A handful of bare minimum Preform syntax.

§1. Nonterminal names. This is a typical internal nonterminal being defined, though it's a bit more meta than most — it's a nonterminal which matches against the name of any nonterminal. (This is used only to parse inclusion requests for the debugging log.)

Note that we use the internal 1 to signal that a correct match must have exactly one word. 

<preform-nonterminal> internal 1 {
    nonterminal *nt = Nonterminals::detect(Lexer::word(Wordings::first_wn(W)));
    if (nt) { ==> { -, nt }; return TRUE; }
    ==> { fail nonterminal };
}

§2. Text positions. A useful nonterminal which matches no text, but detects the position: 

<if-start-of-paragraph> internal 0 {
    int w1 = Wordings::first_wn(W);
    if ((w1 == 0) || (compare_word(w1-1, PARBREAK_V))) return TRUE;
    ==> { fail nonterminal };
}

§3. And another convenience: 

<if-not-cap> internal 0 {
    int w1 = Wordings::first_wn(W);
    if (Word::unexpectedly_upper_case(w1) == FALSE) return TRUE;
    ==> { fail nonterminal };
}

§4. Balancing. The following regular (not internal!) nonterminal matches any text in which braces and brackets are correctly paired. 

<balanced-text> ::=
    ......

§5. Inform contains relatively few syntaxes where commas are actually required, though they can optionally be used in many lists, as here:

parma ham, camembert, grapes

But for when we only want to spot comma placements, this can be used. Note that the comma matches only if not in brackets. 

<list-comma-division> ::=
    ...... , ......

§6. Literal numbers. (Inform itself doesn't use this, but has alternatives for cardinals and ordinals within the VM-representable range.) 

<any-integer> internal 1 {
    if (Vocabulary::test_flags(Wordings::first_wn(W), NUMBER_MC)) {
        int N = Vocabulary::get_literal_number_value(Lexer::word(Wordings::first_wn(W)));
        ==> { N, - };
        return TRUE;
    }
    ==> { fail nonterminal };
}

§7. Literal text. Text is "with substitutions" if it contains square brackets, used in Inform for interpolations called "text substitutions". 

<quoted-text> internal 1 {
    if ((Wordings::nonempty(W)) &&
        (Vocabulary::test_flags(Wordings::first_wn(W), TEXT_MC+TEXTWITHSUBS_MC))) {
        ==> { Wordings::first_wn(W), - }; return TRUE;
    }
    ==> { fail nonterminal };
}

<quoted-text-with-subs> internal 1 {
    if ((Wordings::nonempty(W)) &&
        (Vocabulary::test_flags(Wordings::first_wn(W), TEXTWITHSUBS_MC))) {
        ==> { Wordings::first_wn(W), - }; return TRUE;
    }
    ==> { fail nonterminal };
}

<quoted-text-without-subs> internal 1 {
    if ((Wordings::nonempty(W)) &&
        (Vocabulary::test_flags(Wordings::first_wn(W), TEXT_MC))) {
        ==> { Wordings::first_wn(W), - }; return TRUE;
    }
    ==> { fail nonterminal };
}

§8. For finicky technical reasons the easiest way to detect an empty piece of text "" is to provide a nonterminal matching it: 

<empty-text> internal 1 {
    if ((Wordings::nonempty(W)) &&
        (Word::compare_by_strcmp(Wordings::first_wn(W), U"\"\""))) {
        ==> { Wordings::first_wn(W), - }; return TRUE;
    }
    ==> { fail nonterminal };
}