Skip to main content

Backus-Naur Form

(Content adapted from Critchlow & Eck)

Context-free grammars are used to describe some aspects of the syntax of programming languages. However, a notation that is often used for grammars in the context of programming languages is somewhat different from the notation introduced in the preceding section. The notation that is used is called Backus-Naur Form or BNF. It is named after computer scientists John Backus and Peter Naur, who developed the notation. Actually, several variations of BNF exist. I will discuss one of them here. BNF can be used to describe the syntax of natural languages, as well as programming languages, and some of the examples in this section will deal with the syntax of English.

Like context-free grammars, BNF grammars make use of production rules, non-terminals, and terminals. The non-terminals are usually given meaningful, multi-character names. Here, I will follow a common practice of enclosing non-terminals in angle brackets, so that they can be easily distinguished. For example, ⟨noun⟩\langle\textit{noun}\rangle and ⟨sentence⟩\langle\textit{sentence}\rangle could be non-terminals in a BNF grammar for English, while ⟨program⟩\langle\textit{program}\rangle and ⟨if-statement⟩\langle\textit{if-statement}\rangle might be used in a BNF grammar for a programming language. Note that a BNF non-terminal usually represents a meaningful syntactic category, that is, a certain type of building block in the syntax of the language that is being described, such as an adverb, a prepositional phrase, or a variable declaration statement. The terminals of a BNF grammar are the things that actually appear in the language that is being described. In the case of natural language, the terminals are individual words.

In BNF production rules, I will use the symbol "::=" in place of the "⟶\longrightarrow" that is used in context-free grammars. BNF production rules are more powerful than the production rules in context-free grammars. That is, one BNF rule might be equivalent to several context-free grammar rules. As for context-free grammars, the left-hand side of a BNF production rule is a single non-terminal symbol. The right hand side can include terminals and non-terminals, and can also use the following notations, which should remind you of notations used in regular expressions:

  • A vertical bar, ∣|, indicates a choice of alternatives. For example,

    ⟨digit⟩ ::= 0  ∣  1  ∣  2  ∣  3  ∣  4  ∣  5  ∣  6  ∣  7  ∣  8  ∣  9\langle\textit{digit}\rangle\ ::=\ 0 \;|\; 1 \;|\; 2 \;|\; 3 \;|\; 4 \;|\; 5 \;|\; 6 \;|\; 7 \;|\; 8 \;|\; 9

    indicates that the non-terminal ⟨digit⟩\langle\textit{digit}\rangle can be replaced by any one of the terminal symbols 0, 1, \dots, 9.

  • Items enclosed in brackets are optional. For example,

    ⟨declaration⟩ ::= ⟨type⟩ ⟨variable⟩ [ = ⟨expression⟩ ] ;\langle\textit{declaration}\rangle\ ::=\ \langle\textit{type}\rangle\ \langle\textit{variable}\rangle \ [\ =\ \langle\textit{expression}\rangle\ ]\ ;

    says that ⟨declaration⟩\langle\textit{declaration}\rangle can be replaced either by "⟨type⟩ ⟨variable⟩ ;\langle\textit{type}\rangle\ \langle\textit{variable}\rangle\ ;" or by "⟨type⟩ ⟨variable⟩ = ⟨expression⟩ ;\langle\textit{type}\rangle\ \langle\textit{variable}\rangle\ =\ \langle\textit{expression}\rangle\ ;". (The symbols "=" and ";" are terminal symbols in this rule.)

  • Items enclosed between "[[" and "]…]\ldots" can be repeated zero or more times. (This has the same effect as a "∗*"in a regular expression.) For example,

    ⟨integer⟩ ::= ⟨digit⟩ [ ⟨digit⟩ ]…\langle\textit{integer}\rangle\ ::=\ \langle\textit{digit}\rangle\ [\ \langle\textit{digit}\rangle\ ]\ldots

    says that an ⟨integer⟩\langle\textit{integer}\rangle consists of a ⟨digit⟩\langle\textit{digit}\rangle followed optionally by any number of additional ⟨digit⟩\langle\textit{digit}\rangle's. That is, the non-terminal ⟨integer⟩\langle\textit{integer}\rangle can be replaced by ⟨digit⟩\langle\textit{digit}\rangle or by ⟨digit⟩ ⟨digit⟩\langle\textit{digit}\rangle\ \langle\textit{digit}\rangle or by ⟨digit⟩ ⟨digit⟩ ⟨digit⟩\langle\textit{digit}\rangle\ \langle\textit{digit}\rangle\ \langle\textit{digit}\rangle, and so on.

  • Parentheses can be used as usual, for grouping.

All these notations can be expressed in a context-free grammar by introducing additional production rules. For example, the BNF rule "⟨sign⟩ ::= +  ∣  −\langle\textit{sign}\rangle\ ::=\ +\;|\;-" is equivalent to the two rules, "⟨sign⟩ ::= +\langle\textit{sign}\rangle\ ::=\ +" and "⟨sign⟩ ::= −\langle\textit{sign}\rangle\ ::=\ -". A rule that contains an optional item can also be replaced by two rules. For example,

⟨declaration⟩ ::= ⟨type⟩ ⟨variable⟩ [ = ⟨expression⟩ ] ;\langle\textit{declaration}\rangle\ ::=\ \langle\textit{type}\rangle\ \langle\textit{variable}\rangle \ [\ =\ \langle\textit{expression}\rangle\ ]\ ;

can be replaced by the two rules

⟨declaration⟩ ::= ⟨type⟩ ⟨variable⟩ ;⟨declaration⟩ ::= ⟨type⟩ ⟨variable⟩ = ⟨expression⟩ ;\begin{aligned} &\langle\textit{declaration}\rangle\ ::=\ \langle\textit{type}\rangle\ \langle\textit{variable}\rangle\ ;\\ &\langle\textit{declaration}\rangle\ ::=\ \langle\textit{type}\rangle\ \langle\textit{variable}\rangle \ =\ \langle\textit{expression}\rangle\ ; \end{aligned}

In context-free grammars, repetition can be expressed by using a recursive rule such as "S⟶aSS\longrightarrow aS", in which the same non-terminal symbol appears both on the left-hand side and on the right-hand side of the rule. BNF-style notation using "[[" and "]…]\ldots" can be eliminated by replacing it with a new non-terminal symbol and adding a recursive rule to allow that symbol to repeat zero or more times. For example, the production rule

⟨integer⟩ ::= ⟨digit⟩ [ ⟨digit⟩ ]…\langle\textit{integer}\rangle\ ::=\ \langle\textit{digit}\rangle\ [\ \langle\textit{digit}\rangle\ ]\ldots

can be replaced by three rules using a new non-terminal symbol ⟨digit-list⟩\langle\textit{digit-list}\rangle to represent a string of zero or more ⟨digit⟩\langle\textit{digit}\rangle's:

⟨integer⟩ ::= ⟨digit⟩ ⟨digit-list⟩⟨digit-list⟩ ::= ⟨digit⟩ ⟨digit-list⟩⟨digit-list⟩ ::= ε\begin{aligned} &\langle\textit{integer}\rangle\ ::=\ \langle\textit{digit}\rangle\ \langle\textit{digit-list}\rangle\\ &\langle\textit{digit-list}\rangle\ ::=\ \langle\textit{digit}\rangle\ \langle\textit{digit-list}\rangle\\ &\langle\textit{digit-list}\rangle\ ::=\ \varepsilon \end{aligned}

As an example of a complete BNF grammar, let's look at a BNF grammar for a very small subset of English. The start symbol for the grammar is ⟨sentence⟩\langle\textit{sentence}\rangle, and the terminal symbols are English words. All the sentences that can be produced from this grammar are syntactically correct English sentences, although you wouldn't encounter many of them in conversation. Here is the grammar:

⟨sentence⟩ ::= ⟨simple-sentence⟩ [ and ⟨simple-sentence⟩ ]…⟨simple-sentence⟩ ::= ⟨noun-part⟩ ⟨verb-part⟩⟨noun-part⟩ ::= ⟨article⟩ ⟨noun⟩ [ who ⟨verb-part⟩ ]…⟨verb-part⟩ ::= ⟨intransitive-verb⟩  ∣  ( ⟨transitive-verb⟩ ⟨noun-part⟩ )⟨article⟩ ::= the  ∣  a⟨noun⟩ ::= man  ∣  woman  ∣  dog  ∣  cat  ∣  computer⟨intransitive-verb⟩ ::= runs  ∣  jumps  ∣  hides⟨transitive-verb⟩ ::= knows  ∣  loves  ∣  chases  ∣  owns\begin{aligned} &\langle\textit{sentence}\rangle\ ::=\ \langle\textit{simple-sentence}\rangle\ [\ \textrm{and}\ \langle\textit{simple-sentence}\rangle\ ]\ldots\\ &\langle\textit{simple-sentence}\rangle\ ::=\ \langle\textit{noun-part}\rangle\ \langle\textit{verb-part}\rangle\\ &\langle\textit{noun-part}\rangle\ ::=\ \langle\textit{article}\rangle\ \langle\textit{noun}\rangle\ [\ \textrm{who}\ \langle\textit{verb-part}\rangle\ ]\ldots\\ &\langle\textit{verb-part}\rangle\ ::=\ \langle\textit{intransitive-verb}\rangle \;|\; (\ \langle\textit{transitive-verb}\rangle\ \langle\textit{noun-part}\rangle\ )\\ &\langle\textit{article}\rangle\ ::=\ \textrm{the} \;|\; \textrm{a}\\ &\langle\textit{noun}\rangle\ ::=\ \textrm{man} \;|\; \textrm{woman} \;|\; \textrm{dog} \;|\; \textrm{cat} \;|\; \textrm{computer}\\ &\langle\textit{intransitive-verb}\rangle\ ::=\ \textrm{runs} \;|\; \textrm{jumps} \;|\; \textrm{hides}\\ &\langle\textit{transitive-verb}\rangle\ ::=\ \textrm{knows} \;|\; \textrm{loves} \;|\; \textrm{chases} \;|\; \textrm{owns} \end{aligned}

This grammar can generate sentences such as "A dog chases the cat and the cat hides" and "The man loves a woman who runs." The second sentence, for example, is generated by the derivation

⟨sentence⟩ ⟹ ⟨simple-sentence⟩⟹ ⟨noun-part⟩ ⟨verb-part⟩⟹ ⟨article⟩ ⟨noun⟩ ⟨verb-part⟩⟹ the ⟨noun⟩ ⟨verb-part⟩⟹ the man ⟨verb-part⟩⟹ the man ⟨transitive-verb⟩ ⟨noun-part⟩⟹ the man loves ⟨noun-part⟩⟹ the man loves ⟨article⟩ ⟨noun⟩ who ⟨verb-part⟩⟹ the man loves a ⟨noun⟩ who ⟨verb-part⟩⟹ the man loves a woman who ⟨verb-part⟩⟹ the man loves a woman who ⟨intransitive-verb⟩⟹ the man loves a woman who runs\begin{aligned} \langle\textit{sentence}\rangle\ &\Longrightarrow\ \langle\textit{simple-sentence}\rangle\\ &\Longrightarrow\ \langle\textit{noun-part}\rangle\ \langle\textit{verb-part}\rangle\\ &\Longrightarrow\ \langle\textit{article}\rangle\ \langle\textit{noun}\rangle\ \langle\textit{verb-part}\rangle\\ &\Longrightarrow\ \textrm{the}\ \langle\textit{noun}\rangle\ \langle\textit{verb-part}\rangle\\ &\Longrightarrow\ \textrm{the}\ \textrm{man}\ \langle\textit{verb-part}\rangle\\ &\Longrightarrow\ \textrm{the}\ \textrm{man}\ \langle\textit{transitive-verb}\rangle\ \langle\textit{noun-part}\rangle\\ &\Longrightarrow\ \textrm{the}\ \textrm{man}\ \textrm{loves}\ \langle\textit{noun-part}\rangle\\ &\Longrightarrow\ \textrm{the}\ \textrm{man}\ \textrm{loves}\ \langle\textit{article}\rangle \ \langle\textit{noun}\rangle\ \textrm{who}\ \langle\textit{verb-part}\rangle\\ &\Longrightarrow\ \textrm{the}\ \textrm{man}\ \textrm{loves}\ \textrm{a} \ \langle\textit{noun}\rangle\ \textrm{who}\ \langle\textit{verb-part}\rangle\\ &\Longrightarrow\ \textrm{the}\ \textrm{man}\ \textrm{loves}\ \textrm{a} \ \textrm{woman}\ \textrm{who}\ \langle\textit{verb-part}\rangle\\ &\Longrightarrow\ \textrm{the}\ \textrm{man}\ \textrm{loves}\ \textrm{a} \ \textrm{woman}\ \textrm{who}\ \langle\textit{intransitive-verb}\rangle\\ &\Longrightarrow\ \textrm{the}\ \textrm{man}\ \textrm{loves}\ \textrm{a} \ \textrm{woman}\ \textrm{who}\ \textrm{runs} \end{aligned}

BNF is most often used to specify the syntax of programming languages. Most programming languages are not, in fact, context-free languages, and BNF is not capable of expressing all aspects of their syntax. For example, BNF cannot express the fact that a variable must be declared before it is used or the fact that the number of actual parameters in a subroutine call statement must match the number of formal parameters in the declaration of the subroutine. So BNF is used to express the context-free aspects of the syntax of a programming language, and other restrictions on the syntax—such as the rule about declaring a variable before it is used—are expressed using informal English descriptions.

When BNF is applied to programming languages, the terminal symbols are generally "tokens," which are the minimal meaningful units in a program. For example, the pair of symbols <= constitute a single token, as does a string such as "Hello World". Every number is represented by a single token. (The actual value of the number is stored as a so-called "attribute" of the token, but the value plays no role in the context-free syntax of the language.) I will use the symbol number\textbf{number} to represent a numerical token. Similarly, every variable name, subroutine name, or other identifier in the program is represented by the same token, which I will denote as ident\textbf{ident}. One final complication: Some symbols used in programs, such as "]" and "(", are also used with a special meaning in BNF grammars. When such a symbol occurs as a terminal symbol, I will enclose it in double quotes. For example, in the BNF production rule

⟨array-reference⟩ ::= ident ‘‘[" ⟨expression⟩ ‘‘]"\langle\textit{array-reference}\rangle\ ::=\ \textbf{ident}\ ``["\ \langle\textit{expression}\rangle\ ``]"

the "[" and "]" are terminal symbols in the language that is being described, rather than the BNF notation for an optional item. With this notation, here is part of a BNF grammar that describes statements in the Java programming language:

⟨statement⟩ ::= ⟨block-statement⟩  ∣  ⟨if-statement⟩  ∣  ⟨while-statement⟩  ∣  ⟨assignment-statement⟩  ∣  ⟨null-statement⟩⟨block-statement⟩ ::= { [ ⟨statement⟩ ]… }⟨if-statement⟩ ::= if ‘‘(" ⟨condition⟩ ‘‘)" ⟨statement⟩ [ else ⟨statement⟩ ]⟨while-statement⟩ ::= while ‘‘(" ⟨condition⟩ ‘‘)" ⟨statement⟩⟨assignment-statement⟩ ::= ⟨variable⟩ = ⟨expression⟩ ;⟨null-statement⟩ ::= ε\begin{aligned} &\langle\textit{statement}\rangle\ ::=\ \langle\textit{block-statement}\rangle \;|\; \langle\textit{if-statement}\rangle \;|\; \langle\textit{while-statement}\rangle\\ &\hspace{1.5 in} \;|\; \langle\textit{assignment-statement}\rangle \;|\; \langle\textit{null-statement}\rangle\\ &\langle\textit{block-statement}\rangle\ ::=\ \{\ [\ \langle\textit{statement}\rangle\ ]\dots\ \}\\ &\langle\textit{if-statement}\rangle\ ::=\ \textrm{if}\ ``("\ \langle\textit{condition}\rangle\ ``)"\ \langle\textit{statement}\rangle\ [\ \textrm{else}\ \langle\textit{statement}\rangle\ ]\\ &\langle\textit{while-statement}\rangle\ ::=\ \textrm{while}\ ``("\ \langle\textit{condition}\rangle\ ``)"\ \langle\textit{statement}\rangle\\ &\langle\textit{assignment-statement}\rangle\ ::=\ \langle\textit{variable}\rangle\ =\ \langle\textit{expression}\rangle\ ;\\ &\langle\textit{null-statement}\rangle\ ::=\ \varepsilon \end{aligned}

The non-terminals ⟨condition⟩\langle\textit{condition}\rangle, ⟨variable⟩\langle\textit{variable}\rangle, and ⟨expression⟩\langle\textit{expression}\rangle would, of course, have to be defined by other production rules in the grammar. Here is a set of rules that define simple expressions, made up of numbers, identifiers, parentheses and the arithmetic operators ++, −-, ∗* and //:

⟨expression⟩ ::= ⟨term⟩ [ ( +  ∣  − ) ⟨term⟩ ]…⟨term⟩ ::= ⟨factor⟩ [ ( ∗  ∣  / ) ⟨factor⟩ ]…⟨factor⟩ ::= ident  ∣  number  ∣  ‘‘(" ⟨expression⟩ ‘‘)"\begin{aligned} &\langle\textit{expression}\rangle\ ::=\ \langle\textit{term}\rangle\ [\ (\ + \;|\; -\ )\ \langle\textit{term}\rangle\ ]\ldots\\ &\langle\textit{term}\rangle\ ::=\ \langle\textit{factor}\rangle\ [\ (\ * \;|\; /\ )\ \langle\textit{factor}\rangle\ ]\ldots\\ &\langle\textit{factor}\rangle\ ::=\ \textbf{ident} \;|\; \textbf{number} \;|\; ``("\ \langle\textit{expression}\rangle\ ``)" \end{aligned}

The first rule says that an ⟨expression⟩\langle\textit{expression}\rangle is a sequence of one or more ⟨term⟩\langle\textit{term}\rangle's, separated by plus or minus signs. The second rule defines a ⟨term⟩\langle\textit{term}\rangle to be a sequence of one or more ⟨factor⟩\langle\textit{factor}\rangle's, separated by multiplication or division operators. The last rule says that a ⟨factor⟩\langle\textit{factor}\rangle can be either an identifier or a number or an ⟨expression⟩\langle\textit{expression}\rangle enclosed in parentheses. This small BNF grammar can generate expressions such as "3∗53*5" and "x∗(x+1)−3/(z+2∗(3−x))+7x*(x+1) - 3/(z+2*(3-x)) + 7". The latter expression is made up of three terms: x∗(x+1)x*(x+1), 3/(z+2∗(3−x))3/(z+2*(3-x)), and 77. The first of these terms is made up of two factors, xx and (x+1)(x+1). The factor (x+1)(x+1) consists of the expression x+1x+1 inside a pair of parentheses.

The nice thing about this grammar is that the precedence rules for the operators are implicit in the grammar. For example, according to the grammar, the expression 3+5∗73+5*7 is seen as ⟨term⟩ + ⟨term⟩\langle\textit{term}\rangle\ +\ \langle\textit{term}\rangle where the first term is 33 and the second term is 5∗75*7. The 5∗75*7 occurs as a group, which must be evaluated before the result is added to 33. Parentheses can change the order of evaluation. For example, (3+5)∗7(3+5)*7 is generated by the grammar as a single ⟨term⟩\langle\textit{term}\rangle of the form ⟨factor⟩ ∗ ⟨factor⟩\langle\textit{factor}\rangle\ *\ \langle\textit{factor}\rangle. The first ⟨factor⟩\langle\textit{factor}\rangle is (3+5)(3+5). When (3+5)∗7(3+5)*7 is evaluated, the value of (3+5)(3+5) is computed first and then multiplied by 77. This is an example of how a grammar that describes the syntax of a language can also reflect its meaning.

Although this section has not introduced any really new ideas or theoretical results, I hope it has demonstrated how context-free grammars can be applied in practice.

Exercises​

  1. One of the examples in this section was a grammar for a subset of English. Give five more examples of sentences that can be generated from that grammar. Your examples should, collectively, use all the rules of the grammar.

  2. Rewrite the example BNF grammar for a subset of English as a context-free grammar.

  3. Write a single BNF production rule that is equivalent to the following context-free grammar:

    S⟶aSaS⟶bBB⟶bBB⟶ε\begin{aligned} S &\longrightarrow aSa\\ S &\longrightarrow bB\\ B &\longrightarrow bB\\ B &\longrightarrow \varepsilon \end{aligned}
    Answer
    S ::= a(S  ∣  b[b]…)a\begin{aligned} &S\ ::=\ a(S\;|\;b[b]\ldots)a \end{aligned}

  4. Write a BNF production rule that specifies the syntax of real numbers, as they appear in programming languages such as Java and C. Real numbers can include a sign, a decimal point and an exponential part. Some examples are: 17.3, .73, 23.1e67, -1.34E-12, +0.2, and 100E+100.

  5. Variable references in the Java programming language can be rather complicated. Some examples include: x, list.next, A[7], a.b.c, S[i+1].grid[r][c].red, …. Write a BNF production rule for Java variables. You can use the token ident\textbf{ident} and the non-terminal ⟨expression⟩\langle\textit{expression}\rangle in your rule.

  6. Use BNF to express the syntax of the try ... catch statement in the Java programming language.

  7. Give a BNF grammar for compound propositions made up of propositional variables, parentheses, and the logical operators ∧\land, ∨\lor, and ¬\lnot. Use the non-terminal symbol ⟨pv⟩\langle\textit{pv}\rangle to represent a propositional variable. You do not have to give a definition of ⟨pv⟩\langle\textit{pv}\rangle.

    Answer
    ⟨prop⟩ ::= ⟨term⟩ [∨ ⟨term⟩]…⟨term⟩ ::= ⟨factor⟩ [∧ ⟨factor⟩]…⟨factor⟩ ::= ⟨pv⟩ ∣ ¬ ⟨factor⟩ ∣ ‘‘(" ⟨prop⟩ ‘‘)"\begin{aligned} &\langle\textit{prop}\rangle\ ::=\ \langle\textit{term}\rangle\ [\lor\ \langle\textit{term}\rangle]\ldots\\ &\langle\textit{term}\rangle\ ::=\ \langle\textit{factor}\rangle\ [\land\ \langle\textit{factor}\rangle]\ldots\\ &\langle\textit{factor}\rangle\ ::=\ \langle\textit{pv}\rangle\ |\ \lnot\ \langle\textit{factor}\rangle\ |\ ``("\ \langle\textit{prop}\rangle\ ``)" \end{aligned}