You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@myfaces.apache.org by Volker Weber <us...@weber-oldenburg.de> on 2005/11/16 18:01:38 UTC
Why is verbatim component transient?
Hi,
i just came over a problem with verbatim tags in ajax enabled components.
see:
http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
The UIOutput components created for verbatim tags are not saved to state
because there are explicitly set to transient.
This makes it impossible to use verbatim tags in content of ajax components.
Is there a spec reason for this, then we should have a tomahawk
verbatim, or is this just because of performance?
If not a spec reason this should be changed because seems to be a bug.
Regards
Volker
--
Don't answer to From: address!
Mail to this account are droped if not recieved via mailinglist.
To contact me direct create the mail address by
concatenating my forename to my senders domain.
Re: AJAX & verbatim problem (Was Re: Why is verbatim component transient?)
Posted by Martin Marinschek <ma...@gmail.com>.
Yes, right.
I was talking for my personal usage ;)
regards,
Martin
On 11/17/05, Volker Weber <us...@weber-oldenburg.de> wrote:
> Maybe the loadbundle problem occurs more often, but if we not forbid to
> use verbatim inside of ajax components there is also a problem.
>
> Regards
>
> Volker
>
> Martin Marinschek wrote:
> > not f:verbatim - not using this very much.
> >
> > f:loadBundle is the problem.
> >
> > regards,
> >
> > Martin
> >
> > On 11/17/05, Sean Schofield <se...@gmail.com> wrote:
> >
> >>Can you give an example of how this verbatim thing is causing you
> >>problems? I tend to agree with the reasoning that Adam gave on this
> >>one.
> >>
> >>sean
> >>
> >>On 11/17/05, Martin Marinschek <ma...@gmail.com> wrote:
> >>
> >>>yes, my old friend f:loadBundle.
> >>>
> >>>major PITA, this thing ;)
> >>>
> >>>both that it is only evaluated in the JSP processing, but there only
> >>>in the render response phase - shouldn't that happen much earlier?
> >>>couldn't it be that we need the reference to an I18n text much earlier
> >>>in the lifecycle?
> >>>
> >>>regards,
> >>>
> >>>Martin
> >>>
> >>>On 11/17/05, Volker Weber <vo...@weber-oldenburg.de> wrote:
> >>>
> >>>>I see the problems, but i still think it should be possible to have
> >>>>verbatim content inside of ajax updated components. And i don't see any
> >>>>chance to partialy re-execute the relevant jsp files .
> >>>>
> >>>>
> >>>>
> >>>>> - <c:if> will not re-execute, so if its condition has changed,
> >>>>
> >>>>could be replaced by using rendered attribute on jsf tags
> >>>>
> >>>>
> >>>>> - Resource bundles provided by <f:loadBundle> will not be available
> >>>>
> >>>>could be solved by a session scoped resource bundle tag
> >>>>
> >>>>
> >>>>> - Any EL expressions evaluated inside tags will not re-evaluate
> >>>>
> >>>>I currently don't know of those, but IMHO the evaluation of EL
> >>>>expressions should moved inside component or renderer than.
> >>>>
> >>>>
> >>>>So, to enable ajax updated panels we need custom tags for verbatim and
> >>>>loadbundle.
> >>>>
> >>>>Any more problems or other solutions ?
> >>>>
> >>>>Regards
> >>>> Volker
> >>>>
> >>>>
> >>>>Adam Winer wrote:
> >>>>
> >>>>>This is required by the spec because:
> >>>>>
> >>>>> - Performance - in the general case, saving the state for these is a waste
> >>>>> - Re-evaluation - if you've got, say, ${} EL expressions inside <f:verbatim>,
> >>>>> or any dynamically generated content from a tag, you'd better re-run the
> >>>>> JSP or you'll be showing stale content.
> >>>>>
> >>>>>The second problem is important and cannot be resolved by changing
> >>>>>"transient" on f:verbatim.
> >>>>>
> >>>>>Note that you've got other, likely bigger problems if you attempt to render
> >>>>>without re-executing the JSP:
> >>>>> - <c:if> will not re-execute, so if its condition has changed, ah well
> >>>>> - Resource bundles provided by <f:loadBundle> will not be available
> >>>>> - Any EL expressions evaluated inside tags will not re-evaluate
> >>>>>
> >>>>>-- Adam
> >>>>>
> >>>>>
> >>>>>
> >>>>>On 11/16/05, Volker Weber <us...@weber-oldenburg.de> wrote:
> >>>>>
> >>>>>
> >>>>>>Hi,
> >>>>>>
> >>>>>>i just came over a problem with verbatim tags in ajax enabled components.
> >>>>>>see:
> >>>>>>http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
> >>>>>>
> >>>>>>The UIOutput components created for verbatim tags are not saved to state
> >>>>>>because there are explicitly set to transient.
> >>>>>>
> >>>>>>This makes it impossible to use verbatim tags in content of ajax components.
> >>>>>>
> >>>>>>Is there a spec reason for this, then we should have a tomahawk
> >>>>>>verbatim, or is this just because of performance?
> >>>>>>
> >>>>>>If not a spec reason this should be changed because seems to be a bug.
> >>>>>>
> >>>>>>Regards
> >>>>>> Volker
> >>>>>>--
> >>>>>>Don't answer to From: address!
> >>>>>>Mail to this account are droped if not recieved via mailinglist.
> >>>>>>To contact me direct create the mail address by
> >>>>>>concatenating my forename to my senders domain.
> >>>>>>
> >>>>>
> >>>>>
> >>>>--
> >>>>-------------------------------------------------------------------------
> >>>> Volker Weber Dietrichsweg 38a 26127 Oldenburg Germany
> >>>> MAILTO:Volker@weber-oldenburg.de HTTP://www.weber-oldenburg.de
> >>>>
> >>>
> >>>
> >>>--
> >>>
> >>>http://www.irian.at
> >>>
> >>>Your JSF powerhouse -
> >>>JSF Consulting, Development and
> >>>Courses in English and German
> >>>
> >>>Professional Support for Apache MyFaces
> >>>
> >>
> >
> >
> > --
> >
> > http://www.irian.at
> >
> > Your JSF powerhouse -
> > JSF Consulting, Development and
> > Courses in English and German
> >
> > Professional Support for Apache MyFaces
> >
>
> --
> Don't answer to From: address!
> Mail to this account are droped if not recieved via mailinglist.
> To contact me direct create the mail address by
> concatenating my forename to my senders domain.
>
--
http://www.irian.at
Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German
Professional Support for Apache MyFaces
Re: AJAX & verbatim problem (Was Re: Why is verbatim component transient?)
Posted by Volker Weber <us...@weber-oldenburg.de>.
Maybe the loadbundle problem occurs more often, but if we not forbid to
use verbatim inside of ajax components there is also a problem.
Regards
Volker
Martin Marinschek wrote:
> not f:verbatim - not using this very much.
>
> f:loadBundle is the problem.
>
> regards,
>
> Martin
>
> On 11/17/05, Sean Schofield <se...@gmail.com> wrote:
>
>>Can you give an example of how this verbatim thing is causing you
>>problems? I tend to agree with the reasoning that Adam gave on this
>>one.
>>
>>sean
>>
>>On 11/17/05, Martin Marinschek <ma...@gmail.com> wrote:
>>
>>>yes, my old friend f:loadBundle.
>>>
>>>major PITA, this thing ;)
>>>
>>>both that it is only evaluated in the JSP processing, but there only
>>>in the render response phase - shouldn't that happen much earlier?
>>>couldn't it be that we need the reference to an I18n text much earlier
>>>in the lifecycle?
>>>
>>>regards,
>>>
>>>Martin
>>>
>>>On 11/17/05, Volker Weber <vo...@weber-oldenburg.de> wrote:
>>>
>>>>I see the problems, but i still think it should be possible to have
>>>>verbatim content inside of ajax updated components. And i don't see any
>>>>chance to partialy re-execute the relevant jsp files .
>>>>
>>>>
>>>>
>>>>> - <c:if> will not re-execute, so if its condition has changed,
>>>>
>>>>could be replaced by using rendered attribute on jsf tags
>>>>
>>>>
>>>>> - Resource bundles provided by <f:loadBundle> will not be available
>>>>
>>>>could be solved by a session scoped resource bundle tag
>>>>
>>>>
>>>>> - Any EL expressions evaluated inside tags will not re-evaluate
>>>>
>>>>I currently don't know of those, but IMHO the evaluation of EL
>>>>expressions should moved inside component or renderer than.
>>>>
>>>>
>>>>So, to enable ajax updated panels we need custom tags for verbatim and
>>>>loadbundle.
>>>>
>>>>Any more problems or other solutions ?
>>>>
>>>>Regards
>>>> Volker
>>>>
>>>>
>>>>Adam Winer wrote:
>>>>
>>>>>This is required by the spec because:
>>>>>
>>>>> - Performance - in the general case, saving the state for these is a waste
>>>>> - Re-evaluation - if you've got, say, ${} EL expressions inside <f:verbatim>,
>>>>> or any dynamically generated content from a tag, you'd better re-run the
>>>>> JSP or you'll be showing stale content.
>>>>>
>>>>>The second problem is important and cannot be resolved by changing
>>>>>"transient" on f:verbatim.
>>>>>
>>>>>Note that you've got other, likely bigger problems if you attempt to render
>>>>>without re-executing the JSP:
>>>>> - <c:if> will not re-execute, so if its condition has changed, ah well
>>>>> - Resource bundles provided by <f:loadBundle> will not be available
>>>>> - Any EL expressions evaluated inside tags will not re-evaluate
>>>>>
>>>>>-- Adam
>>>>>
>>>>>
>>>>>
>>>>>On 11/16/05, Volker Weber <us...@weber-oldenburg.de> wrote:
>>>>>
>>>>>
>>>>>>Hi,
>>>>>>
>>>>>>i just came over a problem with verbatim tags in ajax enabled components.
>>>>>>see:
>>>>>>http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
>>>>>>
>>>>>>The UIOutput components created for verbatim tags are not saved to state
>>>>>>because there are explicitly set to transient.
>>>>>>
>>>>>>This makes it impossible to use verbatim tags in content of ajax components.
>>>>>>
>>>>>>Is there a spec reason for this, then we should have a tomahawk
>>>>>>verbatim, or is this just because of performance?
>>>>>>
>>>>>>If not a spec reason this should be changed because seems to be a bug.
>>>>>>
>>>>>>Regards
>>>>>> Volker
>>>>>>--
>>>>>>Don't answer to From: address!
>>>>>>Mail to this account are droped if not recieved via mailinglist.
>>>>>>To contact me direct create the mail address by
>>>>>>concatenating my forename to my senders domain.
>>>>>>
>>>>>
>>>>>
>>>>--
>>>>-------------------------------------------------------------------------
>>>> Volker Weber Dietrichsweg 38a 26127 Oldenburg Germany
>>>> MAILTO:Volker@weber-oldenburg.de HTTP://www.weber-oldenburg.de
>>>>
>>>
>>>
>>>--
>>>
>>>http://www.irian.at
>>>
>>>Your JSF powerhouse -
>>>JSF Consulting, Development and
>>>Courses in English and German
>>>
>>>Professional Support for Apache MyFaces
>>>
>>
>
>
> --
>
> http://www.irian.at
>
> Your JSF powerhouse -
> JSF Consulting, Development and
> Courses in English and German
>
> Professional Support for Apache MyFaces
>
--
Don't answer to From: address!
Mail to this account are droped if not recieved via mailinglist.
To contact me direct create the mail address by
concatenating my forename to my senders domain.
Re: AJAX & verbatim problem (Was Re: Why is verbatim component transient?)
Posted by Martin Marinschek <ma...@gmail.com>.
not f:verbatim - not using this very much.
f:loadBundle is the problem.
regards,
Martin
On 11/17/05, Sean Schofield <se...@gmail.com> wrote:
> Can you give an example of how this verbatim thing is causing you
> problems? I tend to agree with the reasoning that Adam gave on this
> one.
>
> sean
>
> On 11/17/05, Martin Marinschek <ma...@gmail.com> wrote:
> > yes, my old friend f:loadBundle.
> >
> > major PITA, this thing ;)
> >
> > both that it is only evaluated in the JSP processing, but there only
> > in the render response phase - shouldn't that happen much earlier?
> > couldn't it be that we need the reference to an I18n text much earlier
> > in the lifecycle?
> >
> > regards,
> >
> > Martin
> >
> > On 11/17/05, Volker Weber <vo...@weber-oldenburg.de> wrote:
> > > I see the problems, but i still think it should be possible to have
> > > verbatim content inside of ajax updated components. And i don't see any
> > > chance to partialy re-execute the relevant jsp files .
> > >
> > >
> > > > - <c:if> will not re-execute, so if its condition has changed,
> > > could be replaced by using rendered attribute on jsf tags
> > >
> > > > - Resource bundles provided by <f:loadBundle> will not be available
> > > could be solved by a session scoped resource bundle tag
> > >
> > > > - Any EL expressions evaluated inside tags will not re-evaluate
> > > I currently don't know of those, but IMHO the evaluation of EL
> > > expressions should moved inside component or renderer than.
> > >
> > >
> > > So, to enable ajax updated panels we need custom tags for verbatim and
> > > loadbundle.
> > >
> > > Any more problems or other solutions ?
> > >
> > > Regards
> > > Volker
> > >
> > >
> > > Adam Winer wrote:
> > > > This is required by the spec because:
> > > >
> > > > - Performance - in the general case, saving the state for these is a waste
> > > > - Re-evaluation - if you've got, say, ${} EL expressions inside <f:verbatim>,
> > > > or any dynamically generated content from a tag, you'd better re-run the
> > > > JSP or you'll be showing stale content.
> > > >
> > > > The second problem is important and cannot be resolved by changing
> > > > "transient" on f:verbatim.
> > > >
> > > > Note that you've got other, likely bigger problems if you attempt to render
> > > > without re-executing the JSP:
> > > > - <c:if> will not re-execute, so if its condition has changed, ah well
> > > > - Resource bundles provided by <f:loadBundle> will not be available
> > > > - Any EL expressions evaluated inside tags will not re-evaluate
> > > >
> > > > -- Adam
> > > >
> > > >
> > > >
> > > > On 11/16/05, Volker Weber <us...@weber-oldenburg.de> wrote:
> > > >
> > > >>Hi,
> > > >>
> > > >>i just came over a problem with verbatim tags in ajax enabled components.
> > > >>see:
> > > >>http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
> > > >>
> > > >>The UIOutput components created for verbatim tags are not saved to state
> > > >>because there are explicitly set to transient.
> > > >>
> > > >>This makes it impossible to use verbatim tags in content of ajax components.
> > > >>
> > > >>Is there a spec reason for this, then we should have a tomahawk
> > > >>verbatim, or is this just because of performance?
> > > >>
> > > >>If not a spec reason this should be changed because seems to be a bug.
> > > >>
> > > >>Regards
> > > >> Volker
> > > >>--
> > > >>Don't answer to From: address!
> > > >>Mail to this account are droped if not recieved via mailinglist.
> > > >>To contact me direct create the mail address by
> > > >>concatenating my forename to my senders domain.
> > > >>
> > > >
> > > >
> > >
> > > --
> > > -------------------------------------------------------------------------
> > > Volker Weber Dietrichsweg 38a 26127 Oldenburg Germany
> > > MAILTO:Volker@weber-oldenburg.de HTTP://www.weber-oldenburg.de
> > >
> >
> >
> > --
> >
> > http://www.irian.at
> >
> > Your JSF powerhouse -
> > JSF Consulting, Development and
> > Courses in English and German
> >
> > Professional Support for Apache MyFaces
> >
>
--
http://www.irian.at
Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German
Professional Support for Apache MyFaces
Re: AJAX & verbatim problem (Was Re: Why is verbatim component transient?)
Posted by Volker Weber <us...@weber-oldenburg.de>.
see: http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
Volker
Sean Schofield wrote:
> Can you give an example of how this verbatim thing is causing you
> problems? I tend to agree with the reasoning that Adam gave on this
> one.
>
> sean
>
> On 11/17/05, Martin Marinschek <ma...@gmail.com> wrote:
>
>>yes, my old friend f:loadBundle.
>>
>>major PITA, this thing ;)
>>
>>both that it is only evaluated in the JSP processing, but there only
>>in the render response phase - shouldn't that happen much earlier?
>>couldn't it be that we need the reference to an I18n text much earlier
>>in the lifecycle?
>>
>>regards,
>>
>>Martin
>>
>>On 11/17/05, Volker Weber <vo...@weber-oldenburg.de> wrote:
>>
>>>I see the problems, but i still think it should be possible to have
>>>verbatim content inside of ajax updated components. And i don't see any
>>>chance to partialy re-execute the relevant jsp files .
>>>
>>>
>>>
>>>> - <c:if> will not re-execute, so if its condition has changed,
>>>
>>>could be replaced by using rendered attribute on jsf tags
>>>
>>>
>>>> - Resource bundles provided by <f:loadBundle> will not be available
>>>
>>>could be solved by a session scoped resource bundle tag
>>>
>>>
>>>> - Any EL expressions evaluated inside tags will not re-evaluate
>>>
>>>I currently don't know of those, but IMHO the evaluation of EL
>>>expressions should moved inside component or renderer than.
>>>
>>>
>>>So, to enable ajax updated panels we need custom tags for verbatim and
>>>loadbundle.
>>>
>>>Any more problems or other solutions ?
>>>
>>>Regards
>>> Volker
>>>
>>>
>>>Adam Winer wrote:
>>>
>>>>This is required by the spec because:
>>>>
>>>> - Performance - in the general case, saving the state for these is a waste
>>>> - Re-evaluation - if you've got, say, ${} EL expressions inside <f:verbatim>,
>>>> or any dynamically generated content from a tag, you'd better re-run the
>>>> JSP or you'll be showing stale content.
>>>>
>>>>The second problem is important and cannot be resolved by changing
>>>>"transient" on f:verbatim.
>>>>
>>>>Note that you've got other, likely bigger problems if you attempt to render
>>>>without re-executing the JSP:
>>>> - <c:if> will not re-execute, so if its condition has changed, ah well
>>>> - Resource bundles provided by <f:loadBundle> will not be available
>>>> - Any EL expressions evaluated inside tags will not re-evaluate
>>>>
>>>>-- Adam
>>>>
>>>>
>>>>
>>>>On 11/16/05, Volker Weber <us...@weber-oldenburg.de> wrote:
>>>>
>>>>
>>>>>Hi,
>>>>>
>>>>>i just came over a problem with verbatim tags in ajax enabled components.
>>>>>see:
>>>>>http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
>>>>>
>>>>>The UIOutput components created for verbatim tags are not saved to state
>>>>>because there are explicitly set to transient.
>>>>>
>>>>>This makes it impossible to use verbatim tags in content of ajax components.
>>>>>
>>>>>Is there a spec reason for this, then we should have a tomahawk
>>>>>verbatim, or is this just because of performance?
>>>>>
>>>>>If not a spec reason this should be changed because seems to be a bug.
>>>>>
>>>>>Regards
>>>>> Volker
>>>>>--
>>>>>Don't answer to From: address!
>>>>>Mail to this account are droped if not recieved via mailinglist.
>>>>>To contact me direct create the mail address by
>>>>>concatenating my forename to my senders domain.
>>>>>
>>>>
>>>>
>>>--
>>>-------------------------------------------------------------------------
>>> Volker Weber Dietrichsweg 38a 26127 Oldenburg Germany
>>> MAILTO:Volker@weber-oldenburg.de HTTP://www.weber-oldenburg.de
>>>
>>
>>
>>--
>>
>>http://www.irian.at
>>
>>Your JSF powerhouse -
>>JSF Consulting, Development and
>>Courses in English and German
>>
>>Professional Support for Apache MyFaces
>>
>
>
--
Don't answer to From: address!
Mail to this account are droped if not recieved via mailinglist.
To contact me direct create the mail address by
concatenating my forename to my senders domain.
Re: AJAX & verbatim problem (Was Re: Why is verbatim component transient?)
Posted by Sean Schofield <se...@gmail.com>.
Can you give an example of how this verbatim thing is causing you
problems? I tend to agree with the reasoning that Adam gave on this
one.
sean
On 11/17/05, Martin Marinschek <ma...@gmail.com> wrote:
> yes, my old friend f:loadBundle.
>
> major PITA, this thing ;)
>
> both that it is only evaluated in the JSP processing, but there only
> in the render response phase - shouldn't that happen much earlier?
> couldn't it be that we need the reference to an I18n text much earlier
> in the lifecycle?
>
> regards,
>
> Martin
>
> On 11/17/05, Volker Weber <vo...@weber-oldenburg.de> wrote:
> > I see the problems, but i still think it should be possible to have
> > verbatim content inside of ajax updated components. And i don't see any
> > chance to partialy re-execute the relevant jsp files .
> >
> >
> > > - <c:if> will not re-execute, so if its condition has changed,
> > could be replaced by using rendered attribute on jsf tags
> >
> > > - Resource bundles provided by <f:loadBundle> will not be available
> > could be solved by a session scoped resource bundle tag
> >
> > > - Any EL expressions evaluated inside tags will not re-evaluate
> > I currently don't know of those, but IMHO the evaluation of EL
> > expressions should moved inside component or renderer than.
> >
> >
> > So, to enable ajax updated panels we need custom tags for verbatim and
> > loadbundle.
> >
> > Any more problems or other solutions ?
> >
> > Regards
> > Volker
> >
> >
> > Adam Winer wrote:
> > > This is required by the spec because:
> > >
> > > - Performance - in the general case, saving the state for these is a waste
> > > - Re-evaluation - if you've got, say, ${} EL expressions inside <f:verbatim>,
> > > or any dynamically generated content from a tag, you'd better re-run the
> > > JSP or you'll be showing stale content.
> > >
> > > The second problem is important and cannot be resolved by changing
> > > "transient" on f:verbatim.
> > >
> > > Note that you've got other, likely bigger problems if you attempt to render
> > > without re-executing the JSP:
> > > - <c:if> will not re-execute, so if its condition has changed, ah well
> > > - Resource bundles provided by <f:loadBundle> will not be available
> > > - Any EL expressions evaluated inside tags will not re-evaluate
> > >
> > > -- Adam
> > >
> > >
> > >
> > > On 11/16/05, Volker Weber <us...@weber-oldenburg.de> wrote:
> > >
> > >>Hi,
> > >>
> > >>i just came over a problem with verbatim tags in ajax enabled components.
> > >>see:
> > >>http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
> > >>
> > >>The UIOutput components created for verbatim tags are not saved to state
> > >>because there are explicitly set to transient.
> > >>
> > >>This makes it impossible to use verbatim tags in content of ajax components.
> > >>
> > >>Is there a spec reason for this, then we should have a tomahawk
> > >>verbatim, or is this just because of performance?
> > >>
> > >>If not a spec reason this should be changed because seems to be a bug.
> > >>
> > >>Regards
> > >> Volker
> > >>--
> > >>Don't answer to From: address!
> > >>Mail to this account are droped if not recieved via mailinglist.
> > >>To contact me direct create the mail address by
> > >>concatenating my forename to my senders domain.
> > >>
> > >
> > >
> >
> > --
> > -------------------------------------------------------------------------
> > Volker Weber Dietrichsweg 38a 26127 Oldenburg Germany
> > MAILTO:Volker@weber-oldenburg.de HTTP://www.weber-oldenburg.de
> >
>
>
> --
>
> http://www.irian.at
>
> Your JSF powerhouse -
> JSF Consulting, Development and
> Courses in English and German
>
> Professional Support for Apache MyFaces
>
Re: AJAX & verbatim problem (Was Re: Why is verbatim component transient?)
Posted by Martin Marinschek <ma...@gmail.com>.
yes, my old friend f:loadBundle.
major PITA, this thing ;)
both that it is only evaluated in the JSP processing, but there only
in the render response phase - shouldn't that happen much earlier?
couldn't it be that we need the reference to an I18n text much earlier
in the lifecycle?
regards,
Martin
On 11/17/05, Volker Weber <vo...@weber-oldenburg.de> wrote:
> I see the problems, but i still think it should be possible to have
> verbatim content inside of ajax updated components. And i don't see any
> chance to partialy re-execute the relevant jsp files .
>
>
> > - <c:if> will not re-execute, so if its condition has changed,
> could be replaced by using rendered attribute on jsf tags
>
> > - Resource bundles provided by <f:loadBundle> will not be available
> could be solved by a session scoped resource bundle tag
>
> > - Any EL expressions evaluated inside tags will not re-evaluate
> I currently don't know of those, but IMHO the evaluation of EL
> expressions should moved inside component or renderer than.
>
>
> So, to enable ajax updated panels we need custom tags for verbatim and
> loadbundle.
>
> Any more problems or other solutions ?
>
> Regards
> Volker
>
>
> Adam Winer wrote:
> > This is required by the spec because:
> >
> > - Performance - in the general case, saving the state for these is a waste
> > - Re-evaluation - if you've got, say, ${} EL expressions inside <f:verbatim>,
> > or any dynamically generated content from a tag, you'd better re-run the
> > JSP or you'll be showing stale content.
> >
> > The second problem is important and cannot be resolved by changing
> > "transient" on f:verbatim.
> >
> > Note that you've got other, likely bigger problems if you attempt to render
> > without re-executing the JSP:
> > - <c:if> will not re-execute, so if its condition has changed, ah well
> > - Resource bundles provided by <f:loadBundle> will not be available
> > - Any EL expressions evaluated inside tags will not re-evaluate
> >
> > -- Adam
> >
> >
> >
> > On 11/16/05, Volker Weber <us...@weber-oldenburg.de> wrote:
> >
> >>Hi,
> >>
> >>i just came over a problem with verbatim tags in ajax enabled components.
> >>see:
> >>http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
> >>
> >>The UIOutput components created for verbatim tags are not saved to state
> >>because there are explicitly set to transient.
> >>
> >>This makes it impossible to use verbatim tags in content of ajax components.
> >>
> >>Is there a spec reason for this, then we should have a tomahawk
> >>verbatim, or is this just because of performance?
> >>
> >>If not a spec reason this should be changed because seems to be a bug.
> >>
> >>Regards
> >> Volker
> >>--
> >>Don't answer to From: address!
> >>Mail to this account are droped if not recieved via mailinglist.
> >>To contact me direct create the mail address by
> >>concatenating my forename to my senders domain.
> >>
> >
> >
>
> --
> -------------------------------------------------------------------------
> Volker Weber Dietrichsweg 38a 26127 Oldenburg Germany
> MAILTO:Volker@weber-oldenburg.de HTTP://www.weber-oldenburg.de
>
--
http://www.irian.at
Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German
Professional Support for Apache MyFaces
AJAX & verbatim problem (Was Re: Why is verbatim component transient?)
Posted by Volker Weber <vo...@weber-oldenburg.de>.
I see the problems, but i still think it should be possible to have
verbatim content inside of ajax updated components. And i don't see any
chance to partialy re-execute the relevant jsp files .
> - <c:if> will not re-execute, so if its condition has changed,
could be replaced by using rendered attribute on jsf tags
> - Resource bundles provided by <f:loadBundle> will not be available
could be solved by a session scoped resource bundle tag
> - Any EL expressions evaluated inside tags will not re-evaluate
I currently don't know of those, but IMHO the evaluation of EL
expressions should moved inside component or renderer than.
So, to enable ajax updated panels we need custom tags for verbatim and
loadbundle.
Any more problems or other solutions ?
Regards
Volker
Adam Winer wrote:
> This is required by the spec because:
>
> - Performance - in the general case, saving the state for these is a waste
> - Re-evaluation - if you've got, say, ${} EL expressions inside <f:verbatim>,
> or any dynamically generated content from a tag, you'd better re-run the
> JSP or you'll be showing stale content.
>
> The second problem is important and cannot be resolved by changing
> "transient" on f:verbatim.
>
> Note that you've got other, likely bigger problems if you attempt to render
> without re-executing the JSP:
> - <c:if> will not re-execute, so if its condition has changed, ah well
> - Resource bundles provided by <f:loadBundle> will not be available
> - Any EL expressions evaluated inside tags will not re-evaluate
>
> -- Adam
>
>
>
> On 11/16/05, Volker Weber <us...@weber-oldenburg.de> wrote:
>
>>Hi,
>>
>>i just came over a problem with verbatim tags in ajax enabled components.
>>see:
>>http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
>>
>>The UIOutput components created for verbatim tags are not saved to state
>>because there are explicitly set to transient.
>>
>>This makes it impossible to use verbatim tags in content of ajax components.
>>
>>Is there a spec reason for this, then we should have a tomahawk
>>verbatim, or is this just because of performance?
>>
>>If not a spec reason this should be changed because seems to be a bug.
>>
>>Regards
>> Volker
>>--
>>Don't answer to From: address!
>>Mail to this account are droped if not recieved via mailinglist.
>>To contact me direct create the mail address by
>>concatenating my forename to my senders domain.
>>
>
>
--
-------------------------------------------------------------------------
Volker Weber Dietrichsweg 38a 26127 Oldenburg Germany
MAILTO:Volker@weber-oldenburg.de HTTP://www.weber-oldenburg.de
Re: Why is verbatim component transient?
Posted by Adam Winer <aw...@gmail.com>.
This is required by the spec because:
- Performance - in the general case, saving the state for these is a waste
- Re-evaluation - if you've got, say, ${} EL expressions inside <f:verbatim>,
or any dynamically generated content from a tag, you'd better re-run the
JSP or you'll be showing stale content.
The second problem is important and cannot be resolved by changing
"transient" on f:verbatim.
Note that you've got other, likely bigger problems if you attempt to render
without re-executing the JSP:
- <c:if> will not re-execute, so if its condition has changed, ah well
- Resource bundles provided by <f:loadBundle> will not be available
- Any EL expressions evaluated inside tags will not re-evaluate
-- Adam
On 11/16/05, Volker Weber <us...@weber-oldenburg.de> wrote:
> Hi,
>
> i just came over a problem with verbatim tags in ajax enabled components.
> see:
> http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
>
> The UIOutput components created for verbatim tags are not saved to state
> because there are explicitly set to transient.
>
> This makes it impossible to use verbatim tags in content of ajax components.
>
> Is there a spec reason for this, then we should have a tomahawk
> verbatim, or is this just because of performance?
>
> If not a spec reason this should be changed because seems to be a bug.
>
> Regards
> Volker
> --
> Don't answer to From: address!
> Mail to this account are droped if not recieved via mailinglist.
> To contact me direct create the mail address by
> concatenating my forename to my senders domain.
>
Re: Why is verbatim component transient?
Posted by Volker Weber <vo...@weber-oldenburg.de>.
It's in the spec :-(.
9.4.17
... and the transient property must be set to true. ...
Any thoughts about having a <t:verbatim> tag?
Volker Weber wrote:
> Hi,
>
> i just came over a problem with verbatim tags in ajax enabled components.
> see:
> http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
>
> The UIOutput components created for verbatim tags are not saved to state
> because there are explicitly set to transient.
>
> This makes it impossible to use verbatim tags in content of ajax components.
>
> Is there a spec reason for this, then we should have a tomahawk
> verbatim, or is this just because of performance?
>
> If not a spec reason this should be changed because seems to be a bug.
>
> Regards
> Volker
--
-------------------------------------------------------------------------
Volker Weber Dietrichsweg 38a 26127 Oldenburg Germany
MAILTO:Volker@weber-oldenburg.de HTTP://www.weber-oldenburg.de
Re: Why is verbatim component transient?
Posted by Simon Kitching <sk...@obsidium.com>.
Volker Weber wrote:
> Hi,
>
> i just came over a problem with verbatim tags in ajax enabled components.
> see:
> http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
>
> The UIOutput components created for verbatim tags are not saved to state
> because there are explicitly set to transient.
>
> This makes it impossible to use verbatim tags in content of ajax components.
>
> Is there a spec reason for this, then we should have a tomahawk
> verbatim, or is this just because of performance?
>
> If not a spec reason this should be changed because seems to be a bug.
>
I would expect that the reason is for performance/memory/bandwidth
usage. As far as I can see, a UIOutput component (including f:verbatim
contents) can simply be recreated during render and discarded after;
there are never any events or model updates generated by an output
component so the component state is irrelevant.
And an f:verbatim component can potentially be very large. It would be a
great waste to store its entire state in memory - or even worse, to
stream it down to the client then back to the server when using
client-side state saving.
By the way, in most apache projects cross-posting to two different mail
lists is regarded as a bad thing. I've replied only to the user list
here; surely all developers will be subscribed to that.
Regards,
Simon
Re: Why is verbatim component transient?
Posted by Adam Winer <aw...@gmail.com>.
This is required by the spec because:
- Performance - in the general case, saving the state for these is a waste
- Re-evaluation - if you've got, say, ${} EL expressions inside <f:verbatim>,
or any dynamically generated content from a tag, you'd better re-run the
JSP or you'll be showing stale content.
The second problem is important and cannot be resolved by changing
"transient" on f:verbatim.
Note that you've got other, likely bigger problems if you attempt to render
without re-executing the JSP:
- <c:if> will not re-execute, so if its condition has changed, ah well
- Resource bundles provided by <f:loadBundle> will not be available
- Any EL expressions evaluated inside tags will not re-evaluate
-- Adam
On 11/16/05, Volker Weber <us...@weber-oldenburg.de> wrote:
> Hi,
>
> i just came over a problem with verbatim tags in ajax enabled components.
> see:
> http://www.mail-archive.com/users%40myfaces.apache.org/msg12124.html
>
> The UIOutput components created for verbatim tags are not saved to state
> because there are explicitly set to transient.
>
> This makes it impossible to use verbatim tags in content of ajax components.
>
> Is there a spec reason for this, then we should have a tomahawk
> verbatim, or is this just because of performance?
>
> If not a spec reason this should be changed because seems to be a bug.
>
> Regards
> Volker
> --
> Don't answer to From: address!
> Mail to this account are droped if not recieved via mailinglist.
> To contact me direct create the mail address by
> concatenating my forename to my senders domain.
>