leadimage and richtext have factories and point to the wrong marker-interfaces

[iham]

according to the docs on plone (see quote below) a behavior (LeadImage) implementing a factory "needs" a marker interface different from the behavior interface (ILeadImage)

this is currently not registered/implemented and i stumbled into troubles after i wanted to implement a different factory for the ILeadImage behavior for a certain custom contenttype using this behavior.

i will fix and commit that and hope for a fast release on this one, as it will not break any tests

plone.behavior’s marker support can be used in two ways:
As the behavior interface itself. In this case, there is no behavior adapter factory. The behavior interface and the marker interface are one and the same.
As a supplement to a standard behavior adapter. In this case, a factory is provided, and the behavior interface (which the behavior adapter factory implements) is different to the marker interface.

[jensens]

It looks like the LeadImage factory is never instantiated:

plone.app.contenttypes/plone/app/contenttypes/behaviors/leadimage.py

Lines 29 to 34 in e36df79

	@implementer(ILeadImage)
	@adapter(IDexterityContent)
	class LeadImage(object):
	def __init__(self, context):
	self.context = context

Also there are no getters/setters implemented.
Afaik, if it would be used it would not work.

[iham]

i fixed that
look here - getters and setters ;)

tests of leadimage are running fine and also fixed a failing referenceablebehavior import error

i also "fixed" the collection test but this seems kinda odd
look here
and here

i would love to see that merged and released asap, as i would like to move on with tasks i really need working based on that fix

[pbauer]

Great, create a pull-request and run the tests in jenkins. Then we'll review and merge.