Re: Session creation due to session-map access

[Christian Beikov]

I see, sounds great, thank you!

Am 15.06.2020 um 21:24 schrieb Thomas Andraschko:

Cant remember the case but we introduced the config to not break existing
apps. in 2.2 and 2.3 its activated by default.
We deactivated it in trunk however.


Virenfrei.
www.avast.com

<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

Am Mo., 15. Juni 2020 um 21:10 Uhr schrieb Christian Beikov <
christian.bei...@gmail.com>:


Thanks, just found that myself as well. What is the reason for forcing
the session creation by default even if the view is transient? Since
most views are probably non-transient, I'd guess that in these cases
session will be created anyway.

Am 15.06.2020 um 20:57 schrieb Thomas Andraschko:

Hi,

Please see
https://issues.apache.org/jira/plugins/servlet/mobile#issue/MYFACES-4297


Christian Beikov  schrieb am Mo., 15. Juni
2020, 20:19:


Hello,

during debugging, I found out that in


org.apache.myfaces.view.facelets.FaceletViewDeclarationLanguage#getResponseEncoding

the statement sm.put(CHARACTER_ENCODING_KEY, encoding); creates a
session if none exists. I doubt that this is desired, at least in my
case it isn't. Also, I don't see why this has to be done at all.

Any hints? Or is this just a bug?

Regards,

Christian

Re: Session creation due to session-map access

[Thomas Andraschko]

Cant remember the case but we introduced the config to not break existing
apps. in 2.2 and 2.3 its activated by default.
We deactivated it in trunk however.


Virenfrei.
www.avast.com

<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

Am Mo., 15. Juni 2020 um 21:10 Uhr schrieb Christian Beikov <
christian.bei...@gmail.com>:

> Thanks, just found that myself as well. What is the reason for forcing
> the session creation by default even if the view is transient? Since
> most views are probably non-transient, I'd guess that in these cases
> session will be created anyway.
>
> Am 15.06.2020 um 20:57 schrieb Thomas Andraschko:
> > Hi,
> >
> > Please see
> > https://issues.apache.org/jira/plugins/servlet/mobile#issue/MYFACES-4297
> >
> >
> > Christian Beikov  schrieb am Mo., 15. Juni
> > 2020, 20:19:
> >
> >> Hello,
> >>
> >> during debugging, I found out that in
> >>
> org.apache.myfaces.view.facelets.FaceletViewDeclarationLanguage#getResponseEncoding
> >>
> >> the statement sm.put(CHARACTER_ENCODING_KEY, encoding); creates a
> >> session if none exists. I doubt that this is desired, at least in my
> >> case it isn't. Also, I don't see why this has to be done at all.
> >>
> >> Any hints? Or is this just a bug?
> >>
> >> Regards,
> >>
> >> Christian
> >>
> >>
>

Re: Session creation due to session-map access

[Christian Beikov]

Thanks, just found that myself as well. What is the reason for forcing 
the session creation by default even if the view is transient? Since 
most views are probably non-transient, I'd guess that in these cases 
session will be created anyway.


Am 15.06.2020 um 20:57 schrieb Thomas Andraschko:

Hi,

Please see
https://issues.apache.org/jira/plugins/servlet/mobile#issue/MYFACES-4297


Christian Beikov  schrieb am Mo., 15. Juni
2020, 20:19:


Hello,

during debugging, I found out that in
org.apache.myfaces.view.facelets.FaceletViewDeclarationLanguage#getResponseEncoding

the statement sm.put(CHARACTER_ENCODING_KEY, encoding); creates a
session if none exists. I doubt that this is desired, at least in my
case it isn't. Also, I don't see why this has to be done at all.

Any hints? Or is this just a bug?

Regards,

Christian

Re: Session creation due to session-map access

[Thomas Andraschko]

Hi,

Please see
https://issues.apache.org/jira/plugins/servlet/mobile#issue/MYFACES-4297


Christian Beikov  schrieb am Mo., 15. Juni
2020, 20:19:

> Hello,
>
> during debugging, I found out that in
> org.apache.myfaces.view.facelets.FaceletViewDeclarationLanguage#getResponseEncoding
>
> the statement sm.put(CHARACTER_ENCODING_KEY, encoding); creates a
> session if none exists. I doubt that this is desired, at least in my
> case it isn't. Also, I don't see why this has to be done at all.
>
> Any hints? Or is this just a bug?
>
> Regards,
>
> Christian
>
>