general_pref: Utility Functions for Preferences

Description Usage Arguments Details See Also Examples

Description

Collection of some useful functions which are applicable to all preference objects.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
empty()

is.empty_pref(x)

## S4 method for signature 'preference'
length(x)

is.preference(x)

## S4 method for signature 'preference'
as.expression(x, ...)

## S4 method for signature 'preference'
assoc.df(x)

## S4 replacement method for signature 'preference'
assoc.df(x) <- value

Arguments

x

A preference, or, for is.preference, an object to be tested if it is an (empty) preference.

...

Optional arguments passed to as.expression.

value

A data frame to associate with a preference object.

Details

The empty preference empty() is a neutral element for the complex preference compositions {*, &, +}. It holds that empty() * p and empty() & p is equal to p for all preferences p.

The function length(p) returns the term length of the preference term p which is defined as the number of base preferences in a complex preference term. The empty preference empty() has length 0, and all base preferences have length 1.

With as.expression(p) for a preference p the call to the preference is constructed. This means, eval(as.expression(p)) returns the preference p, evaluated in the current environment.

The function is.empty_pref returns TRUE if x is the empty preference object empty() and FALSE otherwise.

With assoc.df the associated data frame of a preference can be retrieved or set. Setting the associated data frame means that a partial evaluation based on this data frame is done. See show.pref for details on partial evaluation of preferences. Next, the preference is linked to that data frame, such that peval(p) can be used instead of psel(df, p). It returns NULL if no data frame is associated. Use set.assoc.df(NULL) to delete an associated data frame.

See Also

See base_pref for the construction of base preferences, and complex_pref for the construction of complex preferences. See show.pref for string output and partial evaluation of preference terms.

Examples

1
2
3
4
5
6
7
8
# Same as low(a) * low(b)
p <- low(a) * low(b) * empty()

# returns 2, as empty() does not count
length(p)

# the preference expression (without empty())
as.expression(p)

rPref documentation built on May 1, 2019, 8:23 p.m.