During an the effect's describe in context action an effect
must
define the clips mandated for that context, it can also define extra clips that it may need for that context. It does this using the
OfxImageEffectSuiteV1::clipDefine
function, the property handle returned by this function is purely for definition purposes only. It has not persistance outside the describe in context action and is distinct to the clip property handles used by instances. The
name
parameter is how you can later access that clip in a plugin instance via the
OfxImageEffectSuiteV1::clipGetHandle
function.
During the describe in context action, the plugin sets properties on a clip to control its use. The properties that can be set during a describe in context call are...
kOfxPropLabel
to give a user readable name to the clip (the host need not use this, for example in a transition it is redundant),
kOfxImageEffectPropSupportedComponents
to specify which components it is willing to accept on that clip,
kOfxImageClipPropOptional
to specify if the clip is optional,
kOfxImageClipPropFieldExtraction
specifies how to extract fielded images from a clip, see \ref ImageEffectsFieldRendering for more details on field and field rendering,
kOfxImageEffectPropTemporalClipAccess
whether the effect wants to access images from the clip at times other that the frame being renderred.
Plugins
must
indicate which pixel depths they can process by setting the
kOfxImageEffectPropSupportedPixelDepths
on the plugin handle during the describe action.
Pixel Aspect Ratios, frame rates, fielding, components and pixel depths are constant for the duration of a clip, they cannot changed from frame to frame.
kOfxImageEffectPropSupportedPixelDepths
plugin property during its describe action
kOfxImageEffectPropSupportedComponents
on an input clip during describe in context