Mailing-List: contact users-help@myfaces.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "MyFaces Discussion" <users@myfaces.apache.org>
Received-SPF: neutral (nike.apache.org: local policy)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
  s=s1024; d=yahoo.de;
  h=X-YMail-OSG:Received:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding;
  b=04OhXB55ieQJoBuTjBByVNYAJL6g2f6nKQCxMw817ExWGv5lMrdInbdtOUPGGDf4SP8XbSzgdEkMl4vgrQ8zSZ9H5KZ8h4Dc67KR87bsBu1OBQGu4bS1m76rqoonvi2iyKeHI7KmOsyQIwLVanndfnQjcIp28xeeKD9aYQhZsBQ=;
References: <4F74640C.7020806@googlemail.com>
 <1333028726.4629.14.camel@localhost> <4F7556D8.8020302@googlemail.com>
 <jl3oje$7s4$1@dough.gmane.org>
 <1335145642.87011.YahooMailNeo@web161706.mail.bf1.yahoo.com>
 <jn36po$v8s$1@dough.gmane.org>
Message-ID: <1335174008.72020.YahooMailNeo@web171502.mail.ir2.yahoo.com>
Date: Mon, 23 Apr 2012 10:40:08 +0100 (BST)
From: Mark Struberg <struberg@yahoo.de>
Reply-To: Mark Struberg <struberg@yahoo.de>
Subject: Re: no parallel ajax requests anymore with JSF 2.1?
To: MyFaces Discussion <users@myfaces.apache.org>
In-Reply-To: <jn36po$v8s$1@dough.gmane.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Not having the request/view token properly updated in a sequential way woul=
d also make it impossible to properly do cross request state forgery preven=
tion.=0A=0ALieGrue,=0Astrub=0A=0A=0A=0A----- Original Message -----=0A> Fro=
m: Werner Punz <werner.punz@gmail.com>=0A> To: users@myfaces.apache.org=0A>=
 Cc: =0A> Sent: Monday, April 23, 2012 11:18 AM=0A> Subject: Re: no paralle=
l ajax requests anymore with JSF 2.1?=0A> =0A> Hi=0A> =0A> Hard to say whet=
her it is a good or bad feature, since I was not part of =0A> the discussio=
n when they went for the serialized model I only can guess =0A> why it is l=
ike that. As I stated before, not being serialized would pose =0A> problems=
 to the backend more precisely to the viewstate history.=0A> =0A> You will =
lose viewstates and in the end will run into a viewexpired =0A> exception. =
Hence my two proposals on how to deal with a long running =0A> operation ei=
ther push it out of the jsf domain or simply poll it while =0A> it performs=
 in the background.=0A> =0A> The Myfaces Queue Control adheres to the spec =
and does not allow =0A> parallel requests, I dont know about the Richfaces =
queue whether it is =0A> more flexible in this area.=0A> =0A> =0A> =0A> Wer=
ner=0A> =0A> =0A> =0A> Am 23.04.12 03:47, schrieb Paul Rivera:=0A>>  Hi Wer=
ner,=0A>> =0A>>  I would have to agree with Michael.=A0 Doesn't this spec c=
onstraint of =0A> having ajax calls serialized make one of the best feature=
s of ajax ineffective?=A0 =0A> Assuming that we can manage these requests p=
roperly, I think that there is a =0A> real advantage of allowing concurrent=
 ajax calls.=A0 Is there any way of =0A> configuring the jsf queue to allow=
 this?=0A>> =0A>> =0A>>  Best Regards,=0A>>  Paul Rivera=0A>> =0A>> =0A>> =
=0A>>  ________________________________=0A>> =A0  From: Werner Punz<werner.=
punz@gmail.com>=0A>>  To: users@myfaces.apache.org=0A>>  Sent: Friday, 30 M=
arch 2012 6:50 PM=0A>>  Subject: Re: no parallel ajax requests anymore with=
 JSF 2.1?=0A>> =0A>>  Speaking of the spec, there is a reason why the reque=
sts are serialized,=0A>>  you simply cannot do it without loosing the views=
tate over time for the=0A>>  long running request. The best bet to workarou=
nd this issue is to=0A>>  offload the long running request to something out=
side of JSF=0A>>  Servlet/Restful service whatever.=0A>> =0A>>  Another sol=
ution would be to trigger a short running request which in=0A>>  the backen=
d triggers a background operation which you can poll for being=0A>>  finish=
ed. That would work within the limits of the JSF spec.=0A>> =0A>> =0A>> =0A=
>>  Werner=0A>> =0A>> =0A>> =0A>>  Am 30.03.12 08:46, schrieb Michael Heine=
n:=0A>>>  Hi Milo,=0A>>> =0A>>>  are you really 100% sure that this is poss=
ible with JSF 2.1 and=0A>>>  Richfaces 4.2?=0A>>>  Did you verify that the =
requests are in parallel via logging or =0A> breakpoins?=0A>>> =0A>>>  I tr=
ied a few combinations of the richfaces queues which were not=0A>>>  workin=
g in parallel.=0A>>>  Afaik the richfaces queues are on top of the JSF queu=
e.=0A>>>  Nick Baleavski from Richfaces said this also (07/2011):=0A>>>  ht=
tps://community.jboss.org/message/614023#614023=0A>>>  "JSF 2 does not allo=
w parallel AJAX requests, they are all being =0A> queued=0A>>>  and then se=
nt in the serial order."=0A>>> =0A>>>  Another comment from Richfaces discu=
ssions, looking for concurrent=0A>>>  requests: https://community.jboss.org=
/message/648601#648601=0A>>> =0A>>> =0A>>>  For me this seems to be a major=
 regression in JSF!=0A>>>  It does support "jax" now but no "Ajax"!=0A>>>  =
Could anybody explain this to me? It worked well with JSF 1.2.=0A>>> =0A>>>=
  Now I have to start new threads manually in a web container, which I=0A>>=
>  really don't like.=0A>>>  And the migration to another JSF version is ag=
ain not estimable at all.=0A>>> =0A>>>  vG=0A>>>  Michael=0A>>> =0A>>> =0A>=
>> =0A>>>  Am 29.03.2012 15:45, schrieb Milo van der Zee:=0A>>>>  Hello Mic=
hael,=0A>>>> =0A>>>>  in RichFaces you could add multiple queues and they w=
on't wait =0A> for=0A>>>>  eachother.=0A>>>> =0A>>>>  MAG,=0A>>>>  Milo van=
 der Zee=0A>>>> =0A>>>> =0A>>>>  On Thu, 2012-03-29 at 15:30 +0200, Michael=
 Heinen wrote:=0A>>>>>  Hi all,=0A>>>>> =0A>>>>>  I'm still converting my a=
pplication (mayfaces, tomahawk and =0A> richfaces)=0A>>>>>  from JSF 1.2 to=
 2.1.=0A>>>>> =0A>>>>>  Now I noticed that parallel ajax requests are not w=
orking at =0A> all!=0A>>>>>  E.g. a long running request which calculates s=
omething and =0A> parallel poll=0A>>>>>  requests to fetch status or partia=
l results until the first =0A> request is=0A>>>>>  finished.=0A>>>>> =0A>>>=
>>  I stumbled over Werner's Blog (at=0A>>>>> =0A> http://werpublogs.blogsp=
ot.de/2011/07/apache-myfaces-jsfjs-queue-control.html)=0A>>>>> =0A>>>>>  wh=
ich contains following statement:=0A>>>>>  "The official spec enforces foll=
owing behavior: if you =0A> submit an Ajax=0A>>>>>  post it is either sent =
directly if no other submit is running =0A> or=0A>>>>>  enqueued until the =
running ajax submit has terminated and then =0A> the=0A>>>>>  submit is iss=
ued."=0A>>>>> =0A>>>>>  Question:=0A>>>>>  Are there any workarounds to all=
ow parallel requests?=0A>>>>>  Or do I have to start new threads manually i=
n my backing beans, =0A> which I=0A>>>>>  really do not like?=0A>>>>> =0A>>=
>>>  Is there something like a migration guide available?=0A>>>>>  I read m=
any documents and ppts about JSF 2 but never read =0A> anything about=0A>>>=
>>  this new queing so far.=0A>>>>> =0A>>>>>  Michael=0A>>>>> =0A>>>>> =0A>=
>>> =0A>>> =0A>>> =0A>>> =0A>