Merge pull request #75 from navytux/y/pydict

Add PyDict mode

.github/workflows/ci.yml

@@ -11,12 +11,6 @@ jobs:
     strategy:
       matrix:
         go:
-          - "1.12.x"
-          - "1.13.x"
-          - "1.14.x"
-          - "1.15.x"
-          - "1.16.x"
-          - "1.17.x"
           - "1.18.x"
           - "1.19.x"
           - "1.20.x"

dict.go

+package ogórek
+// Python-like Dict that handles keys by Python-like equality on access.
+//
+// For example Dict.Get() will access the same element for all keys int(1), float64(1.0) and big.Int(1).
+
+import (
+	"encoding/binary"
+	"fmt"
+	"hash/maphash"
+	"math"
+	"math/big"
+	"reflect"
+	"sort"
+
+	"github.com/aristanetworks/gomap"
+)
+
+// Dict represents dict from Python in PyDict mode.
+//
+// It mirrors Python with respect to which types are allowed to be used as
+// keys, and with respect to keys equality. For example Tuple is allowed to be
+// used as key, and all int(1), float64(1.0) and big.Int(1) are considered to be
+// equal.
+//
+// For strings, similarly to Python3, Bytes and string are considered to be not
+// equal, even if their underlying content is the same. However with same
+// underlying content ByteString, because it represents str type from Python2,
+// is treated equal to both Bytes and string.
+//
+// See PyDict mode documentation in top-level package overview for details.
+//
+// Note: similarly to builtin map Dict is pointer-like type: its zero-value
+// represents nil dictionary that is empty and invalid to use Set on.
+type Dict struct {
+	m *gomap.Map[any, any]
+}
+
+// NewDict returns new empty dictionary.
+func NewDict() Dict {
+	return NewDictWithSizeHint(0)
+}
+
+// NewDictWithSizeHint returns new empty dictionary with preallocated space for size items.
+func NewDictWithSizeHint(size int) Dict {
+	return Dict{m: gomap.NewHint[any, any](size, equal, hash)}
+}
+
+// NewDictWithData returns new dictionary with preset data.
+//
+// kv should be key₁, value₁, key₂, value₂, ...
+func NewDictWithData(kv ...any) Dict {
+	l := len(kv)
+	if l % 2 != 0 {
+		panic("odd number of arguments")
+	}
+	l /= 2
+	d := NewDictWithSizeHint(l)
+	for i := 0; i < l; i++ {
+		k := kv[2*i]
+		v := kv[2*i+1]
+		d.Set(k, v)
+	}
+	return d
+}
+
+// Get returns value associated with equal key.
+//
+// An entry with key equal to the query is looked up and corresponding value
+// is returned.
+//
+// nil is returned if no matching key is present in the dictionary.
+//
+// Get panics if key's type is not allowed to be used as Dict key.
+func (d Dict) Get(key any) any {
+	value, _ := d.Get_(key)
+	return value
+}
+
+// Get_ is comma-ok version of Get.
+func (d Dict) Get_(key any) (value any, ok bool) {
+	return d.m.Get(key)
+}
+
+// Set sets key to be associated with value.
+//
+// Any previous keys, equal to the new key, are removed from the dictionary
+// before the assignment.
+//
+// Set panics if key's type is not allowed to be used as Dict key.
+func (d Dict) Set(key, value any) {
+	// ByteString and container(with ByteString) are non-transitive equal types
+	// so  Set(ByteString)       should first remove Bytes and string,
+	// and Set(Tuple{ByteString) should first remove Tuple{Bytes} and Tuple{string}
+	d.Del(key)
+	d.m.Set(key, value)
+}
+
+// Del removes equal keys from the dictionary.
+//
+// All entries with key equal to the query are looked up and removed.
+//
+// Del panics if key's type is not allowed to be used as Dict key.
+func (d Dict) Del(key any) {
+	// see comment in Set about ByteString and container(with ByteString)
+	for {
+		d.m.Delete(key)
+		_, have := d.Get_(key)
+		if !have {
+			break
+		}
+	}
+}
+
+// Len returns the number of items in the dictionary.
+func (d Dict) Len() int {
+	return d.m.Len()
+}
+
+// Iter returns iterator over all elements in the dictionary.
+//
+// The order to visit entries is arbitrary.
+func (d Dict) Iter() /* iter.Seq2 */ func(yield func(any, any) bool) {
+	it := d.m.Iter()
+	return func(yield func(any, any) bool) {
+		for it.Next() {
+			cont := yield(it.Key(), it.Elem())
+			if !cont {
+				break
+			}
+		}
+	}
+}
+
+// String returns human-readable representation of the dictionary.
+func (d Dict) String() string {
+	return d.sprintf("%v")
+}
+
+// GoString returns detailed human-readable representation of the dictionary.
+func (d Dict) GoString() string {
+	return fmt.Sprintf("%T%s", d, d.sprintf("%#v"))
+}
+
+// sprintf serves String and GoString.
+func (d Dict) sprintf(format string) string {
+	type KV struct { k,v string }
+	vkv := make([]KV, 0, d.Len())
+	d.Iter()(func(k, v any) bool {
+		vkv = append(vkv, KV{
+			k: fmt.Sprintf(format, k),
+			v: fmt.Sprintf(format, v),
+		})
+		return true
+	})
+
+	sort.Slice(vkv, func(i, j int) bool {
+		return vkv[i].k < vkv[j].k
+	})
+
+	s := "{"
+	for i, kv := range vkv {
+		if i > 0 {
+			s += ", "
+		}
+		s += kv.k + ": " + kv.v
+	}
+
+	s += "}"
+	return s
+}
+
+
+// ---- equal ----
+
+// kind represents to which category a type belongs.
+//
+// It primarily classifies bool, numbers, slices, structs and maps, and puts
+// everything else into "other" category.
+type kind uint
+const (
+	kBool = iota
+	kInt     // int + intX
+	kUint    // uint + uintX
+	kFloat   // floatX
+	kComplex // complexX
+	kBigInt  // *big.Int
+
+	kSlice   // slice + array
+	kMap     // map
+	kStruct  // struct
+	kPointer // pointer
+	kOther   // everything else
+)
+
+// kindOf returns kind of x.
+func kindOf(x any) kind {
+	r := reflect.ValueOf(x)
+
+	switch r.Kind() {
+	case reflect.Bool:
+		return kBool
+	case reflect.Int, reflect.Int64, reflect.Int32, reflect.Int16, reflect.Int8:
+		return kInt
+	case reflect.Uint, reflect.Uint64, reflect.Uint32, reflect.Uint16, reflect.Uint8:
+		return kUint
+	case reflect.Float64, reflect.Float32:
+		return kFloat
+	case reflect.Complex128, reflect.Complex64:
+		return kComplex
+
+	case reflect.Slice, reflect.Array:
+		return kSlice
+	case reflect.Map:
+		return kMap
+	case reflect.Struct:
+		return kStruct
+	}
+
+	switch x.(type) {
+	case *big.Int:
+		return kBigInt
+	}
+
+	switch r.Kind() {
+	case reflect.Pointer:
+		return kPointer
+	}
+
+	return kOther
+}
+
+// equal implements equality matching what Python would return for a == b.
+//
+// Equality properties:
+//
+// 1) equality is extension of Go ==
+//
+//	(a == b) ⇒ equal(a,b)
+//
+// 2) self equal:
+//
+//	equal(a,a) = y
+//
+// 3) equality is symmetrical:
+//
+//	equal(a,b) = equal(b,a)
+//
+// 4) equality is mostly transitive:
+//
+//	EqTransitive = set of all x:
+//	  ∀ a,b,c ∈ EqTransitive:
+//	    equal(a,b) ^ equal(b,c) ⇒ equal(a,c)
+//
+//	EqTransitive = all \ {ByteString + containers with ByteString}
+func equal(xa, xb any) bool {
+	// strings/bytes
+	switch a := xa.(type) {
+	case string:
+		switch b := xb.(type) {
+		case string:     return a == b
+		case ByteString: return a == string(b)
+		case Bytes:      return false
+		default:         return false
+		}
+
+	case ByteString:
+		switch b := xb.(type) {
+		case string:     return a == ByteString(b)
+		case ByteString: return a == b
+		case Bytes:      return a == ByteString(b)
+		default:         return false
+		}
+
+	case Bytes:
+		switch b := xb.(type) {
+		case string:     return false
+		case ByteString: return a == Bytes(b)
+		case Bytes:      return a == b
+		default:         return false
+		}
+	}
+
+	// everything else
+	a := reflect.ValueOf(xa)
+	b := reflect.ValueOf(xb)
+
+	ak := kindOf(xa)
+	bk := kindOf(xb)
+
+	// since equality is symmetric, we can implement only half of comparison matrix
+	if ak > bk {
+		a, b = b, a
+		ak, bk = bk, ak
+		xa, xb = xb, xa
+	}
+	// ak ≤ bk
+
+	handled := true
+	switch ak {
+	default:
+		handled = false
+
+	// numbers
+	case kBool:
+		// bool compares to numbers as 1 or 0
+		//
+		// In [1]: 1.0 == True
+		// Out[1]: True
+		//
+		// In [2]: 0.0 == False
+		// Out[2]: True
+		//
+		// In [3]: d = {1: 'abc'}
+		//
+		// In [4]: d[True]
+		// Out[4]: 'abc'
+		abint := bint(a.Bool())
+		switch bk {
+		case kBool:	return eq_Int_Int     (abint, bint(b.Bool()))
+		case kInt:	return eq_Int_Int     (abint, b.Int())
+		case kUint:	return eq_Int_Uint    (abint, b.Uint())
+		case kFloat:	return eq_Int_Float   (abint, b.Float())
+		case kComplex:	return eq_Int_Complex (abint, b.Complex())
+		case kBigInt:	return eq_Int_BigInt  (abint, xb.(*big.Int))
+		}
+
+	case kInt:
+		aint := a.Int()
+		switch bk {
+		// kBool
+		case kInt:	return eq_Int_Int     (aint, b.Int())
+		case kUint:	return eq_Int_Uint    (aint, b.Uint())
+		case kFloat:	return eq_Int_Float   (aint, b.Float())
+		case kComplex:	return eq_Int_Complex (aint, b.Complex())
+		case kBigInt:	return eq_Int_BigInt  (aint, xb.(*big.Int))
+		}
+
+	case kUint:
+		auint := a.Uint()
+		switch bk {
+		// kBool
+		// kInt
+		case kUint:	return eq_Uint_Uint    (auint, b.Uint())
+		case kFloat:	return eq_Uint_Float   (auint, b.Float())
+		case kComplex:	return eq_Uint_Complex (auint, b.Complex())
+		case kBigInt:	return eq_Uint_BigInt  (auint, xb.(*big.Int))
+		}
+
+	case kFloat:
+		afloat := a.Float()
+		switch bk {
+		// kBool
+		// kInt
+		// kUint
+		case kFloat:	return eq_Float_Float   (afloat, b.Float())
+		case kComplex:	return eq_Float_Complex (afloat, b.Complex())
+		case kBigInt:	return eq_Float_BigInt  (afloat, xb.(*big.Int))
+		}
+
+	case kComplex:
+		acomplex := a.Complex()
+		switch bk {
+		// kBool
+		// kInt
+		// kUint
+		// kFloat
+		case kComplex:	return eq_Complex_Complex (acomplex, b.Complex())
+		case kBigInt:	return eq_Complex_BigInt  (acomplex, xb.(*big.Int))
+		}
+
+	case kBigInt:
+		switch bk {
+		// kBool
+		// kInt
+		// kUint
+		// kFloat
+		// kComplex
+		case kBigInt:	return eq_BigInt_BigInt  (xa.(*big.Int), xb.(*big.Int))
+		}
+
+	// slices
+	case kSlice:
+		switch bk {
+		case kSlice:	return eq_Slice_Slice (a, b)
+		}
+
+	// builtin map
+	case kMap:
+		switch bk {
+		case kMap:	return eq_Map_Map  (a, b)
+		}
+		switch b := xb.(type) {
+		case Dict:	return eq_Map_Dict (a, b)
+		}
+	}
+
+	if handled {
+		return false
+	}
+
+	// our types that need special handling
+	switch a := xa.(type) {
+	case Dict:
+		switch b := xb.(type) {
+		case Dict:	return eq_Dict_Dict(a, b)
+		default:        return false
+		}
+	}
+
+	// structs  (also covers None, Class, Call etc...)
+	switch ak {
+	case kStruct:
+		switch bk {
+		case kStruct:	return eq_Struct_Struct (a, b)
+		default:        return false
+		}
+	}
+
+	return (xa == xb) // fallback to builtin equality
+}
+
+
+// equality matrix. nontrivial elements
+
+func eq_Int_Uint(a int64, b uint64) bool {
+	if a >= 0 {
+		return uint64(a) == b
+	}
+	return false
+}
+
+
+func eq_Int_BigInt(a int64, b *big.Int) bool {
+	if b.IsInt64() {
+		return a == b.Int64()
+	}
+	return false
+}
+
+func eq_Uint_BigInt(a uint64, b *big.Int) bool {
+	if b.IsUint64() {
+		return a == b.Uint64()
+	}
+	return false
+}
+
+func eq_Float_BigInt(a float64, b *big.Int) bool {
+	bf, accuracy := bigInt_Float64(b)
+	if accuracy == big.Exact {
+		return a == bf
+	}
+	return false
+}
+
+func eq_Complex_BigInt(a complex128, b *big.Int) bool {
+	if imag(a) == 0 {
+		return eq_Float_BigInt(real(a), b)
+	}
+	return false
+}
+
+func eq_BigInt_BigInt(a, b *big.Int) bool {
+	return (a.Cmp(b) == 0)
+}
+
+func eq_Slice_Slice(a, b reflect.Value) bool {
+	al := a.Len()
+	bl := b.Len()
+	if al != bl {
+		return false
+	}
+	for i := 0; i < al; i++ {
+		if !equal(a.Index(i).Interface(), b.Index(i).Interface()) {
+			return false
+		}
+	}
+	return true
+}
+
+func eq_Struct_Struct(a, b reflect.Value) bool {
+	if a.Type() != b.Type() {
+		return false
+	}
+
+	typ := a.Type()
+	l := typ.NumField()
+	for i := 0; i < l; i++ {
+		af := a.Field(i)
+		bf := b.Field(i)
+
+		// .Interface() is not allowed if the field is private.
+		// Work around the protection via unsafe. We may need to switch
+		// to struct copy if it is not addressable because Addr() is
+		// used in the workaround. https://stackoverflow.com/a/43918797/9456786
+		ftyp := typ.Field(i)
+		if !ftyp.IsExported() {
+			if !af.CanAddr() {
+				// switch a to addressable copy
+				a_ := reflect.New(typ).Elem()
+				a_.Set(a)
+				a = a_
+				af = a.Field(i)
+			}
+
+			if !bf.CanAddr() {
+				// switch b to addressable copy
+				b_ := reflect.New(typ).Elem()
+				b_.Set(b)
+				b = b_
+				bf = b.Field(i)
+			}
+
+			af = reflect.NewAt(ftyp.Type, af.Addr().UnsafePointer()).Elem()
+			bf = reflect.NewAt(ftyp.Type, bf.Addr().UnsafePointer()).Elem()
+		}
+
+		if !equal(af.Interface(), bf.Interface()) {
+			return false
+		}
+	}
+	return true
+}
+
+func eq_Dict_Dict(a Dict, b Dict) bool {
+	// dicts D₁ and D₂ are considered equal if the following is true:
+	//
+	//     - len(D₁) = len(D₂)
+	//     - ∀ k ∈ D₁  equal(D₁[k], D₂[k]) = y
+	//     - ∀ k ∈ D₂  equal(D₁[k], D₂[k]) = y
+	//
+	// this definition is reasonable and fast to implement without additional memory.
+	// Also if D₁ and D₂ have keys only from equal-transitive subset of all
+	// keys (i.e. anything without ByteString), it becomes equivalent to the
+	// following definition:
+	//
+	//     - (k₁i, v₁i) is set of all key/values from D₁
+	//     - (k₂j, v₂j) is set of all key/values from D₂
+	//     - equal(D₁,D₂):
+	//
+	//         ∃ 1-1 mapping in between i<->j: equal(k₁i, k₂j) ^ equal(v₁i, v₂j)
+
+	if a.Len() != b.Len() {
+		return false
+	}
+
+	eq := true
+	a.Iter()(func(k,va any) bool {
+		vb, ok := b.Get_(k)
+		if !ok || !equal(va, vb) {
+			eq = false
+			return false
+		}
+		return true
+	})
+	if !eq {
+		return false
+	}
+
+	b.Iter()(func(k,vb any) bool {
+		va, ok := a.Get_(k)
+		if !ok || !equal(va, vb) {
+			eq = false
+			return false
+		}
+		return true
+	})
+	return eq
+}
+
+// equal(Map, Dict) and equal(Map, Map) follow semantic of equal(Dict, Dict)
+
+func eq_Map_Dict(a reflect.Value, b Dict) bool {
+	if a.Len() != b.Len() {
+		return false
+	}
+
+	aKeyType := a.Type().Key()
+
+	ai := a.MapRange()
+	for ai.Next() {
+		k  := ai.Key().Interface()
+		va := ai.Value().Interface()
+		vb, ok := b.Get_(k)
+		if !ok || !equal(va, vb) {
+			return false
+		}
+	}
+
+	eq := true
+	b.Iter()(func(k,vb any) bool {
+		xk := reflect.ValueOf(k)
+		if !xk.Type().AssignableTo(aKeyType) {
+			eq = false
+			return false
+		}
+		xva := a.MapIndex(xk)
+		if !(xva.IsValid() && equal(xva.Interface(), vb)) {
+			eq = false
+			return false
+		}
+		return true
+	})
+	return eq
+}
+
+func eq_Map_Map(a reflect.Value, b reflect.Value) bool {
+	if a.Len() != b.Len() {
+		return false
+	}
+
+	aKeyType := a.Type().Key()
+	bKeyType := b.Type().Key()
+
+	ai := a.MapRange()
+	for ai.Next() {
+		k  := ai.Key().Interface() // NOTE xk != ai.Key() because that might have type any
+		xk := reflect.ValueOf(k)   //      while xk has type of particular contained value
+		va := ai.Value().Interface()
+		if !xk.Type().AssignableTo(bKeyType) {
+			return false
+		}
+		xvb := b.MapIndex(xk)
+		if !(xvb.IsValid() && equal(va, xvb.Interface())) {
+			return false
+		}
+	}
+
+	bi := b.MapRange()
+	for bi.Next() {
+		k  := bi.Key().Interface() // see ^^^
+		xk := reflect.ValueOf(k)
+		vb := bi.Value().Interface()
+		if !xk.Type().AssignableTo(aKeyType) {
+			return false
+		}
+		xva := a.MapIndex(xk)
+		if !(xva.IsValid() && equal(xva.Interface(), vb)) {
+			return false
+		}
+	}
+
+	return true
+}
+
+
+// equality matrix. trivial elements
+
+func eq_Int_Int     (a int64, b int64)      bool { return a == b }
+func eq_Int_Float   (a int64, b float64)    bool { return float64(a) == b }
+func eq_Int_Complex (a int64, b complex128) bool { return complex(float64(a), 0) == b }
+
+func eq_Uint_Uint    (a uint64, b uint64)     bool { return a == b }
+func eq_Uint_Float   (a uint64, b float64)    bool { return float64(a) == b }
+func eq_Uint_Complex (a uint64, b complex128) bool { return complex(float64(a), 0) == b }
+
+func eq_Float_Float    (a float64, b float64)     bool { return a == b }
+func eq_Float_Complex  (a float64, b complex128)  bool { return complex(a, 0) == b }
+
+func eq_Complex_Complex (a complex128, b complex128)  bool { return a == b }
+
+
+// ---- hash ----
+
+// hash returns hash of x consistent with equality implemented by equal.
+//
+//	equal(a,b)  ⇒  hash(a) = hash(b)
+//
+// hash panics with "unhashable type: ..." if x is not allowed to be used as Dict key.
+func hash(seed maphash.Seed, x any) uint64 {
+	// strings/bytes use standard hash of string
+	switch v := x.(type) {
+	case string:     return maphash_String(seed, v)
+	case ByteString: return maphash_String(seed, string(v))
+	case Bytes:      return maphash_String(seed, string(v))
+	}
+
+	// for everything else we implement custom hashing ourselves to match equal
+	var h maphash.Hash
+	h.SetSeed(seed)
+
+	hash_Uint := func(u uint64) {
+		var b [8]byte
+		binary.BigEndian.PutUint64(b[:], u)
+		h.Write(b[:])
+	}
+
+	hash_Int := func(i int64) {
+		hash_Uint(uint64(i))
+	}
+
+	hash_Float := func(f float64) {
+		// if float is in int range and is integer number - hash it as integer
+		i  := int64(f)
+		f_ := float64(i)
+		if f_ == f {
+			hash_Int(i)
+
+		// else use raw float64 bytes representation for hashing
+		} else {
+			hash_Uint(math.Float64bits(f))
+		}
+	}
+
+
+	// numbers
+	r := reflect.ValueOf(x)
+	k := kindOf(x)
+
+	handled := true
+	switch k {
+	default:
+		handled = false
+
+	case kBool:   hash_Int(bint(r.Bool()))
+	case kInt:    hash_Int(r.Int())
+	case kUint:   hash_Uint(r.Uint())
+	case kFloat:  hash_Float(r.Float())
+
+	case kComplex:
+		c := r.Complex()
+		hash_Float(real(c))
+		if imag(c) != 0 {
+			hash_Float(imag(c))
+		}
+
+	case kBigInt:
+		b := x.(*big.Int)
+		switch {
+		case b.IsInt64():  hash_Int(b.Int64())
+		case b.IsUint64(): hash_Uint(b.Uint64())
+		default:
+			f, accuracy := bigInt_Float64(b)
+			if accuracy == big.Exact {
+				hash_Float(f)
+			} else {
+				h.WriteString("bigInt")
+				h.Write(b.Bytes())
+			}
+		}
+
+	// kSlice  - skip
+	// kStruct - skip
+
+	case kPointer:  hash_Uint(uint64(r.Elem().UnsafeAddr()))
+	}
+
+	if handled {
+		return h.Sum64()
+	}
+
+	// tuple
+	switch v := x.(type) {
+	case Tuple:
+		h.WriteString("tuple")
+		for _, item := range v {
+			hash_Uint(hash(seed, item))
+		}
+		return h.Sum64()
+	}
+
+	// structs  (also covers None, Class, Call etc)
+	switch k {
+	case kStruct:
+		// our types that are handled specially by equal
+		switch x.(type) {
+		case Dict:
+			goto unhashable
+		}
+
+		typ := r.Type()
+		h.WriteString(typ.Name())
+		l := typ.NumField()
+		for i := 0; i < l; i++ {
+			f := r.Field(i)
+
+			// .Interface() is not allowed if the field is private.
+			// Work it around via unsafe. See eq_Struct_Struct for details.
+			ftyp := typ.Field(i)
+			if !ftyp.IsExported() {
+				if !f.CanAddr() {
+					// switch r to addressable copy
+					r_ := reflect.New(typ).Elem()
+					r_.Set(r)
+					r = r_
+					f = r.Field(i)
+				}
+				f = reflect.NewAt(ftyp.Type, f.Addr().UnsafePointer()).Elem()
+			}
+
+			hash_Uint(hash(seed, f.Interface()))
+		}
+		return h.Sum64()
+	}
+
+unhashable:
+	panic(fmt.Sprintf("unhashable type: %T", x))
+}
+
+
+// ---- misc ----
+
+// bint returns int corresponding to bool.
+//
+// true  -> 1
+// false -> 0
+func bint(x bool) int64 {
+	if x {
+		return 1
+	}
+	return 0
+}

dict_test.go

+package ogórek
+
+import (
+	"fmt"
+	"hash/maphash"
+	"reflect"
+	"strings"
+	"testing"
+)
+
+
+// tStructWithPrivate is used by tests to verify handing of struct with private fields.
+type tStructWithPrivate struct {
+	x, y any
+}
+
+
+// TestEqual verifies equal and hash.
+func TestEqual(t *testing.T) {
+	// tEqualSet represents tested set of values:
+	// ∀ a ∈ tEqualSet:
+	//   ∀ b ∈ tEqualSet ⇒ equal(a,b) = y
+	//   ∀ c ∉ tEqualSet ⇒ equal(a,c) = n
+	//
+	// Intersection in between different tEqualSets is mostly empty: such
+	// intersections can contain elements only from all \ EqTransitive, i.e. only ByteString.
+	type tAllEqual []any
+
+	// E is shortcut to create tEqualSet
+	E := func(v ...any) tAllEqual { return tAllEqual(v) }
+
+	// D and M are shortcuts to create Dict and map[any]any
+	D := NewDictWithData
+	type M = map[any]any
+
+	// i1  and i1_  are two integer variables equal to 1 but with different address
+	// obj and obj_ are similar equal structures located at different memory regions
+	i1  := 1;  i1_ := 1
+	obj := &Class{"a","b"};  obj_ := &Class{"a","b"}
+
+	// testv is vector of all test-cases
+	testv := []tAllEqual {
+		// numbers
+		E(int(0),
+		   int64(0),  int32(0),  int16(0),  int8(0),
+		  uint64(0), uint32(0), uint16(0), uint8(0),
+		  bigInt("0"),
+		  false,
+		  float32  (0), float64   (0),
+		  complex64(0), complex128(0)),
+
+		E(int(1),
+		  int64 (1),  int32(1),  int16(1),  int8(1),
+		  uint64(1), uint32(1), uint16(1), uint8(1),
+		  bigInt("1"),
+		  true,
+		  float32  (1), float64   (1),
+		  complex64(1), complex128(1)),
+
+		E(int(-1),
+		  int64(-1), int32(-1), int16(-1),  int8(-1),
+		  // NOTE no uintX because they ≥ 0 only
+		  bigInt("-1"),
+		  // NOTE no bool because it ∈ {0,1}
+		  float32  (-1), float64   (-1),
+		  complex64(-1), complex128(-1)),
+
+		// intX/uintX different range
+		E(int(0xff),
+		   int64(0xff),  int32(0xff),  int16(0xff), // int8(overflow),
+		  uint64(0xff), uint32(0xff), uint16(0xff), // uint8(overflow),
+		  bigInt("255"),
+		  bigInt("255"), // two different *big.Int instances
+		  float32  (0xff), float64   (0xff),
+		  complex64(0xff), complex128(0xff)),
+
+		E(int(-0x80),
+		  int64(-0x80), int32(-0x80), int16(-0x80), int8(-0x80),
+		  //uint64(), uint32(), uint16(), uint8(), ≥ 0 only
+		  bigInt("-128"),
+		  float32  (-0x80), float64   (-0x80),
+		  complex64(-0x80), complex128(-0x80)),
+
+		E(int(0xffff),
+		   int64(0xffff),  int32(0xffff), // int16(overflow), int8(overflow),
+		  uint64(0xffff), uint32(0xffff), uint16(0xffff), // uint8(overflow),
+		  bigInt("65535"),
+		  float32  (0xffff), float64   (0xffff),
+		  complex64(0xffff), complex128(0xffff)),
+
+		E(int(-0x8000),
+		  int64(-0x8000), int32(-0x8000), int16(-0x8000), // int8(overflow),
+		  //uint64(), uint32(), uint16(), uint8(), ≥ 0 only
+		  bigInt("-32768"),
+		  float32  (-0x8000), float64   (-0x8000),
+		  complex64(-0x8000), complex128(-0x8000)),
+
+		E(int(0xffffffff),
+		   int64(0xffffffff), // int32(overflow), int16(overflow), int8(overflow),
+		  uint64(0xffffffff), uint32(0xffffffff), // uint16(overflow), uint8(overflow),
+		  bigInt("4294967295"),
+		  /* float32  (precision loss), */ float64   (0xffffffff),
+		  /* complex64(precision loss), */ complex128(0xffffffff)),
+
+		E(int(-0x80000000),
+		  int64(-0x80000000), int32(-0x80000000), // int16(overflow), int8(overflow),
+		  //uint64(), uint32(), uint16(), uint8(), ≥ 0 only
+		  bigInt("-2147483648"),
+		  float32  (-0x80000000), float64   (-0x80000000),
+		  complex64(-0x80000000), complex128(-0x80000000)),
+
+		E(// int(overflow),
+		  // int64(overflow), int32(overflow), int16(overflow), int8(overflow),
+		  uint64(0xffffffffffffffff), // uint32(overflow), uint16(overflow), uint8(overflow),
+		  bigInt("18446744073709551615")),
+		  // float32  (precision loss), float64   (precision loss),
+		  // complex64(precision loss), complex128(precision loss)),
+
+		E(int(-0x8000000000000000),
+		  int64(-0x8000000000000000), // int32(overflow), int16(overflow), int8(overflow),
+		  //uint64(), uint32(), uint16(), uint8(), ≥ 0 only
+		  bigInt("-9223372036854775808"),
+		  float32  (-0x8000000000000000), float64   (-0x8000000000000000),
+		  complex64(-0x8000000000000000), complex128(-0x8000000000000000)),
+
+		E(bigInt("1"+strings.Repeat("0",22)), float64(1e22), complex128(complex(1e22,0))),
+		E(complex64(complex(0,1)), complex128(complex(0,1))),
+		E(float64(1.25), float32(1.25), complex64(complex(1.25,0)), complex128(complex(1.25,0))),
+
+		// strings/bytes
+		E("",    ByteString("")),	E(ByteString(""),    Bytes("")),
+		E("a",   ByteString("a")),	E(ByteString("a"),   Bytes("a")),
+		E("мир", ByteString("мир")),	E(ByteString("мир"), Bytes("мир")),
+
+		// none / empty tuple|list
+		E(None{}),
+		E(Tuple{}, []any{}),
+
+		// sequences
+		E([]int{}, []float32{}, []any{}, Tuple{}, [0]float64{}),
+		E([]int{1,2}, []float32{1,2}, []any{1,2}, Tuple{1,2}, [2]float64{1,2}),
+		E([]any{1,"a"}, Tuple{1,"a"}, [2]any{1,"a"}, Tuple{1,ByteString("a")}),
+
+		// Dict, map
+		E(D(),
+		  M{}, map[int]bool{}),
+		E(D(1,bigInt("2")),
+		  M{1:2.0}, map[int]int{1:2}),
+		E(D(1,"a"),
+		  M{1:"a"}, map[int]string{1:"a"}),
+		E(D("a",1),
+		  M{"a":1}),
+		E(D("a",1, None{},2),
+		  M{"a":1, None{}:2}),
+		E(D("a",1, Bytes("a"),1),
+		  M{"a":1, Bytes("a"):1}),
+		E(D("a",1, Bytes("a"),2),
+		  M{"a":1, Bytes("a"):2}),
+		E(D("a",1), D(ByteString("a"),1)),  E(D(ByteString("a"),1), D(Bytes("a"),1)),
+		E(D("a",1, Bytes("a"),1, ByteString("b"),2),
+		  D(ByteString("a"),1, "b",2, Bytes("b"),2)),
+
+		// structs
+		E(Class{"mod","cls"}, Class{"mod","cls"}),
+		E(Call{Class{"mod","cls"}, Tuple{"a","b",3}},
+		  Call{Class{"mod","cls"}, Tuple{ByteString("a"),"b",bigInt("3")}}),
+		E(Ref{1}, Ref{bigInt("1")}, Ref{1.0}),
+		E(tStructWithPrivate{"a",1}, tStructWithPrivate{ByteString("a"),bigInt("1")}),
+		E(tStructWithPrivate{"b",2}, tStructWithPrivate{"b",2.0}),
+
+		// pointers, as in builtin ==, are compared only by address
+		E(&i1), E(&i1_), E(&obj), E(&obj_),
+
+		// nil
+		E(nil),
+	}
+	// automatically test equality on Tuples/list from ^^^ data
+	testvAddSequences := func() {
+		l := len(testv)
+		for i := 0; i < l; i++ {
+			Ex := testv[i]
+			Ey := testv[(i+1)%l]
+
+			x0 := Ex[0];  x1 := Ex[1%len(Ex)]
+			y0 := Ey[0];  y1 := Ey[1%len(Ey)]
+
+			t1 := Tuple{x0,y0};  l1 := []any{x0,y0}
+			t2 := Tuple{x1,y1};  l2 := []any{x1,y1}
+
+			testv = append(testv, E(t1, t2, l1, l2))
+		}
+	}
+	testvAddSequences()
+	// and sequences of sequences
+	testvAddSequences()
+
+	// thash is used to invoke hash.
+	// if x is not hashable ok=false is returned instead of panic.
+	tseed := maphash.MakeSeed()
+	thash := func(x any) (h uint64, ok bool) {
+		defer func() {
+			r := recover()
+			if r != nil {
+				s, sok := r.(string)
+				if sok && strings.HasPrefix(s, "unhashable type: ") {
+					ok = false
+					h  = 0
+				} else {
+					panic(r)
+				}
+			}
+		}()
+
+		return hash(tseed, x), true
+	}
+
+	// tequal is used to invoke equal.
+	// it automatically checks Go-extension, self-equal, symmetry and hash invariants:
+	//
+	//	a==b        ⇒  equal(a,b)
+	//	equal(a,a)  =  y
+	//	equal(a,b)  =  equal(b,a)
+	//	equal(a,b)  ⇒  hash(a) = hash(b)
+	tequal := func(a, b any) bool {
+		aa := equal(a, a)
+		bb := equal(b, b)
+		if !aa {
+			t.Errorf("not self-equal  %T %#v", a,a)
+		}
+		if !bb {
+			t.Errorf("not self-equal  %T %#v", b,b)
+		}
+
+		eq := equal(a, b)
+		qe := equal(b, a)
+
+		if eq != qe {
+			t.Errorf("equal not symmetric:  %T %#v  %T %#v;  a == b: %v  b == a: %v",
+				 a,a, b,b, eq, qe)
+		}
+
+		ah, ahOk := thash(a)
+		bh, bhOk := thash(b)
+		if eq && ahOk && bhOk && !(ah == bh) {
+			t.Errorf("hash different of equal  %T %#v hash:%x  %T %#v hash:%x",
+				 a,a,ah, b,b,bh)
+		}
+
+		goeq := false
+		func() {
+			// a == b can trigger "comparing uncomparable type ..."
+			// even if reflect reports both types as comparable
+			// (see mapTryAssign for details)
+			defer func() {
+				recover()
+			}()
+
+			goeq = (a == b)
+		}()
+
+		if goeq && !eq {
+			t.Errorf("equal is not extension of ==  %T %#v  %T %#v",
+				a,a, b,b)
+		}
+
+		return eq
+	}
+
+	// EHas returns whether x ∈ E.
+	EHas := func(E tAllEqual, x any) bool {
+		for _, a := range E {
+			if tequal(a, x) {
+				return true
+			}
+		}
+		return false;
+	}
+
+
+	// do the tests
+	for i, E1 := range testv {
+		// ∀ a,b ∈ tEqualSet ⇒ equal(a,b) = y
+		for _, a := range E1 {
+			for _, b := range E1 {
+				if !tequal(a,b) {
+					t.Errorf("not equal  %T %#v  %T %#v", a,a, b,b)
+				}
+			}
+		}
+
+		// ∀ a ∈ tEqualSet
+		// ∀ c ∉ tEqualSet ⇒ equal(a,c) = n
+		for j, E2 := range testv {
+			if j == i {
+				continue
+			}
+
+			for _, a := range E1 {
+				for _, c := range E2 {
+					if EHas(E1, c) {
+						continue
+					}
+
+					if tequal(a,c) {
+						t.Errorf("equal  %T %#v  %T %#v", a,a, c,c)
+					}
+				}
+			}
+		}
+	}
+}
+
+
+// TestDict verifies Dict.
+func TestDict(t *testing.T) {
+	d := NewDict()
+
+	// assertData asserts that d has data exactly as specified by provided key,value pairs.
+	assertData := func(kvok ...any) {
+		t.Helper()
+
+		if len(kvok) % 2 != 0 {
+			panic("kvok % 2 != 0")
+		}
+		lok := len(kvok)/2
+		kvokGet := func(k any) (any, bool) {
+			t.Helper()
+			for i := 0; i < lok; i++ {
+				kok := kvok[2*i]
+				vok := kvok[2*i+1]
+				if reflect.TypeOf(k) == reflect.TypeOf(kok) &&
+				   equal(k, kok) {
+					return vok, true
+				}
+			}
+			return nil, false
+		}
+
+		bad := false
+		badf := func(format string, argv ...any) {
+			t.Helper()
+			bad = true
+			t.Errorf(format, argv...)
+		}
+
+		l := d.Len()
+		if l != lok {
+			badf("len: have: %d  want: %d", l, lok)
+		}
+
+		d.Iter()(func(k,v any) bool {
+			t.Helper()
+			vok, ok := kvokGet(k)
+			if !ok {
+				badf("unexpected key %#v", k)
+			}
+			if v != vok {
+				badf("key %T %#v -> value %#T %#v  ;  want %T %#v", k,k, v,v, vok,vok)
+			}
+			return true
+		})
+
+		if bad {
+			t.Fatalf("\nd:   %#v\nkvok: %#v", d, kvok)
+		}
+	}
+
+	// assertGet asserts that d.Get(k) results in exactly vok or any element from vokExtra.
+	assertGet := func(k any, vok any, vokExtra ...any) {
+		t.Helper()
+		v := d.Get(k)
+		if v == vok {
+			return
+		}
+		for _, eok := range vokExtra {
+			if v == eok {
+				return
+			}
+		}
+
+		emsg := fmt.Sprintf("get %#v: have: %#v  want: %#v", k, v, vok)
+		for _, eok := range vokExtra {
+			emsg += fmt.Sprintf(" ∪ %#v", eok)
+		}
+		emsg += fmt.Sprintf("\nd: %#v", d)
+		t.Fatal(emsg)
+	}
+
+	// numbers
+	assertData()
+
+	d.Set(1, "x")
+	assertData(1,"x")
+	assertGet(1,            "x")
+	assertGet(1.0,          "x")
+	assertGet(bigInt("1"),  "x")
+	assertGet(complex(1,0), "x")
+
+	d.Del(7)
+	assertData(1,"x")
+	assertGet(1,            "x")
+	assertGet(1.0,          "x")
+	assertGet(bigInt("1"),  "x")
+	assertGet(complex(1,0), "x")
+
+	d.Set(2.5, "y")
+	assertData(1,"x", 2.5,"y")
+	assertGet(1,              "x")
+	assertGet(1.0,            "x")
+	assertGet(bigInt("1"),    "x")
+	assertGet(complex(1,0),   "x")
+	assertGet(2,              nil)
+	assertGet(2.5,            "y")
+	assertGet(bigInt("2"),    nil)
+	assertGet(complex(2.5,0), "y")
+
+	d.Del(1)
+	assertData(2.5,"y")
+	assertGet(1,              nil)
+	assertGet(1.0,            nil)
+	assertGet(bigInt("1"),    nil)
+	assertGet(complex(1,0),   nil)
+	assertGet(2,              nil)
+	assertGet(2.5,            "y")
+	assertGet(bigInt("2"),    nil)
+	assertGet(complex(2.5,0), "y")
+
+	d.Del(2.5)
+	assertData()
+	assertGet(1,              nil)
+	assertGet(1.0,            nil)
+	assertGet(bigInt("1"),    nil)
+	assertGet(complex(1,0),   nil)
+	assertGet(2,              nil)
+	assertGet(2.5,            nil)
+	assertGet(bigInt("2"),    nil)
+	assertGet(complex(2.5,0), nil)
+
+	// strings/bytes
+	assertData()
+	assertGet("abc", nil)
+
+	d.Set("abc", "a")
+	assertData("abc","a")
+	assertGet("abc",             "a")
+	assertGet(Bytes("abc"),      nil)
+	assertGet(ByteString("abc"), "a")
+
+	d.Set(Bytes("abc"), "b")
+	assertData("abc","a", Bytes("abc"),"b")
+	assertGet("abc",             "a")
+	assertGet(Bytes("abc"),      "b")
+	assertGet(ByteString("abc"), "a", "b")
+
+	d.Set(ByteString("abc"), "c")
+	assertData(ByteString("abc"),"c")
+	assertGet("abc",             "c")
+	assertGet(Bytes("abc"),      "c")
+	assertGet(ByteString("abc"), "c")
+
+	d.Del("abc")
+	assertData()
+	assertGet("abc",             nil)
+	assertGet(Bytes("abc"),      nil)
+	assertGet(ByteString("abc"), nil)
+
+	d.Set("abc", "a")
+	assertData("abc","a")
+	assertGet("abc",             "a")
+	assertGet(Bytes("abc"),      nil)
+	assertGet(ByteString("abc"), "a")
+
+	d.Set(Bytes("abc"), "b")
+	assertData("abc","a", Bytes("abc"),"b")
+	assertGet("abc",             "a")
+	assertGet(Bytes("abc"),      "b")
+	assertGet(ByteString("abc"), "a", "b")
+
+	d.Del(ByteString("abc"))
+	assertData()
+	assertGet("abc",             nil)
+	assertGet(Bytes("abc"),      nil)
+	assertGet(ByteString("abc"), nil)
+
+	// None, tuple
+	assertData()
+
+	d.Set(None{}, "n")
+	assertData(None{},"n")
+	assertGet(None{},  "n")
+	assertGet(Tuple{}, nil)
+
+	d.Set(Tuple{}, "t")
+	assertData(None{},"n", Tuple{},"t")
+	assertGet(None{},  "n")
+	assertGet(Tuple{}, "t")
+
+	d.Set(Tuple{1,2,"a"}, "t12a")
+	assertData(None{},"n", Tuple{},"t", Tuple{1,2,"a"},"t12a")
+	assertGet(None{},       "n")
+	assertGet(Tuple{},      "t")
+	assertGet(Tuple{1,2},   nil)
+	assertGet(Tuple{1,2,"a"},             "t12a")
+	assertGet(Tuple{1,2,Bytes("a")},      nil)
+	assertGet(Tuple{1,2,ByteString("a")}, "t12a")
+
+	d.Set(Tuple{1,2,Bytes("a")}, "t12b")
+	assertData(None{},"n", Tuple{},"t", Tuple{1,2,"a"},"t12a", Tuple{1,2,Bytes("a")},"t12b")
+	assertGet(None{},       "n")
+	assertGet(Tuple{},      "t")
+	assertGet(Tuple{1,2},   nil)
+	assertGet(Tuple{1,2,"a"},             "t12a")
+	assertGet(Tuple{1,2,Bytes("a")},      "t12b")
+	assertGet(Tuple{1,2,ByteString("a")}, "t12a", "t12b")
+
+	d.Set(Tuple{1,2,ByteString("a")}, "t12c")
+	assertData(None{},"n", Tuple{},"t", Tuple{1,2,ByteString("a")},"t12c")
+	assertGet(None{},       "n")
+	assertGet(Tuple{},      "t")
+	assertGet(Tuple{1,2},   nil)
+	assertGet(Tuple{1,2,"a"},             "t12c")
+	assertGet(Tuple{1,2,Bytes("a")},      "t12c")
+	assertGet(Tuple{1,2,ByteString("a")}, "t12c")
+
+	d.Set(Tuple{1,2,"a"}, "t12a")
+	assertData(None{},"n", Tuple{},"t", Tuple{1,2,"a"},"t12a")
+	assertGet(None{},       "n")
+	assertGet(Tuple{},      "t")
+	assertGet(Tuple{1,2},   nil)
+	assertGet(Tuple{1,2,"a"},             "t12a")
+	assertGet(Tuple{1,2,Bytes("a")},      nil)
+	assertGet(Tuple{1,2,ByteString("a")}, "t12a")
+
+	d.Set(Tuple{1,2,Bytes("a")}, "t12b")
+	assertData(None{},"n", Tuple{},"t", Tuple{1,2,"a"},"t12a", Tuple{1,2,Bytes("a")},"t12b")
+	assertGet(None{},       "n")
+	assertGet(Tuple{},      "t")
+	assertGet(Tuple{1,2},   nil)
+	assertGet(Tuple{1,2,"a"},             "t12a")
+	assertGet(Tuple{1,2,Bytes("a")},      "t12b")
+	assertGet(Tuple{1,2,ByteString("a")}, "t12a", "t12b")
+
+	d.Del(Tuple{1,2,ByteString("a")})
+	assertData(None{},"n", Tuple{},"t")
+	assertGet(None{},       "n")
+	assertGet(Tuple{},      "t")
+	assertGet(Tuple{1,2},   nil)
+	assertGet(Tuple{1,2,"a"},             nil)
+	assertGet(Tuple{1,2,Bytes("a")},      nil)
+	assertGet(Tuple{1,2,ByteString("a")}, nil)
+
+	// structs
+	d = NewDict()
+	d.Set(Class{"a","b"}, 1)
+	d.Set(Class{"c","d"}, 2)
+	d.Set(Ref{"a"}, 3)
+	d.Set(tStructWithPrivate{"x","y"}, 4)
+	assertData(Class{"a","b"},1, Class{"c","d"},2, Ref{"a"},3, tStructWithPrivate{"x","y"},4)
+	assertGet(Class{"a","b"},               1)
+	assertGet(Class{"c","d"},               2)
+	assertGet(Class{"x","y"},               nil)
+	assertGet(Ref{"a"},                     3)
+	assertGet(Ref{"x"},                     nil)
+	assertGet(tStructWithPrivate{"x","y"},  4)
+	assertGet(tStructWithPrivate{"p","q"},  nil)
+
+	// pointers
+	i := 1
+	j := 1
+	k := 1
+	x := Class{"a","b"}
+	y := Class{"a","b"}
+	z := Class{"a","b"}
+	d = NewDict()
+	d.Set(&i, 1)
+	d.Set(&j, 2)
+	d.Set(&x, 3)
+	d.Set(&y, 4)
+	assertData(&i,1, &j,2, &x,3, &y,4)
+	assertGet(&i, 1)
+	assertGet(&j, 2)
+	assertGet(&k, nil)
+	assertGet(&x, 3)
+	assertGet(&y, 4)
+	assertGet(&z, nil)
+
+	// NewDictWithSizeHint
+	d = NewDictWithSizeHint(100)
+	assertData()
+	assertGet(1,   nil)
+	assertGet(2,   nil)
+	assertGet("a", nil)
+	assertGet("b", nil)
+
+	// NewDictWithData
+	d = NewDictWithData("a",1, 2,"b")
+	assertData("a",1, 2,"b")
+	assertGet(1,   nil)
+	assertGet(2,   "b")
+	assertGet("a", 1)
+	assertGet("b", nil)
+
+	// unhashable types
+	vbad := []any{
+		[]any{},
+		[]any{1,2,3},
+		[]int{},
+		[]int{1,2,3},
+		NewDict(),
+		map[any]any{},
+		map[int]bool{},
+		Ref{[]any{}},
+		tStructWithPrivate{1,[]any{}},
+		tStructWithPrivate{[]any{},1},
+		tStructWithPrivate{[]any{},[]any{}},
+	}
+
+	assertPanics := func(subj any, errPrefix string, f func()) {
+		t.Helper()
+		defer func() {
+			t.Helper()
+			r := recover()
+			if r == nil {
+				t.Errorf("%#v: no panic", subj)
+				return
+			}
+			s, ok := r.(string)
+			if ok && strings.HasPrefix(s, errPrefix) {
+				// ok
+			} else {
+				panic(r)
+			}
+
+		}()
+
+		f()
+	}
+
+	for _, k := range vbad {
+		assertUnhashable := func(f func()) {
+			t.Helper()
+			assertPanics(k, "unhashable type: ", f)
+		}
+
+		assertUnhashable(func() { d.Get(k) })
+		assertUnhashable(func() { d.Set(k, 1) })
+		assertUnhashable(func() { d.Del(k) })
+		assertUnhashable(func() { NewDictWithData(k,1) })
+	}
+
+	// = ~nil
+	d = Dict{}
+	assertData()
+	assertGet(1,   nil)
+	assertGet(2,   nil)
+	assertGet("a", nil)
+	assertGet("b", nil)
+	d.Del(1)
+	assertData()
+	assertGet(1,   nil)
+	assertGet(2,   nil)
+	assertGet("a", nil)
+	assertGet("b", nil)
+
+	assertPanics("nil.Set", "Set called on nil map", func() { d.Set(1, "x") })
+}
+
+
+// benchmarks for map and Dict compare them from performance point of view.
+
+func BenchmarkMapGet(b *testing.B) {
+	m := map[any]any{}
+	for i := 0; i < 100; i++ {
+		m[i] = i
+	}
+	m["abc"] = 777
+
+	b.Run("string", func(b *testing.B) {
+		for i := 0; i < b.N; i++ {
+			_ = m["abc"]
+		}
+	})
+
+	b.Run("int", func(b *testing.B) {
+		for i := 0; i < b.N; i++ {
+			_ = m[77]
+		}
+	})
+}
+
+func BenchmarkDictGet(b *testing.B) {
+	d := NewDict()
+	for i := 0; i < 100; i++ {
+		d.Set(i, i)
+	}
+	d.Set("abc", 777)
+
+	b.Run("string", func(b *testing.B) {
+		for i := 0; i < b.N; i++ {
+			_ = d.Get("abc")
+		}
+	})
+
+	b.Run("int", func(b *testing.B) {
+		for i := 0; i < b.N; i++ {
+			_ = d.Get(77)
+		}
+	})
+}

doc.go

@@ -22,12 +22,29 @@
 //	long	↔  *big.Int
 //	float	↔  float64
 //	float	←  floatX
-//	list	↔  []interface{}
+//	list	↔  []any
 //	tuple	↔  ogórek.Tuple
-//	dict	↔  map[interface{}]interface{}
 //
 //
-// For strings there are two modes. In the first, default, mode both py2/py3
+// For dicts there are two modes. In the first, default, mode Python dicts are
+// decoded into standard Go map. This mode tries to use builtin Go type, but
+// cannot mirror py behaviour fully because e.g. int(1), big.Int(1) and
+// float64(1.0) are all treated as different keys by Go, while Python treats
+// them as being equal. It also does not support decoding dicts with tuple
+// used in keys:
+//
+//      dict    ↔  map[any]any                       PyDict=n mode, default
+//              ←  ogórek.Dict
+//
+// With PyDict=y mode, however, Python dicts are decoded as ogórek.Dict which
+// mirrors behaviour of Python dict with respect to keys equality, and with
+// respect to which types are allowed to be used as keys.
+//
+//      dict    ↔  ogórek.Dict                       PyDict=y mode
+//              ←  map[any]any
+//
+//
+// For strings there are also two modes. In the first, default, mode both py2/py3
 // str and py2 unicode are decoded into string with py2 str being considered
 // as UTF-8 encoded. Correspondingly for protocol ≤ 2 Go string is encoded as
 // UTF-8 encoded py2 str, and for protocol ≥ 3 as py3 str / py2 unicode.
@@ -155,6 +172,11 @@
 // Using the helpers fits into Python3 strings/bytes model but also allows to
 // handle the data generated from under Python2.
 //
+// Similarly Dict considers ByteString to be equal to both string and Bytes
+// with the same underlying content. This allows programs to access Dict via
+// string/bytes keys following Python3 model, while still being able to handle
+// dictionaries generated from under Python2.
+//
 //
 // --------
 //

encode.go

@@ -505,6 +505,41 @@ func (e *Encoder) encodeMap(m reflect.Value) error {
 	return e.emit(opDict)
 }
 
+func (e *Encoder) encodeDict(d Dict) error {
+	l := d.Len()
+
+	// protocol >= 1: ø dict -> EMPTY_DICT
+	if e.config.Protocol >= 1 && l == 0 {
+		return e.emit(opEmptyDict)
+	}
+
+	// MARK + ... + DICT
+	// TODO cycles + sort keys (see encodeMap for details)
+	err := e.emit(opMark)
+	if err != nil {
+		return err
+	}
+
+	d.Iter()(func(k, v any) bool {
+		err = e.encode(reflectValueOf(k))
+		if err != nil {
+			return false
+		}
+
+		err = e.encode(reflectValueOf(v))
+		if err != nil {
+			return false
+		}
+
+		return true
+	})
+	if err != nil {
+		return err
+	}
+
+	return e.emit(opDict)
+}
+
 func (e *Encoder) encodeCall(v *Call) error {
 	err := e.encodeClass(&v.Callable)
 	if err != nil {
@@ -578,6 +613,8 @@ func (e *Encoder) encodeStruct(st reflect.Value) error {
 		return e.encodeRef(&v)
 	case big.Int:
 		return e.encodeLong(&v)
+	case Dict:
+		return e.encodeDict(v)
 	}
 
 	structTags := getStructTags(st)

fuzz.go

@@ -5,24 +5,27 @@ package ogórek
 import (
 	"bytes"
 	"fmt"
-	"reflect"
 )
 
 func Fuzz(data []byte) int {
-	f1 := fuzz(data, false)
-	f2 := fuzz(data, true)
+	f := 0
+
+	f += fuzz(data, false, false)
+	f += fuzz(data, false, true)
+	f += fuzz(data, true,  false)
+	f += fuzz(data, true,  true)
 
-	f := f1+f2
 	if f > 1 {
 		f = 1
 	}
 	return f
 }
 
-func fuzz(data []byte, strictUnicode bool) int {
+func fuzz(data []byte, pyDict, strictUnicode bool) int {
 	// obj = decode(data) - this tests things like stack overflow in Decoder
 	buf := bytes.NewBuffer(data)
 	dec := NewDecoderWithConfig(buf, &DecoderConfig{
+		PyDict:        pyDict,
 		StrictUnicode: strictUnicode,
 	})
 	obj, err := dec.Decode()
@@ -38,7 +41,7 @@ func fuzz(data []byte, strictUnicode bool) int {
 	// because obj - as we got it as decoding from input - is known not to
 	// contain arbitrary Go structs.
 	for proto := 0; proto <= highestProtocol; proto++ {
-		subj := fmt.Sprintf("strictUnicode %v: protocol %d", strictUnicode, proto)
+		subj := fmt.Sprintf("pyDict %v: strictUnicode %v: protocol %d", pyDict, strictUnicode, proto)
 
 		buf.Reset()
 		enc := NewEncoderWithConfig(buf, &EncoderConfig{
@@ -67,6 +70,7 @@ func fuzz(data []byte, strictUnicode bool) int {
 		encoded := buf.String()
 
 		dec = NewDecoderWithConfig(bytes.NewBufferString(encoded), &DecoderConfig{
+			PyDict:        pyDict,
 			StrictUnicode: strictUnicode,
 		})
 		obj2, err := dec.Decode()
@@ -75,7 +79,7 @@ func fuzz(data []byte, strictUnicode bool) int {
 			panic(fmt.Sprintf("%s: decode back error: %s\npickle: %q", subj, err, encoded))
 		}
 
-		if !reflect.DeepEqual(obj, obj2) {
+		if !deepEqual(obj, obj2) {
 			panic(fmt.Sprintf("%s: decode·encode != identity:\nhave: %#v\nwant: %#v", subj, obj2, obj))
 		}
 	}

fuzz/corpus/test-226742f92b8d32c29e3e40915db82adfa33bfa08.pickle

+((tI0
+d.
\ No newline at end of file

fuzz/corpus/test-67c9063161f712ed30c0edf3782bc585a3bd1ccf.pickle

+((I1
+I2
+tI0
+d.
\ No newline at end of file

fuzz/corpus/test-ba9950130dfd69a59235c78dfcdedeb4e8713f79.pickle

+(L123L
+I0
+d.
\ No newline at end of file

go.mod

 module github.com/kisielk/og-rek
 
-go 1.12
+go 1.18
+
+require github.com/aristanetworks/gomap v0.0.0-20230726210543-f4e41046dced
+
+require golang.org/x/exp v0.0.0-20230725093048-515e97ebf090 // indirect

go.sum

+github.com/aristanetworks/gomap v0.0.0-20230726210543-f4e41046dced h1:HxlRMDx/VeRqzj3nvqX9k4tjeBcEIkoNHDJPsS389hs=
+github.com/aristanetworks/gomap v0.0.0-20230726210543-f4e41046dced/go.mod h1:p7lmI+ecoe1RTyD11SPXWsSQ3H+pJ4cp5y7vtKW4QdM=
+golang.org/x/exp v0.0.0-20230725093048-515e97ebf090 h1:Di6/M8l0O2lCLc6VVRWhgCiApHV8MnQurBnFSHsQtNY=
+golang.org/x/exp v0.0.0-20230725093048-515e97ebf090/go.mod h1:FXUEEKJgO7OQYeo8N01OfiKP8RXMtf6e8aTskBGqWdc=

ogorek.go

@@ -189,6 +189,11 @@ type DecoderConfig struct {
 	// decoded into ByteString in this mode. See StrictUnicode mode
 	// documentation in top-level package overview for details.
 	StrictUnicode bool
+
+	// PyDict, when true, requests to decode Python dicts as ogórek.Dict
+	// instead of builtin map. See PyDict mode documentation in top-level
+	// package overview for details.
+	PyDict bool
 }
 
 // NewDecoder returns a new Decoder with the default configuration.
@@ -1009,29 +1014,75 @@ func mapTryAssign(m map[interface{}]interface{}, key, value interface{}) (ok boo
 	return
 }
 
+// dictTryAssign is like mapTryAssign but for Dict.
+func dictTryAssign(d Dict, key, value interface{}) (ok bool) {
+	defer func() {
+		if r := recover(); r != nil {
+			ok = false
+		}
+	}()
+
+	d.Set(key, value)
+	ok = true
+	return
+}
+
 func (d *Decoder) loadDict() error {
 	k, err := d.marker()
 	if err != nil {
 		return err
 	}
 
-	m := make(map[interface{}]interface{}, 0)
 	items := d.stack[k+1:]
 	if len(items) % 2 != 0 {
 		return fmt.Errorf("pickle: loadDict: odd # of elements")
 	}
+
+	var m interface{}
+	if d.config.PyDict {
+		m, err = d.loadDictDict(items)
+	} else {
+		m, err = d.loadDictMap(items)
+	}
+
+	if err != nil {
+		return err
+	}
+
+	d.stack = append(d.stack[:k], m)
+	return nil
+}
+
+func (d *Decoder) loadDictMap(items []interface{}) (map[interface{}]interface{}, error) {
+	m := make(map[interface{}]interface{}, len(items)/2)
 	for i := 0; i < len(items); i += 2 {
 		key := items[i]
 		if !mapTryAssign(m, key, items[i+1]) {
-			return fmt.Errorf("pickle: loadDict: invalid key type %T", key)
+			return nil, fmt.Errorf("pickle: loadDict: map: invalid key type %T", key)
 		}
 	}
-	d.stack = append(d.stack[:k], m)
-	return nil
+	return m, nil
 }
 
+func (d *Decoder) loadDictDict(items []interface{}) (Dict, error) {
+	m := NewDictWithSizeHint(len(items)/2)
+	for i := 0; i < len(items); i += 2 {
+		key := items[i]
+		if !dictTryAssign(m, key, items[i+1]) {
+			return Dict{}, fmt.Errorf("pickle: loadDict: Dict: invalid key type %T", key)
+		}
+	}
+	return m, nil
+}
+
+
 func (d *Decoder) loadEmptyDict() error {
-	m := make(map[interface{}]interface{}, 0)
+	var m interface{}
+	if d.config.PyDict {
+		m = NewDict()
+	} else {
+		m = make(map[interface{}]interface{}, 0)
+	}
 	d.push(m)
 	return nil
 }
@@ -1218,10 +1269,14 @@ func (d *Decoder) loadSetItem() error {
 	switch m := m.(type) {
 	case map[interface{}]interface{}:
 		if !mapTryAssign(m, k, v) {
-			return fmt.Errorf("pickle: loadSetItem: invalid key type %T", k)
+			return fmt.Errorf("pickle: loadSetItem: map: invalid key type %T", k)
+		}
+	case Dict:
+		if !dictTryAssign(m, k, v) {
+			return fmt.Errorf("pickle: loadSetItem: Dict: invalid key type %T", k)
 		}
 	default:
-		return fmt.Errorf("pickle: loadSetItem: expected a map, got %T", m)
+		return fmt.Errorf("pickle: loadSetItem: expected a map or Dict, got %T", m)
 	}
 	return nil
 }
@@ -1234,23 +1289,31 @@ func (d *Decoder) loadSetItems() error {
 	if k < 1 {
 		return errStackUnderflow
 	}
+	if (len(d.stack) - (k + 1)) % 2 != 0 {
+		return fmt.Errorf("pickle: loadSetItems: odd # of elements")
+	}
 
 	l := d.stack[k-1]
 	switch m := l.(type) {
 	case map[interface{}]interface{}:
-		if (len(d.stack) - (k + 1)) % 2 != 0 {
-			return fmt.Errorf("pickle: loadSetItems: odd # of elements")
-		}
 		for i := k + 1; i < len(d.stack); i += 2 {
 			key := d.stack[i]
 			if !mapTryAssign(m, key, d.stack[i+1]) {
-				return fmt.Errorf("pickle: loadSetItems: invalid key type %T", key)
+				return fmt.Errorf("pickle: loadSetItems: map: invalid key type %T", key)
 			}
 		}
-		d.stack = append(d.stack[:k-1], m)
+	case Dict:
+		for i := k + 1; i < len(d.stack); i += 2 {
+			key := d.stack[i]
+			if !dictTryAssign(m, key, d.stack[i+1]) {
+				return fmt.Errorf("pickle: loadSetItems: Dict: invalid key type %T", key)
+			}
+		}
+
 	default:
-		return fmt.Errorf("pickle: loadSetItems: expected a map, got %T", m)
+		return fmt.Errorf("pickle: loadSetItems: expected a map or Dict, got %T", m)
 	}
+	d.stack = append(d.stack[:k-1], l)
 	return nil
 }
 

ogorek_test.go

@@ -7,7 +7,6 @@ import (
 	"fmt"
 	"io"
 	"math/big"
-	"reflect"
 	"strconv"
 	"strings"
 	"testing"
@@ -86,6 +85,9 @@ type TestEntry struct {
 
 	strictUnicodeN bool // whether to test with StrictUnicode=n while decoding/encoding
 	strictUnicodeY bool // whether to test with StrictUnicode=y while decoding/encoding
+
+	pyDictN bool // whether to test with PyDict=n while decoding/encoding
+	pyDictY bool // ----//----           PyDict=y
 }
 
 // X, I, P0, P1, P* form a language to describe decode/encode tests:
@@ -105,7 +107,8 @@ type TestEntry struct {
 // the entry is tested under both StrictUnicode=n and StrictUnicode=y modes.
 func X(name string, object interface{}, picklev ...TestPickle) TestEntry {
 	return TestEntry{name: name, objectIn: object, objectOut: object, picklev: picklev,
-			 strictUnicodeN: true, strictUnicodeY: true}
+			 strictUnicodeN: true, strictUnicodeY: true,
+			 pyDictN: true, pyDictY: true}
 }
 
 // Xuauto is syntactic sugar to prepare one TestEntry that is tested only under StrictUnicode=n mode.
@@ -122,6 +125,38 @@ func Xustrict(name string, object interface{}, picklev ...TestPickle) TestEntry
 	return x
 }
 
+// Xdgo is syntactic sugar to prepare one TestEntry that is tested only under PyDict=n mode.
+func Xdgo(name string, object interface{}, picklev ...TestPickle) TestEntry {
+	x := X(name, object, picklev...)
+	x.pyDictY = false
+	return x
+}
+
+// Xdpy is syntactic sugar to prepare one TestEntry that is tested only under PyDict=y mode.
+func Xdpy(name string, object interface{}, picklev ...TestPickle) TestEntry {
+	x := X(name, object, picklev...)
+	x.pyDictN = false
+	return x
+}
+
+// Xuauto_dgo is syntactic sugar to prepare one TestEntry that is tested only
+// under StrictUnicode=n ^ pyDict=n mode.
+func Xuauto_dgo(name string, object interface{}, picklev ...TestPickle) TestEntry {
+	x := X(name, object, picklev...)
+	x.strictUnicodeY = false
+	x.pyDictY = false
+	return x
+}
+
+// Xuauto_dpy is syntactic sugar to prepare one TestEntry that is tested only
+// under StrictUnicode=n ^ pyDict=y mode.
+func Xuauto_dpy(name string, object interface{}, picklev ...TestPickle) TestEntry {
+	x := X(name, object, picklev...)
+	x.strictUnicodeY = false
+	x.pyDictN = false
+	return x
+}
+
 // Xloosy is syntactic sugar to prepare one TestEntry with loosy encoding.
 func Xloosy(name string, objectIn, objectOut interface{}, picklev ...TestPickle) TestEntry {
 	x := X(name, objectIn, picklev...)
@@ -129,9 +164,9 @@ func Xloosy(name string, objectIn, objectOut interface{}, picklev ...TestPickle)
 	return x
 }
 
-// Xloosy_uauto is like Xuauto but for Xloosy.
-func Xloosy_uauto(name string, objectIn, objectOut interface{}, picklev ...TestPickle) TestEntry {
-	x := Xuauto(name, objectIn, picklev...)
+// Xloosy_uauto_dgo is like Xuauto_dgo but for Xloosy.
+func Xloosy_uauto_dgo(name string, objectIn, objectOut interface{}, picklev ...TestPickle) TestEntry {
+	x := Xuauto_dgo(name, objectIn, picklev...)
 	x.objectOut = objectOut
 	return x
 }
@@ -416,19 +451,46 @@ var tests = []TestEntry{
 		// bytearray(text, encoding); GLOBAL + BINUNICODE + TUPLE + REDUCE
 		I("c__builtin__\nbytearray\nq\x00(X\x13\x00\x00\x00hello\n\xc3\x90\xc2\xbc\xc3\x90\xc2\xb8\xc3\x91\xc2\x80\x01q\x01X\x07\x00\x00\x00latin-1q\x02tq\x03Rq\x04.")),
 
+	// dicts in default PyDict=n mode
+
+	Xdgo("dict({})", make(map[interface{}]interface{}),
+		P0("(d."), // MARK + DICT
+		P1_("}."), // EMPTY_DICT
+		I("(dp0\n.")),
+
+	Xuauto_dgo("dict({'a': '1'})", map[interface{}]interface{}{"a": "1"},
+		P0("(S\"a\"\nS\"1\"\nd."),                     // MARK + STRING + DICT
+		P12("(U\x01aU\x011d."),                        // MARK + SHORT_BINSTRING + DICT
+		P3("(X\x01\x00\x00\x00aX\x01\x00\x00\x001d."), // MARK + BINUNICODE + DICT
+		P4_("(\x8c\x01a\x8c\x011d.")),                 // MARK + SHORT_BINUNICODE + DICT
+
+	Xuauto_dgo("dict({'a': '1', 'b': '2'})", map[interface{}]interface{}{"a": "1", "b": "2"},
+		// map iteration order is not stable - test only decoding
+		I("(S\"a\"\nS\"1\"\nS\"b\"\nS\"2\"\nd."), // P0: MARK + STRING + DICT
+		I("(U\x01aU\x011U\x01bU\x012d."),         // P12: MARK + SHORT_BINSTRING + DICT
+
+		// P3: MARK + BINUNICODE + DICT
+		I("(X\x01\x00\x00\x00aX\x01\x00\x00\x001X\x01\x00\x00\x00bX\x01\x00\x00\x002d."),
+
+		I("(\x8c\x01a\x8c\x011\x8c\x01b\x8c\x012d."), // P4_: MARK + SHORT_BINUNICODE + DICT
+		I("(dS'a'\nS'1'\nsS'b'\nS'2'\ns."),           // MARK + DICT + STRING + SETITEM
+		I("}(U\x01aU\x011U\x01bU\x012u."),            // EMPTY_DICT + MARK + SHORT_BINSTRING + SETITEMS
+		I("(dp0\nS'a'\np1\nS'1'\np2\nsS'b'\np3\nS'2'\np4\ns.")),
+
+	// dicts in PyDict=y mode
 
-	X("dict({})", make(map[interface{}]interface{}),
+	Xdpy("dict({})", NewDict(),
 		P0("(d."), // MARK + DICT
 		P1_("}."), // EMPTY_DICT
 		I("(dp0\n.")),
 
-	Xuauto("dict({'a': '1'})", map[interface{}]interface{}{"a": "1"},
+	Xuauto_dpy("dict({'a': '1'})", NewDictWithData("a","1"),
 		P0("(S\"a\"\nS\"1\"\nd."),                     // MARK + STRING + DICT
 		P12("(U\x01aU\x011d."),                        // MARK + SHORT_BINSTRING + DICT
 		P3("(X\x01\x00\x00\x00aX\x01\x00\x00\x001d."), // MARK + BINUNICODE + DICT
 		P4_("(\x8c\x01a\x8c\x011d.")),                 // MARK + SHORT_BINUNICODE + DICT
 
-	Xuauto("dict({'a': '1', 'b': '2'})", map[interface{}]interface{}{"a": "1", "b": "2"},
+	Xuauto_dpy("dict({'a': '1', 'b': '2'})", NewDictWithData("a","1", "b","2"),
 		// map iteration order is not stable - test only decoding
 		I("(S\"a\"\nS\"1\"\nS\"b\"\nS\"2\"\nd."), // P0: MARK + STRING + DICT
 		I("(U\x01aU\x011U\x01bU\x012d."),         // P12: MARK + SHORT_BINSTRING + DICT
@@ -441,6 +503,21 @@ var tests = []TestEntry{
 		I("}(U\x01aU\x011U\x01bU\x012u."),            // EMPTY_DICT + MARK + SHORT_BINSTRING + SETITEMS
 		I("(dp0\nS'a'\np1\nS'1'\np2\nsS'b'\np3\nS'2'\np4\ns.")),
 
+	Xdpy("dict({123L: 0})", NewDictWithData(bigInt("123"), int64(0)),
+		P0("(L123L\nI0\nd."),    // MARK + LONG + INT + DICT
+		P1("(L123L\nK\x00d."),   // MARK + LONG + BININT1 + DICT
+		I("(\x8a\x01{K\x00d.")), // MARK + LONG1 + BININT1 + DICT
+
+	Xdpy("dict(tuple(): 0)", NewDictWithData(Tuple{}, int64(0)),
+		P0("((tI0\nd."),   // MARK + MARK + TUPLE + INT + DICT
+		P1_("()K\x00d.")), // MARK + EMPTY_TUPLE + BININT1 + DICT
+
+	Xdpy("dict(tuple(1,2): 0)", NewDictWithData(Tuple{int64(1), int64(2)}, int64(0)),
+		P0("((I1\nI2\ntI0\nd."),        // MARK + MARK + INT + INT + TUPLE + INT + DICT
+		P1("((K\x01K\x02tK\x00d."),     // MARK + MARK + BININT1 + BININT1 + TUPLE + BININT1 + DICT
+		P2_("(K\x01K\x02\x86K\x00d.")), // MARK + BININT1 + BININT1 + TUPLE2 + BININT1 + DICT
+
+
 	Xuauto("foo.bar  # global", Class{Module: "foo", Name: "bar"},
 		P0123("cfoo\nbar\n."),              // GLOBAL
 		P4_("\x8c\x03foo\x8c\x03bar\x93."), // SHORT_BINUNICODE + STACK_GLOBAL
@@ -480,9 +557,9 @@ var tests = []TestEntry{
 	X("LONG_BINPUT", []interface{}{int64(17)},
 		I("(lr0000I17\na.")),
 
-	Xuauto("graphite message1", graphiteObject1, graphitePickle1),
-	Xuauto("graphite message2", graphiteObject2, graphitePickle2),
-	Xuauto("graphite message3", graphiteObject3, graphitePickle3),
+	Xuauto_dgo("graphite message1", graphiteObject1, graphitePickle1),
+	Xuauto_dgo("graphite message2", graphiteObject2, graphitePickle2),
+	Xuauto_dgo("graphite message3", graphiteObject3, graphitePickle3),
 	Xuauto("too long line", longLine, I("V" + longLine + "\n.")),
 
 	// opcodes from protocol 4
@@ -492,7 +569,7 @@ var tests = []TestEntry{
 
 	// loosy encode: decoding back gives another object.
 	// the only case where ogórek encoding is loosy is for Go struct types.
-	Xloosy_uauto("[]ogórek.foo{\"Qux\", 4}", []foo{{"Qux", 4}},
+	Xloosy_uauto_dgo("[]ogórek.foo{\"Qux\", 4}", []foo{{"Qux", 4}},
 		[]interface{}{map[interface{}]interface{}{"Foo": "Qux", "Bar": int64(4)}},
 
 		// MARK + STRING + INT + DICT + LIST
@@ -519,9 +596,16 @@ type foo struct {
 // protocol prefix is always automatically prepended and is always concrete.
 var protoPrefixTemplate = string([]byte{opProto, 0xff})
 
-// TestDecode verifies ogórek decoder.
-func TestDecode(t *testing.T) {
-	for _, test := range tests {
+// WithEachMode runs f under all decoding/encoding modes covered by test entry.
+func (test TestEntry) WithEachMode(t *testing.T, f func(t *testing.T, decConfig DecoderConfig, encConfig EncoderConfig)) {
+	for _, pyDict := range []bool{false, true} {
+		if pyDict && !test.pyDictY {
+			continue
+		}
+		if !pyDict && !test.pyDictN {
+			continue
+		}
+
 		for _, strictUnicode := range []bool{false, true} {
 			if  strictUnicode && !test.strictUnicodeY {
 				continue
@@ -529,8 +613,28 @@ func TestDecode(t *testing.T) {
 			if !strictUnicode && !test.strictUnicodeN {
 				continue
 			}
-			testname := fmt.Sprintf("%s/StrictUnicode=%s", test.name, yn(strictUnicode))
 
+			t.Run(fmt.Sprintf("%s/PyDict=%s/StrictUnicode=%s", test.name, yn(pyDict), yn(strictUnicode)),
+			      func(t *testing.T) {
+				decConfig := DecoderConfig{
+					PyDict:        pyDict,
+					StrictUnicode: strictUnicode,
+				}
+				encConfig := EncoderConfig{
+					// no PyDict setting for encoder
+					StrictUnicode: strictUnicode,
+				}
+				f(t, decConfig, encConfig)
+			})
+
+		}
+	}
+}
+
+// TestDecode verifies ogórek decoder.
+func TestDecode(t *testing.T) {
+	for _, test := range tests {
+		test.WithEachMode(t, func(t *testing.T, decConfig DecoderConfig, encConfig EncoderConfig) {
 			for _, pickle := range test.picklev {
 				if pickle.err != nil {
 					continue
@@ -543,32 +647,24 @@ func TestDecode(t *testing.T) {
 						data := string([]byte{opProto, byte(proto)}) +
 							pickle.data[len(protoPrefixTemplate):]
 
-						t.Run(fmt.Sprintf("%s/%q/proto=%d", testname, data, proto), func(t *testing.T) {
-							testDecode(t, strictUnicode, test.objectOut, data)
+						t.Run(fmt.Sprintf("%q/proto=%d", data, proto), func(t *testing.T) {
+							testDecode(t, decConfig, test.objectOut, data)
 						})
 					}
 				} else {
-					t.Run(fmt.Sprintf("%s/%q", testname, pickle.data), func(t *testing.T) {
-						testDecode(t, strictUnicode, test.objectOut, pickle.data)
+					t.Run(fmt.Sprintf("%q", pickle.data), func(t *testing.T) {
+						testDecode(t, decConfig, test.objectOut, pickle.data)
 					})
 				}
 			}
-		}
+		})
 	}
 }
 
 // TestEncode verifies ogórek encoder.
 func TestEncode(t *testing.T) {
 	for _, test := range tests {
-		for _, strictUnicode := range []bool{false, true} {
-			if  strictUnicode && !test.strictUnicodeY {
-				continue
-			}
-			if !strictUnicode && !test.strictUnicodeN {
-				continue
-			}
-			testname := fmt.Sprintf("%s/StrictUnicode=%s", test.name, yn(strictUnicode))
-
+		test.WithEachMode(t, func(t *testing.T, decConfig DecoderConfig, encConfig EncoderConfig) {
 			alreadyTested := make(map[int]bool) // protocols we tested encode with so far
 			for _, pickle := range test.picklev {
 				for _, proto := range pickle.protov {
@@ -578,8 +674,8 @@ func TestEncode(t *testing.T) {
 						dataOk = string([]byte{opProto, byte(proto)}) + dataOk
 					}
 
-					t.Run(fmt.Sprintf("%s/proto=%d", testname, proto), func(t *testing.T) {
-						testEncode(t, proto, strictUnicode, test.objectIn, test.objectOut, dataOk, pickle.err)
+					t.Run(fmt.Sprintf("proto=%d", proto), func(t *testing.T) {
+						testEncode(t, proto, encConfig, decConfig, test.objectIn, test.objectOut, dataOk, pickle.err)
 					})
 
 					alreadyTested[proto] = true
@@ -592,11 +688,11 @@ func TestEncode(t *testing.T) {
 					continue
 				}
 
-				t.Run(fmt.Sprintf("%s/proto=%d(roundtrip)", testname, proto), func(t *testing.T) {
-					testEncode(t, proto, strictUnicode, test.objectIn, test.objectOut, "", nil)
+				t.Run(fmt.Sprintf("proto=%d(roundtrip)", proto), func(t *testing.T) {
+					testEncode(t, proto, encConfig, decConfig, test.objectIn, test.objectOut, "", nil)
 				})
 			}
-		}
+		})
 	}
 }
 
@@ -604,11 +700,9 @@ func TestEncode(t *testing.T) {
 //
 // It also verifies decoder robustness - via feeding it various kinds of
 // corrupt data derived from input.
-func testDecode(t *testing.T, strictUnicode bool, object interface{}, input string) {
+func testDecode(t *testing.T, decConfig DecoderConfig, object interface{}, input string) {
 	newDecoder := func(r io.Reader) *Decoder {
-		return NewDecoderWithConfig(r, &DecoderConfig{
-			StrictUnicode: strictUnicode,
-		})
+		return NewDecoderWithConfig(r, &decConfig)
 	}
 
 	// decode(input) -> expected
@@ -619,7 +713,7 @@ func testDecode(t *testing.T, strictUnicode bool, object interface{}, input stri
 		t.Error(err)
 	}
 
-	if !reflect.DeepEqual(v, object) {
+	if !deepEqual(v, object) {
 		t.Errorf("decode:\nhave: %#v\nwant: %#v", v, object)
 	}
 
@@ -665,18 +759,16 @@ func testDecode(t *testing.T, strictUnicode bool, object interface{}, input stri
 // encode-back tests are still performed.
 //
 // If errOk != nil, object encoding must produce that error.
-func testEncode(t *testing.T, proto int, strictUnicode bool, object, objectDecodedBack interface{}, dataOk string, errOk error) {
+func testEncode(t *testing.T, proto int, encConfig EncoderConfig, decConfig DecoderConfig, object, objectDecodedBack interface{}, dataOk string, errOk error) {
 	newEncoder := func(w io.Writer) *Encoder {
-		return NewEncoderWithConfig(w, &EncoderConfig{
-			Protocol:      proto,
-			StrictUnicode: strictUnicode,
-		})
+		econf := EncoderConfig{}
+		econf = encConfig
+		econf.Protocol = proto
+		return NewEncoderWithConfig(w, &econf)
 	}
 
 	newDecoder := func(r io.Reader) *Decoder {
-		return NewDecoderWithConfig(r, &DecoderConfig{
-			StrictUnicode: strictUnicode,
-		})
+		return NewDecoderWithConfig(r, &decConfig)
 	}
 
 	buf := &bytes.Buffer{}
@@ -716,9 +808,9 @@ func testEncode(t *testing.T, proto int, strictUnicode bool, object, objectDecod
 	if err != nil {
 		t.Errorf("encode -> decode -> error: %s", err)
 	} else {
-		if !reflect.DeepEqual(v, objectDecodedBack) {
+		if !deepEqual(v, objectDecodedBack) {
 			what := "identity"
-			if !reflect.DeepEqual(object, objectDecodedBack) {
+			if !deepEqual(object, objectDecodedBack) {
 				what = "expected object"
 			}
 			t.Errorf("encode -> decode != %s\nhave: %#v\nwant: %#v", what, v, objectDecodedBack)
@@ -742,7 +834,7 @@ func TestDecodeMultiple(t *testing.T) {
 			t.Errorf("step #%v: %v", i, err)
 		}
 
-		if !reflect.DeepEqual(obj, objOk) {
+		if !deepEqual(obj, objOk) {
 			t.Errorf("step #%v: %q  ; want %q", i, obj, objOk)
 		}
 	}
@@ -929,7 +1021,7 @@ func TestPersistentRefs(t *testing.T) {
 			errExpect = "pickle: handleRef: " + e.Error()
 		}
 
-		if !(reflect.DeepEqual(v, expected) &&
+		if !(deepEqual(v, expected) &&
 			((err == nil && errExpect == "") || err.Error() == errExpect)) {
 			t.Errorf("%q: decode -> %#v, %q; want %#v, %q",
 				tt.input, v, err, expected, errExpect)
@@ -955,7 +1047,7 @@ func TestPersistentRefs(t *testing.T) {
 			continue
 		}
 
-		if !reflect.DeepEqual(v, tt.expected) {
+		if !deepEqual(v, tt.expected) {
 			t.Errorf("%q: expected -> encode -> decode != identity\nhave: %#v\nwant: %#v",
 				tt.input, v, tt.expected)
 		}
@@ -1012,7 +1104,9 @@ func BenchmarkDecode(b *testing.B) {
 	b.ResetTimer()
 	for i := 0; i < b.N; i++ {
 		buf := bytes.NewBuffer(input)
-		dec := NewDecoder(buf)
+		dec := NewDecoderWithConfig(buf, &DecoderConfig{
+			PyDict: true, // so that decoding e.g. {(): 0} does not fail
+		})
 
 		j := 0
 		for ; ; j++ {

typeconv_test.go

@@ -45,7 +45,7 @@ func TestAsInt64(t *testing.T) {
 			}
 		}
 
-		if !reflect.DeepEqual(out, tt.outOK) {
+		if !deepEqual(out, tt.outOK) {
 			t.Errorf("%T %#v -> %T %#v  ; want %T %#v",
 				tt.in, tt.in, out, out, tt.outOK, tt.outOK)
 		}
@@ -100,12 +100,12 @@ func TestAsBytesString(t *testing.T) {
 			serrOK = Estring(tt.in)
 		}
 
-		if !(bout == boutOK && reflect.DeepEqual(berr, berrOK)) {
+		if !(bout == boutOK && deepEqual(berr, berrOK)) {
 			t.Errorf("%#v: AsBytes:\nhave %#v %#v\nwant %#v %#v",
 				tt.in, bout, berr, boutOK, berrOK)
 		}
 
-		if !(sout == soutOK && reflect.DeepEqual(serr, serrOK)) {
+		if !(sout == soutOK && deepEqual(serr, serrOK)) {
 			t.Errorf("%#v: AsString:\nhave %#v %#v\nwant %#v %#v",
 				tt.in, sout, serr, soutOK, serrOK)
 		}

xbig_120.go

+//go:build !go1.21
+
+package ogórek
+
+import (
+	"math/big"
+)
+
+func bigInt_Float64(b *big.Int) (float64, big.Accuracy) {
+	return new(big.Float).SetInt(b).Float64()
+}

xbig_121.go

+//go:build go1.21
+
+package ogórek
+
+import (
+	"math/big"
+)
+
+func bigInt_Float64(b *big.Int) (float64, big.Accuracy) {
+	return b.Float64()
+}

xmaphash_118.go

+//go:build !go1.19
+
+package ogórek
+
+import (
+	"hash/maphash"
+)
+
+func maphash_String(seed maphash.Seed, s string) uint64 {
+	var h maphash.Hash
+	h.SetSeed(seed)
+	h.WriteString(s)
+	return h.Sum64()
+}

xmaphash_119.go

+//go:build go1.19
+
+package ogórek
+
+import (
+	"hash/maphash"
+)
+
+func maphash_String(seed maphash.Seed, s string) uint64 {
+	return maphash.String(seed, s)
+}

xreflect.go

+package ogórek
+// Utilities that complement std reflect package.
+
+import (
+	"reflect"
+)
+
+
+// deepEqual is like reflect.DeepEqual but also supports Dict.
+//
+// It is needed because reflect.DeepEqual considers two Dicts not-equal because
+// each Dict is made with its own seed.
+//
+// XXX only top-level Dict is supported currently.
+//     For example comparing Dict inside list with the same won't work.
+func deepEqual(a, b any) bool {
+	da, ok := a.(Dict)
+	if !ok {
+		return reflect.DeepEqual(a, b)
+	}
+	db, ok := b.(Dict)
+	if !ok {
+		return false // Dict != non-dict
+	}
+
+	if da.Len() != db.Len() {
+		return false
+	}
+
+	// XXX O(n^2) because we want to compare keys exactly and so cannot use
+	//     db.Get(ka) because Dict.Get uses general equality that would match e.g. int == int64
+	eq := true
+	da.Iter()(func(ka, va any) bool {
+		keq := false
+		db.Iter()(func(kb, vb any) bool {
+			// NOTE don't use reflect.Equal(ka,kb) because it does not handle e.g. big.Int
+			if reflect.TypeOf(ka) == reflect.TypeOf(kb) && equal(ka,kb) {
+				if reflect.DeepEqual(va, vb) {
+					keq = true
+				}
+				return false
+			}
+			return true
+		})
+		if !keq {
+			eq = false
+			return false
+		}
+		return true
+	})
+
+	return eq
+}