</DIV>
<DIV>&nbsp;</DIV>
<DIV>&nbsp;</DIV>
<DIV>What Jim might be interesed in doing is seeing if there is any way to generate protocol buffer code from Java annotations.&nbsp; I kind of doubt this will be the case (if I recall correctly,&nbsp;the idea behind protocol buffers was to have a language neutral data transfer definition that then compiles to objects in different languages) - but Google is into annotations these days, so maybe they have something like that in the works.</DIV>
<DIV>&nbsp;</DIV>
<DIV>- K</DIV>
<DIV>&nbsp;</DIV>
<DIV>&nbsp;</DIV>
<DIV>&nbsp;</DIV>
<DIV></DIV>
<DIV>&nbsp;</DIV></FONT>
<DIV style="FONT-SIZE: x-small; FONT-FAMILY: Tahoma">
<DIV>----------------------- <B>Original Message</B> -----------------------</DIV>
<DIV>&nbsp;&nbsp;</DIV>
<DIV><B>From:</B>&nbsp;Jim Sermersheim <A href="mailto:ji...@fu..."><FONT color=#0000ff>&lt;ji...@fu...&gt;</FONT></A></DIV>
<DIV><B>To:</B>&nbsp;<A href="mailto:Nia...@ub..."><FONT color=#0000ff>Nia...@ub...</FONT></A></DIV>
<DIV><B>Cc:</B>&nbsp;<A href="mailto:sim...@li..."><FONT color=#0000ff>sim...@li...</FONT></A></DIV>
<DIV><B>Date:</B>&nbsp;Tue, 03 Mar 2009 09:24:59 -0700</DIV>
<DIV><B>Subject:&nbsp;<U>Re: [Simple-support] any performance tricks?</U></B></DIV>
<DIV>&nbsp;&nbsp;</DIV></DIV>Thanks Niall,<BR><BR>No, by default protobuf serializes into a binary format (<A class=moz-txt-link-freetext href="http://code.google.com/apis/protocolbuffers/docs/encoding.html">http://code.google.com/apis/protocolbuffers/docs/encoding.html</A>). One could implement a formatter callback and cause the output to be XML, but for my needs binary is fine. I know that creates an unfair apples to oranges comparison in looking at Simple vs protobuf. Mostly I was just wondering if there were any known performance tricks I may have overlooked. FWIW, I'm transferring the resulting data over the wire and even if I serialized to XML, I'd end up zipping it to reduce the payload size.<BR><BR>Jim<BR><BR><A class=moz-txt-link-abbreviated href="mailto:Nia...@ub...">Nia...@ub...</A> wrote: 
<BLOCKQUOTE cite=mid:9BB...@NL... type="cite"><PRE wrap="">Hi,

Does protobuf generate XML as a result of the serialization process?
There is definitely room for performance improvements in Simple. Its
been a while since I have done a release focused on such enhancements.
Since the introduction of namespaces there has been a slight lag in
performance.

However, I have already begun a scanner framework based around Javassist
which will hopefully provide a signifigant performance improvement over
reflection. 

I think though it will be weeks rather than days before any signifigant
improvements in this area are released.

Niall

-----Original Message-----
From: Jim Sermersheim [<A class=moz-txt-link-freetext href="mailto:ji...@fu...">mailto:ji...@fu...</A>] 
Sent: 03 March 2009 15:53
To: Gallagher, Niall-N
Cc: <A class=moz-txt-link-abbreviated href="mailto:sim...@li...">sim...@li...</A>
Subject: Re: [Simple-support] any performance tricks?

You're right, the first serialization takes hundreds of ms.

I should have mentioned that my test serializes the object a few hundred
times (just to make sure *everything* is primed -- cache, hotspot,
etc.), then starts the timer and serializes it 10000 more times and
calculates an average time per serialization.

protobuf is a different beast in that you build a template and it
generates your code for you. I prefer the more flexible annotation-based
method of Simple, but might have to go with the speed of protobuf :(

<A class=moz-txt-link-abbreviated href="mailto:Nia...@ub...">Nia...@ub...</A> wrote:
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">Hi,

The first time you serialize or deserialize an object will be much 
slower that subsequent times. The best thing to do is write the object
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">once, then start your timer and write it again. You will find the 
second time is much faster than the first.

There is also a planned performance improvement for serialization, 
where I intend to reduce the impact of namespaces, which have added 
about 20% more overhead than needed. With regard to protobuf I am not 
familiar with it. However you will find Simple as fast as JAXB or
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->Xstream.
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">Niall

-----Original Message-----
From: Jim Sermersheim [<A class=moz-txt-link-freetext href="mailto:ji...@fu...">mailto:ji...@fu...</A>]
Sent: 02 March 2009 19:07
To: <A class=moz-txt-link-abbreviated href="mailto:sim...@li...">sim...@li...</A>
Subject: [Simple-support] any performance tricks?

Hi,

I'm evaluating Simple along with some other tools that will serialize 
data for purposes of data transfer across a network. I much prefer the
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">annotations-based way that Simple lets me prepare data for 
serialization to the other tool I'm evaluating (Google protobuf), but 
the performance is becoming a determining factor. For example, a 
sample data structure which takes 1.53ms to encode with Simple only 
takes 0.048ms to encode with protobuf.

I'm just doing a very simple standard annotation markup with 
@Attribute and @Element (the outer structure has a dozen fields and a 
list of two inner structures, each having about a half-dozen fields). 
Then I just use persister.write to serialize.

I'm guessing maybe the reflection is taking a lot of time?  Not sure. 
But are there any known performance tips/tricks to know about with 
Simple?

Thanks,

Jim



CONFIDENTIAL

This document and attachments contain information from Fusion-io, Inc.
which is confidential and/or legally privileged. 
The information is intended only for the use of the individual or 
entity named on this transmission.
If you are not the intended recipient, you are hereby notified that 
any disclosure, copying, distribution or taking of any action in 
reliance on the contents of this emailed information is strictly 
prohibited, and that the documents should be returned to Fusion-io,
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->Inc. immediately.
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">In this regard, if you have received this email in error, please 
notify us by return email immediately.

----------------------------------------------------------------------
--
------
Open Source Business Conference (OSBC), March 24-25, 2009, San 
Francisco, CA -OSBC tackles the biggest issue in open source: Open 
Sourcing the Enterprise -Strategies to boost innovation and cut costs 
with open source participation -Receive a $600 discount off the 
registration fee with the source code: SFAD 
<A class=moz-txt-link-freetext href="http://p.sf.net/sfu/XcvMzF8H">http://p.sf.net/sfu/XcvMzF8H</A> 
_______________________________________________
Simple-support mailing list
<A class=moz-txt-link-abbreviated href="mailto:Sim...@li...">Sim...@li...</A>
<A class=moz-txt-link-freetext href="https://lists.sourceforge.net/lists/listinfo/simple-support">https://lists.sourceforge.net/lists/listinfo/simple-support</A>
Visit our website at <A class=moz-txt-link-freetext href="http://www.ubs.com">http://www.ubs.com</A>

This message contains confidential information and is intended only 
for the individual named.  If you are not the named addressee you 
should not disseminate, distribute or copy this e-mail.  Please notify
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">the sender immediately by e-mail if you have received this e-mail by 
mistake and delete this e-mail from your system.
	
E-mails are not encrypted and cannot be guaranteed to be secure or 
error-free as information could be intercepted, corrupted, lost, 
destroyed, arrive late or incomplete, or contain viruses.  The sender 
therefore does not accept liability for any errors or omissions in the
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">contents of this message which arise as a result of e-mail
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->transmission.
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">If verification is required please request a hard-copy version.  This 
message is provided for informational purposes and should not be 
construed as a solicitation or offer to buy or sell any securities or 
related financial instruments.

UBS Limited is a company registered in England &amp; Wales under company 
number 2035362, whose registered office is at 1 Finsbury Avenue, 
London, EC2M 2PP, United Kingdom.

UBS AG (London Branch) is registered as a branch of a foreign company 
under number BR004507, whose registered office is at
1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

UBS Clearing and Execution Services Limited is a company registered in
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">England &amp; Wales under company number 03123037, whose registered office
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->
  </PRE>
<BLOCKQUOTE type="cite"><PRE wrap="">is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
  
    </PRE></BLOCKQUOTE><PRE wrap=""><!---->

CONFIDENTIAL

This document and attachments contain information from Fusion-io, Inc.
which is confidential and/or legally privileged. 
The information is intended only for the use of the individual or entity
named on this transmission. 
If you are not the intended recipient, you are hereby notified that any
disclosure, copying, distribution or taking of any action in reliance on
the contents of this emailed information is strictly prohibited, and
that the documents should be returned to Fusion-io, Inc. immediately. 
In this regard, if you have received this email in error, please notify
us by return email immediately.
Visit our website at <A class=moz-txt-link-freetext href="http://www.ubs.com">http://www.ubs.com</A>

This message contains confidential information and is intended only 
for the individual named.  If you are not the named addressee you 
should not disseminate, distribute or copy this e-mail.  Please 
notify the sender immediately by e-mail if you have received this 
e-mail by mistake and delete this e-mail from your system.
	
E-mails are not encrypted and cannot be guaranteed to be secure or 
error-free as information could be intercepted, corrupted, lost, 
destroyed, arrive late or incomplete, or contain viruses.  The sender 
therefore does not accept liability for any errors or omissions in the 
contents of this message which arise as a result of e-mail transmission.  
If verification is required please request a hard-copy version.  This 
message is provided for informational purposes and should not be 
construed as a solicitation or offer to buy or sell any securities 
or related financial instruments.

UBS Limited is a company registered in England &amp; Wales under company
number 2035362, whose registered office is at 1 Finsbury Avenue,
London, EC2M 2PP, United Kingdom.

UBS AG (London Branch) is registered as a branch of a foreign company
under number BR004507, whose registered office is at
1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

UBS Clearing and Execution Services Limited is a company registered
in England &amp; Wales under company number 03123037, whose registered
office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
  </PRE></BLOCKQUOTE><BR><BR>CONFIDENTIAL<BR><BR>This document and attachments contain information from Fusion-io, Inc. which is confidential and/or legally privileged.<BR>The information is intended only for the use of the individual or entity named on this transmission.<BR>If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution or taking of any<BR>action in reliance on the contents of this emailed information is strictly prohibited, and that the documents should be returned to Fusion-io, Inc. immediately.<BR>In this regard, if you have received this email in error, please notify us by return email immediately.<BR><BR>
<STYLE type=text/css> P, UL, OL, DL, DIR, MENU, PRE { margin: 0 auto;}</STYLE>
<FONT face=Tahoma size=2>
<DIV>------------------------------------------------------------------------------<BR>Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA<BR>-OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise<BR>-Strategies to boost innovation and cut costs with open source participation<BR>-Receive a $600 discount off the registration fee with the source code: SFAD<BR>http://p.sf.net/sfu/XcvMzF8H<BR><BR></DIV></FONT>
<STYLE type=text/css> P, UL, OL, DL, DIR, MENU, PRE { margin: 0 auto;}</STYLE>
<FONT face=Tahoma size=2>
<DIV>_______________________________________________<BR>Simple-support mailing list<BR>Sim...@li...<BR>https://lists.sourceforge.net/lists/listinfo/simple-support<BR><BR><BR></DIV></FONT></BODY></HTML>

Re: [Simple-support] any performance tricks?

[<Nia...@ub...>]

Visit our website at http://www.ubs.com

This message contains confidential information and is intended only 
for the individual named.  If you are not the named addressee you 
should not disseminate, distribute or copy this e-mail.  Please 
notify the sender immediately by e-mail if you have received this 
e-mail by mistake and delete this e-mail from your system.
	
E-mails are not encrypted and cannot be guaranteed to be secure or 
error-free as information could be intercepted, corrupted, lost, 
destroyed, arrive late or incomplete, or contain viruses.  The sender 
therefore does not accept liability for any errors or omissions in the 
contents of this message which arise as a result of e-mail transmission.  
If verification is required please request a hard-copy version.  This 
message is provided for informational purposes and should not be 
construed as a solicitation or offer to buy or sell any securities 
or related financial instruments.

UBS Limited is a company registered in England & Wales under company
number 2035362, whose registered office is at 1 Finsbury Avenue,
London, EC2M 2PP, United Kingdom.

UBS AG (London Branch) is registered as a branch of a foreign company
under number BR004507, whose registered office is at
1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

UBS Clearing and Execution Services Limited is a company registered
in England & Wales under company number 03123037, whose registered
office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

Re: [Simple-support] any performance tricks?

[Jim S. <ji...@fu...>]

That's what I was initially thinking, but after stepping into the code, 
it looks to me like this is already being done.  There's a cache of 
org.simpleframework.xml.core.Schema objects if I'm not wrong.


Quirino Zagarese wrote:
> Hi,
> I was reading about performance improvements and I was thinking:
> considered simple makes calls to various getAnnotation methods in 
> order to achieve serialization infos at each persister.write/read 
> call(if I'm not wrong), could caching annotations infos once at 
> classloading time be a solution?
> Regards
>
> 2009/3/3 <Nia...@ub... <mailto:Nia...@ub...>>
>
>     Hi,
>
>     Does protobuf generate XML as a result of the serialization process?
>     There is definitely room for performance improvements in Simple. Its
>     been a while since I have done a release focused on such enhancements.
>     Since the introduction of namespaces there has been a slight lag in
>     performance.
>
>     However, I have already begun a scanner framework based around
>     Javassist
>     which will hopefully provide a signifigant performance improvement
>     over
>     reflection.
>
>     I think though it will be weeks rather than days before any
>     signifigant
>     improvements in this area are released.
>
>     Niall
>
>     -----Original Message-----
>     From: Jim Sermersheim [mailto:ji...@fu...
>     <mailto:ji...@fu...>]
>     Sent: 03 March 2009 15:53
>     To: Gallagher, Niall-N
>     Cc: sim...@li...
>     <mailto:sim...@li...>
>     Subject: Re: [Simple-support] any performance tricks?
>
>     You're right, the first serialization takes hundreds of ms.
>
>     I should have mentioned that my test serializes the object a few
>     hundred
>     times (just to make sure *everything* is primed -- cache, hotspot,
>     etc.), then starts the timer and serializes it 10000 more times and
>     calculates an average time per serialization.
>
>     protobuf is a different beast in that you build a template and it
>     generates your code for you. I prefer the more flexible
>     annotation-based
>     method of Simple, but might have to go with the speed of protobuf :(
>
>     Nia...@ub... <mailto:Nia...@ub...> wrote:
>     > Hi,
>     >
>     > The first time you serialize or deserialize an object will be much
>     > slower that subsequent times. The best thing to do is write the
>     object
>
>     > once, then start your timer and write it again. You will find the
>     > second time is much faster than the first.
>     >
>     > There is also a planned performance improvement for serialization,
>     > where I intend to reduce the impact of namespaces, which have added
>     > about 20% more overhead than needed. With regard to protobuf I
>     am not
>     > familiar with it. However you will find Simple as fast as JAXB or
>     Xstream.
>     >
>     > Niall
>     >
>     > -----Original Message-----
>     > From: Jim Sermersheim [mailto:ji...@fu...
>     <mailto:ji...@fu...>]
>     > Sent: 02 March 2009 19:07
>     > To: sim...@li...
>     <mailto:sim...@li...>
>     > Subject: [Simple-support] any performance tricks?
>     >
>     > Hi,
>     >
>     > I'm evaluating Simple along with some other tools that will
>     serialize
>     > data for purposes of data transfer across a network. I much
>     prefer the
>
>     > annotations-based way that Simple lets me prepare data for
>     > serialization to the other tool I'm evaluating (Google
>     protobuf), but
>     > the performance is becoming a determining factor. For example, a
>     > sample data structure which takes 1.53ms to encode with Simple only
>     > takes 0.048ms to encode with protobuf.
>     >
>     > I'm just doing a very simple standard annotation markup with
>     > @Attribute and @Element (the outer structure has a dozen fields
>     and a
>     > list of two inner structures, each having about a half-dozen
>     fields).
>     > Then I just use persister.write to serialize.
>     >
>     > I'm guessing maybe the reflection is taking a lot of time?  Not
>     sure.
>     > But are there any known performance tips/tricks to know about with
>     > Simple?
>     >
>     > Thanks,
>     >
>     > Jim
>     >
>     >
>     >
>     > CONFIDENTIAL
>     >
>     > This document and attachments contain information from
>     Fusion-io, Inc.
>     > which is confidential and/or legally privileged.
>     > The information is intended only for the use of the individual or
>     > entity named on this transmission.
>     > If you are not the intended recipient, you are hereby notified that
>     > any disclosure, copying, distribution or taking of any action in
>     > reliance on the contents of this emailed information is strictly
>     > prohibited, and that the documents should be returned to Fusion-io,
>     Inc. immediately.
>     > In this regard, if you have received this email in error, please
>     > notify us by return email immediately.
>     >
>     >
>     ----------------------------------------------------------------------
>     > --
>     > ------
>     > Open Source Business Conference (OSBC), March 24-25, 2009, San
>     > Francisco, CA -OSBC tackles the biggest issue in open source: Open
>     > Sourcing the Enterprise -Strategies to boost innovation and cut
>     costs
>     > with open source participation -Receive a $600 discount off the
>     > registration fee with the source code: SFAD
>     > http://p.sf.net/sfu/XcvMzF8H
>     > _______________________________________________
>     > Simple-support mailing list
>     > Sim...@li...
>     <mailto:Sim...@li...>
>     > https://lists.sourceforge.net/lists/listinfo/simple-support
>     > Visit our website at http://www.ubs.com
>     >
>     > This message contains confidential information and is intended only
>     > for the individual named.  If you are not the named addressee you
>     > should not disseminate, distribute or copy this e-mail.  Please
>     notify
>
>     > the sender immediately by e-mail if you have received this e-mail by
>     > mistake and delete this e-mail from your system.
>     >
>     > E-mails are not encrypted and cannot be guaranteed to be secure or
>     > error-free as information could be intercepted, corrupted, lost,
>     > destroyed, arrive late or incomplete, or contain viruses.  The
>     sender
>     > therefore does not accept liability for any errors or omissions
>     in the
>
>     > contents of this message which arise as a result of e-mail
>     transmission.
>     > If verification is required please request a hard-copy version.
>      This
>     > message is provided for informational purposes and should not be
>     > construed as a solicitation or offer to buy or sell any
>     securities or
>     > related financial instruments.
>     >
>     > UBS Limited is a company registered in England & Wales under company
>     > number 2035362, whose registered office is at 1 Finsbury Avenue,
>     > London, EC2M 2PP, United Kingdom.
>     >
>     > UBS AG (London Branch) is registered as a branch of a foreign
>     company
>     > under number BR004507, whose registered office is at
>     > 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>     >
>     > UBS Clearing and Execution Services Limited is a company
>     registered in
>
>     > England & Wales under company number 03123037, whose registered
>     office
>
>     > is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>     >
>
>
>     CONFIDENTIAL
>
>     This document and attachments contain information from Fusion-io, Inc.
>     which is confidential and/or legally privileged.
>     The information is intended only for the use of the individual or
>     entity
>     named on this transmission.
>     If you are not the intended recipient, you are hereby notified
>     that any
>     disclosure, copying, distribution or taking of any action in
>     reliance on
>     the contents of this emailed information is strictly prohibited, and
>     that the documents should be returned to Fusion-io, Inc. immediately.
>     In this regard, if you have received this email in error, please
>     notify
>     us by return email immediately.
>     Visit our website at http://www.ubs.com
>
>     This message contains confidential information and is intended only
>     for the individual named.  If you are not the named addressee you
>     should not disseminate, distribute or copy this e-mail.  Please
>     notify the sender immediately by e-mail if you have received this
>     e-mail by mistake and delete this e-mail from your system.
>
>     E-mails are not encrypted and cannot be guaranteed to be secure or
>     error-free as information could be intercepted, corrupted, lost,
>     destroyed, arrive late or incomplete, or contain viruses.  The sender
>     therefore does not accept liability for any errors or omissions in the
>     contents of this message which arise as a result of e-mail
>     transmission.
>     If verification is required please request a hard-copy version.  This
>     message is provided for informational purposes and should not be
>     construed as a solicitation or offer to buy or sell any securities
>     or related financial instruments.
>
>     UBS Limited is a company registered in England & Wales under company
>     number 2035362, whose registered office is at 1 Finsbury Avenue,
>     London, EC2M 2PP, United Kingdom.
>
>     UBS AG (London Branch) is registered as a branch of a foreign company
>     under number BR004507, whose registered office is at
>     1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>
>     UBS Clearing and Execution Services Limited is a company registered
>     in England & Wales under company number 03123037, whose registered
>     office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>
>     ------------------------------------------------------------------------------
>     Open Source Business Conference (OSBC), March 24-25, 2009, San
>     Francisco, CA
>     -OSBC tackles the biggest issue in open source: Open Sourcing the
>     Enterprise
>     -Strategies to boost innovation and cut costs with open source
>     participation
>     -Receive a $600 discount off the registration fee with the source
>     code: SFAD
>     http://p.sf.net/sfu/XcvMzF8H
>     _______________________________________________
>     Simple-support mailing list
>     Sim...@li...
>     <mailto:Sim...@li...>
>     https://lists.sourceforge.net/lists/listinfo/simple-support
>
>
>
>
> -- 
> Quirino Zagarese
>
> LaszloItalia Founder (www.laszloitalia.org <http://www.laszloitalia.org>)
> Software Development Manager - Galilaeus s.r.l.


CONFIDENTIAL

This document and attachments contain information from Fusion-io, Inc. which is confidential and/or legally privileged. 
The information is intended only for the use of the individual or entity named on this transmission. 
If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution or taking of any 
action in reliance on the contents of this emailed information is strictly prohibited, and that the documents should be returned to Fusion-io, Inc. immediately. 
In this regard, if you have received this email in error, please notify us by return email immediately.

Re: [Simple-support] any performance tricks?

[Jim S. <ji...@fu...>]

Thanks Niall,

No, by default protobuf serializes into a binary format 
(http://code.google.com/apis/protocolbuffers/docs/encoding.html). One 
could implement a formatter callback and cause the output to be XML, but 
for my needs binary is fine. I know that creates an unfair apples to 
oranges comparison in looking at Simple vs protobuf. Mostly I was just 
wondering if there were any known performance tricks I may have 
overlooked. FWIW, I'm transferring the resulting data over the wire and 
even if I serialized to XML, I'd end up zipping it to reduce the payload 
size.

Jim

Nia...@ub... wrote:
> Hi,
>
> Does protobuf generate XML as a result of the serialization process?
> There is definitely room for performance improvements in Simple. Its
> been a while since I have done a release focused on such enhancements.
> Since the introduction of namespaces there has been a slight lag in
> performance.
>
> However, I have already begun a scanner framework based around Javassist
> which will hopefully provide a signifigant performance improvement over
> reflection. 
>
> I think though it will be weeks rather than days before any signifigant
> improvements in this area are released.
>
> Niall
>
> -----Original Message-----
> From: Jim Sermersheim [mailto:ji...@fu...] 
> Sent: 03 March 2009 15:53
> To: Gallagher, Niall-N
> Cc: sim...@li...
> Subject: Re: [Simple-support] any performance tricks?
>
> You're right, the first serialization takes hundreds of ms.
>
> I should have mentioned that my test serializes the object a few hundred
> times (just to make sure *everything* is primed -- cache, hotspot,
> etc.), then starts the timer and serializes it 10000 more times and
> calculates an average time per serialization.
>
> protobuf is a different beast in that you build a template and it
> generates your code for you. I prefer the more flexible annotation-based
> method of Simple, but might have to go with the speed of protobuf :(
>
> Nia...@ub... wrote:
>   
>> Hi,
>>
>> The first time you serialize or deserialize an object will be much 
>> slower that subsequent times. The best thing to do is write the object
>>     
>
>   
>> once, then start your timer and write it again. You will find the 
>> second time is much faster than the first.
>>
>> There is also a planned performance improvement for serialization, 
>> where I intend to reduce the impact of namespaces, which have added 
>> about 20% more overhead than needed. With regard to protobuf I am not 
>> familiar with it. However you will find Simple as fast as JAXB or
>>     
> Xstream.
>   
>> Niall
>>
>> -----Original Message-----
>> From: Jim Sermersheim [mailto:ji...@fu...]
>> Sent: 02 March 2009 19:07
>> To: sim...@li...
>> Subject: [Simple-support] any performance tricks?
>>
>> Hi,
>>
>> I'm evaluating Simple along with some other tools that will serialize 
>> data for purposes of data transfer across a network. I much prefer the
>>     
>
>   
>> annotations-based way that Simple lets me prepare data for 
>> serialization to the other tool I'm evaluating (Google protobuf), but 
>> the performance is becoming a determining factor. For example, a 
>> sample data structure which takes 1.53ms to encode with Simple only 
>> takes 0.048ms to encode with protobuf.
>>
>> I'm just doing a very simple standard annotation markup with 
>> @Attribute and @Element (the outer structure has a dozen fields and a 
>> list of two inner structures, each having about a half-dozen fields). 
>> Then I just use persister.write to serialize.
>>
>> I'm guessing maybe the reflection is taking a lot of time?  Not sure. 
>> But are there any known performance tips/tricks to know about with 
>> Simple?
>>
>> Thanks,
>>
>> Jim
>>
>>
>>
>> CONFIDENTIAL
>>
>> This document and attachments contain information from Fusion-io, Inc.
>> which is confidential and/or legally privileged. 
>> The information is intended only for the use of the individual or 
>> entity named on this transmission.
>> If you are not the intended recipient, you are hereby notified that 
>> any disclosure, copying, distribution or taking of any action in 
>> reliance on the contents of this emailed information is strictly 
>> prohibited, and that the documents should be returned to Fusion-io,
>>     
> Inc. immediately.
>   
>> In this regard, if you have received this email in error, please 
>> notify us by return email immediately.
>>
>> ----------------------------------------------------------------------
>> --
>> ------
>> Open Source Business Conference (OSBC), March 24-25, 2009, San 
>> Francisco, CA -OSBC tackles the biggest issue in open source: Open 
>> Sourcing the Enterprise -Strategies to boost innovation and cut costs 
>> with open source participation -Receive a $600 discount off the 
>> registration fee with the source code: SFAD 
>> http://p.sf.net/sfu/XcvMzF8H 
>> _______________________________________________
>> Simple-support mailing list
>> Sim...@li...
>> https://lists.sourceforge.net/lists/listinfo/simple-support
>> Visit our website at http://www.ubs.com
>>
>> This message contains confidential information and is intended only 
>> for the individual named.  If you are not the named addressee you 
>> should not disseminate, distribute or copy this e-mail.  Please notify
>>     
>
>   
>> the sender immediately by e-mail if you have received this e-mail by 
>> mistake and delete this e-mail from your system.
>> 	
>> E-mails are not encrypted and cannot be guaranteed to be secure or 
>> error-free as information could be intercepted, corrupted, lost, 
>> destroyed, arrive late or incomplete, or contain viruses.  The sender 
>> therefore does not accept liability for any errors or omissions in the
>>     
>
>   
>> contents of this message which arise as a result of e-mail
>>     
> transmission.
>   
>> If verification is required please request a hard-copy version.  This 
>> message is provided for informational purposes and should not be 
>> construed as a solicitation or offer to buy or sell any securities or 
>> related financial instruments.
>>
>> UBS Limited is a company registered in England & Wales under company 
>> number 2035362, whose registered office is at 1 Finsbury Avenue, 
>> London, EC2M 2PP, United Kingdom.
>>
>> UBS AG (London Branch) is registered as a branch of a foreign company 
>> under number BR004507, whose registered office is at
>> 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>>
>> UBS Clearing and Execution Services Limited is a company registered in
>>     
>
>   
>> England & Wales under company number 03123037, whose registered office
>>     
>
>   
>> is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>>   
>>     
>
>
> CONFIDENTIAL
>
> This document and attachments contain information from Fusion-io, Inc.
> which is confidential and/or legally privileged. 
> The information is intended only for the use of the individual or entity
> named on this transmission. 
> If you are not the intended recipient, you are hereby notified that any
> disclosure, copying, distribution or taking of any action in reliance on
> the contents of this emailed information is strictly prohibited, and
> that the documents should be returned to Fusion-io, Inc. immediately. 
> In this regard, if you have received this email in error, please notify
> us by return email immediately.
> Visit our website at http://www.ubs.com
>
> This message contains confidential information and is intended only 
> for the individual named.  If you are not the named addressee you 
> should not disseminate, distribute or copy this e-mail.  Please 
> notify the sender immediately by e-mail if you have received this 
> e-mail by mistake and delete this e-mail from your system.
> 	
> E-mails are not encrypted and cannot be guaranteed to be secure or 
> error-free as information could be intercepted, corrupted, lost, 
> destroyed, arrive late or incomplete, or contain viruses.  The sender 
> therefore does not accept liability for any errors or omissions in the 
> contents of this message which arise as a result of e-mail transmission.  
> If verification is required please request a hard-copy version.  This 
> message is provided for informational purposes and should not be 
> construed as a solicitation or offer to buy or sell any securities 
> or related financial instruments.
>
> UBS Limited is a company registered in England & Wales under company
> number 2035362, whose registered office is at 1 Finsbury Avenue,
> London, EC2M 2PP, United Kingdom.
>
> UBS AG (London Branch) is registered as a branch of a foreign company
> under number BR004507, whose registered office is at
> 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>
> UBS Clearing and Execution Services Limited is a company registered
> in England & Wales under company number 03123037, whose registered
> office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>   


CONFIDENTIAL

This document and attachments contain information from Fusion-io, Inc. which is confidential and/or legally privileged. 
The information is intended only for the use of the individual or entity named on this transmission. 
If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution or taking of any 
action in reliance on the contents of this emailed information is strictly prohibited, and that the documents should be returned to Fusion-io, Inc. immediately. 
In this regard, if you have received this email in error, please notify us by return email immediately.

Re: [Simple-support] any performance tricks?

[Quirino Z. <qu...@la...>]

Hi,
I was reading about performance improvements and I was thinking:
considered simple makes calls to various getAnnotation methods in order to
achieve serialization infos at each persister.write/read call(if I'm not
wrong), could caching annotations infos once at classloading time be a
solution?
Regards

2009/3/3 <Nia...@ub...>

> Hi,
>
> Does protobuf generate XML as a result of the serialization process?
> There is definitely room for performance improvements in Simple. Its
> been a while since I have done a release focused on such enhancements.
> Since the introduction of namespaces there has been a slight lag in
> performance.
>
> However, I have already begun a scanner framework based around Javassist
> which will hopefully provide a signifigant performance improvement over
> reflection.
>
> I think though it will be weeks rather than days before any signifigant
> improvements in this area are released.
>
> Niall
>
> -----Original Message-----
> From: Jim Sermersheim [mailto:ji...@fu...]
> Sent: 03 March 2009 15:53
> To: Gallagher, Niall-N
> Cc: sim...@li...
> Subject: Re: [Simple-support] any performance tricks?
>
> You're right, the first serialization takes hundreds of ms.
>
> I should have mentioned that my test serializes the object a few hundred
> times (just to make sure *everything* is primed -- cache, hotspot,
> etc.), then starts the timer and serializes it 10000 more times and
> calculates an average time per serialization.
>
> protobuf is a different beast in that you build a template and it
> generates your code for you. I prefer the more flexible annotation-based
> method of Simple, but might have to go with the speed of protobuf :(
>
> Nia...@ub... wrote:
> > Hi,
> >
> > The first time you serialize or deserialize an object will be much
> > slower that subsequent times. The best thing to do is write the object
>
> > once, then start your timer and write it again. You will find the
> > second time is much faster than the first.
> >
> > There is also a planned performance improvement for serialization,
> > where I intend to reduce the impact of namespaces, which have added
> > about 20% more overhead than needed. With regard to protobuf I am not
> > familiar with it. However you will find Simple as fast as JAXB or
> Xstream.
> >
> > Niall
> >
> > -----Original Message-----
> > From: Jim Sermersheim [mailto:ji...@fu...]
> > Sent: 02 March 2009 19:07
> > To: sim...@li...
> > Subject: [Simple-support] any performance tricks?
> >
> > Hi,
> >
> > I'm evaluating Simple along with some other tools that will serialize
> > data for purposes of data transfer across a network. I much prefer the
>
> > annotations-based way that Simple lets me prepare data for
> > serialization to the other tool I'm evaluating (Google protobuf), but
> > the performance is becoming a determining factor. For example, a
> > sample data structure which takes 1.53ms to encode with Simple only
> > takes 0.048ms to encode with protobuf.
> >
> > I'm just doing a very simple standard annotation markup with
> > @Attribute and @Element (the outer structure has a dozen fields and a
> > list of two inner structures, each having about a half-dozen fields).
> > Then I just use persister.write to serialize.
> >
> > I'm guessing maybe the reflection is taking a lot of time?  Not sure.
> > But are there any known performance tips/tricks to know about with
> > Simple?
> >
> > Thanks,
> >
> > Jim
> >
> >
> >
> > CONFIDENTIAL
> >
> > This document and attachments contain information from Fusion-io, Inc.
> > which is confidential and/or legally privileged.
> > The information is intended only for the use of the individual or
> > entity named on this transmission.
> > If you are not the intended recipient, you are hereby notified that
> > any disclosure, copying, distribution or taking of any action in
> > reliance on the contents of this emailed information is strictly
> > prohibited, and that the documents should be returned to Fusion-io,
> Inc. immediately.
> > In this regard, if you have received this email in error, please
> > notify us by return email immediately.
> >
> > ----------------------------------------------------------------------
> > --
> > ------
> > Open Source Business Conference (OSBC), March 24-25, 2009, San
> > Francisco, CA -OSBC tackles the biggest issue in open source: Open
> > Sourcing the Enterprise -Strategies to boost innovation and cut costs
> > with open source participation -Receive a $600 discount off the
> > registration fee with the source code: SFAD
> > http://p.sf.net/sfu/XcvMzF8H
> > _______________________________________________
> > Simple-support mailing list
> > Sim...@li...
> > https://lists.sourceforge.net/lists/listinfo/simple-support
> > Visit our website at http://www.ubs.com
> >
> > This message contains confidential information and is intended only
> > for the individual named.  If you are not the named addressee you
> > should not disseminate, distribute or copy this e-mail.  Please notify
>
> > the sender immediately by e-mail if you have received this e-mail by
> > mistake and delete this e-mail from your system.
> >
> > E-mails are not encrypted and cannot be guaranteed to be secure or
> > error-free as information could be intercepted, corrupted, lost,
> > destroyed, arrive late or incomplete, or contain viruses.  The sender
> > therefore does not accept liability for any errors or omissions in the
>
> > contents of this message which arise as a result of e-mail
> transmission.
> > If verification is required please request a hard-copy version.  This
> > message is provided for informational purposes and should not be
> > construed as a solicitation or offer to buy or sell any securities or
> > related financial instruments.
> >
> > UBS Limited is a company registered in England & Wales under company
> > number 2035362, whose registered office is at 1 Finsbury Avenue,
> > London, EC2M 2PP, United Kingdom.
> >
> > UBS AG (London Branch) is registered as a branch of a foreign company
> > under number BR004507, whose registered office is at
> > 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
> >
> > UBS Clearing and Execution Services Limited is a company registered in
>
> > England & Wales under company number 03123037, whose registered office
>
> > is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
> >
>
>
> CONFIDENTIAL
>
> This document and attachments contain information from Fusion-io, Inc.
> which is confidential and/or legally privileged.
> The information is intended only for the use of the individual or entity
> named on this transmission.
> If you are not the intended recipient, you are hereby notified that any
> disclosure, copying, distribution or taking of any action in reliance on
> the contents of this emailed information is strictly prohibited, and
> that the documents should be returned to Fusion-io, Inc. immediately.
> In this regard, if you have received this email in error, please notify
> us by return email immediately.
> Visit our website at http://www.ubs.com
>
> This message contains confidential information and is intended only
> for the individual named.  If you are not the named addressee you
> should not disseminate, distribute or copy this e-mail.  Please
> notify the sender immediately by e-mail if you have received this
> e-mail by mistake and delete this e-mail from your system.
>
> E-mails are not encrypted and cannot be guaranteed to be secure or
> error-free as information could be intercepted, corrupted, lost,
> destroyed, arrive late or incomplete, or contain viruses.  The sender
> therefore does not accept liability for any errors or omissions in the
> contents of this message which arise as a result of e-mail transmission.
> If verification is required please request a hard-copy version.  This
> message is provided for informational purposes and should not be
> construed as a solicitation or offer to buy or sell any securities
> or related financial instruments.
>
> UBS Limited is a company registered in England & Wales under company
> number 2035362, whose registered office is at 1 Finsbury Avenue,
> London, EC2M 2PP, United Kingdom.
>
> UBS AG (London Branch) is registered as a branch of a foreign company
> under number BR004507, whose registered office is at
> 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>
> UBS Clearing and Execution Services Limited is a company registered
> in England & Wales under company number 03123037, whose registered
> office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>
>
> ------------------------------------------------------------------------------
> Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco,
> CA
> -OSBC tackles the biggest issue in open source: Open Sourcing the
> Enterprise
> -Strategies to boost innovation and cut costs with open source
> participation
> -Receive a $600 discount off the registration fee with the source code:
> SFAD
> http://p.sf.net/sfu/XcvMzF8H
> _______________________________________________
> Simple-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simple-support
>



-- 
Quirino Zagarese

LaszloItalia Founder (www.laszloitalia.org)
Software Development Manager - Galilaeus s.r.l.

Re: [Simple-support] any performance tricks?

[<Nia...@ub...>]

Hi,

Does protobuf generate XML as a result of the serialization process?
There is definitely room for performance improvements in Simple. Its
been a while since I have done a release focused on such enhancements.
Since the introduction of namespaces there has been a slight lag in
performance.

However, I have already begun a scanner framework based around Javassist
which will hopefully provide a signifigant performance improvement over
reflection. 

I think though it will be weeks rather than days before any signifigant
improvements in this area are released.

Niall

-----Original Message-----
From: Jim Sermersheim [mailto:ji...@fu...] 
Sent: 03 March 2009 15:53
To: Gallagher, Niall-N
Cc: sim...@li...
Subject: Re: [Simple-support] any performance tricks?

You're right, the first serialization takes hundreds of ms.

I should have mentioned that my test serializes the object a few hundred
times (just to make sure *everything* is primed -- cache, hotspot,
etc.), then starts the timer and serializes it 10000 more times and
calculates an average time per serialization.

protobuf is a different beast in that you build a template and it
generates your code for you. I prefer the more flexible annotation-based
method of Simple, but might have to go with the speed of protobuf :(

Nia...@ub... wrote:
> Hi,
>
> The first time you serialize or deserialize an object will be much 
> slower that subsequent times. The best thing to do is write the object

> once, then start your timer and write it again. You will find the 
> second time is much faster than the first.
>
> There is also a planned performance improvement for serialization, 
> where I intend to reduce the impact of namespaces, which have added 
> about 20% more overhead than needed. With regard to protobuf I am not 
> familiar with it. However you will find Simple as fast as JAXB or
Xstream.
>
> Niall
>
> -----Original Message-----
> From: Jim Sermersheim [mailto:ji...@fu...]
> Sent: 02 March 2009 19:07
> To: sim...@li...
> Subject: [Simple-support] any performance tricks?
>
> Hi,
>
> I'm evaluating Simple along with some other tools that will serialize 
> data for purposes of data transfer across a network. I much prefer the

> annotations-based way that Simple lets me prepare data for 
> serialization to the other tool I'm evaluating (Google protobuf), but 
> the performance is becoming a determining factor. For example, a 
> sample data structure which takes 1.53ms to encode with Simple only 
> takes 0.048ms to encode with protobuf.
>
> I'm just doing a very simple standard annotation markup with 
> @Attribute and @Element (the outer structure has a dozen fields and a 
> list of two inner structures, each having about a half-dozen fields). 
> Then I just use persister.write to serialize.
>
> I'm guessing maybe the reflection is taking a lot of time?  Not sure. 
> But are there any known performance tips/tricks to know about with 
> Simple?
>
> Thanks,
>
> Jim
>
>
>
> CONFIDENTIAL
>
> This document and attachments contain information from Fusion-io, Inc.
> which is confidential and/or legally privileged. 
> The information is intended only for the use of the individual or 
> entity named on this transmission.
> If you are not the intended recipient, you are hereby notified that 
> any disclosure, copying, distribution or taking of any action in 
> reliance on the contents of this emailed information is strictly 
> prohibited, and that the documents should be returned to Fusion-io,
Inc. immediately.
> In this regard, if you have received this email in error, please 
> notify us by return email immediately.
>
> ----------------------------------------------------------------------
> --
> ------
> Open Source Business Conference (OSBC), March 24-25, 2009, San 
> Francisco, CA -OSBC tackles the biggest issue in open source: Open 
> Sourcing the Enterprise -Strategies to boost innovation and cut costs 
> with open source participation -Receive a $600 discount off the 
> registration fee with the source code: SFAD 
> http://p.sf.net/sfu/XcvMzF8H 
> _______________________________________________
> Simple-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simple-support
> Visit our website at http://www.ubs.com
>
> This message contains confidential information and is intended only 
> for the individual named.  If you are not the named addressee you 
> should not disseminate, distribute or copy this e-mail.  Please notify

> the sender immediately by e-mail if you have received this e-mail by 
> mistake and delete this e-mail from your system.
> 	
> E-mails are not encrypted and cannot be guaranteed to be secure or 
> error-free as information could be intercepted, corrupted, lost, 
> destroyed, arrive late or incomplete, or contain viruses.  The sender 
> therefore does not accept liability for any errors or omissions in the

> contents of this message which arise as a result of e-mail
transmission.
> If verification is required please request a hard-copy version.  This 
> message is provided for informational purposes and should not be 
> construed as a solicitation or offer to buy or sell any securities or 
> related financial instruments.
>
> UBS Limited is a company registered in England & Wales under company 
> number 2035362, whose registered office is at 1 Finsbury Avenue, 
> London, EC2M 2PP, United Kingdom.
>
> UBS AG (London Branch) is registered as a branch of a foreign company 
> under number BR004507, whose registered office is at
> 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>
> UBS Clearing and Execution Services Limited is a company registered in

> England & Wales under company number 03123037, whose registered office

> is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>   


CONFIDENTIAL

This document and attachments contain information from Fusion-io, Inc.
which is confidential and/or legally privileged. 
The information is intended only for the use of the individual or entity
named on this transmission. 
If you are not the intended recipient, you are hereby notified that any
disclosure, copying, distribution or taking of any action in reliance on
the contents of this emailed information is strictly prohibited, and
that the documents should be returned to Fusion-io, Inc. immediately. 
In this regard, if you have received this email in error, please notify
us by return email immediately.
Visit our website at http://www.ubs.com

This message contains confidential information and is intended only 
for the individual named.  If you are not the named addressee you 
should not disseminate, distribute or copy this e-mail.  Please 
notify the sender immediately by e-mail if you have received this 
e-mail by mistake and delete this e-mail from your system.
	
E-mails are not encrypted and cannot be guaranteed to be secure or 
error-free as information could be intercepted, corrupted, lost, 
destroyed, arrive late or incomplete, or contain viruses.  The sender 
therefore does not accept liability for any errors or omissions in the 
contents of this message which arise as a result of e-mail transmission.  
If verification is required please request a hard-copy version.  This 
message is provided for informational purposes and should not be 
construed as a solicitation or offer to buy or sell any securities 
or related financial instruments.

UBS Limited is a company registered in England & Wales under company
number 2035362, whose registered office is at 1 Finsbury Avenue,
London, EC2M 2PP, United Kingdom.

UBS AG (London Branch) is registered as a branch of a foreign company
under number BR004507, whose registered office is at
1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

UBS Clearing and Execution Services Limited is a company registered
in England & Wales under company number 03123037, whose registered
office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

Re: [Simple-support] any performance tricks?

[Jim S. <ji...@fu...>]

You're right, the first serialization takes hundreds of ms.

I should have mentioned that my test serializes the object a few hundred 
times (just to make sure *everything* is primed -- cache, hotspot, 
etc.), then starts the timer and serializes it 10000 more times and 
calculates an average time per serialization.

protobuf is a different beast in that you build a template and it 
generates your code for you. I prefer the more flexible annotation-based 
method of Simple, but might have to go with the speed of protobuf :(

Nia...@ub... wrote:
> Hi,
>
> The first time you serialize or deserialize an object will be much
> slower that subsequent times. The best thing to do is write the object
> once, then start your timer and write it again. You will find the second
> time is much faster than the first.
>
> There is also a planned performance improvement for serialization, where
> I intend to reduce the impact of namespaces, which have added about 20%
> more overhead than needed. With regard to protobuf I am not familiar
> with it. However you will find Simple as fast as JAXB or Xstream.
>
> Niall
>
> -----Original Message-----
> From: Jim Sermersheim [mailto:ji...@fu...] 
> Sent: 02 March 2009 19:07
> To: sim...@li...
> Subject: [Simple-support] any performance tricks?
>
> Hi,
>
> I'm evaluating Simple along with some other tools that will serialize
> data for purposes of data transfer across a network. I much prefer the
> annotations-based way that Simple lets me prepare data for serialization
> to the other tool I'm evaluating (Google protobuf), but the performance
> is becoming a determining factor. For example, a sample data structure
> which takes 1.53ms to encode with Simple only takes 0.048ms to encode
> with protobuf.
>
> I'm just doing a very simple standard annotation markup with @Attribute
> and @Element (the outer structure has a dozen fields and a list of two
> inner structures, each having about a half-dozen fields). Then I just
> use persister.write to serialize.
>
> I'm guessing maybe the reflection is taking a lot of time?  Not sure. 
> But are there any known performance tips/tricks to know about with
> Simple?
>
> Thanks,
>
> Jim
>
>
>
> CONFIDENTIAL
>
> This document and attachments contain information from Fusion-io, Inc.
> which is confidential and/or legally privileged. 
> The information is intended only for the use of the individual or entity
> named on this transmission. 
> If you are not the intended recipient, you are hereby notified that any
> disclosure, copying, distribution or taking of any action in reliance on
> the contents of this emailed information is strictly prohibited, and
> that the documents should be returned to Fusion-io, Inc. immediately. 
> In this regard, if you have received this email in error, please notify
> us by return email immediately.
>
> ------------------------------------------------------------------------
> ------
> Open Source Business Conference (OSBC), March 24-25, 2009, San
> Francisco, CA -OSBC tackles the biggest issue in open source: Open
> Sourcing the Enterprise -Strategies to boost innovation and cut costs
> with open source participation -Receive a $600 discount off the
> registration fee with the source code: SFAD http://p.sf.net/sfu/XcvMzF8H
> _______________________________________________
> Simple-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simple-support
> Visit our website at http://www.ubs.com
>
> This message contains confidential information and is intended only 
> for the individual named.  If you are not the named addressee you 
> should not disseminate, distribute or copy this e-mail.  Please 
> notify the sender immediately by e-mail if you have received this 
> e-mail by mistake and delete this e-mail from your system.
> 	
> E-mails are not encrypted and cannot be guaranteed to be secure or 
> error-free as information could be intercepted, corrupted, lost, 
> destroyed, arrive late or incomplete, or contain viruses.  The sender 
> therefore does not accept liability for any errors or omissions in the 
> contents of this message which arise as a result of e-mail transmission.  
> If verification is required please request a hard-copy version.  This 
> message is provided for informational purposes and should not be 
> construed as a solicitation or offer to buy or sell any securities 
> or related financial instruments.
>
> UBS Limited is a company registered in England & Wales under company
> number 2035362, whose registered office is at 1 Finsbury Avenue,
> London, EC2M 2PP, United Kingdom.
>
> UBS AG (London Branch) is registered as a branch of a foreign company
> under number BR004507, whose registered office is at
> 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>
> UBS Clearing and Execution Services Limited is a company registered
> in England & Wales under company number 03123037, whose registered
> office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.
>   


CONFIDENTIAL

This document and attachments contain information from Fusion-io, Inc. which is confidential and/or legally privileged. 
The information is intended only for the use of the individual or entity named on this transmission. 
If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution or taking of any 
action in reliance on the contents of this emailed information is strictly prohibited, and that the documents should be returned to Fusion-io, Inc. immediately. 
In this regard, if you have received this email in error, please notify us by return email immediately.

Re: [Simple-support] Problem with override

[saurabh s. <sau...@ya...>]

Hi,

Here is a sample how JAXB does it...


public class Root{    @XmlElementRefs
    ( 
            @XmlElementRef(type=Truck.class),
            @XmlElementRef(type=Motorcycle.class)        {        }    )}
 public abstract class 
}Automobile { 
public class 
}Truck extends Automobile { 
public class
}Motorcyle extends Automobile {
    List<Automobile> vehicles;


Regards,
Saurabh.


________________________________
From: "Nia...@ub..." <Nia...@ub...>
To: sau...@ya...; gal...@ya...; sim...@li...; gio...@te...
Sent: Tuesday, 3 March, 2009 2:55:09 PM
Subject: RE: [Simple-support] Problem with override


Hi,
 
This looks like a nice idea, some had raised a similar issue some time ago. Do you know what annotation is used in JAXB to achieve this? I've not seen it before.
 
Thanks,
Niall


________________________________
From: saurabh sule [mailto:sau...@ya...] 
Sent: 02 March 2009 06:41
To: gal...@ya...; sim...@li...; Sacchi Giovanna
Subject: Re: [Simple-support] Problem with override


One way to achieve this is on the same lines of JAXB.

You can have an annotation like this:

@ElementRef(class="<comma-separared list of all the implementing classes>"). Now when you serialize, you can use the name of the actual implementing class itself that is contained within the vairable or use the name from the annotation on that class.

While deserializing, you can match the name of the tag from the xml to a name from the list or their annotations and create an instance of the actual implementing class.

E.g

public
    @ElementRef(class="Impl1.class, Impl2.class")
    public Iface iface;
}
public class Impl1 implements Iface {
}
public class Impl2 implementsIface {Regards,
}
 
XML - 1:
 <?
<xml version="1.0" encoding="UTF-8"?>root>
<impl1></impl1>root>
</
 
XML - 2:
 <?
<xml version="1.0" encoding="UTF-8"?>root>
<impl2></impl2>root>
</
  class Root {
Saurabh.




________________________________
From: Niall Gallagher <gal...@ya...>
To: "sim...@li..." <sim...@li...>; Sacchi Giovanna <gio...@te...>; saurabh sule <sau...@ya...>
Sent: Friday, 27 February, 2009 6:15:31 PM
Subject: Re: [Simple-support] Problem with override


Hi,

Actually, you are right, it will call all instances "reference". Did not see the different types mentioned. Here the problem is how to match the element with the field. If the names do not match, either by field name or by the name attribute on the annotation then there is no real way to match the element or attribute with the field or method.

However, I welcome any proposals for how to do this, either via an extra annotation or some other scheme.

Thanks,
Niall

--- On Thu, 2/26/09, saurabh sule <sau...@ya...> wrote:

From: saurabh sule <sau...@ya...>
Subject: Re: [Simple-support] Problem with override
To: gal...@ya..., "sim...@li..." <sim...@li...>, "Sacchi Giovanna" <gio...@te...>
Date: Thursday, February 26, 2009, 6:12 PM


Hi,
But this solution will always serialize the Formula to "reference", am I right?
What if one wants to have "reference" , "constant" or "function" depending on the actual implementation?
Is this possible?

Regards,
Saurabh.




________________________________
From: Niall Gallagher <gal...@ya...>
To: "sim...@li..." <sim...@li...>; Sacchi Giovanna <gio...@te...>
Sent: Thursday, 26 February, 2009 6:20:32 PM
Subject: Re: [Simple-support] Problem with override


Hi,

Just name the @Element like so.

@Element(name="reference")
Formula formula;

This will serialize the Formula class as "reference".

Hope this helps,
Niall



--- On Thu, 2/26/09, Sacchi Giovanna <gio...@te...> wrote:

From: Sacchi Giovanna <gio...@te...>
Subject: [Simple-support] Problem with override
To: "sim...@li..." <sim...@li...>
Date: Thursday, February 26, 2009, 3:13 AM


 
Hi,
I’m trying to serialize a class similar to this one:
 
publicclassBinding {
      
      @Attribute
      privateType type;
      
      @Attribute
      privateString name;
      
      @Attribute
      privateSide side;
      
      @Attribute(required=false)
      privateString part;
      
      @Element
      privateFormula formula;
...
}
 
Where Formula is an interface implemented by three classes: Constant, Function and Reference. For instance, I have annotated the Reference class, like that:
 
publicclassReference implementsFormula {
      
      @Attribute
      privateString activity;
      
      @Attribute
      privateType type;
      
      @Attribute
      privateString name;
      
      @Attribute
      privateSide side;
...
}
 
When the Binding class is serialized the following xml file is produced:
<bindingtype="NORMAL" name="par21" side="IN">
        <formulaclass="com.tilab.wemash.model.Reference" activity="webServiceActivity1" type="NORMAL" name="par11" side="OUT"/>
</binding>
 
I would like to know if it there is any way to produce, instead, the following xml:
<bindingtype="NORMAL" name="par21" side="IN">
        <referenceclass="com.tilab.wemash.model.Reference" activity="webServiceActivity1" type="NORMAL" name="par11" side="OUT"/>
</binding>
 
That is: when the element formula of the Binding is serialized, I want that the name of the class Reference is used as element name instead of the name of the interface implemented by the class Reference.
 
Regards,
Giovanna 
 
 
------------------------------------------------------------------
Telecom Italia
Giovanna Sacchi
Information Technology
Innovation, Architecture & Quality
IT Innovation
Via G. Reiss Romoli, 274 - 10148 Torino
+ 39 011 2288040
+ 39 331 6001656
 
Questo messaggio e i suoi allegati sono indirizzati esclusivamente alle persone indicate. La diffusione, copia o qualsiasi altra azione derivante dalla conoscenza di queste informazioni sono rigorosamente vietate. Qualora abbiate ricevuto questo documento per errore siete cortesemente pregati di darne immediata comunicazione al mittente e di provvedere alla sua distruzione, Grazie. 
This e-mail and any attachments is confidential and may contain privileged information intended for the addressee(s) only. Dissemination, copying, printing or use by anybody else is unauthorised. If you are not the intended recipient, please delete this message and any attachments and advise the sender by return e-mail, Thanks.Rispetta l'ambiente. Non stampare questa mail se non è necessario.  
------------------------------------------------------------------------------
Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
-OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
-Strategies to boost innovation and cut costs with open source participation
-Receive a $600 discount off the registration fee with the source code: SFAD
http://p.sf.net/sfu/XcvMzF8H
_______________________________________________
Simple-support mailing list
Sim...@li...
https://lists.sourceforge.net/lists/listinfo/simple-support
 

________________________________
Add more friends to your messenger and enjoy! Invite them now. 

________________________________
Add more friends to your messenger and enjoy! Invite them now.


      Download prohibited? No problem. CHAT from any browser, without download. Go to http://in.webmessenger.yahoo.com/

Re: [Simple-support] any performance tricks?

[<Nia...@ub...>]

Hi,

The first time you serialize or deserialize an object will be much
slower that subsequent times. The best thing to do is write the object
once, then start your timer and write it again. You will find the second
time is much faster than the first.

There is also a planned performance improvement for serialization, where
I intend to reduce the impact of namespaces, which have added about 20%
more overhead than needed. With regard to protobuf I am not familiar
with it. However you will find Simple as fast as JAXB or Xstream.

Niall

-----Original Message-----
From: Jim Sermersheim [mailto:ji...@fu...] 
Sent: 02 March 2009 19:07
To: sim...@li...
Subject: [Simple-support] any performance tricks?

Hi,

I'm evaluating Simple along with some other tools that will serialize
data for purposes of data transfer across a network. I much prefer the
annotations-based way that Simple lets me prepare data for serialization
to the other tool I'm evaluating (Google protobuf), but the performance
is becoming a determining factor. For example, a sample data structure
which takes 1.53ms to encode with Simple only takes 0.048ms to encode
with protobuf.

I'm just doing a very simple standard annotation markup with @Attribute
and @Element (the outer structure has a dozen fields and a list of two
inner structures, each having about a half-dozen fields). Then I just
use persister.write to serialize.

I'm guessing maybe the reflection is taking a lot of time?  Not sure. 
But are there any known performance tips/tricks to know about with
Simple?

Thanks,

Jim



CONFIDENTIAL

This document and attachments contain information from Fusion-io, Inc.
which is confidential and/or legally privileged. 
The information is intended only for the use of the individual or entity
named on this transmission. 
If you are not the intended recipient, you are hereby notified that any
disclosure, copying, distribution or taking of any action in reliance on
the contents of this emailed information is strictly prohibited, and
that the documents should be returned to Fusion-io, Inc. immediately. 
In this regard, if you have received this email in error, please notify
us by return email immediately.

------------------------------------------------------------------------
------
Open Source Business Conference (OSBC), March 24-25, 2009, San
Francisco, CA -OSBC tackles the biggest issue in open source: Open
Sourcing the Enterprise -Strategies to boost innovation and cut costs
with open source participation -Receive a $600 discount off the
registration fee with the source code: SFAD http://p.sf.net/sfu/XcvMzF8H
_______________________________________________
Simple-support mailing list
Sim...@li...
https://lists.sourceforge.net/lists/listinfo/simple-support
Visit our website at http://www.ubs.com

This message contains confidential information and is intended only 
for the individual named.  If you are not the named addressee you 
should not disseminate, distribute or copy this e-mail.  Please 
notify the sender immediately by e-mail if you have received this 
e-mail by mistake and delete this e-mail from your system.
	
E-mails are not encrypted and cannot be guaranteed to be secure or 
error-free as information could be intercepted, corrupted, lost, 
destroyed, arrive late or incomplete, or contain viruses.  The sender 
therefore does not accept liability for any errors or omissions in the 
contents of this message which arise as a result of e-mail transmission.  
If verification is required please request a hard-copy version.  This 
message is provided for informational purposes and should not be 
construed as a solicitation or offer to buy or sell any securities 
or related financial instruments.

UBS Limited is a company registered in England & Wales under company
number 2035362, whose registered office is at 1 Finsbury Avenue,
London, EC2M 2PP, United Kingdom.

UBS AG (London Branch) is registered as a branch of a foreign company
under number BR004507, whose registered office is at
1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

UBS Clearing and Execution Services Limited is a company registered
in England & Wales under company number 03123037, whose registered
office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

Re: [Simple-support] Problem with override

[<Nia...@ub...>]

Visit our website at http://www.ubs.com

This message contains confidential information and is intended only 
for the individual named.  If you are not the named addressee you 
should not disseminate, distribute or copy this e-mail.  Please 
notify the sender immediately by e-mail if you have received this 
e-mail by mistake and delete this e-mail from your system.
	
E-mails are not encrypted and cannot be guaranteed to be secure or 
error-free as information could be intercepted, corrupted, lost, 
destroyed, arrive late or incomplete, or contain viruses.  The sender 
therefore does not accept liability for any errors or omissions in the 
contents of this message which arise as a result of e-mail transmission.  
If verification is required please request a hard-copy version.  This 
message is provided for informational purposes and should not be 
construed as a solicitation or offer to buy or sell any securities 
or related financial instruments.

UBS Limited is a company registered in England & Wales under company
number 2035362, whose registered office is at 1 Finsbury Avenue,
London, EC2M 2PP, United Kingdom.

UBS AG (London Branch) is registered as a branch of a foreign company
under number BR004507, whose registered office is at
1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

UBS Clearing and Execution Services Limited is a company registered
in England & Wales under company number 03123037, whose registered
office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.