Re: [Framework-Team] Re: hard dependency on PIL?

2006-09-12 Thread Helge Tesdal
On 3:12 pm 09/12/06 Martin Aspeli <[EMAIL PROTECTED]> wrote:
> I think a dependency is probably OK, even though it would annoy me to
> have to do it for all development instances. The swinger for me is
> the recent security problems that we've had to use PIL to get around.

Is it possible to disable member portraits if PIL is not installed?


___
Framework-Team mailing list
Framework-Team@lists.plone.org
http://lists.plone.org/mailman/listinfo/framework-team


Re: [Framework-Team] Re: hard dependency on PIL?

2006-09-12 Thread Alec Mitchell

On 9/12/06, Helge Tesdal <[EMAIL PROTECTED]> wrote:

On 3:12 pm 09/12/06 Martin Aspeli <[EMAIL PROTECTED]> wrote:
> I think a dependency is probably OK, even though it would annoy me to
> have to do it for all development instances. The swinger for me is
> the recent security problems that we've had to use PIL to get around.

Is it possible to disable member portraits if PIL is not installed?


It's not possible to start Plone if PIL is not installed currently
(due to the member image fix).  PIL is included in all the installers
AFAIK, and is a package in every distro I've known.  So installing PIL
is generally as easy as installing python (whether you use your distro
packages or sompile from source).  It can also be as easy as doing:

easy_install -f http://www.pythonware.com/products/pil/ Imaging

If you don't want to use distro packages.  If someone can point to a
sane way to make the import conditional without reopening the spam
hole, that would be great.  But, it's not really possible AFAICT to
tell whether member portraits are going to be a problem or not at
startup time.  Perhaps we can have a config flag that when disabled
makes the image check/transform a no-op, but that doesn't seem too
helpful as it still requies monkeying with the source.  The spam issue
is too serious to allow the fix to be disabled automatically if PIL is
not present (whatever warning we show during startup will likely be
ignored).

Alec

___
Framework-Team mailing list
Framework-Team@lists.plone.org
http://lists.plone.org/mailman/listinfo/framework-team


Re: [Framework-Team] Re: hard dependency on PIL?

2006-09-12 Thread Wichert Akkerman
Previously Alec Mitchell wrote:
> It's not possible to start Plone if PIL is not installed currently
> (due to the member image fix).  PIL is included in all the installers
> AFAIK, and is a package in every distro I've known.  So installing PIL
> is generally as easy as installing python (whether you use your distro
> packages or sompile from source).  It can also be as easy as doing:
> 
> easy_install -f http://www.pythonware.com/products/pil/ Imaging

That is actually not true: the success of that command is bound to be
highly dependant on the development packages of the various graphic
toolks you have installed.

Wichert.

-- 
Wichert Akkerman <[EMAIL PROTECTED]>It is simple to make things.
http://www.wiggy.net/   It is hard to make things simple.

___
Framework-Team mailing list
Framework-Team@lists.plone.org
http://lists.plone.org/mailman/listinfo/framework-team


Re: [Framework-Team] Re: hard dependency on PIL?

2006-09-13 Thread Alec Mitchell

On 9/13/06, Raphael Ritz <[EMAIL PROTECTED]> wrote:

Wichert Akkerman schrieb:
> Previously Alec Mitchell wrote:
>> It's not possible to start Plone if PIL is not installed currently
>> (due to the member image fix).  PIL is included in all the installers
>> AFAIK, and is a package in every distro I've known.  So installing PIL
>> is generally as easy as installing python (whether you use your distro
>> packages or sompile from source).  It can also be as easy as doing:
>>
>> easy_install -f http://www.pythonware.com/products/pil/ Imaging
>
> That is actually not true: the success of that command is bound to be
> highly dependant on the development packages of the various graphic
> toolks you have installed.
>

that's what I meant by saying earlier that PIL isn't necessarily
trivial to install. But anyway, I consider my original question
answered: it wasn't introduced on purpose in the first place but
now that a security-related issue depends on it anyway (the
portrait checking) people are willing to accept this.


When I introduced the dependency for the portrait fix in 2.5 and 2.1
it was fully intentional.

Alec

___
Framework-Team mailing list
Framework-Team@lists.plone.org
http://lists.plone.org/mailman/listinfo/framework-team