ginger/gg/term.go

package gg

import (
	"errors"
	"fmt"
	"io"
	"strconv"
	"strings"
	"unicode"

	"github.com/mediocregopher/ginger/graph"
	"golang.org/x/exp/slices"
)

var (
	errNoMatch = errors.New("not found")
)

type stringerFn func() string

func (fn stringerFn) String() string {
	return fn()
}

type stringerStr string

func (str stringerStr) String() string {
	return string(str)
}

type term[T any] struct {
	name     fmt.Stringer
	decodeFn func(d *Decoder) (T, error)
}

func (t term[T]) String() string {
	return t.name.String()
}

func firstOf[T any](terms ...*term[T]) *term[T] {
	if len(terms) < 2 {
		panic("firstOfTerms requires at least 2 terms")
	}

	return &term[T]{
		name: stringerFn(func() string {
			descrs := make([]string, len(terms))
			for i := range terms {
				descrs[i] = terms[i].String()
			}
			return strings.Join(descrs, " or ")
		}),
		decodeFn: func(d *Decoder) (T, error) {
			var zero T
			for _, t := range terms {
				v, err := t.decodeFn(d)
				if errors.Is(err, errNoMatch) {
					continue
				} else if err != nil {
					return zero, err
				}

				return v, nil
			}

			return zero, errNoMatch
		},
	}
}

func seq[Ta, Tb, Tc any](
	name fmt.Stringer,
	termA *term[Ta],
	termB *term[Tb],
	fn func(Ta, Tb) Tc,
) *term[Tc] {
	return &term[Tc]{
		name: name,
		decodeFn: func(d *Decoder) (Tc, error) {
			var zero Tc

			va, err := termA.decodeFn(d)
			if err != nil {
				return zero, err
			}

			vb, err := termB.decodeFn(d)
			if errors.Is(err, errNoMatch) {
				return zero, d.nextLoc().errf("expected %v", termB)
			} else if err != nil {
				return zero, err
			}

			return fn(va, vb), nil
		},
	}
}

func prefixed[Ta, Tb any](termA *term[Ta], termB *term[Tb]) *term[Tb] {
	return seq(termA, termA, termB, func(_ Ta, b Tb) Tb {
		return b
	})
}

func prefixIgnored[Ta, Tb any](termA *term[Ta], termB *term[Tb]) *term[Tb] {
	return &term[Tb]{
		name: termB,
		decodeFn: func(d *Decoder) (Tb, error) {
			var zero Tb
			if _, err := termA.decodeFn(d); err != nil {
				return zero, err
			}
			return termB.decodeFn(d)
		},
	}
}

func suffixIgnored[Ta, Tb any](
	termA *term[Ta], termB *term[Tb],
) *term[Ta] {
	return seq(termA, termA, termB, func(a Ta, _ Tb) Ta {
		return a
	})
}

func oneOrMore[T any](t *term[T]) *term[[]T] {
	return &term[[]T]{
		name: stringerFn(func() string {
			return fmt.Sprintf("one or more %v", t)
		}),
		decodeFn: func(d *Decoder) ([]T, error) {
			var vv []T
			for {
				v, err := t.decodeFn(d)
				if errors.Is(err, errNoMatch) {
					break
				} else if err != nil {
					return nil, err
				}

				vv = append(vv, v)
			}

			if len(vv) == 0 {
				return nil, errNoMatch
			}

			return vv, nil
		},
	}
}

func zeroOrMore[T any](t *term[T]) *term[[]T] {
	return &term[[]T]{
		name: stringerFn(func() string {
			return fmt.Sprintf("zero or more %v", t)
		}),
		decodeFn: func(d *Decoder) ([]T, error) {
			var vv []T
			for {
				v, err := t.decodeFn(d)
				if errors.Is(err, errNoMatch) {
					break
				} else if err != nil {
					return nil, err
				}

				vv = append(vv, v)
			}

			return vv, nil
		},
	}
}

func mapTerm[Ta, Tb any](
	name fmt.Stringer, t *term[Ta], fn func(Ta) Tb,
) *term[Tb] {
	return &term[Tb]{
		name: name,
		decodeFn: func(d *Decoder) (Tb, error) {
			var zero Tb
			va, err := t.decodeFn(d)
			if err != nil {
				return zero, err
			}
			return fn(va), nil
		},
	}
}

func runePredTerm(
	name fmt.Stringer, pred func(rune) bool,
) *term[locatableRune] {
	return &term[locatableRune]{
		name: name,
		decodeFn: func(d *Decoder) (locatableRune, error) {
			lr, err := d.readRune()
			if errors.Is(err, io.EOF) {
				return locatableRune{}, errNoMatch
			} else if err != nil {
				return locatableRune{}, err
			}

			if !pred(lr.r) {
				d.unreadRune(lr)
				return locatableRune{}, errNoMatch
			}

			return lr, nil
		},
	}
}

func runeTerm(r rune) *term[locatableRune] {
	return runePredTerm(
		stringerStr(fmt.Sprintf("'%c'", r)),
		func(r2 rune) bool { return r2 == r },
	)
}

func locatableRunesToString(rr []locatableRune) string {
	str := make([]rune, len(rr))
	for i := range rr {
		str[i] = rr[i].r
	}
	return string(str)
}

func runesToStringTerm(
	t *term[[]locatableRune],
) *term[locatableString] {
	return mapTerm(
		t, t, func(rr []locatableRune) locatableString {
			return locatableString{rr[0].locate(), locatableRunesToString(rr)}
		},
	)
}

func discard[T any](t *term[T]) *term[struct{}] {
	return mapTerm(t, t, func(_ T) struct{} { return struct{}{} })
}

var (
	notNewlineTerm = runePredTerm(
		stringerStr("not-newline"),
		func(r rune) bool { return r != '\n' },
	)

	commentTerm = prefixed(
		prefixed(runeTerm('*'), zeroOrMore(notNewlineTerm)),
		runeTerm('\n'),
	)

	whitespaceTerm = zeroOrMore(firstOf(
		discard(runePredTerm(stringerStr("whitespace"), unicode.IsSpace)),
		discard(commentTerm),
	))
)

func trimmedTerm[T any](t *term[T]) *term[T] {
	t = prefixIgnored(whitespaceTerm, t)
	t = suffixIgnored(t, whitespaceTerm)
	return t
}

func trimmedRuneTerm(r rune) *term[locatableRune] {
	return trimmedTerm(runeTerm(r))
}

var (
	digitTerm = runePredTerm(
		stringerStr("digit"),
		func(r rune) bool { return '0' <= r && r <= '9' },
	)

	positiveNumberTerm = runesToStringTerm(oneOrMore(digitTerm))

	negativeNumberTerm = seq(
		stringerStr("negative-number"),
		runeTerm('-'),
		positiveNumberTerm,
		func(neg locatableRune, posNum locatableString) locatableString {
			return locatableString{neg.locate(), string(neg.r) + posNum.str}
		},
	)

	numberTerm = mapTerm(
		stringerStr("number"),
		firstOf(negativeNumberTerm, positiveNumberTerm),
		func(str locatableString) Value {
			i, err := strconv.ParseInt(str.str, 10, 64)
			if err != nil {
				panic(fmt.Errorf("parsing %q as int: %w", str, err))
			}

			return Value{Number: &i, Location: str.locate()}
		},
	)
)

var (
	letterTerm = runePredTerm(
		stringerStr("letter"),
		func(r rune) bool {
			return unicode.In(r, unicode.Letter, unicode.Mark)
		},
	)

	letterTailTerm = zeroOrMore(firstOf(letterTerm, digitTerm))

	nameTerm = seq(
		stringerStr("name"),
		letterTerm,
		letterTailTerm,
		func(head locatableRune, tail []locatableRune) Value {
			name := string(head.r) + locatableRunesToString(tail)
			return Value{Name: &name, Location: head.locate()}
		},
	)
)

func openEdgeIntoValue(val Value, oe *OpenEdge) *OpenEdge {
	switch {
	case oe == nil:
		return graph.ValueOut(None, val)
	case !oe.EdgeValue().Valid:
		return oe.WithEdgeValue(Some(val))
	default:
		return graph.TupleOut(Some(val), oe)
	}
}

var graphTerm, valueTerm = func() (*term[Value], *term[Value]) {
	type tupleState struct {
		ins []*OpenEdge
		oe  *OpenEdge
	}

	type graphState struct {
		g  *Graph
		oe *OpenEdge
	}

	var (
		rightParenthesis = trimmedRuneTerm(')')
		tupleEndTerm     = mapTerm(
			rightParenthesis,
			rightParenthesis,
			func(lr locatableRune) tupleState {
				// if ')', then map that to an empty state. This acts as a
				// sentinel value to indicate "end of tuple".
				return tupleState{}
			},
		)

		rightCurlyBrace = trimmedRuneTerm('}')
		graphEndTerm    = mapTerm(
			rightCurlyBrace,
			rightCurlyBrace,
			func(lr locatableRune) graphState {
				// if '}', then map that to an empty state. This acts as a
				// sentinel value to indicate "end of graph".
				return graphState{}
			},
		)
	)

	var (
		// pre-define these, and then fill in the pointers after, in order to
		// deal with recursive dependencies between them.
		valueTerm = new(term[Value])

		tupleTerm                  = new(term[*OpenEdge])
		tupleTailTerm              = new(term[tupleState])
		tupleOpenEdgeTerm          = new(term[tupleState])
		tupleOpenEdgeTailTerm      = new(term[tupleState])
		tupleOpenEdgeValueTailTerm = new(term[tupleState])

		graphTerm                  = new(term[Value])
		graphTailTerm              = new(term[graphState])
		graphOpenEdgeTerm          = new(term[graphState])
		graphOpenEdgeTailTerm      = new(term[graphState])
		graphOpenEdgeValueTailTerm = new(term[graphState])
	)

	*tupleTerm = *seq(
		stringerStr("tuple"),
		trimmedRuneTerm('('),
		tupleTailTerm,
		func(lr locatableRune, ts tupleState) *OpenEdge {
			slices.Reverse(ts.ins)
			return graph.TupleOut(None, ts.ins...)
		},
	)

	*tupleTailTerm = *firstOf(
		tupleEndTerm,
		mapTerm(
			tupleOpenEdgeTerm,
			tupleOpenEdgeTerm,
			func(ts tupleState) tupleState {
				ts.ins = append(ts.ins, ts.oe)
				ts.oe = nil
				return ts
			},
		),
	)

	*tupleOpenEdgeTerm = *firstOf(
		seq(
			valueTerm,
			valueTerm,
			tupleOpenEdgeValueTailTerm,
			func(val Value, ts tupleState) tupleState {
				ts.oe = openEdgeIntoValue(val, ts.oe)
				return ts
			},
		),
		seq(
			tupleTerm,
			tupleTerm,
			tupleOpenEdgeTailTerm,
			func(oe *OpenEdge, ts tupleState) tupleState {
				ts.oe = oe
				return ts
			},
		),
	)

	*tupleOpenEdgeTailTerm = *firstOf(
		tupleEndTerm,
		prefixed(trimmedRuneTerm(','), tupleTailTerm),
	)

	*tupleOpenEdgeValueTailTerm = *firstOf(
		tupleOpenEdgeTailTerm,
		prefixed(trimmedRuneTerm('<'), tupleOpenEdgeTerm),
	)

	*graphTerm = *seq(
		stringerStr("graph"),
		trimmedRuneTerm('{'),
		graphTailTerm,
		func(lr locatableRune, gs graphState) Value {
			if gs.g == nil {
				gs.g = new(Graph)
			}

			return Value{Graph: gs.g, Location: lr.locate()}
		},
	)

	*graphTailTerm = *firstOf(
		graphEndTerm,
		seq(
			nameTerm,
			nameTerm,
			prefixed(trimmedRuneTerm('='), graphOpenEdgeTerm),
			func(name Value, gs graphState) graphState {
				if gs.g == nil {
					gs.g = new(Graph)
				}

				gs.g = gs.g.AddValueIn(name, gs.oe)
				gs.oe = nil
				return gs
			},
		),
	)

	*graphOpenEdgeTerm = *firstOf(
		seq(
			valueTerm,
			valueTerm,
			graphOpenEdgeValueTailTerm,
			func(val Value, gs graphState) graphState {
				gs.oe = openEdgeIntoValue(val, gs.oe)
				return gs
			},
		),
		seq(
			tupleTerm,
			tupleTerm,
			graphOpenEdgeTailTerm,
			func(oe *OpenEdge, gs graphState) graphState {
				gs.oe = oe
				return gs
			},
		),
	)

	*graphOpenEdgeTailTerm = *firstOf(
		graphEndTerm,
		prefixed(trimmedRuneTerm(';'), graphTailTerm),
	)

	*graphOpenEdgeValueTailTerm = *firstOf(
		graphOpenEdgeTailTerm,
		prefixed(trimmedRuneTerm('<'), graphOpenEdgeTerm),
	)

	*valueTerm = *firstOf(nameTerm, numberTerm, graphTerm)

	return graphTerm, valueTerm
}()

var topLevelTerm = trimmedTerm(valueTerm)