Span: Span objects

In NLP: Natural Language Processing Infrastructure

Span objects

Description

Creation and manipulation of span objects.

Usage

Span(start, end)
as.Span(x)
is.Span(x)

Arguments

start, end	integer vectors giving the start and end positions of the spans.
x	an R object.

Details

A single span is a pair with “slots” ‘start’ and ‘end’, giving the start and end positions of the span.

Span objects provide sequences (allowing positional access) of single spans. They have class "Span". Span objects can be coerced to annotation objects via as.Annotation() (which of course is only appropriate provided that the spans are character spans of the natural language text being annotated), and annotation objects can be coerced to span objects via as.Span() (giving the character spans of the annotations).

Subscripting span objects via [ extracts subsets of spans; subscripting via $ extracts integer vectors with the sequence of values of the named slot.

There are several additional methods for class "Span": print() and format(); c() combines spans (or objects coercible to these using as.Span()), and as.list() and as.data.frame() coerce, respectively, to lists (of single span objects) and data frames (with spans and slots corresponding to rows and columns). Finally, one can add a scalar and a span object (resulting in shifting the start and end positions by the scalar).

Span() creates span objects from the given sequences of start and end positions, which must have the same length.

as.Span() coerces to span objects, with a method for annotation objects.

is.Span() tests whether an object inherits from class "Span" (and hence returns TRUE for both span and annotation objects).

Value

For Span() and as.Span(), a span object (of class "Span").

For is.Span(), a logical.

NLP documentation built on Sept. 11, 2024, 6:59 p.m.