Return-Path: Delivered-To: apmail-myfaces-dev-archive@www.apache.org Received: (qmail 92573 invoked from network); 26 Jul 2005 16:21:17 -0000 Received: from hermes.apache.org (HELO mail.apache.org) (209.237.227.199) by minotaur.apache.org with SMTP; 26 Jul 2005 16:21:17 -0000 Received: (qmail 67705 invoked by uid 500); 26 Jul 2005 16:21:15 -0000 Delivered-To: apmail-myfaces-dev-archive@myfaces.apache.org Received: (qmail 67658 invoked by uid 500); 26 Jul 2005 16:21:14 -0000 Mailing-List: contact dev-help@myfaces.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: "MyFaces Development" Delivered-To: mailing list dev@myfaces.apache.org Received: (qmail 67645 invoked by uid 99); 26 Jul 2005 16:21:14 -0000 Received: from asf.osuosl.org (HELO asf.osuosl.org) (140.211.166.49) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 26 Jul 2005 09:21:14 -0700 X-ASF-Spam-Status: No, hits=0.1 required=10.0 tests=HTML_40_50,HTML_MESSAGE X-Spam-Check-By: apache.org Received-SPF: pass (asf.osuosl.org: local policy) Received: from [17.250.248.85] (HELO smtpout.mac.com) (17.250.248.85) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 26 Jul 2005 09:21:07 -0700 Received: from mac.com (smtpin02-en2 [10.13.10.147]) by smtpout.mac.com (Xserve/8.12.11/smtpout03/MantshX 4.0) with ESMTP id j6QGLCw1005123 for ; Tue, 26 Jul 2005 09:21:12 -0700 (PDT) Received: from [192.168.0.3] (71-208-19-104.hlrn.qwest.net [71.208.19.104]) (authenticated bits=0) by mac.com (Xserve/smtpin02/MantshX 4.0) with ESMTP id j6QGLAWk007061 for ; Tue, 26 Jul 2005 09:21:11 -0700 (PDT) Mime-Version: 1.0 (Apple Message framework v733) In-Reply-To: <5a99335f0507260053b5577cd@mail.gmail.com> References: <200507222016.CGB48492@ms10.netsolmail.com> <2387fbc50507221319186b953@mail.gmail.com> <073F196B-859F-45D1-A808-DBB7BFC4D389@mac.com> <5a99335f050724030868e63283@mail.gmail.com> <2F1705B3-B0EA-4A10-80B0-D85C1114FDA0@mac.com> <5a99335f0507260053b5577cd@mail.gmail.com> Content-Type: multipart/alternative; boundary=Apple-Mail-35-206296242 Message-Id: <4A609122-F7ED-48EA-B235-02798D026A9B@mac.com> From: Bill Dudney Subject: Re: [proposal] serialVersionUID? Date: Tue, 26 Jul 2005 10:21:40 -0600 To: MyFaces Development X-Mailer: Apple Mail (2.733) X-Virus-Checked: Checked by ClamAV on apache.org X-Spam-Rating: minotaur.apache.org 1.6.2 0/1000/N --Apple-Mail-35-206296242 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed OK, The first round of updates is checked in for the serialVersionUID. here is an old but useful brief write up on how to deal with serialized objects and their versions. http://www.ajug.org/info/tech/serial/serial5.html Since we are removing the compiler/runtime from dealing with the serialVersionUID we will have to update the seriaVersionUID when we change the class in an incompatible way. I'll be committing the rest shortly. TTFN, -bd- On Jul 26, 2005, at 1:53 AM, Martin Marinschek wrote: > cool! > > regards, > > Martin > > On 7/25/05, Bill Dudney wrote: > Hi Martin, > > This would make a difference in development. If the UID is the same > then you won't get the incompatible message on de-serialization. > > Wow an actual use case :-) > > TTFN, > > -bd- > > On Jul 24, 2005, at 4:08 AM, Martin Marinschek wrote: > >> Well, I do have that during development. I change a class, and of >> course the client side serialized version of it is not compatible >> anymore. >> >> Does it help anything having this UID in there with this usecase? >> I mean the compiler would still not be able to do anything about >> instantiating the correct class, cause it is not on the classpath >> anymore, right? >> >> I don't have anything against adding the UID. >> >> regards, >> >> Martin >> >> On 7/22/05, Bill Dudney < bdudney@mac.com> wrote: >> Hi Sean, >> >> yeah I guess 'poor form' is a bit over stated. >> >> I honestly can't think of a situation where it would actually be >> important to have it (user requests a page, the admin upgrades to a >> new version of myfaces, the user clicks a button and the serialized >> page state now has a binary incompatibility seems very far >> fetched :-) but I'm tired of the warning messages from Eclipse >> (similar to the 'unused import' warning) so I'll fix it as I go >> along. >> >> Have a great weekend. >> >> TTFN, >> >> -bd- >> >> On Jul 22, 2005, at 2:19 PM, Sean Schofield wrote: >> >> > I know what it does. We used it in an application when serializing >> > between client and server and where the two had different VM's, >> etc. >> > I was just curious as to how Eclipse fits into all of this. >> > >> > No problem adding it. I would disagree that its poor form not >> having >> > it for classes used in a webapp but if you have a preference to add >> > it, no big deal. >> > >> > sean >> > >> >> > > --Apple-Mail-35-206296242 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=ISO-8859-1 OK,=A0

The first round of updates = is checked in for the serialVersionUID.

here is an old but useful = brief write up on how to deal with serialized objects and their = versions.


Since we are removing the = compiler/runtime from dealing with the serialVersionUID we will have to = update the seriaVersionUID when we change the class in an=A0incompatible = way.

I'll be = committing the rest shortly.

TTFN,

-bd-

= On Jul 26, 2005, at 1:53 AM, Martin Marinschek wrote:

cool!
=
regards,

Martin

On = 7/25/05, Bill Dudney <bdudney@mac.com> = wrote:
Hi Martin,

This would make a difference in = development. If the UID is the same then you won't get the incompatible = message on de-serialization.

Wow an actual use = case :-)

TTFN,

-bd-

On Jul 24, 2005, at 4:08 = AM, Martin Marinschek wrote:

Well, I = do have that during development. I change a class, and of course the = client side serialized version of it is not compatible anymore.

= Does it help anything having this UID in there with this usecase? I mean = the compiler would still not be able to do anything about instantiating = the correct class, cause it is not on the classpath anymore, right?
=
I don't have anything against adding the UID.

regards,
=
Martin

On 7/22/05, Bill Dudney < bdudney@mac.com> = wrote:
Hi Sean,

yeah I guess 'poor form' is a bit over stated. =

I honestly can't think of a situation where it would actually = be
important to have it (user requests a page, the admin upgrades to = a
new version of myfaces, the user clicks a button and the serialized =
page state now has a binary incompatibility seems very far =
fetched :-) but I'm tired of the warning messages from = Eclipse
(similar to the 'unused import' warning) so I'll fix it as I = go along.

Have a great weekend. =

TTFN,

-bd-

On Jul 22, 2005, at 2:19 PM, Sean = Schofield wrote:

> I know what it does.=A0=A0We used it in an = application when serializing
> between client and server and where = the two had different VM's, etc.
> I was just curious as to how = Eclipse fits into all of this.
>
> No problem adding it.=A0=A0= I would disagree that its poor form not having
> it for classes = used in a webapp but if you have a preference to add
> it, no big = deal.
>
> sean
>
=


=


= --Apple-Mail-35-206296242--