General OLE Concepts
One of the most attractive aspects of OLE is that it is completely modular. It
is designed so that each component can exist, for the most part, on its own.
For example, you can add drag and drop support to your application without
adding in-place activation, automation, or compound storage.
Even if your application only uses a small part of OLE, the Component Object
Model (COM) enables other applications to know what your application's OLE
capabilities are. COM is the fundamental, underlying model that OLE is based upon;
all OLE objects are also component objects.
COM stipulates that any component object must control its own life span and be
able to tell other objects about its capabilities in a strictly defined
manner. To control its life span, a component object maintains a reference count.
Capabilities are grouped into logical sets called interfaces; each interface is a
set of member functions necessary to support a certain capability. The
"strictly defined manner" that component objects must use is itself an interface, which
is called
IUnknown. Because all OLE interfaces are derived from
IUnknown, they are component objects.
IUnknown has three member functions:
QueryInterface,
AddRef, and
Release.
An object uses
QueryInterface to tell other objects about its capabilities. If the object implements the
requested interface, it returns a pointer to the interface. If it does not
implement it, it returns the E_NOTIMPL error value stating that the object does not
support the requested interface.
AddRef and
Release are used to control the object's life span. An object's
AddRef member function is called when another object holds a pointer to the object,
and the
Release member function is called when the pointer is no longer needed. If a call to
Release causes the object's reference count to go to zero, the object can safely
unload itself.
COM provides a couple of immediate benefits:
- An object can determine in advance if another object supports a certain
feature. If the other object does not support the feature, the calling object can
react accordingly.
- Objects do not remain in memory longer (or shorter) than necessary, and they
do not rely on the user to launch or close them.
The OLE data transfer mechanism is a crucial element of drag and drop support.
Data transfer in OLE allows objects to be very specific about the data that
they transfer. Instead of simply being able to transfer a plain old bitmap, an
object can now transfer a bitmap of the object's contents rendered for a printer
device and stored in a stream to be released by OLE.
To accomplish this, OLE uses the
IDataObject interface and the
FORMATETC and
STGMEDIUM structures. Applications implement
IDataObject to accomplish all data transfer in OLE; it includes member functions that set
and retrieve an object's data, enumerate the available data formats, and
receive data change notifications.
FORMATETC and
STGMEDIUM provide the specific details about the data that is being transferred
that is, the target device, aspect, storage medium, and release method.
Every drag and drop operation involves two objects: a source and a target. The
source object contains the data to be dragged, and the target object accepts
the dragged data.
- Software for developers
-
Delphi Components
.Net Components
Software for Android Developers
- More information resources
-
MegaDetailed.Net
Unix Manual Pages
Delphi Examples
- Databases for Amazon shops developers
-
Amazon Categories Database
Browse Nodes Database
