CairoPathData
is used to represent the path data inside a
CairoPath
. The data structure is designed to try to balance the demands of
efficiency and ease-of-use. A path is represented as a list of
CairoPathData
, which is a union of headers and points. Each portion of the path is represented by one or more elements in
the list, (one header followed by 0 or more points). The length
value of the header is the number of list elements for the current
portion including the header, (ie. length == 1 + # of points), and
where the number of points for each element type is as follows:
%CAIRO_PATH_MOVE_TO: 1 point
%CAIRO_PATH_LINE_TO: 1 point
%CAIRO_PATH_CURVE_TO: 3 points
%CAIRO_PATH_CLOSE_PATH: 0 points
The semantics and ordering of the coordinate values are consistent
with cairoMoveTo
, cairoLineTo
, cairoCurveTo
, and
cairoClosePath
. Here is sample code for iterating through a ""
path <- cr$copyPath()$data for (data in path) {
switch(CairoPathDataType[attr(data, "type") + 1L],
"move-to" = do_move_to_things(data[1], data[2]),
"line-to" = do_line_to_things(data[1], data[2]),
"curve-to" = do_curve_to_things(data[1], data[2], data[3], data[4],
data[5], data[6]),
"close-path" = do_close_path_things())
}
As of cairo 1.4, cairo does not mind if there are more elements in
a portion of the path than needed. Such elements can be used by
users of the cairo API to hold extra values in the path data
structure. For this reason, it is recommended that applications
always use data->header.length
to
iterate over the path data, instead of hardcoding the number of
elements for each element type.
CairoPathData
is a transparent-type.