Package 'iplots'

Add a straight line to the current iPlot.

Description

This function adds one straight line through the current iPlot.

Usage

iabline(a = NULL, b = NULL, reg = NULL, coef = NULL, ..., plot = iplot.cur())

Arguments

a, b	the intercept and slope
coef	a vector of length two giving the intercept and slope
reg	an object with a coef component
...	object options for iobj.opt
plot	plot to which this line is to be added

Details

Typical usages are

  iabline(a, b, ...)
  iabline(coef=, ...)
  iabline(reg=, ...)

Value

Resulting iObject.

See Also

ilines, iobj.opt

---

Interactive Bar Chart

Description

This function creates a new interactive bar chart from the given data.

Usage

ibar(var, ...)

Arguments

var	Factor, vector of strings or variable to use.
...	All additional parameters are passed to iplot.opt.

Details

Creates an interactive barchart or spline plot.

Additional parameters (also accessible via iplot.opt):

drawTicks

Whether ticks should be drawn.

isSpine

Whether a spineplot should be created instead of a normal barchart.

borderColorSel

Here this color defaults to black. See iplot.opt for details.

Value

Resulting plot object.

See Also

iplot, ihist, iplot.list, iplot.opt

Examples

data(iris)
attach(iris)
ibar(Species)

---

Interactive Box Plot

Description

This function creates a new interactive box plot from the given data.

Usage

ibox(x, y=NULL, ...)

Arguments

x	Vector of numbers or data frame containing the variables
y	A factor to specify groups for y-by-x boxplot
...	All additional parameters are passed to iplot.opt. Many of the parameters used in plot are supported.

Details

Creates either a regular boxplot (if y is not specified) or 'x-by-y boxplot (if y is specified). In the latter case both x and y must be of the same length.

Additional parameters:

alterningLabels

Whether lables should be placed alternately at the top and at the bottom.

Value

Resulting plot object.

See Also

ihist, ibar, iplot.list, iplot.opt

Examples

data(iris)
attach(iris)
ibox(Petal.Length)
ibox(Sepal.Length,Species)

---

Interactive events

Description

This function provides a way of building interactive event loops in R. Currently SJava interface has no stable callbacks, therefore this is the only way of defining new interactions in iPlots.

ievent.wait waits until some iPlots event (iEvent) occurs. The most commonly used events include selection change and the "break" event. It is possible to generate user-definable events (such as events sent by custom buttons or menu entries) with low-level API (via SJava)

Note that in the current implementation the Windows GUI "freezes" until the function returns.

Usage

ievent.wait()

Value

NULL is the break event occured or a ievent object if some other event occured.

Examples

data(iris)
  attach(iris)
  iplot(Sepal.Length, Petal.Length)

  d<-iplot.data()

  iabline(lm(d$y ~ d$x), col = "red")
  ilines(lowess(d$x,d$y), col = "#0000c0")
  ilines(c(0,0),c(0,0), col = "marked", visible = FALSE)

  cat("Select 'Break' from the menu of any plot to return back to R.\n")

  while (!is.null(ievent.wait())) {
    if (iset.sel.changed()) {
      s <- iset.selected()
      if (length(s) > 1) 
        iobj.opt(x=lowess(d$x[s],d$y[s]),visible = TRUE)
      else iobj.opt(visible = FALSE)
    }
  }
  for(i in 1:3) iobj.rm()
  iplot.off()

---

Interactive Hammock Plot

Description

This function creates a new interactive hammock plot. Please note that a hammock plot was not necessarily designed to support highlighting, so it may be of limited use. It was created as a proof of the iBase concept which makes it very easy to implement new interactive plots.

Usage

ihammock(...)

Arguments

...	All unnamed parameters are treated as variables to display. All additional parameters are passed to iplot.opt.

Details

The plot can be used in either in the form ihammock(df) to plot a list or data frame of variables or ihammock(x, y, z) to plot factors x, y and z.

Value

Resulting plot object.

See Also

ihist, ibar, iplot, iplot.opt

Examples

library(MASS)
data(Cars93)
attach(Cars93)
ihammock(AirBags, Cylinders, Origin)

---

Interactive Histogram

Description

This function creates a new interactive histogram from the given data.

Usage

ihist(var, ...)

Arguments

var	Vector of numbers or variable to use.
...	All additional parameters are passed to iplot.opt.

Details

Creates an interactive histogram.

Additional parameters (also available via iplot.opt):

anchor

Anchor point for the histogram.

binw

Bin width.

autoScaleXAxis

Whether the x axis should be automatically rescaled.

autoScaleYAxis

Whether the y axis should be automatically rescaled.

Value

Resulting plot object.

See Also

iplot, ibar, iplot.list, iplot.opt

Examples

data(iris)
ihist(iris$Sepal.Width)

---

Add connected lines or polygon to the current iPlot.

Description

A generic function taking coordinates of points in data space and creating corresponding connected lines or polygon in the current iPlot.

Usage

ilines(x, y, col=NULL, fill=NULL, visible=NULL, plot = iplot.cur())

Arguments

x, y	Coordinate vectors of points to join. xy.coords is used to obtain the coordinates for plotting.
col	Drawing color of the lines. Currently only "#rrggbb" notation and named colors are supported. NULL means that color of the object is not explicitely specified.
fill	Color of the polygon area or NA if no filling should be performed.
visible	If set to FALSE the lines/polygon won't be visible initially.
plot	parent plot for the lines

Details

The point vectors x and y can contain NA values, in which case each sequence of points separated by NAs will be treated as a separate polygon. NAs must be present in both coordinates at the same index, otherwise the behavior is undefined.

Value

Resulting iObject.

See Also

ihist, ibar, iplot.list, iobj.opt

Examples

data(iris)
attach(iris)
iplot(Sepal.Width,Petal.Width)
l<-lowess(Sepal.Width,Petal.Width)
ilines(l)

---

Interactive Map

Description

This function creates a new interactive map from the given data.

Usage

imap(x, y=NULL, ...)

Arguments

x	either an object of the class "map" as created by the map function or a vector of the x-coordinates of the map polygons
y	y-coordinates of the map polygons
...	All additional parameters are passed to iplot.opt.

Details

Creates an interactive map plot.

The input can be either an object of the class "map" passed directly to the x parameter or two vectors of matching coordinates passed to x and y. The format for polygons in the same as used by the map function.

Each polygon should correspond to a case and it will be linked correspondingly. See ivar.new.map for details on map variables.

Note: this function is currently experimental and it may change in the future.

Value

Resulting plot object.

See Also

iplot, ivar.new.map

Examples

library(maps)
m <- map('state', plot=FALSE)
imap(m)

---

Interactive Mosaic Plot

Description

This function creates a new interactive mosaic plot from the given data.

Usage

imosaic(...)

Arguments

...

All unnamed parameters are treated as variables to display. Additional parameters are passed to iplot.opt. The parameter "type" selects the variation of the mosaic plot. Valid values are "observed", "expected", "fluctuation", "same.bin.size" and "multiple.barchart". Partial matching is used, so first letter is sufficient.

Details

The plot can be used either in the form imosaic(df) to plot a list or data frame of variables contained in df or imosaic(x, y, z) to plot variables x, y and z.

Additional parameters:

rotateYLabelsBy

The default value of this variable is changed here. For a description see iplot.opt.

Value

Resulting plot object.

See Also

ihist, ibar, iplot.list, iplot.opt

Examples

library(MASS)
data(Cars93)
attach(Cars93)
imosaic(AirBags,Cylinders,Origin)
imosaic(AirBags,Cylinders,Origin,type="mul")
iplot.location(300,100,TRUE)

---

Interactive objects (iObjects) management functions.

Description

These functions are used to manage iObjects of an iPlot. Exactly one of the iObjects is the current one (for each iPlot). Every newly created iObject automatically becomes current.

Please note that both iPlots and iObjects can be also used directly - each function creating an iObject or iPlot returns the newly created object which can be used in calls to functions requiring plot parameter (for iPlots) or iobj.opt, iobj.rm and similar (iObjects). The object list management functions below are provided for convenience only - the direct use of objects is encouraged instead.

iobj.list returns all iObjects of the current iPlot.

iobj.cur returns the current iObject.

iobj.next and iobj.prev return the ID of the next resp. previous object in the list relative to the object specified by the argument.

iobj.set makes the object with the specified ID current.

iobj.get returns the object specified by its ID.

iobj.rm removes the object

== and != operators can be used to compare two iObjects

Usage

iobj.list(plot = iplot.cur())
iobj.cur(plot = iplot.cur())
iobj.next(which=iobj.cur(), plot = iplot.cur())
iobj.prev(which=iobj.cur(), plot = iplot.cur())
iobj.set(which=iobj.next())
iobj.get(pos, plot=iplot.cur())
iobj.rm(which=iobj.cur(), plot = iplot.cur())
## S3 method for class 'iobj'
a == b
## S3 method for class 'iobj'
a != b

Arguments

which	An object or an integer specifying the object number.
pos	ID of an object
plot	plot to operate on (either as integer ID or plot object itself).
a	object to compare
b	object to compare

See Also

ilines, iabline

---

Modify parameters of an iObject

Description

This function modifies parameters of an iObject.

Usage

iobj.opt(o=iobj.cur(),...)

Arguments

o	object whose options are to be set
...	options to be set

Value

If no parameters (except for the object) are specified, a list of the current parameters is returned (if supported by the object).

The following common parameters are used by most iObjects:

visible visibility flag. The default is TRUE.

layer layer in which the object is placed. Currently the following layers are supported: 0 - background, 1 - data points, 2 - selection, 3 - drag boxes. The value -1 has a special meaning by denoting the topmost layer (which is the default)

col drawing color of the object

fill filling color of the object (where applicable). If set to NA then no filling is performed.

coord coordinates system to use. 0=graphical coordinates, 1=data space coordinates (default), 2=relative coordinates.

update if set to FALSE then no plot refresh is done after updating the options. When modifying multiple objects it is common practice to set update to FALSE for all but the last updated object. The default is TRUE.

See Also

iobj.list

---

Interactive Parallel Coordinates Plot

Description

This function creates a new interactive parallel coordinates plot from the given data.

Usage

ipcp(...)

Arguments

...	All unnamed parameters are treated as variables to display. All additional parameters are passed to iplot.opt.

Details

Creates an interactive parallel coordinates plot.

The plot can be used either in the form ipcp(df) to plot a list or data frame of variables contained in df or ipcp(x, y, z) to plot variables x, y and z.

Additional parameters:

alterningLabels

Whether lables should be placed alternately at the top and at the bottom.

COL_AXES

Color of the (optional) axes.

Value

Resulting plot object.

See Also

ihist, ibar, iplot.list, iplot.opt

Examples

data(iris)
ipcp(iris)

---

Interactive Scatterplot

Description

This function creates a new interactive scatterplot from the given data.

Usage

iplot(x, y=NULL, xlab=NULL, ylab=NULL, ...)

Arguments

x	Data for the x axis. It can be either a vector of values or a variable of an iset. If y is not given, this must be a list of one of these.
y	Data for the y axis. It can be either a vector of values or a variable of an iset.
xlab	Name for x variable.
ylab	Name for y variable.
...	All additional parameters are passed to iplot.opt. Many of the parameters used in plot are supported.

Details

Creates an interactive scatterplot.

Additional parameters:

changePtDiamBy

Number of pixels the point diameter should be changed when in-/decreasing it via keyboard or menu.

customFieldBg

Whether to use a custom background color.

COL_CUSTOMBG

The custom background color.

drawAxes

Whether axes should be drawn.

equiscale

Whether the same scale should be applied to both axes.

minimalDiam

Minimal point diameter.

ptDiam

Point diameter.

spaceprop

Gives the amount of space around the data points. 1.0 means no space, 1.5 means half as much space around the data as is used for the data itself. This resets zoom.

Default values:

    changePtDiamBy=2, customFieldBg=FALSE, COL_CUSTOMBG="white",
    drawAxes=TRUE, equiscale=FALSE, minimalDiam=1, ptDiam=3,
    spaceprop=1.1
  

Value

Resulting plot object.

See Also

ihist, ibar, iplot.list, iplot.opt

Examples

data(iris)
attach(iris)
iplot(Sepal.Width,Petal.Width)
iplot(Sepal.Width/Sepal.Length, Species)

---

Retireve data from a plot.

Description

This function retrieves the associated data from the current plot.

Usage

iplot.data(id=NULL)

Arguments

id	number of the variable to retrieve. If omitted a list of all associated varaibles (and their contents) is returned.

Value

A vector representaing a variable or a list of variable contents. The first two vecctors in the list are traditionally named x and y. The number of variables depends on the plot used, e.g. scatter plot has two, histogram or bar chart return one.

See Also

iplot, ihist, ibar

Examples

data(iris)
attach(iris)
iplot(Sepal.Width,Petal.Width)
iplot(Sepal.Width/Sepal.Length, Species)

---

Interactive plots management functions.

Description

These functions are used to manage currently open iPlots. Exactly one of the open iPlots is the current plot. Every newly created iPlot automatically becomes the current plot. Any plot specific functions, such as ilines operate on the current plot.

Please note that the functions below are provided for convenience only. It is also possible to use plot objects directly without using the plot list. Each function creating a new iPlot directly returns the plot object which can then be used to any subsequent calls to ilines, iplot.opt etc.

iplot.list returns all currently registered iPlots (even if they are hidden).

iplot.cur returns the ID of the current plot.

iplot.next and iplot.prev return the ID of the next resp. previous plot in the list relative to the plot specified by the argument.

iplot.set makes the plot with the specified ID current.

iplot.off closes the plot.

Usage

iplot.list()
iplot.cur()
iplot.next(which=iplot.cur())
iplot.prev(which=iplot.cur())
iplot.set(which=iplot.next())
iplot.off(plot=iplot.cur())

Arguments

which	An integer specifying a plot number.
plot	Plot object or plot number of a plot to close.

See Also

ilines, iplot, ihist, ibar

Examples

data(iris)
attach(iris)
iplot(Sepal.Width,Petal.Width)
ibar(Species)
iplot.list()

---

iPlot manipulation functions

Description

The following functions are used to mainpulate iplots. They are NOT part of the official API and may disappear without warning. Most of them are legacy functions introduced before iplot.opt was available.

iplot.backend retrieves or sets the iPlots back-end.

iplot.resetZoom reset zoom

iplot.rotate set plot rotation

iplot.setExtendedQuery set text for extended query

iplot.zoomIn zoom into specified area

iplot.zoomOut zoom out (the coordinates are ignored as the zoom is heararchical)

iplot.location get or set the location of the iplot

iplot.size get or set the size of an iplot plot

Usage

iplot.backend(type = NULL)
iplot.resetZoom() 
iplot.rotate(i) 
iplot.setExtendedQuery(str, plotID=.iplot.curid) 
iplot.zoomIn(x1, y1, x2, y2) 
iplot.zoomOut(x, y)
iplot.location(x, y, relative=FALSE, plot=iplot.cur())
iplot.size(width, height, plot=iplot.cur())
## S3 method for class 'iobj'
print(x, ...) 
## S3 method for class 'iplot'
print(x, ...)

Arguments

i	rotation orientation
plotID	plot ID (number)
str	string to show on extended query or FALSE or NULL to disable extended query
type	back-end type - one of "awt", "swing" or "opengl" (or any unambiguous first part hereof) to set the type or NULL to retrieve the current back-end type
x1	basis coordinate for the x axis
y1	basis coordinate for the y axis
x2	edge coordinate for the x axis
y2	edge coordinate for the y axis
x	object to print or x-coordinate
y	y-coordinate
plot	plot to query or move (id or object)
relative	can be TRUE, FALSE or another plot
width	width of the plot (in pixels)
height	height of the plot (in pixels)
...	additional parameters

Details

iplot.location, iplot.size: those functions either query or set the location or size of the plot. If either of the coordinates is missing, the size/location in that coordinate will not be changed. If both coordinates are missing, the functions have no side effect and just return the current size (iplot.size) or location and size (iplot.location) as named vectors (x, y for location and width, height for size).

iplot.backend determines the back-end used by iplots. The choices are awt, swing an dopengl. Not all back-ends are available on all platforms. AWT is the most compatible back-end and available on all platforms, Swing has more modern look and feel and OpenGL sacrifices text rendering quality for speed (i.e. OpenGL backend is best for large data). When no backend is specified, the function has no side effect and just returns the currently requested back-end. Currently there is no indication whether the back-end request was honored or not. If a backend is not supported, the request is silently ignored.

---

Modify parameters of an interactive plot

Description

This function modifies parameters of an iPlot.

Usage

iplot.opt(..., plot=iplot.cur())

Arguments

...	Parameters to modify
plot	Plot whose parameters are to be modified

Details

The following parameters are common to all plots. See help pages of individual plots for plot-specific parameters.

xlim

Range of the X axis (vector of two numbers).

ylim

Range of the Y axis (vector of two numbers).

col

Colors of the points. See iset.brush for details.

autoAdjustMargins

Whether iPlots should try to determine appropriate margin sizes. Set this to FALSE if you want to set the margins manually.

defaultMargins

Vector of default values for left, right, top and bottom margins.

mLeft

Size of left margin.

mRight

Size of right margin.

mTop

Size of top margin.

mBottom

Size of bottom margin.

fillColor

Fill color.

borderColor

Border color.

fillColorSel

Fill color when selected.

borderColorSel

Border color when selected.

fillColorDrag

Fill color of dragged bars.

COL_INVALID

Color of invalid elements.

COL_OUTLINE

Default line color.

COL_SELBG

Background color of selection rectangle. Defaults to selection color with alpha value 0.298.

COL_ZOOMBG

Background color of selection rectangle.

horizontalMedDist

Mean horizontal label distance.

horizontalMinDist

Minimal horizontal label distance.

verticalMedDist

Mean vertical label distance.

verticalMinDist

Minimal vertical label distance.

extQueryString

Extended query string.

rotateYLabels

Whether labels for the y axis should be rotated by the angle given by rotateYLabelsBy.

rotateYLabelsBy

The labels for the y axis are rotated by this amount of degrees if rotateYLabels is set to true.

title

Frame title.

Value

If no parameters (except for the plot) are specified, a list of the current parameters is returned.

See Also

iplot.list

Examples

data(iris)
attach(iris)
iplot(Sepal.Width,Petal.Width)
iplot.opt(xlim=c(1.5,5),col=Species)

---

Add a bitmap (raster) image to the current iPlot.

Description

iraster adds a raster image as an iObj to the given iPlot. Position in the plot is specified by bottom-left and top-right points of the image.

Usage

iraster(x1, y1, x2, y2, img, ..., plot = iplot.cur())

Arguments

x1, y1, x2, y2	coordinates of the bottom-left (x1, y1) and top-right (x2, y2) corner. Alternatively, x1 can be a vector of length 4 specifying the four values, or x1 and x2 can be vectors of length 2 specifying one point each
img	image to draw. It can be either a raster, file name, binary connection or a raw vector containing an image in a common image format such as PNG or JPEG. If it is a raster, then the raster is first encoded into PNG format and passed as raw vector.
...	additional arguments that will be passed to iobj.opt if present.
plot	parent plot for the image

Details

The current implementation uses Java's ImageIO API to read the image, so the supported formats will depend on your Java implemetation. Raster objects (i.e., of class "raster", "nativeRaster"), matrices and arrays) are simply passed to png::writePNG(img) so for anything other than computationally constructed objects it is more efficient to use the encoded image.

Value

Resulting iObject.

See Also

iobj.opt

Examples

## very silly example ...
iplot(0:20/20, 0:20/20)
## get a sample image (R logo) from the png package
fn <- system.file("img", "Rlogo.png", package="png")
## put this image behind all points
iraster(0, 0, 1, 1, fn, layer=-2)
## you can use a raster but it's less eficient
## this one goes to the top layer where iObjs reside normally
iraster(0, 0, 0.5, 0.5, png::readPNG(fn))

---

iSet and iVar - managing data in iPlots

Description

iPlots maintain a separate copy of all data that are displayed in the iPlots. This allows iPlots to opearate even after the underlying data has been deleted in R. It also allows iPlots to use hot linking in all aspects, including update of plots on data changes.

This data management consists mainly of two classes: iset and ivar. The iset object encapsuletes an iSets which can be thought of as a kind of special data frame that allows linking of all variables contained therein. Each variable (or column in data.frame-speech) is represented by an ivar object encapsulating an iVar.

An instance of a ivar class can be use to create new iplots or update data in the existing iplots. Convenience operators on both ivar and iset objects include those such as length, subsetting and subassignment and therefor from user's point of view it is possible to use them transparently a data frames (iSet) or vectors (iVar). In addition, iSet supports methods such as names or dim.

iset returns the object corresponding to a given iSet.

isets returs objects for all iSets

Usage

iset(which=iset.cur())
isets()
## S3 method for class 'iset'
x[[i]]
## S3 method for class 'iset'
x$name
## S3 replacement method for class 'iset'
x$name <- value
## S3 replacement method for class 'iset'
x[[i]] <- value
## S3 method for class 'iset'
x[i = 1:(dim(x)[1]), j = 1:length(x)]
## S3 replacement method for class 'iset'
x[i = 1:(dim(x)[1]), j = 1:length(x)] <- value
## S3 replacement method for class 'iset'
names(x) <- value
## S3 method for class 'iset'
names(x)
## S3 method for class 'iset'
length(x)
## S3 method for class 'iset'
dim(x)

Arguments

x	iset object
i	row (case) index
j	column (variable) index
value	replacement value
which	An integer specifying an iSet ID or a string specifying a name of the iSet.
name	name of the variable

See Also

ivar, iset.cur

Examples

data(iris)
s <- iset.new("iris", iris)
rm(iris)
s[]
names(s)
dim(s)
length(s)
s[[1]]
s$Species
s[1:5,c("Sepal.Length","Petal.Length")]
rm(s)
# even if we get rid of the reference object,
# we can get it back
s <- iset()
s
# see ?ivar for more examples

---

Set color of cases in an iSet.

Description

This function sets the colors of cases of an iSet. Every iPlot is free to use the corresponding representation individually. Color of a case is a iSet-global property, that is all iPlots associated with an iSet will reflect the change.

Usage

iset.col(col)

Arguments

col	Vector of integers specifying colors or a factor. The values are re-cycled if necessary. 0 has a special meaning denoting "no color", i.e. the plot is free to use the default color. This is also the default state of a newly created iSet.

See Also

iset.new

Examples

data(iris)
attach(iris)
iplot(Sepal.Width,Petal.Width)
ibar(Species)
iset.col(Species)

---

iSet management functions.

Description

These functions are used to manage iSets. An iSets groups all iPlots which use the same data as a basis. All iPlots belonging to an iSet are linked, that is selecting cases in one iPlot cause the same cases to be highlighted in all other iPlots based on the same iSet. The linking is done on case-level, therefore the variables should have the same number of cases.

You can get the iset object by calling the iset function. An iset object can be used in a way similar to data frames, so subsetting and other operations work as expected. Subassignment is always done "live", so changeing parts of the data will cause plots to be updated accordingly.

There is always exactly one current iSet. An initial iSet called "default" is created when the iplots library is loaded. All new iPlots are created using the current iSet. The set of functions described here allow the manipulation of iSets.

Note that changing the iSet also has an effect on iplot.list and iobj.list since iPlots are linked to their iSets.

iset.new creates a new iSet and makes it current

iset.list returns all iSets

iset.cur returns the ID of the current iSet

iset.next and iset.prev return the ID of the next resp. previous iSet in the list relative to the iSet specified by the argument.

iset.set makes the iSet with the specified ID current.

iset.rm removes the given iSet and all associated plots. Note that this may cause the current set ID may change (even if some other iSet is removed).

print.iset prints an iSet object.

Usage

iset.new(name=NULL, payload=NULL)
iset.list()
iset.cur()
iset.next(which=iset.cur())
iset.prev(which=iset.cur())
iset.set(which=iset.next())
iset.rm(which=iset.cur())
## S3 method for class 'iset'
print(x, ...)

Arguments

name	Name of the new iSet. If no name is specified, an automatic name of the form "data.X" is created where "X" is an integer. The name can be used instead on an ID in many cases.
payload	Initial contents of the iSet in the form of a list/data.frame or NULL if an empty iSet is desired.
which	An integer specifying an iSet ID or a string specifying a name of the iSet.
x	An iSet object to print
...	further parameters to be passed to print

Methods

show

signature(object = "iset"): is mapped to print.iset

See Also

iset, iplot.list, iobj.list

---

Selection / linked highlighting

Description

These functions modify the selection or return the currently selected (highlighted) cases.

iset.selected returns a vector of IDs of all currently selected cases (in the current iSet)

iset.select changes the selection of the current iSet. All plots are updated immediately to reflect the new selection.

iset.sel.changed returns TRUE if the selection has changed since last call to this function.

iset.selectAll selects everything

iset.selectNone clears all selections (yes, not a good name)

Usage

iset.selected()
iset.select(what, mode="replace", mark=TRUE, batch=FALSE)
iset.sel.changed(iset=iset.cur())
iset.selectAll(batch=FALSE)
iset.selectNone(batch=FALSE)

Arguments

what	specification of the new selection. This can be either a vector of case IDs or a logical vector.
mode	mode to be used when combining the previous selection and teh surrent one. Supported modes are "replace", "union" and "intersect".
mark	mark to be used.
iset	iSet to query
batch	when set to TRUE dependents (e.g. plots) are not notified. This is useful for performing many complex updates at once without the need to re-draw all plots. Use with care as the system usually relies on event propagation.

Value

List of IDs of selected cases (iset.selected), boolean value (iset.sel.changed).

See Also

iset.list

Examples

data(iris)
attach(iris)
iplot(Sepal.Length,Petal.Length)
iset.select(Species=="virginica")

---

Add text to the current iPlot.

Description

text adds the strings given in the vector labels at the coordinates given by x and y to the current iPlot. y may be missing since xy.coords(x,y) is used for construction of the coordinates.

Usage

itext(x, y=NULL, labels=seq(along=x), ax=NULL, ay=NULL, ..., plot = iplot.cur())

Arguments

x, y	Coordinate vectors of the text. xy.coords is used to obtain the coordinates.
labels	one or more character strings or expressions specifying the text to be written.
ax, ay	anchor coordinates of the text. They specify where is the point specified by x and y relative to the text. 0 means left resp. top and 1 means right/bottom. Therefore vertically and horizontally centered text can be obtained by using ax=0.5 and ay=0.5.
...	any additional options are passed to iobj.opt
plot	parent plot

Value

Resulting iObject.

See Also

ilines, iobj.opt

---

Data (ivar) manipulation functions

Description

The following functions are used to mainpulate variables for iplots:

ivar.data returns the data associated with an iVar variable

ivar.new creates a new iVar variable in the current iSet using the supplied data

ivar.new.map same as ivar.new but for map (polygon) data

ivar.update replaces the content of a variable

iVar variables (deprecated - use iset.new instead)

[.ivar allows data subsetting of ivar contents

[<-.ivar allows partial replacement of ivar contents

length.ivar returns the length of a variable

iset.updateVars notifies all plots that variables have been changed. This function should not be normally used, because notification happens automatically, unless batch updates are performed.

Usage

ivar.data(var)
ivar.new(name = deparse(substitute(cont)), cont)
ivar.new.map(name, x, y)
ivar.update(var, cont, batch = FALSE)
## S3 method for class 'ivar'
length(x)
## S3 method for class 'ivar'
x[i, ...]
## S3 replacement method for class 'ivar'
x[...] <- value
## S3 method for class 'ivar'
print(x, ...)

iset.updateVars()

Arguments

var	an existing iVar variable
cont	desired contents - usually a numeric vector or a factor
name	variable name in an iSet
batch	if set to TRUE then plots are not automatically notified about the change. This allows an update of multiple variables in batches without constant re-painting of the plots.
x	object to subset or x-coordinate of the map data
y	y-coordinate of the map data
i	index - if missing all data are returned
...	indices
value	replacement value

Details

All data that will are displayed in an iPlot are organized in iSets. Each iSet contains variables, called iVars, that represent the displayed data. Typically an iSet corresponds to a data frame and an iVar corresponds to a column in a data frame. All variables in one iSet must have the same length and indexing order. This allows iPlots to perform proper linking of all plots.

Before some data can be displayed in an iPlot, they are put into an ivar using the ivar.new function. Each variable has a name that is unique within an iSet. This process is automatic if you pass arbitrary vectors to the iPlots plotting functions. However, it is possible (and desirable) to register variables beforehand using iset.new. It is much faster to use iVars in the plots instead of the raw data.

Once an iVar is created, it is merely a reference to the data in the iSet. The regular subsetting operators [ or ivar.data function can be used to retrieve the data back to R.

In addition, it is possible to update the contents of an iVar using the ivar.update function. Once the variable is updated, all plots that use the variable will be updated, too, and reflect the change, unless batch was set to TRUE. The notification is performed using the iset.updateVars function.

Methods

show

signature(object = "ivar"): is mapped to print.ivar

See Also

iset.new, ibar, iplot.list, iplot.opt

Examples

data(iris)
s <- iset.new("iris", iris)
rm(iris) # we don't need iris anymore
s[1:5,]
s$Species
s$Species[]
s[1:5,c("Sepal.Width","Sepal.Length")]
s$Test <- factor(c("yes","no")[runif(dim(s)[1])+1.5])
s[1:5,]
iplot(s$Sepal.Length, s$Petal.Length)
ibar(s$Species)
iplot.location(400,,TRUE)
iset.select(s$Test[]=="yes")
ibar(s$Test)
iplot.location(400,250,TRUE)
# use iset.rm() to remove everything

---