About Segments and Retained Graphics (1.2)
◄Using Section► ◄Function Group► ◄Up► ◄Next► ◄Previous►
────────────────────────────────────────────────────────────────────────────
About Segments and Retained Graphics
This topic describes retained graphics and graphics segments. You should
also be familiar with the following topics:
◄Presentation spaces and device contexts►
◄Coordinate spaces and transformations►
◄Line and arc primitives►
◄Color and mix modes►
There are two kinds of graphics output in MS OS/2: retained and nonretained.
Most developers are familiar with the latter. When an application draws
non-retained output by calling a graphics function, the output appears
immediately on a video display. If part of the picture in the display is
erased, or if part of the picture is repeated in another location on the
display, the application calls the same graphics functions a second time.
The fundamental drawback of non-retained graphics in drawing or
computer-aided design (CAD) applications is obvious: These applications
frequently draw pictures and parts of pictures repeatedly, so it is
important that they be able to store the primitives used to draw the
pictures. Using retained graphics, however, the developer can store graphics
primitives and redraw their associated output as necessary. Rather than
calling the individual graphics functions each time it is necessary to
redraw the picture, the application can make a single call to display the
retained graphics. In MS OS/2, applications store retained graphics in
segments.
Segments, Elements, and Graphics Orders
A segment is a collection of elements, each element containing one or more
graphics orders. A presentation space can have no more than 16,378 segments.
A graphics order is a low-level graphics command that corresponds to a
graphics function or an attribute function. There are four graphics-order
sizes in MS OS/2:
♦ 1-byte orders
♦ 2-byte orders
♦ Long orders (maximum length of 256 bytes)
♦ Very long orders (maximum length of 64K bytes)
A 1-byte order contains a hexadecimal identifier corresponding to a graphics
function or an attribute function. A 2-byte order contains a hexadecimal
identifier in the first byte and data in the second byte. The long and very
long orders each contain a hexadecimal identifier that corresponds to a
graphics-drawing or attribute function, a length value that specifies how
many bytes are used by the graphics-order arguments, and the actual
arguments. The following example shows a long graphics order that
corresponds to the GpiLine function:
81 8 100 0 0 0 100 0 0 0
The first number, 81, is the hexadecimal identifier that corresponds to
GpiLine. The second number, 8, is the length value that specifies how many
bytes are used by the graphics-order arguments. The next eight bytes contain
the arguments for GpiLine. In this case, these arguments specify the line's
end point at (100,100).
Segments and Subpictures
In most cases, graphics orders in a segment correspond to a subpicture,
which is part of a complete, more complex picture. For example, an
architectural drafting application that draws the layout of a room could
store the graphics orders for a desk in one segment, the graphics orders for
a chair in another segment, the graphics orders for a plant in a third
segment, and so on. Each of these segments contains a subpicture, which the
application combines with other subpictures to form the complete room
diagram.
Types of Segments
There are two types of segments: chained segments and called segments.
MS OS/2 links chained segments together in a segment chain. Chained segments
are sometimes called root segments. Each presentation space can contain only
one segment chain. Applications use chains to generate complete pictures
from subpictures. When MS OS/2 draws subpictures in a chain, it draws the
subpicture for the first segment in the chain, followed by the subpicture
for the second segment in the chain, and so on. Sometimes, it is necessary
to alter the order in which the subpictures are drawn. You can do this by
calling the GpiSetSegmentPriority function. You can also draw the entire
chain, part of the chain, or a single segment in the chain by calling,
respectively, the GpiDrawChain, GpiDrawFrom, or GpiDrawSegment function.
A dynamic segment is a chained segment that possesses special properties.
When MS OS/2 draws subpictures associated with dynamic segments, the XOR
raster operation is set. An application can move the subpicture associated
with a dynamic segment without destroying other subpictures in nondynamic
segments. Applications draw dynamic segments by calling the GpiDrawDynamics
function. Applications can remove the subpictures associated with dynamic
segments by calling the GpiRemoveDynamics function.
Called segments are not linked to the chain. Instead, applications draw them
by calling the GpiCallSegmentMatrix function from within a chained segment
or another called segment.
Segment Attributes
Segments, whether chained or called, have characteristics, called
attributes, that you can set and change according to what your application
needs. There are seven segment attributes, each described in the following
list:
Attribute Description
────────────────────────────────────────────────────────────────────────────
Detectable If the detectable and visible attributes are set, your
application can perform correlation operations on
segments created in its presentation space.
Visible If the visible attribute is set, the GpiDrawChain,
GpiDrawFrom, and GpiDrawSegment functions will
generate output on a device.
Chained If the chained attribute is set, MS OS/2 adds each new
segment in your application's presentation space to
the segment chain.
Dynamic If the dynamic attribute is set, MS OS/2 draws segment
output using the XOR raster operation.
Fast chain If the fast-chain attribute is set, MS OS/2 does not
reset the primitive attributes to their default values
before drawing the segment.
Propagate detectable If the propagate-detectable attribute is set, MS OS/2
treats any called segments as though the detectable
attribute were set for those segments.
Propagate visible If the propagate-visible attribute is set, MS OS/2
treats any called segments as though the visible
attribute were set for those segments.
When an application creates a segment in a presentation space, MS OS/2
assigns initial attributes to it. If you haven't altered the initial
attributes with the GpiSetInitialSegmentAttrs function, five of the
attributes will be set and two of the attributes will not be set. The
following list describes which attributes are set and which are not:
Attribute Default setting
────────────────────────────────────────────────────────────────────────────
Detectable OFF
Visible ON
Chained ON
Dynamic OFF
Fast chain ON
Propagate detectable ON
Propagate visible ON
You can retrieve the values of the current initial attributes by calling the
GpiQueryInitialSegmentAttrs function.
After you create a segment, you may need to alter its attributes. For
example, if you created a segment using the default attributes and you want
to perform a correlation operation on the subpicture in that segment, you'll
need to set the detectable attribute by calling the GpiSetSegmentAttrs
function. You can retrieve the values of the attributes for any segment by
calling the GpiQuerySegmentAttrs function.
Storing Graphics Orders in Segments
There are three drawing modes that affect how MS OS/2 stores graphics orders
in segments. These modes are described in the following list:
Drawing mode Description
────────────────────────────────────────────────────────────────────────────
Draw When the draw mode is set, it is not possible to store
graphics orders in a chained segment.
Retain When the retain mode is set, your application can store
graphics orders in chained and unchained segments.
Draw-and-retain When the draw-and-retain mode is set, your application can
store graphics orders in chained and unchained segments. In
this mode, output intended for a chained segment is both
drawn on the device and stored in a segment.
When you create a presentation space, the drawing mode is set to draw. You
can set the drawing mode to retain or draw-and-retain by calling the
GpiSetDrawingMode function. You can determine which drawing mode is set by
calling the GpiQueryDrawingMode function.
Creating Segments
In MS OS/2, applications identify segments with long integer values greater
than zero. You can determine which values have already been assigned to
segments by calling the GpiQuerySegmentNames function. This function
retrieves the current segment identifiers that your application is using.
Once you determine which values are assigned, you can choose a new value for
the segment you are about to create. The first function you should call
after you choose a segment identifier is GpiOpenSegment, which defines the
beginning of a segment bracket. A segment bracket is a collection of
graphics-drawing and attribute functions that MS OS/2 converts into graphics
orders and stores in the segment. The graphics orders in a segment are
organized into elements, each containing one or more graphics orders. Once
you have called the necessary primitives and attribute functions, you should
close the segment bracket by calling the GpiCloseSegment function.
Destroying Segments
Once you are through drawing the subpicture associated with a segment, you
should delete the segment by calling the GpiDeleteSegment function. If you
are through drawing a number of subpictures in the segment chain, you can
delete an entire range of segments by calling the GpiDeleteSegments
function. In both cases, you use the segment identifiers to identify the
segment or range of segments that are no longer useful.
Correlation
Correlation is the process of determining whether graphic output appears in
a particular region, called a pick aperture, in your application's page
space. The pick aperture is a rectangular region that you can use to isolate
part or all of a subpicture or group of subpictures. To perform correlation
operations, you must set the segment attributes to detectable and visible by
calling the GpiSetSegmentAttrs function.
If the drawing mode is retain or draw-and-retain, you can set the
pick-aperture position by calling the GpiCorrelateSegment, GpiCorrelateFrom,
or GpiCorrelateChain function. If the drawing mode is nonretained, you can
set the pick-aperture position by calling the GpiSetPickAperturePosition
function, retrieve the page-space coordinates of the center of the aperture
by calling the GpiQueryPickAperturePosition function, increase or reduce the
pick-aperture dimensions by calling the GpiSetPickApertureSize function, and
determine the current pick-aperture dimensions by calling the
GpiQueryPickApertureSize function. When you perform a correlation operation,
MS OS/2 returns an array of long-integer pairs. The first value in the
long-integer pair identifies a segment containing a subpicture that
intersects the pick aperture. The second value in this pair identifies a
tag. A tag identifies the specific primitives and associated graphics orders
that intersect the pick aperture.
If your application will perform correlation operations, you must assign
tags by calling the GpiSetTag function. Typically, applications assign tags
only to elements that correspond to primitives. You can determine the value
of the last tag assigned to an element by calling the GpiQueryTag function.
You can perform a correlation operation on the entire segment chain, part of
the chain, or a single segment in the chain by calling, respectively, the
GpiCorrelateChain, GpiCorrelateFrom, or GpiCorrelateSegment function.
Editing Segments
MS OS/2 provides segment-editing functions you can use to write applications
that allow users to edit segments or elements in a segment. After performing
a correlation operation using your application, a user may need to alter the
elements that intersected the pick aperture. If your application takes
advantage of the segment-editing capabilities, it should assign a label (in
addition to a tag) to each element that corresponds to a graphics primitive
in the segment. You assign a label, which is a long integer value, by
calling the GpiLabel function. You can access elements in a segment with the
element pointer by setting the element pointer so that it points to an
element identified by a label. You can set the element pointer so that it
points to an element identified by a label by calling the
GpiSetElementPointerAtLabel function. You can also set the element pointer
by adding or subtracting an offset from its current location by calling the
GpiOffsetElementPointer function. Or you can set the element pointer by
specifying an element number in the segment and calling the
GpiSetElementPointer function. You can determine the current location of the
element pointer by calling the GpiQueryElementPointer function.
There are two edit modes in MS OS/2: insert mode and replace mode. You can
set the edit modes by calling the GpiSetEditMode function, or you can
determine which mode is currently set by calling the GpiQueryEditMode
function. If the edit mode is set to insert, you can insert an element at
the current location of the element pointer. MS OS/2 shifts the element that
was previously at that location into the next slot, and so on, until the
last element is shifted into a new, final slot.
If replace mode is set, you can replace the element at the current pointer
location with a new element.
In addition to inserting and replacing elements in a segment, you can delete
elements from a segment. You can delete a single element at the current
element-pointer location by calling the GpiDeleteElement function. You can
delete a range of elements in a segment by calling the
GpiDeleteElementRange function, and you can delete a series of elements
between two labels by calling the GpiDeleteElementsBetweenLabels function.
You can copy elements from one segment to another or from one position in a
segment to another position in the same segment by calling the GpiGetData
and the GpiPutData functions. GpiGetData copies one or more graphics orders
from one or more elements in a segment into a buffer of bytes. You can then
copy the order from the buffer to a new location in a segment by calling
GpiPutData. In addition to copying multiple elements from one segment to
another, or from one segment to another location in the same segment, you
can copy the graphics orders from a single element by calling the
GpiQueryElement and GpiElement functions. GpiQueryElement copies the
graphics orders from an element into an array of bytes. GpiElement copies
the orders from the array back into a segment.
♦