The QCanvasItem class provides an abstract graphic object on a QCanvas.
.PP
A variety of QCanvasItem subclasses provide immediately usable behaviour. This class is a pure abstract superclass providing the behaviour that is shared among all the concrete canvas item classes. QCanvasItem is not intended for direct subclassing. It is much easier to subclass one of its subclasses, e.g. QCanvasPolygonalItem (the commonest base class), QCanvasRectangle, QCanvasSprite, QCanvasEllipse or QCanvasText.
.PP
Canvas items are added to a canvas by constructing them and passing the canvas to the canvas item's constructor. An item can be moved to a different canvas using setCanvas().
.PP
Items appear on the canvas after their show() function has been called (or setVisible(TRUE)), and \fIafter\fR update() has been called. The canvas only shows items that are visible, and then only if update() is called. If you created the canvas without passing a width and height to the constructor you'll also need to call resize(). Since the canvas background defaults to white and canvas items default to white, you may need to change colors to see your items.
.PP
A QCanvasItem object can be moved in the x(), y() and z() dimensions using functions such as move(), moveBy(), setX(), setY() and setZ(). A canvas item can be set in motion, `animated', using setAnimated() and given a velocity in the x and y directions with setXVelocity() and setYVelocity() -- the same effect can be achieved by calling setVelocity(). Use the collidesWith() function to see if the canvas item will collide on the \fInext\fR advance(1) and use collisions() to see what collisions have occurred.
.PP
Use QCanvasSprite or your own subclass of QCanvasSprite to create canvas items which are animated, i.e. which change over time.
.PP
The size of a canvas item is given by boundingRect(). Use boundingRectAdvanced() to see what the size of the canvas item will be \fIafter\fR the next advance(1) call.
.PP
The rtti() function is used for identifying subclasses of QCanvasItem. The canvas() function returns a pointer to the canvas which contains the canvas item.
.PP
QCanvasItem provides the show() and isVisible() functions like those in QWidget.
.PP
QCanvasItem also provides the setEnabled(), setActive() and setSelected() functions; these functions set the relevant boolean and cause a repaint but the boolean values they set are not used in QCanvasItem itself. You can make use of these booleans in your subclasses.
By default, canvas items have no velocity, no size, and are not in motion. The subclasses provided in TQt do not change these defaults except where noted.
Destroys the QCanvasItem and removes it from its canvas.
.SH "bool QCanvasItem::active () const"
\fBThis function is obsolete.\fR It is provided to keep old source working. We strongly advise against using it in new code.
.PP
Use isActive() instead.
.SH "void QCanvasItem::advance ( int phase )\fC [virtual]\fR"
The default implementation moves the canvas item, if it is animated(), by the preset velocity if \fIphase\fR is 1, and does nothing if \fIphase\fR is 0.
.PP
Note that if you reimplement this function, the reimplementation must not change the canvas in any way, for example it must not add or remove items.
.PP
See also QCanvas::advance() and setVelocity().
.PP
Example: canvas/canvas.cpp.
.PP
Reimplemented in QCanvasSprite.
.SH "bool QCanvasItem::animated () const"
Returns TRUE if the canvas item is in motion; otherwise returns FALSE.
Returns TRUE if the canvas item will collide with the \fIother\fR item \fIafter\fR they have moved by their current velocities; otherwise returns FALSE.
Returns the list of canvas items that this canvas item has collided with.
.PP
A collision is generally defined as occurring when the pixels of one item draw on the pixels of another item, but not all subclasses are so precise. Also, since pixel-wise collision detection can be slow, this function works in either exact or inexact mode, according to the \fIexact\fR parameter.
.PP
If \fIexact\fR is TRUE, the canvas items returned have been accurately tested for collision with the canvas item.
.PP
If \fIexact\fR is FALSE, the canvas items returned are \fInear\fR the canvas item. You can test the canvas items returned using collidesWith() if any are interesting collision candidates. By using this approach, you can ignore some canvas items for which collisions are not relevant.
.PP
The returned list is a list of QCanvasItems, but often you will need to cast the items to their subclass types. The safe way to do this is to use rtti() before casting. This provides some of the functionality of the standard C++ dynamic cast operation even on compilers where dynamic casts are not available.
.PP
Note that a canvas item may be `on' a canvas, e.g. it was created with the canvas as parameter, even though its coordinates place it beyond the edge of the canvas's area. Collision detection only works for canvas items which are wholly or partly within the canvas's area.
.PP
Note that if items have a velocity (see setVelocity()), then collision testing is done based on where the item \fIwill\fR be when it moves, not its current location. For example, a "ball" item doesn't need to actually embed into a "wall" item before a collision is detected. For items without velocity, plain intersection is used.
This abstract virtual function draws the canvas item using \fIpainter\fR.
.PP
\fBWarning:\fR When you reimplement this function, make sure that you leave the painter in the same state as you found it. For example, if you start by calling QPainter::translate(50, 50), end your code by calling QPainter::translate(-50, -50). Be also aware that the painter might already have some transformations set (i.e., don't call QPainter::resetXForm() when you're done).
.PP
Reimplemented in QCanvasSprite, QCanvasPolygonalItem, and QCanvasText.
.SH "bool QCanvasItem::enabled () const"
\fBThis function is obsolete.\fR It is provided to keep old source working. We strongly advise against using it in new code.
.PP
Use isEnabled() instead.
.SH "void QCanvasItem::hide ()"
Shorthand for setVisible(FALSE).
.SH "bool QCanvasItem::isActive () const"
Returns TRUE if the QCanvasItem is active; otherwise returns FALSE.
.SH "bool QCanvasItem::isEnabled () const"
Returns TRUE if the QCanvasItem is enabled; otherwise returns FALSE.
.SH "bool QCanvasItem::isSelected () const"
Returns TRUE if the canvas item is selected; otherwise returns FALSE.
.SH "bool QCanvasItem::isVisible () const"
Returns TRUE if the canvas item is visible; otherwise returns FALSE.
.PP
Note that in this context TRUE does \fInot\fR mean that the canvas item is currently in a view, merely that if a view is showing the area where the canvas item is positioned, and the item is not obscured by items with higher z values, and the view is not obscured by overlaying windows, it would be visible.
.PP
See also setVisible() and z().
.SH "void QCanvasItem::move ( double x, double y )"
Moves the canvas item to the absolute position (\fIx\fR, \fIy\fR).
.PP
Example: canvas/canvas.cpp.
.SH "void QCanvasItem::moveBy ( double dx, double dy )\fC [virtual]\fR"
Moves the canvas item relative to its current position by (\fIdx\fR, \fIdy\fR).
Make your derived classes return their own values for rtti(), so that you can distinguish between objects returned by QCanvas::at(). You should use values greater than 1000 to allow for extensions to this class.
.PP
Overuse of this functionality can damage it's extensibility. For example, once you have identified a base class of a QCanvasItem found by QCanvas::at(), cast it to that type and call meaningful methods rather than acting upon the object based on its rtti value.
.PP
For example:
.PP
.nf
.br
QCanvasItem* item;
.br
// Find an item, e.g. with QCanvasItem::collisions().
.br
...
.br
if (item->rtti() == MySprite::RTTI ) {
.br
MySprite* s = (MySprite*)item;
.br
if (s->isDamagable()) s->loseHitPoints(1000);
.br
if (s->isHot()) myself->loseHitPoints(1000);
.br
...
.br
}
.br
.fi
.PP
Example: canvas/canvas.cpp.
.PP
Reimplemented in QCanvasSprite, QCanvasPolygonalItem, and QCanvasText.
.SH "bool QCanvasItem::selected () const"
\fBThis function is obsolete.\fR It is provided to keep old source working. We strongly advise against using it in new code.
Sets the active flag of the item to \fIyes\fR. If this changes the item's active state the item will be redrawn when QCanvas::update() is next called.
.PP
The QCanvas, QCanvasItem and the Qt-supplied QCanvasItem subclasses do not make use of this value. The setActive() function is supplied because many applications need it, but it is up to you how you use the isActive() value.
.SH "void QCanvasItem::setAnimated ( bool y )\fC [virtual]\fR"
Sets the canvas item to be in motion if \fIy\fR is TRUE, or not if \fIy\fR is FALSE. The speed and direction of the motion is set with setVelocity(), or with setXVelocity() and setYVelocity().
.PP
See also advance() and QCanvas::advance().
.SH "void QCanvasItem::setCanvas ( QCanvas * c )\fC [virtual]\fR"
Sets the QCanvas upon which the canvas item is to be drawn to \fIc\fR.
Sets the enabled flag of the item to \fIyes\fR. If this changes the item's enabled state the item will be redrawn when QCanvas::update() is next called.
.PP
The QCanvas, QCanvasItem and the Qt-supplied QCanvasItem subclasses do not make use of this value. The setEnabled() function is supplied because many applications need it, but it is up to you how you use the isEnabled() value.
Sets the selected flag of the item to \fIyes\fR. If this changes the item's selected state the item will be redrawn when QCanvas::update() is next called.
.PP
The QCanvas, QCanvasItem and the Qt-supplied QCanvasItem subclasses do not make use of this value. The setSelected() function is supplied because many applications need it, but it is up to you how you use the isSelected() value.