You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cocoon.apache.org by Issac Goldstand <ne...@writeme.com> on 2001/05/10 08:05:27 UTC
Cocoon in native C?
Are there any plans to port Cocoon (or a similar XML translater) for Apache to native C? The reason I ask is that it simply seems rather silly to have the portion of the webserver which delivers dynamic and device-specific content require a (rather heavy) device specific code interpreter plus interaction with a secondary back-end servlet engine just to translate my XML output.
Please don't misunderstand - I'm a huge JAVA/XML/device independance fan. But I still think it would be a lot simpler and faster to have Cocoon compiled directly into the httpd binary (or at worst into a DSO module) using native C. I think that this would drasticly improve performance on the webserver. After all, remember that we know exactly which platform our webserver is running, and that (the webserver), and not the unknown end-user's device, is where Cocoon will be running.
The only logical JAVA based XML translaters that I can see use for would be one that's targeted at a command line interface environment, or the planned XSLTC generated translaters (I think Sun's working on that, but I could be wrong) which, unlike Cocoon, are supposed to run on the target devices.
Any comments? Maybe I've overlooked something here?
Thanks in advance,
Issac
Internet is a wonderful mechanism for making a fool of
yourself in front of a very large audience.
--Anonymous
Moving the mouse won't get you into trouble... Clicking it might.
--Anonymous
PGP Key 0xE0FA561B - Fingerprint:
7E18 C018 D623 A57B 7F37 D902 8C84 7675 E0FA 561B
Re: Cocoon in native C?
Posted by Ulrich Mayring <ul...@denic.de>.
Echoes wrote:
>
> I never said that cocoon was bad and that java was bad. I use booth.
> But i'll be glad to se the come back of C in this kind of application
> (nobody's perfect). When it comes to interoperability, i don't really
> understand your remark. Are you talking about libraries, or about data ?
> If about libraries, i'm quite ok, but not about data. For data, you
> just have to transmit it as XML ;-)
First you take your sourcecode from platform A to platform B and it
won't compile. If it compiles, it may not run. In Java it's clear which
library to use for what and for most imaginable tasks out there, you can
find a library for free. There's just so much more stuff available for
Java and you can download it and be sure it'll compile and work.
In C you don't really know what to do. Need a GUI? Well, we could take
QT or perhaps we take GNOME... well, both may not run on Windows, hmm...
there's this cool portable UI toolkit, we should use that. Unfortunately
none of our users have it, so we have to distribute it with our app.
Hmm... how come we get so many nasty support calls from our customers?
Don't they know they should call the manufacturer of that UI toolkit?
Ulrich
--
Ulrich Mayring
DENIC eG, Systementwicklung
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: Cocoon in native C?
Posted by Echoes <ec...@free.fr>.
Ulrich Mayring wrote:
>
> Echoes wrote:
> >
> > Puting all these beasts together is the hard work. But software
> > worked before java came in...
>
> Sure. But this particular software happens to be written in Java. If you
> think it should be ported to C and find people to do it, then we all
> would be very happy for the additional option. But it's a lot of work
> and I don't think it can be done, because most people (like us) nowadays
> use Java as their house standard and so are not really interested to
> help in a C project. C certainly has its merits, but interoperability is
> bad and that is the killer argument for most business scenarios.
I never said that cocoon was bad and that java was bad. I use booth.
But i'll be glad to se the come back of C in this kind of application
(nobody's perfect). When it comes to interoperability, i don't really
understand your remark. Are you talking about libraries, or about data ?
If about libraries, i'm quite ok, but not about data. For data, you
just have to transmit it as XML ;-)
Eugène
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: Cocoon in native C?
Posted by Ulrich Mayring <ul...@denic.de>.
Echoes wrote:
>
> Puting all these beasts together is the hard work. But software
> worked before java came in...
Sure. But this particular software happens to be written in Java. If you
think it should be ported to C and find people to do it, then we all
would be very happy for the additional option. But it's a lot of work
and I don't think it can be done, because most people (like us) nowadays
use Java as their house standard and so are not really interested to
help in a C project. C certainly has its merits, but interoperability is
bad and that is the killer argument for most business scenarios.
Ulrich
--
Ulrich Mayring
DENIC eG, Systementwicklung
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: Cocoon in native C?
Posted by Echoes <ec...@free.fr>.
Berin Loritsch wrote:
>
>
> The "simultaneous" users were 20 hitting the server every 100ms.
> Currently, I don't have the facilities to test beyond this--separate
> machines, etc.
Yep. I understand your testing problems, but 20 users is not a great
deal. So please don't just speak in hits/day; the real mark is data
troughput in Mbytes/s and how many users can comfortably use your
service simultaneously.
> I am aware that Xalan C is not the only XSLT game in town--but one
> thing that Java affords us over C/C++ is the ability easily plug
> and play transformation engines due to JAXP 1.1. There is no
> equivalent (that I am aware of) for C/C++ XSLT/Parser engines.
> Love it or Hate it, Java is here to stay for the duration.
What is the percentage of coccon users that use something else
than xalan for XSLT ? I guess it's a low number, even if XT or
Saxon happen to do a better job (speed wise at least).
> BTW:
> You can pool Tomcat instances (load balancing accross multiple VMs
> on one machine) with Apache at the helm. In fact, you can load
> balance multiple physical servers with Apache httpd, each with
> multiple Tomcat instances running in their own VM, and session
> management is maintained by that infrastucture. This bypasses the
> problems of multiple simultaneous users.
I'll be curious to see how many MBytes/user are consumed by such
a set up. I know hardware is cheap, but is really so good way to
go. People tend no to care coding and configuring carefully (aka
tuning) their software, just putting in more RAM and M(G)Hz...
> Let's face it, in order to leverage _all_ the benefits of _all_
> the standards that Java affords us (JDBC, EJB, JMS, JavaMail, etc),
> there would be an incredible number of adapters to give Cocoon-C
> users a standard interface to develop against. Then there is the
> problem of XSP--compiled serverpages that are compiled at runtime.
JDBC : SQLRelay
EJB : shared libraries
JMS : Jabber ?
javaMail : sendmail & al
Puting all these beasts together is the hard work. But software
worked before java came in...
> There are a number of factors that go into this, and overall,
> Java has the best environment if for no other reason than the
> standards that all developers can take advantage of.
This is a good point : use what you a confortable with.
And be happy !!
Eugène
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: Cocoon in native C?
Posted by Berin Loritsch <bl...@apache.org>.
Echoes wrote:
>
> Berin Loritsch wrote:
> >
> >
> > At this time, I would venture to say no. I would encourage you to check out Cocoon 2, and run
> > some load tests. I slammed my modest workstation (acting as a server) with the equivalent load
> > of 17 million+ hits a day (even load) and Cocoon processed all the requests with sub-second response
> > times. I'd say that at this point the C/C++ conversion would be a huge effort for not alot of
> > real gain--Especially when XSLTMark tests show Xalan C to be slower than Xalan J!
>
> Xalan C is not the only C XSLT processor available. Take a look at
> libxslt (http://www.xmlsoft.org/),
> it is quite impressive (far better than Xalan C at XSLTMark).
> As for your tests, they are not impressive because the main factor is
> not hits/day, but simultaneous
> users. And java VM tend to be memory hogs in these conditions. I'm
> inclined to believe that an
> apache module would behave better. And load management is also doable
> with apache.
The "simultaneous" users were 20 hitting the server every 100ms.
Currently, I don't have the facilities to test beyond this--separate
machines, etc.
I am aware that Xalan C is not the only XSLT game in town--but one
thing that Java affords us over C/C++ is the ability easily plug
and play transformation engines due to JAXP 1.1. There is no
equivalent (that I am aware of) for C/C++ XSLT/Parser engines.
Love it or Hate it, Java is here to stay for the duration.
BTW:
You can pool Tomcat instances (load balancing accross multiple VMs
on one machine) with Apache at the helm. In fact, you can load
balance multiple physical servers with Apache httpd, each with
multiple Tomcat instances running in their own VM, and session
management is maintained by that infrastucture. This bypasses the
problems of multiple simultaneous users.
In fact, your biggest limiting factor is the JDBC driver--if it
can handle the load, you are in good shape. This is also true
of ODBC or native drivers.
Let's face it, in order to leverage _all_ the benefits of _all_
the standards that Java affords us (JDBC, EJB, JMS, JavaMail, etc),
there would be an incredible number of adapters to give Cocoon-C
users a standard interface to develop against. Then there is the
problem of XSP--compiled serverpages that are compiled at runtime.
There are a number of factors that go into this, and overall,
Java has the best environment if for no other reason than the
standards that all developers can take advantage of.
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: Cocoon in native C?
Posted by Echoes <ec...@free.fr>.
Berin Loritsch wrote:
>
>
> At this time, I would venture to say no. I would encourage you to check out Cocoon 2, and run
> some load tests. I slammed my modest workstation (acting as a server) with the equivalent load
> of 17 million+ hits a day (even load) and Cocoon processed all the requests with sub-second response
> times. I'd say that at this point the C/C++ conversion would be a huge effort for not alot of
> real gain--Especially when XSLTMark tests show Xalan C to be slower than Xalan J!
Xalan C is not the only C XSLT processor available. Take a look at
libxslt (http://www.xmlsoft.org/),
it is quite impressive (far better than Xalan C at XSLTMark).
As for your tests, they are not impressive because the main factor is
not hits/day, but simultaneous
users. And java VM tend to be memory hogs in these conditions. I'm
inclined to believe that an
apache module would behave better. And load management is also doable
with apache.
> Cocoon really is only a framework or pipeline manager if you will. It depends on a large number
> of 3rd party libraries. You would have to convert each of the additional libraries to C as well
> to gain the benefits. Especially with SAX, as calling Java from C is slower than calling Java
> from Java and SAX is a series of method calls to represent the structure of the document.
You're right here, but again I think that all this stuff causes to much
memory consumption compared
to the benefits you obtain.
So who volunteers for a Cocoon-like in C ?
Eugène
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: Cocoon in native C?
Posted by Berin Loritsch <bl...@apache.org>.
> Issac Goldstand wrote:
>
> Are there any plans to port Cocoon (or a similar XML translater) for Apache to native C? The reason I ask is that it simply seems
> rather silly to have the portion of the webserver which delivers dynamic and device-specific content require a (rather heavy)
> device specific code interpreter plus interaction with a secondary back-end servlet engine just to translate my XML output.
> Please don't misunderstand - I'm a huge JAVA/XML/device independance fan. But I still think it would be a lot simpler and faster
> to have Cocoon compiled directly into the httpd binary (or at worst into a DSO module) using native C. I think that this would
> drasticly improve performance on the webserver. After all, remember that we know exactly which platform our webserver is running,
> and that (the webserver), and not the unknown end-user's device, is where Cocoon will be running.
At this time, I would venture to say no. I would encourage you to check out Cocoon 2, and run
some load tests. I slammed my modest workstation (acting as a server) with the equivalent load
of 17 million+ hits a day (even load) and Cocoon processed all the requests with sub-second response
times. I'd say that at this point the C/C++ conversion would be a huge effort for not alot of
real gain--Especially when XSLTMark tests show Xalan C to be slower than Xalan J!
Cocoon really is only a framework or pipeline manager if you will. It depends on a large number
of 3rd party libraries. You would have to convert each of the additional libraries to C as well
to gain the benefits. Especially with SAX, as calling Java from C is slower than calling Java
from Java and SAX is a series of method calls to represent the structure of the document.
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: Cocoon in native C?
Posted by Donald Ball <ba...@webslingerZ.com>.
just my 2c on this. i think we're likely to see a mod_xslt in apache
before too long, now that we've got good clean fast xslt libraries in c to
work with. i don't think cocoon itself is ever likely to be rewritten in c
- the ability to write xsp pages in java and plug in custom java
components is just too compelling, and the avalon port would be difficult
and time consuming. i don't think it's worth it - but to each his own.
- donald
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: [C2] repsonse.sendRedirect
Posted by Donald Ball <ba...@webslingerZ.com>.
On Thu, 10 May 2001, Andreas Neuenschwander wrote:
> I tried to use the repsonse.sendRedirect function, but it doesn't seem
> to work in Cocoon2 any longer.
>
> Is there an alternative way to send a redirect to the browser?
>
> I used this function in C1 without any problems...
some developers removed this since they felt it violated the seperation of
concerns. you can send redirects using the sitemap redirect elements, but
you cannot do it using the response object. if you have a use case that
you cannot accomodate using sitemap redirects, please, post it to the
cocoon-dev list - this discussion hasn't been settled yet.
- donald
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: [C2] repsonse.sendRedirect
Posted by Martin Man <Ma...@seznam.cz>.
On Fri, May 11, 2001 at 10:12:00AM +0200, Andreas Neuenschwander wrote:
> Thanks a lot to Donald and Giacomo for their helpful informations.
>
> I started to adapt Ulrich's authentication logicsheet to be compatible with
> Cocoon 2, but I felt into this trap.
>
> The lack of this function in C2 reduces the power of the logicsheet concept
> significantly.
>
> I don't understand right now why the SAX pipeline prevents the sendRedirect
> function to work correctly.
>
> But I found out that I can't build an authentication application like
> Ulrich's auth lib,
> without the capability to send redirects within logicsheets.
> I like the way of Ulrich authentication taglib, because the authors of xml
> pages can
> easily include a password protection. But I see that this affects the
> sitemap concept
> a little bit.
>
> My question is now, will it be possible in near future that this function is
> also save
> with the SAX pipeline concept, or isn't it possible at all?
IMO
as said ten times before, would be worth porting auth taglib to the action,
because it can do much more and is much more clean solution (at least for me,
using the declarative way to protect resources from sitemap, unlike the
procedural one from xsps itself)
>
>
>
> - andi
>
>
>
--
-------------------------------------------------------------------------------
"Only dead fish swims with a stream"
gpg_key_available: http://globales.cz/~mman/martin.man.gpg
gpg_key_fingerprint: 2CC0 4AF6 92DA 5CBF 5F09 7BCB 6202 7024 6E06 0223
---------------------------------------------------------------------
To unsubscribe, e-mail: cocoon-dev-unsubscribe@xml.apache.org
For additional commands, email: cocoon-dev-help@xml.apache.org
Re: [C2] repsonse.sendRedirect
Posted by Andreas Neuenschwander <an...@andi.ch>.
Thanks a lot to Donald and Giacomo for their helpful informations.
I started to adapt Ulrich's authentication logicsheet to be compatible with
Cocoon 2, but I felt into this trap.
The lack of this function in C2 reduces the power of the logicsheet concept
significantly.
I don't understand right now why the SAX pipeline prevents the sendRedirect
function to work correctly.
But I found out that I can't build an authentication application like
Ulrich's auth lib,
without the capability to send redirects within logicsheets.
I like the way of Ulrich authentication taglib, because the authors of xml
pages can
easily include a password protection. But I see that this affects the
sitemap concept
a little bit.
My question is now, will it be possible in near future that this function is
also save
with the SAX pipeline concept, or isn't it possible at all?
- andi
> > Hi,
> >
> > I tried to use the repsonse.sendRedirect function, but it doesn't seem
> > to work in Cocoon2 any longer.
> >
> > Is there an alternative way to send a redirect to the browser?
> >
> > I used this function in C1 without any problems...
>
> In C1 it was save to use sendredirect. In C2 because of the SAX pipeline
> concept it is not save anymore. Use sitemap capabilities to do redirect
> or write Actions for it.
>
> Giacomo
>
> some developers removed this since they felt it violated the seperation of
> concerns. you can send redirects using the sitemap redirect elements, but
> you cannot do it using the response object. if you have a use case that
> you cannot accomodate using sitemap redirects, please, post it to the
> cocoon-dev list - this discussion hasn't been settled yet.
>
> - donald
---------------------------------------------------------------------
To unsubscribe, e-mail: cocoon-dev-unsubscribe@xml.apache.org
For additional commands, email: cocoon-dev-help@xml.apache.org
Re: [C2] repsonse.sendRedirect
Posted by giacomo <gi...@apache.org>.
On Thu, 10 May 2001, Andreas Neuenschwander wrote:
> Hi,
>
> I tried to use the repsonse.sendRedirect function, but it doesn't seem
> to work in Cocoon2 any longer.
>
> Is there an alternative way to send a redirect to the browser?
>
> I used this function in C1 without any problems...
In C1 it was save to use sendredirect. In C2 because of the SAX pipeline
concept it is not save anymore. Use sitemap capabilities to do redirect
or write Actions for it.
Giacomo
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
[C2] repsonse.sendRedirect
Posted by Andreas Neuenschwander <an...@andi.ch>.
Hi,
I tried to use the repsonse.sendRedirect function, but it doesn't seem
to work in Cocoon2 any longer.
Is there an alternative way to send a redirect to the browser?
I used this function in C1 without any problems...
Andi
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
[OT] RE: Cocoon in native C?
Posted by Vadim Gritsenko <vg...@hns.com>.
Hi,
> What I would really like to see though, is that you can choose
> between java byte compiling and also have the opportunity to choose
> to compile to CPU native code.. I heard there is a java-to-cpu-code
> compiler for windows..
Try TowerJ JVM (http://www.towerj.com/), it translates Java -> C, and have ability to
dynamically load classes too (so XSPs should work also).
If you publish some benchmark results it would be great.
Good luck,
Vadim
> -----Original Message-----
> From: forge@neonics.com [mailto:forge@neonics.com]
> Sent: Thursday, May 10, 2001 8:00
> To: cocoon-users@xml.apache.org
> Subject: Re: Cocoon in native C?
>
>
>
> Just some remarks:
>
> I've read on an apache site that the java version of Xalan and/or Xerces
> is faster than the C version.
>
> Also, when you use JIT, eventually, (after running for a while), the code
> is faster than binary compiled C code (at least that is true for
> Micro$oft Visual C).
>
> What I would really like to see though, is that you can choose
> between java byte compiling and also have the opportunity to choose
> to compile to CPU native code.. I heard there is a java-to-cpu-code
> compiler for windows..
>
> But I'm drifting away from the main topic of this newsgroup :)
>
> Anyway, you should also keep in mind that the Servlet engine (Tomcat
> for instance) does not have to run on the same server apache runs on..
> which takes a big byte out of your argument ;)
> (although almost always the servlet engine runs on the apache server..)
>
>
> My 2 cents ;)
>
>
> > Are there any plans to port Cocoon (or a similar XML translater) for Apache to native C? The reason I ask
> is that it simply seems rather silly to have the portion of the webserver which delivers dynamic and
> device-specific content require a (rather heavy) device specific code interpreter plus interaction with a
> secondary back-end servlet engine just to translate my XML output.
> > Please don't misunderstand - I'm a huge JAVA/XML/device independance fan. But I still think it would be a
> lot simpler and faster to have Cocoon compiled directly into the httpd binary (or at worst into a DSO module)
> using native C. I think that this would drasticly improve performance on the webserver. After all, remember
> that we know exactly which platform our webserver is running, and that (the webserver), and not the unknown
> end-user's device, is where Cocoon will be running.
> >
> > The only logical JAVA based XML translaters that I can see use for would be one that's targeted at a
> command line interface environment, or the planned XSLTC generated translaters (I think Sun's working on
> that, but I could be wrong) which, unlike Cocoon, are supposed to run on the target devices.
> >
> > Any comments? Maybe I've overlooked something here?
> >
> > Thanks in advance,
> > Issac
> >
> > Internet is a wonderful mechanism for making a fool of
> > yourself in front of a very large audience.
> > --Anonymous
> >
> > Moving the mouse won't get you into trouble... Clicking it might.
> > --Anonymous
> >
> > PGP Key 0xE0FA561B - Fingerprint:
> > 7E18 C018 D623 A57B 7F37 D902 8C84 7675 E0FA 561B
> >
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>
Re: Cocoon in native C?
Posted by fo...@neonics.com.
Just some remarks:
I've read on an apache site that the java version of Xalan and/or Xerces
is faster than the C version.
Also, when you use JIT, eventually, (after running for a while), the code
is faster than binary compiled C code (at least that is true for
Micro$oft Visual C).
What I would really like to see though, is that you can choose
between java byte compiling and also have the opportunity to choose
to compile to CPU native code.. I heard there is a java-to-cpu-code
compiler for windows..
But I'm drifting away from the main topic of this newsgroup :)
Anyway, you should also keep in mind that the Servlet engine (Tomcat
for instance) does not have to run on the same server apache runs on..
which takes a big byte out of your argument ;)
(although almost always the servlet engine runs on the apache server..)
My 2 cents ;)
> Are there any plans to port Cocoon (or a similar XML translater) for Apache to native C? The reason I ask is that it simply seems rather silly to have the portion of the webserver which delivers dynamic and device-specific content require a (rather heavy) device specific code interpreter plus interaction with a secondary back-end servlet engine just to translate my XML output.
> Please don't misunderstand - I'm a huge JAVA/XML/device independance fan. But I still think it would be a lot simpler and faster to have Cocoon compiled directly into the httpd binary (or at worst into a DSO module) using native C. I think that this would drasticly improve performance on the webserver. After all, remember that we know exactly which platform our webserver is running, and that (the webserver), and not the unknown end-user's device, is where Cocoon will be running.
>
> The only logical JAVA based XML translaters that I can see use for would be one that's targeted at a command line interface environment, or the planned XSLTC generated translaters (I think Sun's working on that, but I could be wrong) which, unlike Cocoon, are supposed to run on the target devices.
>
> Any comments? Maybe I've overlooked something here?
>
> Thanks in advance,
> Issac
>
> Internet is a wonderful mechanism for making a fool of
> yourself in front of a very large audience.
> --Anonymous
>
> Moving the mouse won't get you into trouble... Clicking it might.
> --Anonymous
>
> PGP Key 0xE0FA561B - Fingerprint:
> 7E18 C018 D623 A57B 7F37 D902 8C84 7675 E0FA 561B
>
>
>
>
>
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faqs.html>
To unsubscribe, e-mail: <co...@xml.apache.org>
For additional commands, e-mail: <co...@xml.apache.org>