GrammarRules

✖
`GrammarRules`

✖

GrammarRules[rules]

represents grammar rules to be deployed to a cloud object that implements the grammar in a form suitable for use with functions like GrammarApply and Interpreter.

✖

GrammarRules[rules,defs]

uses grammar definitions defs.

Details and Options

Grammar rules are of the form pattern->expr or pattern:>expr.
The pattern in a grammar rule must be a combination of the following elements:

	"string"	literal string
	StringExpression[…]	arbitrary string pattern
	RegularExpression[…]	regular expression
	form₁\|form₂\|…	alternative forms
	OptionalElement[form,def]	optional form, with default
	FixedOrder[form₁,form₂,…]	forms that must occur in a fixed order
	AnyOrder[form₁,form₂,…]	forms that can occur in any order
	form..	form repeated any number of times
	DelimitedSequence[form,…]	form repeated with delimiters in between
	GrammarToken["name"]	built-in or defined domain
	CaseSensitive[form]	form that must match case sensitively
	x:form	form named to be used

CloudDeploy[GrammarRules[…]] deploys a grammar and returns a cloud object, which can be used in GrammarApply and Interpreter.
The rules associated with the grammar in effect specify how inputs should be transformed to expressions.
In GrammarRules[rules,defs], the list defs can give subsidiary rules for grammar domains that can be used in the main grammar rules.
GrammarRules supports the following options:

	AllowLooseGrammar	Automatic	whether to allow extra grammatical "fluff"
	IgnoreDiacritics	True	whether to ignore diacritics when applying grammar rules
	IgnoreCase	True	default for case insensitivity
	AmbiguityFunction	Automatic	default function for handling ambiguity
	SaveDefinitions	False	whether to save all definitions associated with rules and defs
	Initialization	None	an expression to be evaluated when the grammar is applied

Examples

open allclose all

Basic Examples (2)Summary of the most common use cases

Deploy a simple grammar for adding numbers:

In[1]:=1

✖

https://wolfram.com/xid/01y9o8z44y6-vpwyom

Out[1]=1

Apply the deployed grammar to a natural language string:

In[2]:=2

✖

https://wolfram.com/xid/01y9o8z44y6-6zvjkl

Out[2]=2

Define rules for custom GrammarToken objects:

In[1]:=1

✖

https://wolfram.com/xid/01y9o8z44y6-ps886q

Out[1]=1

Apply the grammar:

In[2]:=2

✖

https://wolfram.com/xid/01y9o8z44y6-vp3fjk

Out[2]=2

In[3]:=3

✖

https://wolfram.com/xid/01y9o8z44y6-6elnrq

Out[3]=3

Properties & Relations (1)Properties of the function, and connections to other functions

Repeated and DelimitedSequence are equivalent to recursive grammar rules:

In[1]:=1

✖

https://wolfram.com/xid/01y9o8z44y6-8hue5c

Out[1]=1

In[2]:=2

✖

https://wolfram.com/xid/01y9o8z44y6-78897f

Out[2]=2

In[3]:=3

✖

https://wolfram.com/xid/01y9o8z44y6-zqjjqn

Out[3]=3

In[4]:=4

✖

https://wolfram.com/xid/01y9o8z44y6-p6byl8

Out[4]=4

Possible Issues (1)Common pitfalls and unexpected behavior

Expressions on the right-hand side of a grammar rule evaluate during GrammarApply:

In[1]:=1

✖

https://wolfram.com/xid/01y9o8z44y6-xp1xzp

Out[1]=1

To keep an expression unevaluated, use a wrapper like HoldComplete or Inactivate:

In[2]:=2

✖

https://wolfram.com/xid/01y9o8z44y6-s45zz7

Out[2]=2

Top