From f6bddb669ff4854d5556181597aa3bba8d6f704e Mon Sep 17 00:00:00 2001 From: Daniel Liu Date: Wed, 15 May 2024 19:28:59 +0800 Subject: [PATCH] rlp: upgarde package rlp to 2024-05-15 --- rlp/decode.go | 693 ++++++++++++++--------- rlp/decode_test.go | 519 ++++++++++++++++- rlp/doc.go | 149 ++++- rlp/encbuffer.go | 423 ++++++++++++++ rlp/encbuffer_example_test.go | 45 ++ rlp/encode.go | 597 ++++++++------------ rlp/encode_test.go | 332 ++++++++++- rlp/encoder_example_test.go | 20 +- rlp/internal/rlpstruct/rlpstruct.go | 213 +++++++ rlp/iterator.go | 59 ++ rlp/iterator_test.go | 59 ++ rlp/raw.go | 67 +++ rlp/raw_test.go | 160 +++++- rlp/rlpgen/gen.go | 800 +++++++++++++++++++++++++++ rlp/rlpgen/gen_test.go | 107 ++++ rlp/rlpgen/main.go | 144 +++++ rlp/rlpgen/testdata/bigint.in.txt | 10 + rlp/rlpgen/testdata/bigint.out.txt | 49 ++ rlp/rlpgen/testdata/nil.in.txt | 30 + rlp/rlpgen/testdata/nil.out.txt | 289 ++++++++++ rlp/rlpgen/testdata/optional.in.txt | 17 + rlp/rlpgen/testdata/optional.out.txt | 153 +++++ rlp/rlpgen/testdata/rawvalue.in.txt | 11 + rlp/rlpgen/testdata/rawvalue.out.txt | 64 +++ rlp/rlpgen/testdata/uint256.in.txt | 10 + rlp/rlpgen/testdata/uint256.out.txt | 44 ++ rlp/rlpgen/testdata/uints.in.txt | 10 + rlp/rlpgen/testdata/uints.out.txt | 53 ++ rlp/rlpgen/types.go | 124 +++++ rlp/safe.go | 27 + rlp/typecache.go | 266 ++++++--- rlp/unsafe.go | 30 + 32 files changed, 4762 insertions(+), 812 deletions(-) create mode 100644 rlp/encbuffer.go create mode 100644 rlp/encbuffer_example_test.go create mode 100644 rlp/internal/rlpstruct/rlpstruct.go create mode 100644 rlp/iterator.go create mode 100644 rlp/iterator_test.go create mode 100644 rlp/rlpgen/gen.go create mode 100644 rlp/rlpgen/gen_test.go create mode 100644 rlp/rlpgen/main.go create mode 100644 rlp/rlpgen/testdata/bigint.in.txt create mode 100644 rlp/rlpgen/testdata/bigint.out.txt create mode 100644 rlp/rlpgen/testdata/nil.in.txt create mode 100644 rlp/rlpgen/testdata/nil.out.txt create mode 100644 rlp/rlpgen/testdata/optional.in.txt create mode 100644 rlp/rlpgen/testdata/optional.out.txt create mode 100644 rlp/rlpgen/testdata/rawvalue.in.txt create mode 100644 rlp/rlpgen/testdata/rawvalue.out.txt create mode 100644 rlp/rlpgen/testdata/uint256.in.txt create mode 100644 rlp/rlpgen/testdata/uint256.out.txt create mode 100644 rlp/rlpgen/testdata/uints.in.txt create mode 100644 rlp/rlpgen/testdata/uints.out.txt create mode 100644 rlp/rlpgen/types.go create mode 100644 rlp/safe.go create mode 100644 rlp/unsafe.go diff --git a/rlp/decode.go b/rlp/decode.go index 60d9dab2b5c4..1d32ff8523d4 100644 --- a/rlp/decode.go +++ b/rlp/decode.go @@ -26,103 +26,80 @@ import ( "math/big" "reflect" "strings" + "sync" + + "github.com/XinFinOrg/XDPoSChain/rlp/internal/rlpstruct" + "github.com/holiman/uint256" ) +//lint:ignore ST1012 EOL is not an error. + +// EOL is returned when the end of the current list +// has been reached during streaming. +var EOL = errors.New("rlp: end of list") + var ( + ErrExpectedString = errors.New("rlp: expected String or Byte") + ErrExpectedList = errors.New("rlp: expected List") + ErrCanonInt = errors.New("rlp: non-canonical integer format") + ErrCanonSize = errors.New("rlp: non-canonical size information") + ErrElemTooLarge = errors.New("rlp: element is larger than containing list") + ErrValueTooLarge = errors.New("rlp: value size exceeds available input length") + ErrMoreThanOneValue = errors.New("rlp: input contains more than one value") + + // internal errors + errNotInList = errors.New("rlp: call of ListEnd outside of any list") + errNotAtEOL = errors.New("rlp: call of ListEnd not positioned at EOL") + errUintOverflow = errors.New("rlp: uint overflow") errNoPointer = errors.New("rlp: interface given to Decode must be a pointer") errDecodeIntoNil = errors.New("rlp: pointer given to Decode must not be nil") + errUint256Large = errors.New("rlp: value too large for uint256") + + streamPool = sync.Pool{ + New: func() interface{} { return new(Stream) }, + } ) -// Decoder is implemented by types that require custom RLP -// decoding rules or need to decode into private fields. +// Decoder is implemented by types that require custom RLP decoding rules or need to decode +// into private fields. // -// The DecodeRLP method should read one value from the given -// Stream. It is not forbidden to read less or more, but it might -// be confusing. +// The DecodeRLP method should read one value from the given Stream. It is not forbidden to +// read less or more, but it might be confusing. type Decoder interface { DecodeRLP(*Stream) error } -// Decode parses RLP-encoded data from r and stores the result in the -// value pointed to by val. Val must be a non-nil pointer. If r does -// not implement ByteReader, Decode will do its own buffering. -// -// Decode uses the following type-dependent decoding rules: -// -// If the type implements the Decoder interface, decode calls -// DecodeRLP. -// -// To decode into a pointer, Decode will decode into the value pointed -// to. If the pointer is nil, a new value of the pointer's element -// type is allocated. If the pointer is non-nil, the existing value -// will be reused. -// -// To decode into a struct, Decode expects the input to be an RLP -// list. The decoded elements of the list are assigned to each public -// field in the order given by the struct's definition. The input list -// must contain an element for each decoded field. Decode returns an -// error if there are too few or too many elements. +// Decode parses RLP-encoded data from r and stores the result in the value pointed to by +// val. Please see package-level documentation for the decoding rules. Val must be a +// non-nil pointer. // -// The decoding of struct fields honours certain struct tags, "tail", -// "nil" and "-". +// If r does not implement ByteReader, Decode will do its own buffering. // -// The "-" tag ignores fields. +// Note that Decode does not set an input limit for all readers and may be vulnerable to +// panics cause by huge value sizes. If you need an input limit, use // -// For an explanation of "tail", see the example. -// -// The "nil" tag applies to pointer-typed fields and changes the decoding -// rules for the field such that input values of size zero decode as a nil -// pointer. This tag can be useful when decoding recursive types. -// -// type StructWithEmptyOK struct { -// Foo *[20]byte `rlp:"nil"` -// } -// -// To decode into a slice, the input must be a list and the resulting -// slice will contain the input elements in order. For byte slices, -// the input must be an RLP string. Array types decode similarly, with -// the additional restriction that the number of input elements (or -// bytes) must match the array's length. -// -// To decode into a Go string, the input must be an RLP string. The -// input bytes are taken as-is and will not necessarily be valid UTF-8. -// -// To decode into an unsigned integer type, the input must also be an RLP -// string. The bytes are interpreted as a big endian representation of -// the integer. If the RLP string is larger than the bit size of the -// type, Decode will return an error. Decode also supports *big.Int. -// There is no size limit for big integers. -// -// To decode into an interface value, Decode stores one of these -// in the value: -// -// []interface{}, for RLP lists -// []byte, for RLP strings -// -// Non-empty interface types are not supported, nor are booleans, -// signed integers, floating point numbers, maps, channels and -// functions. -// -// Note that Decode does not set an input limit for all readers -// and may be vulnerable to panics cause by huge value sizes. If -// you need an input limit, use -// -// NewStream(r, limit).Decode(val) +// NewStream(r, limit).Decode(val) func Decode(r io.Reader, val interface{}) error { - // TODO: this could use a Stream from a pool. - return NewStream(r, 0).Decode(val) + stream := streamPool.Get().(*Stream) + defer streamPool.Put(stream) + + stream.Reset(r, 0) + return stream.Decode(val) } -// DecodeBytes parses RLP data from b into val. -// Please see the documentation of Decode for the decoding rules. -// The input must contain exactly one value and no trailing data. +// DecodeBytes parses RLP data from b into val. Please see package-level documentation for +// the decoding rules. The input must contain exactly one value and no trailing data. func DecodeBytes(b []byte, val interface{}) error { - // TODO: this could use a Stream from a pool. - r := bytes.NewReader(b) - if err := NewStream(r, uint64(len(b))).Decode(val); err != nil { + r := (*sliceReader)(&b) + + stream := streamPool.Get().(*Stream) + defer streamPool.Put(stream) + + stream.Reset(r, uint64(len(b))) + if err := stream.Decode(val); err != nil { return err } - if r.Len() > 0 { + if len(b) > 0 { return ErrMoreThanOneValue } return nil @@ -173,21 +150,26 @@ func addErrorContext(err error, ctx string) error { var ( decoderInterface = reflect.TypeOf(new(Decoder)).Elem() bigInt = reflect.TypeOf(big.Int{}) + u256Int = reflect.TypeOf(uint256.Int{}) ) -func makeDecoder(typ reflect.Type, tags tags) (dec decoder, err error) { +func makeDecoder(typ reflect.Type, tags rlpstruct.Tags) (dec decoder, err error) { kind := typ.Kind() switch { case typ == rawValueType: return decodeRawValue, nil - case typ.Implements(decoderInterface): - return decodeDecoder, nil - case kind != reflect.Ptr && reflect.PtrTo(typ).Implements(decoderInterface): - return decodeDecoderNoPtr, nil - case typ.AssignableTo(reflect.PtrTo(bigInt)): + case typ.AssignableTo(reflect.PointerTo(bigInt)): return decodeBigInt, nil case typ.AssignableTo(bigInt): return decodeBigIntNoPtr, nil + case typ == reflect.PointerTo(u256Int): + return decodeU256, nil + case typ == u256Int: + return decodeU256NoPtr, nil + case kind == reflect.Ptr: + return makePtrDecoder(typ, tags) + case reflect.PointerTo(typ).Implements(decoderInterface): + return decodeDecoder, nil case isUint(kind): return decodeUint, nil case kind == reflect.Bool: @@ -198,11 +180,6 @@ func makeDecoder(typ reflect.Type, tags tags) (dec decoder, err error) { return makeListDecoder(typ, tags) case kind == reflect.Struct: return makeStructDecoder(typ) - case kind == reflect.Ptr: - if tags.nilOK { - return makeOptionalPtrDecoder(typ) - } - return makePtrDecoder(typ) case kind == reflect.Interface: return decodeInterface, nil default: @@ -252,35 +229,48 @@ func decodeBigIntNoPtr(s *Stream, val reflect.Value) error { } func decodeBigInt(s *Stream, val reflect.Value) error { - b, err := s.Bytes() + i := val.Interface().(*big.Int) + if i == nil { + i = new(big.Int) + val.Set(reflect.ValueOf(i)) + } + + err := s.decodeBigInt(i) if err != nil { return wrapStreamError(err, val.Type()) } - i := val.Interface().(*big.Int) + return nil +} + +func decodeU256NoPtr(s *Stream, val reflect.Value) error { + return decodeU256(s, val.Addr()) +} + +func decodeU256(s *Stream, val reflect.Value) error { + i := val.Interface().(*uint256.Int) if i == nil { - i = new(big.Int) + i = new(uint256.Int) val.Set(reflect.ValueOf(i)) } - // Reject leading zero bytes - if len(b) > 0 && b[0] == 0 { - return wrapStreamError(ErrCanonInt, val.Type()) + + err := s.ReadUint256(i) + if err != nil { + return wrapStreamError(err, val.Type()) } - i.SetBytes(b) return nil } -func makeListDecoder(typ reflect.Type, tag tags) (decoder, error) { +func makeListDecoder(typ reflect.Type, tag rlpstruct.Tags) (decoder, error) { etype := typ.Elem() - if etype.Kind() == reflect.Uint8 && !reflect.PtrTo(etype).Implements(decoderInterface) { + if etype.Kind() == reflect.Uint8 && !reflect.PointerTo(etype).Implements(decoderInterface) { if typ.Kind() == reflect.Array { return decodeByteArray, nil - } else { - return decodeByteSlice, nil } + return decodeByteSlice, nil } - etypeinfo, err := cachedTypeInfo1(etype, tags{}) - if err != nil { - return nil, err + etypeinfo := theTC.infoWhileGenerating(etype, rlpstruct.Tags{}) + if etypeinfo.decoderErr != nil { + return nil, etypeinfo.decoderErr } var dec decoder switch { @@ -288,7 +278,7 @@ func makeListDecoder(typ reflect.Type, tag tags) (decoder, error) { dec = func(s *Stream, val reflect.Value) error { return decodeListArray(s, val, etypeinfo.decoder) } - case tag.tail: + case tag.Tail: // A slice with "tail" tag can occur as the last field // of a struct and is supposed to swallow all remaining // list elements. The struct decoder already called s.List, @@ -381,25 +371,23 @@ func decodeByteArray(s *Stream, val reflect.Value) error { if err != nil { return err } - vlen := val.Len() + slice := byteArrayBytes(val, val.Len()) switch kind { case Byte: - if vlen == 0 { + if len(slice) == 0 { return &decodeError{msg: "input string too long", typ: val.Type()} - } - if vlen > 1 { + } else if len(slice) > 1 { return &decodeError{msg: "input string too short", typ: val.Type()} } - bv, _ := s.Uint() - val.Index(0).SetUint(bv) + slice[0] = s.byteval + s.kind = -1 case String: - if uint64(vlen) < size { + if uint64(len(slice)) < size { return &decodeError{msg: "input string too long", typ: val.Type()} } - if uint64(vlen) > size { + if uint64(len(slice)) > size { return &decodeError{msg: "input string too short", typ: val.Type()} } - slice := val.Slice(0, vlen).Interface().([]byte) if err := s.readFull(slice); err != nil { return err } @@ -418,13 +406,25 @@ func makeStructDecoder(typ reflect.Type) (decoder, error) { if err != nil { return nil, err } + for _, f := range fields { + if f.info.decoderErr != nil { + return nil, structFieldError{typ, f.index, f.info.decoderErr} + } + } dec := func(s *Stream, val reflect.Value) (err error) { if _, err := s.List(); err != nil { return wrapStreamError(err, typ) } - for _, f := range fields { + for i, f := range fields { err := f.info.decoder(s, val.Field(f.index)) if err == EOL { + if f.optional { + // The field is optional, so reaching the end of the list before + // reaching the last field is acceptable. All remaining undecoded + // fields are zeroed. + zeroFields(val, fields[i:]) + break + } return &decodeError{msg: "too few elements", typ: typ} } else if err != nil { return addErrorContext(err, "."+typ.Field(f.index).Name) @@ -435,15 +435,29 @@ func makeStructDecoder(typ reflect.Type) (decoder, error) { return dec, nil } -// makePtrDecoder creates a decoder that decodes into -// the pointer's element type. -func makePtrDecoder(typ reflect.Type) (decoder, error) { +func zeroFields(structval reflect.Value, fields []field) { + for _, f := range fields { + fv := structval.Field(f.index) + fv.Set(reflect.Zero(fv.Type())) + } +} + +// makePtrDecoder creates a decoder that decodes into the pointer's element type. +func makePtrDecoder(typ reflect.Type, tag rlpstruct.Tags) (decoder, error) { etype := typ.Elem() - etypeinfo, err := cachedTypeInfo1(etype, tags{}) - if err != nil { - return nil, err + etypeinfo := theTC.infoWhileGenerating(etype, rlpstruct.Tags{}) + switch { + case etypeinfo.decoderErr != nil: + return nil, etypeinfo.decoderErr + case !tag.NilOK: + return makeSimplePtrDecoder(etype, etypeinfo), nil + default: + return makeNilPtrDecoder(etype, etypeinfo, tag), nil } - dec := func(s *Stream, val reflect.Value) (err error) { +} + +func makeSimplePtrDecoder(etype reflect.Type, etypeinfo *typeinfo) decoder { + return func(s *Stream, val reflect.Value) (err error) { newval := val if val.IsNil() { newval = reflect.New(etype) @@ -453,30 +467,39 @@ func makePtrDecoder(typ reflect.Type) (decoder, error) { } return err } - return dec, nil } -// makeOptionalPtrDecoder creates a decoder that decodes empty values -// as nil. Non-empty values are decoded into a value of the element type, -// just like makePtrDecoder does. +// makeNilPtrDecoder creates a decoder that decodes empty values as nil. Non-empty +// values are decoded into a value of the element type, just like makePtrDecoder does. // // This decoder is used for pointer-typed struct fields with struct tag "nil". -func makeOptionalPtrDecoder(typ reflect.Type) (decoder, error) { - etype := typ.Elem() - etypeinfo, err := cachedTypeInfo1(etype, tags{}) - if err != nil { - return nil, err - } - dec := func(s *Stream, val reflect.Value) (err error) { +func makeNilPtrDecoder(etype reflect.Type, etypeinfo *typeinfo, ts rlpstruct.Tags) decoder { + typ := reflect.PointerTo(etype) + nilPtr := reflect.Zero(typ) + + // Determine the value kind that results in nil pointer. + nilKind := typeNilKind(etype, ts) + + return func(s *Stream, val reflect.Value) (err error) { kind, size, err := s.Kind() - if err != nil || size == 0 && kind != Byte { + if err != nil { + val.Set(nilPtr) + return wrapStreamError(err, typ) + } + // Handle empty values as a nil pointer. + if kind != Byte && size == 0 { + if kind != nilKind { + return &decodeError{ + msg: fmt.Sprintf("wrong kind of empty value (got %v, want %v)", kind, nilKind), + typ: typ, + } + } // rearm s.Kind. This is important because the input // position must advance to the next value even though // we don't read anything. s.kind = -1 - // set the pointer to nil. - val.Set(reflect.Zero(typ)) - return err + val.Set(nilPtr) + return nil } newval := val if val.IsNil() { @@ -487,7 +510,6 @@ func makeOptionalPtrDecoder(typ reflect.Type) (decoder, error) { } return err } - return dec, nil } var ifsliceType = reflect.TypeOf([]interface{}{}) @@ -516,25 +538,12 @@ func decodeInterface(s *Stream, val reflect.Value) error { return nil } -// This decoder is used for non-pointer values of types -// that implement the Decoder interface using a pointer receiver. -func decodeDecoderNoPtr(s *Stream, val reflect.Value) error { - return val.Addr().Interface().(Decoder).DecodeRLP(s) -} - func decodeDecoder(s *Stream, val reflect.Value) error { - // Decoder instances are not handled using the pointer rule if the type - // implements Decoder with pointer receiver (i.e. always) - // because it might handle empty values specially. - // We need to allocate one here in this case, like makePtrDecoder does. - if val.Kind() == reflect.Ptr && val.IsNil() { - val.Set(reflect.New(val.Type().Elem())) - } - return val.Interface().(Decoder).DecodeRLP(s) + return val.Addr().Interface().(Decoder).DecodeRLP(s) } // Kind represents the kind of value contained in an RLP stream. -type Kind int +type Kind int8 const ( Byte Kind = iota @@ -555,29 +564,6 @@ func (k Kind) String() string { } } -var ( - // EOL is returned when the end of the current list - // has been reached during streaming. - EOL = errors.New("rlp: end of list") - - // Actual Errors - ErrExpectedString = errors.New("rlp: expected String or Byte") - ErrExpectedList = errors.New("rlp: expected List") - ErrCanonInt = errors.New("rlp: non-canonical integer format") - ErrCanonSize = errors.New("rlp: non-canonical size information") - ErrElemTooLarge = errors.New("rlp: element is larger than containing list") - ErrValueTooLarge = errors.New("rlp: value size exceeds available input length") - - // This error is reported by DecodeBytes if the slice contains - // additional data after the first RLP value. - ErrMoreThanOneValue = errors.New("rlp: input contains more than one value") - - // internal errors - errNotInList = errors.New("rlp: call of ListEnd outside of any list") - errNotAtEOL = errors.New("rlp: call of ListEnd not positioned at EOL") - errUintOverflow = errors.New("rlp: uint overflow") -) - // ByteReader must be implemented by any input reader for a Stream. It // is implemented by e.g. bufio.Reader and bytes.Reader. type ByteReader interface { @@ -600,22 +586,16 @@ type ByteReader interface { type Stream struct { r ByteReader - // number of bytes remaining to be read from r. - remaining uint64 - limited bool - - // auxiliary buffer for integer decoding - uintbuf []byte - - kind Kind // kind of value ahead - size uint64 // size of value ahead - byteval byte // value of single byte in type tag - kinderr error // error from last readKind - stack []listpos + remaining uint64 // number of bytes remaining to be read from r + size uint64 // size of value ahead + kinderr error // error from last readKind + stack []uint64 // list sizes + uintbuf [32]byte // auxiliary buffer for integer decoding + kind Kind // kind of value ahead + byteval byte // value of single byte in type tag + limited bool // true if input limit is in effect } -type listpos struct{ pos, size uint64 } - // NewStream creates a new decoding stream reading from r. // // If r implements the ByteReader interface, Stream will @@ -675,6 +655,37 @@ func (s *Stream) Bytes() ([]byte, error) { } } +// ReadBytes decodes the next RLP value and stores the result in b. +// The value size must match len(b) exactly. +func (s *Stream) ReadBytes(b []byte) error { + kind, size, err := s.Kind() + if err != nil { + return err + } + switch kind { + case Byte: + if len(b) != 1 { + return fmt.Errorf("input value has wrong size 1, want %d", len(b)) + } + b[0] = s.byteval + s.kind = -1 // rearm Kind + return nil + case String: + if uint64(len(b)) != size { + return fmt.Errorf("input value has wrong size %d, want %d", size, len(b)) + } + if err = s.readFull(b); err != nil { + return err + } + if size == 1 && b[0] < 128 { + return ErrCanonSize + } + return nil + default: + return ErrExpectedString + } +} + // Raw reads a raw encoded value including RLP type information. func (s *Stream) Raw() ([]byte, error) { kind, size, err := s.Kind() @@ -685,8 +696,8 @@ func (s *Stream) Raw() ([]byte, error) { s.kind = -1 // rearm Kind return []byte{s.byteval}, nil } - // the original header has already been read and is no longer - // available. read content and put a new header in front of it. + // The original header has already been read and is no longer + // available. Read content and put a new header in front of it. start := headsize(size) buf := make([]byte, uint64(start)+size) if err := s.readFull(buf[start:]); err != nil { @@ -703,10 +714,31 @@ func (s *Stream) Raw() ([]byte, error) { // Uint reads an RLP string of up to 8 bytes and returns its contents // as an unsigned integer. If the input does not contain an RLP string, the // returned error will be ErrExpectedString. +// +// Deprecated: use s.Uint64 instead. func (s *Stream) Uint() (uint64, error) { return s.uint(64) } +func (s *Stream) Uint64() (uint64, error) { + return s.uint(64) +} + +func (s *Stream) Uint32() (uint32, error) { + i, err := s.uint(32) + return uint32(i), err +} + +func (s *Stream) Uint16() (uint16, error) { + i, err := s.uint(16) + return uint16(i), err +} + +func (s *Stream) Uint8() (uint8, error) { + i, err := s.uint(8) + return uint8(i), err +} + func (s *Stream) uint(maxbits int) (uint64, error) { kind, size, err := s.Kind() if err != nil { @@ -769,7 +801,14 @@ func (s *Stream) List() (size uint64, err error) { if kind != List { return 0, ErrExpectedList } - s.stack = append(s.stack, listpos{0, size}) + + // Remove size of inner list from outer list before pushing the new size + // onto the stack. This ensures that the remaining outer list size will + // be correct after the matching call to ListEnd. + if inList, limit := s.listLimit(); inList { + s.stack[len(s.stack)-1] = limit - size + } + s.stack = append(s.stack, size) s.kind = -1 s.size = 0 return size, nil @@ -778,22 +817,116 @@ func (s *Stream) List() (size uint64, err error) { // ListEnd returns to the enclosing list. // The input reader must be positioned at the end of a list. func (s *Stream) ListEnd() error { - if len(s.stack) == 0 { + // Ensure that no more data is remaining in the current list. + if inList, listLimit := s.listLimit(); !inList { return errNotInList - } - tos := s.stack[len(s.stack)-1] - if tos.pos != tos.size { + } else if listLimit > 0 { return errNotAtEOL } s.stack = s.stack[:len(s.stack)-1] // pop - if len(s.stack) > 0 { - s.stack[len(s.stack)-1].pos += tos.size - } s.kind = -1 s.size = 0 return nil } +// MoreDataInList reports whether the current list context contains +// more data to be read. +func (s *Stream) MoreDataInList() bool { + _, listLimit := s.listLimit() + return listLimit > 0 +} + +// BigInt decodes an arbitrary-size integer value. +func (s *Stream) BigInt() (*big.Int, error) { + i := new(big.Int) + if err := s.decodeBigInt(i); err != nil { + return nil, err + } + return i, nil +} + +func (s *Stream) decodeBigInt(dst *big.Int) error { + var buffer []byte + kind, size, err := s.Kind() + switch { + case err != nil: + return err + case kind == List: + return ErrExpectedString + case kind == Byte: + buffer = s.uintbuf[:1] + buffer[0] = s.byteval + s.kind = -1 // re-arm Kind + case size == 0: + // Avoid zero-length read. + s.kind = -1 + case size <= uint64(len(s.uintbuf)): + // For integers smaller than s.uintbuf, allocating a buffer + // can be avoided. + buffer = s.uintbuf[:size] + if err := s.readFull(buffer); err != nil { + return err + } + // Reject inputs where single byte encoding should have been used. + if size == 1 && buffer[0] < 128 { + return ErrCanonSize + } + default: + // For large integers, a temporary buffer is needed. + buffer = make([]byte, size) + if err := s.readFull(buffer); err != nil { + return err + } + } + + // Reject leading zero bytes. + if len(buffer) > 0 && buffer[0] == 0 { + return ErrCanonInt + } + // Set the integer bytes. + dst.SetBytes(buffer) + return nil +} + +// ReadUint256 decodes the next value as a uint256. +func (s *Stream) ReadUint256(dst *uint256.Int) error { + var buffer []byte + kind, size, err := s.Kind() + switch { + case err != nil: + return err + case kind == List: + return ErrExpectedString + case kind == Byte: + buffer = s.uintbuf[:1] + buffer[0] = s.byteval + s.kind = -1 // re-arm Kind + case size == 0: + // Avoid zero-length read. + s.kind = -1 + case size <= uint64(len(s.uintbuf)): + // All possible uint256 values fit into s.uintbuf. + buffer = s.uintbuf[:size] + if err := s.readFull(buffer); err != nil { + return err + } + // Reject inputs where single byte encoding should have been used. + if size == 1 && buffer[0] < 128 { + return ErrCanonSize + } + default: + return errUint256Large + } + + // Reject leading zero bytes. + if len(buffer) > 0 && buffer[0] == 0 { + return ErrCanonInt + } + // Set the integer bytes. + dst.SetBytes(buffer) + return nil +} + // Decode decodes a value and stores the result in the value pointed // to by val. Please see the documentation for the Decode function // to learn about the decoding rules. @@ -809,14 +942,14 @@ func (s *Stream) Decode(val interface{}) error { if rval.IsNil() { return errDecodeIntoNil } - info, err := cachedTypeInfo(rtyp.Elem(), tags{}) + decoder, err := cachedDecoder(rtyp.Elem()) if err != nil { return err } - err = info.decoder(s, rval.Elem()) + err = decoder(s, rval.Elem()) if decErr, ok := err.(*decodeError); ok && len(decErr.ctx) > 0 { - // add decode target type to error so context has more meaning + // Add decode target type to error so context has more meaning. decErr.ctx = append(decErr.ctx, fmt.Sprint("(", rtyp.Elem(), ")")) } return err @@ -839,6 +972,9 @@ func (s *Stream) Reset(r io.Reader, inputLimit uint64) { case *bytes.Reader: s.remaining = uint64(br.Len()) s.limited = true + case *bytes.Buffer: + s.remaining = uint64(br.Len()) + s.limited = true case *strings.Reader: s.remaining = uint64(br.Len()) s.limited = true @@ -857,9 +993,8 @@ func (s *Stream) Reset(r io.Reader, inputLimit uint64) { s.size = 0 s.kind = -1 s.kinderr = nil - if s.uintbuf == nil { - s.uintbuf = make([]byte, 8) - } + s.byteval = 0 + s.uintbuf = [32]byte{} } // Kind returns the kind and size of the next value in the @@ -874,35 +1009,29 @@ func (s *Stream) Reset(r io.Reader, inputLimit uint64) { // the value. Subsequent calls to Kind (until the value is decoded) // will not advance the input reader and return cached information. func (s *Stream) Kind() (kind Kind, size uint64, err error) { - var tos *listpos - if len(s.stack) > 0 { - tos = &s.stack[len(s.stack)-1] - } - if s.kind < 0 { - s.kinderr = nil - // Don't read further if we're at the end of the - // innermost list. - if tos != nil && tos.pos == tos.size { - return 0, 0, EOL - } - s.kind, s.size, s.kinderr = s.readKind() - if s.kinderr == nil { - if tos == nil { - // At toplevel, check that the value is smaller - // than the remaining input length. - if s.limited && s.size > s.remaining { - s.kinderr = ErrValueTooLarge - } - } else { - // Inside a list, check that the value doesn't overflow the list. - if s.size > tos.size-tos.pos { - s.kinderr = ErrElemTooLarge - } - } + if s.kind >= 0 { + return s.kind, s.size, s.kinderr + } + + // Check for end of list. This needs to be done here because readKind + // checks against the list size, and would return the wrong error. + inList, listLimit := s.listLimit() + if inList && listLimit == 0 { + return 0, 0, EOL + } + // Read the actual size tag. + s.kind, s.size, s.kinderr = s.readKind() + if s.kinderr == nil { + // Check the data size of the value ahead against input limits. This + // is done here because many decoders require allocating an input + // buffer matching the value size. Checking it here protects those + // decoders from inputs declaring very large value size. + if inList && s.size > listLimit { + s.kinderr = ErrElemTooLarge + } else if s.limited && s.size > s.remaining { + s.kinderr = ErrValueTooLarge } } - // Note: this might return a sticky error generated - // by an earlier call to readKind. return s.kind, s.size, s.kinderr } @@ -929,37 +1058,35 @@ func (s *Stream) readKind() (kind Kind, size uint64, err error) { s.byteval = b return Byte, 0, nil case b < 0xB8: - // Otherwise, if a string is 0-55 bytes long, - // the RLP encoding consists of a single byte with value 0x80 plus the - // length of the string followed by the string. The range of the first - // byte is thus [0x80, 0xB7]. + // Otherwise, if a string is 0-55 bytes long, the RLP encoding consists + // of a single byte with value 0x80 plus the length of the string + // followed by the string. The range of the first byte is thus [0x80, 0xB7]. return String, uint64(b - 0x80), nil case b < 0xC0: - // If a string is more than 55 bytes long, the - // RLP encoding consists of a single byte with value 0xB7 plus the length - // of the length of the string in binary form, followed by the length of - // the string, followed by the string. For example, a length-1024 string - // would be encoded as 0xB90400 followed by the string. The range of - // the first byte is thus [0xB8, 0xBF]. + // If a string is more than 55 bytes long, the RLP encoding consists of a + // single byte with value 0xB7 plus the length of the length of the + // string in binary form, followed by the length of the string, followed + // by the string. For example, a length-1024 string would be encoded as + // 0xB90400 followed by the string. The range of the first byte is thus + // [0xB8, 0xBF]. size, err = s.readUint(b - 0xB7) if err == nil && size < 56 { err = ErrCanonSize } return String, size, err case b < 0xF8: - // If the total payload of a list - // (i.e. the combined length of all its items) is 0-55 bytes long, the - // RLP encoding consists of a single byte with value 0xC0 plus the length - // of the list followed by the concatenation of the RLP encodings of the - // items. The range of the first byte is thus [0xC0, 0xF7]. + // If the total payload of a list (i.e. the combined length of all its + // items) is 0-55 bytes long, the RLP encoding consists of a single byte + // with value 0xC0 plus the length of the list followed by the + // concatenation of the RLP encodings of the items. The range of the + // first byte is thus [0xC0, 0xF7]. return List, uint64(b - 0xC0), nil default: - // If the total payload of a list is more than 55 bytes long, - // the RLP encoding consists of a single byte with value 0xF7 - // plus the length of the length of the payload in binary - // form, followed by the length of the payload, followed by - // the concatenation of the RLP encodings of the items. The - // range of the first byte is thus [0xF8, 0xFF]. + // If the total payload of a list is more than 55 bytes long, the RLP + // encoding consists of a single byte with value 0xF7 plus the length of + // the length of the payload in binary form, followed by the length of + // the payload, followed by the concatenation of the RLP encodings of + // the items. The range of the first byte is thus [0xF8, 0xFF]. size, err = s.readUint(b - 0xF7) if err == nil && size < 56 { err = ErrCanonSize @@ -977,23 +1104,22 @@ func (s *Stream) readUint(size byte) (uint64, error) { b, err := s.readByte() return uint64(b), err default: + buffer := s.uintbuf[:8] + clear(buffer) start := int(8 - size) - for i := 0; i < start; i++ { - s.uintbuf[i] = 0 - } - if err := s.readFull(s.uintbuf[start:]); err != nil { + if err := s.readFull(buffer[start:]); err != nil { return 0, err } - if s.uintbuf[start] == 0 { - // Note: readUint is also used to decode integer - // values. The error needs to be adjusted to become - // ErrCanonInt in this case. + if buffer[start] == 0 { + // Note: readUint is also used to decode integer values. + // The error needs to be adjusted to become ErrCanonInt in this case. return 0, ErrCanonSize } - return binary.BigEndian.Uint64(s.uintbuf), nil + return binary.BigEndian.Uint64(buffer[:]), nil } } +// readFull reads into buf from the underlying stream. func (s *Stream) readFull(buf []byte) (err error) { if err := s.willRead(uint64(len(buf))); err != nil { return err @@ -1004,11 +1130,18 @@ func (s *Stream) readFull(buf []byte) (err error) { n += nn } if err == io.EOF { - err = io.ErrUnexpectedEOF + if n < len(buf) { + err = io.ErrUnexpectedEOF + } else { + // Readers are allowed to give EOF even though the read succeeded. + // In such cases, we discard the EOF, like io.ReadFull() does. + err = nil + } } return err } +// readByte reads a single byte from the underlying stream. func (s *Stream) readByte() (byte, error) { if err := s.willRead(1); err != nil { return 0, err @@ -1020,16 +1153,16 @@ func (s *Stream) readByte() (byte, error) { return b, err } +// willRead is called before any read from the underlying stream. It checks +// n against size limits, and updates the limits if n doesn't overflow them. func (s *Stream) willRead(n uint64) error { s.kind = -1 // rearm Kind - if len(s.stack) > 0 { - // check list overflow - tos := s.stack[len(s.stack)-1] - if n > tos.size-tos.pos { + if inList, limit := s.listLimit(); inList { + if n > limit { return ErrElemTooLarge } - s.stack[len(s.stack)-1].pos += n + s.stack[len(s.stack)-1] = limit - n } if s.limited { if n > s.remaining { @@ -1039,3 +1172,31 @@ func (s *Stream) willRead(n uint64) error { } return nil } + +// listLimit returns the amount of data remaining in the innermost list. +func (s *Stream) listLimit() (inList bool, limit uint64) { + if len(s.stack) == 0 { + return false, 0 + } + return true, s.stack[len(s.stack)-1] +} + +type sliceReader []byte + +func (sr *sliceReader) Read(b []byte) (int, error) { + if len(*sr) == 0 { + return 0, io.EOF + } + n := copy(b, *sr) + *sr = (*sr)[n:] + return n, nil +} + +func (sr *sliceReader) ReadByte() (byte, error) { + if len(*sr) == 0 { + return 0, io.EOF + } + b := (*sr)[0] + *sr = (*sr)[1:] + return b, nil +} diff --git a/rlp/decode_test.go b/rlp/decode_test.go index 4d8abd001281..38cca38aa548 100644 --- a/rlp/decode_test.go +++ b/rlp/decode_test.go @@ -26,6 +26,9 @@ import ( "reflect" "strings" "testing" + + "github.com/XinFinOrg/XDPoSChain/common/math" + "github.com/holiman/uint256" ) func TestStreamKind(t *testing.T) { @@ -284,6 +287,47 @@ func TestStreamRaw(t *testing.T) { } } +func TestStreamReadBytes(t *testing.T) { + tests := []struct { + input string + size int + err string + }{ + // kind List + {input: "C0", size: 1, err: "rlp: expected String or Byte"}, + // kind Byte + {input: "04", size: 0, err: "input value has wrong size 1, want 0"}, + {input: "04", size: 1}, + {input: "04", size: 2, err: "input value has wrong size 1, want 2"}, + // kind String + {input: "820102", size: 0, err: "input value has wrong size 2, want 0"}, + {input: "820102", size: 1, err: "input value has wrong size 2, want 1"}, + {input: "820102", size: 2}, + {input: "820102", size: 3, err: "input value has wrong size 2, want 3"}, + } + + for _, test := range tests { + test := test + name := fmt.Sprintf("input_%s/size_%d", test.input, test.size) + t.Run(name, func(t *testing.T) { + s := NewStream(bytes.NewReader(unhex(test.input)), 0) + b := make([]byte, test.size) + err := s.ReadBytes(b) + if test.err == "" { + if err != nil { + t.Errorf("unexpected error %q", err) + } + } else { + if err == nil { + t.Errorf("expected error, got nil") + } else if err.Error() != test.err { + t.Errorf("wrong error %q", err) + } + } + }) + } +} + func TestDecodeErrors(t *testing.T) { r := bytes.NewReader(nil) @@ -327,6 +371,15 @@ type recstruct struct { Child *recstruct `rlp:"nil"` } +type bigIntStruct struct { + I *big.Int + B string +} + +type invalidNilTag struct { + X []byte `rlp:"nil"` +} + type invalidTail1 struct { A uint `rlp:"tail"` B string @@ -347,19 +400,79 @@ type tailUint struct { Tail []uint `rlp:"tail"` } -var ( - veryBigInt = big.NewInt(0).Add( - big.NewInt(0).Lsh(big.NewInt(0xFFFFFFFFFFFFFF), 16), - big.NewInt(0xFFFF), - ) -) +type tailPrivateFields struct { + A uint + Tail []uint `rlp:"tail"` + x, y bool //lint:ignore U1000 unused fields required for testing purposes. +} + +type nilListUint struct { + X *uint `rlp:"nilList"` +} + +type nilStringSlice struct { + X *[]uint `rlp:"nilString"` +} + +type intField struct { + X int +} + +type optionalFields struct { + A uint + B uint `rlp:"optional"` + C uint `rlp:"optional"` +} + +type optionalAndTailField struct { + A uint + B uint `rlp:"optional"` + Tail []uint `rlp:"tail"` +} + +type optionalBigIntField struct { + A uint + B *big.Int `rlp:"optional"` +} + +type optionalPtrField struct { + A uint + B *[3]byte `rlp:"optional"` +} + +type nonOptionalPtrField struct { + A uint + B *[3]byte +} -type hasIgnoredField struct { +type multipleOptionalFields struct { + A *[3]byte `rlp:"optional"` + B *[3]byte `rlp:"optional"` +} + +type optionalPtrFieldNil struct { + A uint + B *[3]byte `rlp:"optional,nil"` +} + +type ignoredField struct { A uint B uint `rlp:"-"` C uint } +var ( + veryBigInt = new(big.Int).Add( + new(big.Int).Lsh(big.NewInt(0xFFFFFFFFFFFFFF), 16), + big.NewInt(0xFFFF), + ) + veryVeryBigInt = new(big.Int).Exp(veryBigInt, big.NewInt(8), nil) +) + +var ( + veryBigInt256, _ = uint256.FromBig(veryBigInt) +) + var decodeTests = []decodeTest{ // booleans {input: "01", ptr: new(bool), value: true}, @@ -428,12 +541,31 @@ var decodeTests = []decodeTest{ {input: "C0", ptr: new(string), error: "rlp: expected input string or byte for string"}, // big ints + {input: "80", ptr: new(*big.Int), value: big.NewInt(0)}, {input: "01", ptr: new(*big.Int), value: big.NewInt(1)}, {input: "89FFFFFFFFFFFFFFFFFF", ptr: new(*big.Int), value: veryBigInt}, + {input: "B848FFFFFFFFFFFFFFFFF800000000000000001BFFFFFFFFFFFFFFFFC8000000000000000045FFFFFFFFFFFFFFFFC800000000000000001BFFFFFFFFFFFFFFFFF8000000000000000001", ptr: new(*big.Int), value: veryVeryBigInt}, {input: "10", ptr: new(big.Int), value: *big.NewInt(16)}, // non-pointer also works + + // big int errors {input: "C0", ptr: new(*big.Int), error: "rlp: expected input string or byte for *big.Int"}, - {input: "820001", ptr: new(big.Int), error: "rlp: non-canonical integer (leading zero bytes) for *big.Int"}, - {input: "8105", ptr: new(big.Int), error: "rlp: non-canonical size information for *big.Int"}, + {input: "00", ptr: new(*big.Int), error: "rlp: non-canonical integer (leading zero bytes) for *big.Int"}, + {input: "820001", ptr: new(*big.Int), error: "rlp: non-canonical integer (leading zero bytes) for *big.Int"}, + {input: "8105", ptr: new(*big.Int), error: "rlp: non-canonical size information for *big.Int"}, + + // uint256 + {input: "80", ptr: new(*uint256.Int), value: uint256.NewInt(0)}, + {input: "01", ptr: new(*uint256.Int), value: uint256.NewInt(1)}, + {input: "88FFFFFFFFFFFFFFFF", ptr: new(*uint256.Int), value: uint256.NewInt(math.MaxUint64)}, + {input: "89FFFFFFFFFFFFFFFFFF", ptr: new(*uint256.Int), value: veryBigInt256}, + {input: "10", ptr: new(uint256.Int), value: *uint256.NewInt(16)}, // non-pointer also works + + // uint256 errors + {input: "C0", ptr: new(*uint256.Int), error: "rlp: expected input string or byte for *uint256.Int"}, + {input: "00", ptr: new(*uint256.Int), error: "rlp: non-canonical integer (leading zero bytes) for *uint256.Int"}, + {input: "820001", ptr: new(*uint256.Int), error: "rlp: non-canonical integer (leading zero bytes) for *uint256.Int"}, + {input: "8105", ptr: new(*uint256.Int), error: "rlp: non-canonical size information for *uint256.Int"}, + {input: "A1FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF00", ptr: new(*uint256.Int), error: "rlp: value too large for uint256"}, // structs { @@ -446,6 +578,13 @@ var decodeTests = []decodeTest{ ptr: new(recstruct), value: recstruct{1, &recstruct{2, &recstruct{3, nil}}}, }, + { + // This checks that empty big.Int works correctly in struct context. It's easy to + // miss the update of s.kind for this case, so it needs its own test. + input: "C58083343434", + ptr: new(bigIntStruct), + value: bigIntStruct{new(big.Int), "444"}, + }, // struct errors { @@ -479,20 +618,20 @@ var decodeTests = []decodeTest{ error: "rlp: expected input string or byte for uint, decoding into (rlp.recstruct).Child.I", }, { - input: "C0", - ptr: new(invalidTail1), - error: "rlp: invalid struct tag \"tail\" for rlp.invalidTail1.A (must be on last field)", - }, - { - input: "C0", - ptr: new(invalidTail2), - error: "rlp: invalid struct tag \"tail\" for rlp.invalidTail2.B (field type is not slice)", + input: "C103", + ptr: new(intField), + error: "rlp: type int is not RLP-serializable (struct field rlp.intField.X)", }, { input: "C50102C20102", ptr: new(tailUint), error: "rlp: expected input string or byte for uint, decoding into (rlp.tailUint).Tail[1]", }, + { + input: "C0", + ptr: new(invalidNilTag), + error: `rlp: invalid struct tag "nil" for rlp.invalidNilTag.X (field is not a pointer)`, + }, // struct tag "tail" { @@ -510,12 +649,192 @@ var decodeTests = []decodeTest{ ptr: new(tailRaw), value: tailRaw{A: 1, Tail: []RawValue{}}, }, + { + input: "C3010203", + ptr: new(tailPrivateFields), + value: tailPrivateFields{A: 1, Tail: []uint{2, 3}}, + }, + { + input: "C0", + ptr: new(invalidTail1), + error: `rlp: invalid struct tag "tail" for rlp.invalidTail1.A (must be on last field)`, + }, + { + input: "C0", + ptr: new(invalidTail2), + error: `rlp: invalid struct tag "tail" for rlp.invalidTail2.B (field type is not slice)`, + }, // struct tag "-" { input: "C20102", - ptr: new(hasIgnoredField), - value: hasIgnoredField{A: 1, C: 2}, + ptr: new(ignoredField), + value: ignoredField{A: 1, C: 2}, + }, + + // struct tag "nilList" + { + input: "C180", + ptr: new(nilListUint), + error: "rlp: wrong kind of empty value (got String, want List) for *uint, decoding into (rlp.nilListUint).X", + }, + { + input: "C1C0", + ptr: new(nilListUint), + value: nilListUint{}, + }, + { + input: "C103", + ptr: new(nilListUint), + value: func() interface{} { + v := uint(3) + return nilListUint{X: &v} + }(), + }, + + // struct tag "nilString" + { + input: "C1C0", + ptr: new(nilStringSlice), + error: "rlp: wrong kind of empty value (got List, want String) for *[]uint, decoding into (rlp.nilStringSlice).X", + }, + { + input: "C180", + ptr: new(nilStringSlice), + value: nilStringSlice{}, + }, + { + input: "C2C103", + ptr: new(nilStringSlice), + value: nilStringSlice{X: &[]uint{3}}, + }, + + // struct tag "optional" + { + input: "C101", + ptr: new(optionalFields), + value: optionalFields{1, 0, 0}, + }, + { + input: "C20102", + ptr: new(optionalFields), + value: optionalFields{1, 2, 0}, + }, + { + input: "C3010203", + ptr: new(optionalFields), + value: optionalFields{1, 2, 3}, + }, + { + input: "C401020304", + ptr: new(optionalFields), + error: "rlp: input list has too many elements for rlp.optionalFields", + }, + { + input: "C101", + ptr: new(optionalAndTailField), + value: optionalAndTailField{A: 1}, + }, + { + input: "C20102", + ptr: new(optionalAndTailField), + value: optionalAndTailField{A: 1, B: 2, Tail: []uint{}}, + }, + { + input: "C401020304", + ptr: new(optionalAndTailField), + value: optionalAndTailField{A: 1, B: 2, Tail: []uint{3, 4}}, + }, + { + input: "C101", + ptr: new(optionalBigIntField), + value: optionalBigIntField{A: 1, B: nil}, + }, + { + input: "C20102", + ptr: new(optionalBigIntField), + value: optionalBigIntField{A: 1, B: big.NewInt(2)}, + }, + { + input: "C101", + ptr: new(optionalPtrField), + value: optionalPtrField{A: 1}, + }, + { + input: "C20180", // not accepted because "optional" doesn't enable "nil" + ptr: new(optionalPtrField), + error: "rlp: input string too short for [3]uint8, decoding into (rlp.optionalPtrField).B", + }, + { + input: "C20102", + ptr: new(optionalPtrField), + error: "rlp: input string too short for [3]uint8, decoding into (rlp.optionalPtrField).B", + }, + { + input: "C50183010203", + ptr: new(optionalPtrField), + value: optionalPtrField{A: 1, B: &[3]byte{1, 2, 3}}, + }, + { + // all optional fields nil + input: "C0", + ptr: new(multipleOptionalFields), + value: multipleOptionalFields{A: nil, B: nil}, + }, + { + // all optional fields set + input: "C88301020383010203", + ptr: new(multipleOptionalFields), + value: multipleOptionalFields{A: &[3]byte{1, 2, 3}, B: &[3]byte{1, 2, 3}}, + }, + { + // nil optional field appears before a non-nil one + input: "C58083010203", + ptr: new(multipleOptionalFields), + error: "rlp: input string too short for [3]uint8, decoding into (rlp.multipleOptionalFields).A", + }, + { + // decode a nil ptr into a ptr that is not nil or not optional + input: "C20180", + ptr: new(nonOptionalPtrField), + error: "rlp: input string too short for [3]uint8, decoding into (rlp.nonOptionalPtrField).B", + }, + { + input: "C101", + ptr: new(optionalPtrFieldNil), + value: optionalPtrFieldNil{A: 1}, + }, + { + input: "C20180", // accepted because "nil" tag allows empty input + ptr: new(optionalPtrFieldNil), + value: optionalPtrFieldNil{A: 1}, + }, + { + input: "C20102", + ptr: new(optionalPtrFieldNil), + error: "rlp: input string too short for [3]uint8, decoding into (rlp.optionalPtrFieldNil).B", + }, + + // struct tag "optional" field clearing + { + input: "C101", + ptr: &optionalFields{A: 9, B: 8, C: 7}, + value: optionalFields{A: 1, B: 0, C: 0}, + }, + { + input: "C20102", + ptr: &optionalFields{A: 9, B: 8, C: 7}, + value: optionalFields{A: 1, B: 2, C: 0}, + }, + { + input: "C20102", + ptr: &optionalAndTailField{A: 9, B: 8, Tail: []uint{7, 6, 5}}, + value: optionalAndTailField{A: 1, B: 2, Tail: []uint{}}, + }, + { + input: "C101", + ptr: &optionalPtrField{A: 9, B: &[3]byte{8, 7, 6}}, + value: optionalPtrField{A: 1}, }, // RawValue @@ -591,6 +910,26 @@ func TestDecodeWithByteReader(t *testing.T) { }) } +func testDecodeWithEncReader(t *testing.T, n int) { + s := strings.Repeat("0", n) + _, r, _ := EncodeToReader(s) + var decoded string + err := Decode(r, &decoded) + if err != nil { + t.Errorf("Unexpected decode error with n=%v: %v", n, err) + } + if decoded != s { + t.Errorf("Decode mismatch with n=%v", n) + } +} + +// This is a regression test checking that decoding from encReader +// works for RLP values of size 8192 bytes or more. +func TestDecodeWithEncReader(t *testing.T) { + testDecodeWithEncReader(t, 8188) // length with header is 8191 + testDecodeWithEncReader(t, 8189) // length with header is 8192 +} + // plainReader reads from a byte slice but does not // implement ReadByte. It is also not recognized by the // size validation. This is useful to test how the decoder @@ -661,6 +1000,22 @@ func TestDecodeDecoder(t *testing.T) { } } +func TestDecodeDecoderNilPointer(t *testing.T) { + var s struct { + T1 *testDecoder `rlp:"nil"` + T2 *testDecoder + } + if err := Decode(bytes.NewReader(unhex("C2C002")), &s); err != nil { + t.Fatalf("Decode error: %v", err) + } + if s.T1 != nil { + t.Errorf("decoder T1 allocated for empty input (called: %v)", s.T1.called) + } + if s.T2 == nil || !s.T2.called { + t.Errorf("decoder T2 not allocated/called") + } +} + type byteDecoder byte func (bd *byteDecoder) DecodeRLP(s *Stream) error { @@ -691,13 +1046,66 @@ func TestDecoderInByteSlice(t *testing.T) { } } +type unencodableDecoder func() + +func (f *unencodableDecoder) DecodeRLP(s *Stream) error { + if _, err := s.List(); err != nil { + return err + } + if err := s.ListEnd(); err != nil { + return err + } + *f = func() {} + return nil +} + +func TestDecoderFunc(t *testing.T) { + var x func() + if err := DecodeBytes([]byte{0xC0}, (*unencodableDecoder)(&x)); err != nil { + t.Fatal(err) + } + x() +} + +// This tests the validity checks for fields with struct tag "optional". +func TestInvalidOptionalField(t *testing.T) { + type ( + invalid1 struct { + A uint `rlp:"optional"` + B uint + } + invalid2 struct { + T []uint `rlp:"tail,optional"` + } + invalid3 struct { + T []uint `rlp:"optional,tail"` + } + ) + + tests := []struct { + v interface{} + err string + }{ + {v: new(invalid1), err: `rlp: invalid struct tag "" for rlp.invalid1.B (must be optional because preceding field "A" is optional)`}, + {v: new(invalid2), err: `rlp: invalid struct tag "optional" for rlp.invalid2.T (also has "tail" tag)`}, + {v: new(invalid3), err: `rlp: invalid struct tag "tail" for rlp.invalid3.T (also has "optional" tag)`}, + } + for _, test := range tests { + err := DecodeBytes(unhex("C20102"), test.v) + if err == nil { + t.Errorf("no error for %T", test.v) + } else if err.Error() != test.err { + t.Errorf("wrong error for %T: %v", test.v, err.Error()) + } + } +} + func ExampleDecode() { input, _ := hex.DecodeString("C90A1486666F6F626172") type example struct { - A, B uint - private uint // private fields are ignored - String string + A, B uint + String string } var s example @@ -708,7 +1116,7 @@ func ExampleDecode() { fmt.Printf("Decoded value: %#v\n", s) } // Output: - // Decoded value: rlp.example{A:0xa, B:0x14, private:0x0, String:"foobar"} + // Decoded value: rlp.example{A:0xa, B:0x14, String:"foobar"} } func ExampleDecode_structTagNil() { @@ -768,7 +1176,7 @@ func ExampleStream() { // [102 111 111 98 97 114] } -func BenchmarkDecode(b *testing.B) { +func BenchmarkDecodeUints(b *testing.B) { enc := encodeTestSlice(90000) b.SetBytes(int64(len(enc))) b.ReportAllocs() @@ -783,7 +1191,7 @@ func BenchmarkDecode(b *testing.B) { } } -func BenchmarkDecodeIntSliceReuse(b *testing.B) { +func BenchmarkDecodeUintsReused(b *testing.B) { enc := encodeTestSlice(100000) b.SetBytes(int64(len(enc))) b.ReportAllocs() @@ -798,6 +1206,65 @@ func BenchmarkDecodeIntSliceReuse(b *testing.B) { } } +func BenchmarkDecodeByteArrayStruct(b *testing.B) { + enc, err := EncodeToBytes(&byteArrayStruct{}) + if err != nil { + b.Fatal(err) + } + b.SetBytes(int64(len(enc))) + b.ReportAllocs() + b.ResetTimer() + + var out byteArrayStruct + for i := 0; i < b.N; i++ { + if err := DecodeBytes(enc, &out); err != nil { + b.Fatal(err) + } + } +} + +func BenchmarkDecodeBigInts(b *testing.B) { + ints := make([]*big.Int, 200) + for i := range ints { + ints[i] = math.BigPow(2, int64(i)) + } + enc, err := EncodeToBytes(ints) + if err != nil { + b.Fatal(err) + } + b.SetBytes(int64(len(enc))) + b.ReportAllocs() + b.ResetTimer() + + var out []*big.Int + for i := 0; i < b.N; i++ { + if err := DecodeBytes(enc, &out); err != nil { + b.Fatal(err) + } + } +} + +func BenchmarkDecodeU256Ints(b *testing.B) { + ints := make([]*uint256.Int, 200) + for i := range ints { + ints[i], _ = uint256.FromBig(math.BigPow(2, int64(i))) + } + enc, err := EncodeToBytes(ints) + if err != nil { + b.Fatal(err) + } + b.SetBytes(int64(len(enc))) + b.ReportAllocs() + b.ResetTimer() + + var out []*uint256.Int + for i := 0; i < b.N; i++ { + if err := DecodeBytes(enc, &out); err != nil { + b.Fatal(err) + } + } +} + func encodeTestSlice(n uint) []byte { s := make([]uint, n) for i := uint(0); i < n; i++ { @@ -811,7 +1278,7 @@ func encodeTestSlice(n uint) []byte { } func unhex(str string) []byte { - b, err := hex.DecodeString(strings.Replace(str, " ", "", -1)) + b, err := hex.DecodeString(strings.ReplaceAll(str, " ", "")) if err != nil { panic(fmt.Sprintf("invalid hex string: %q", str)) } diff --git a/rlp/doc.go b/rlp/doc.go index b3a81fe2326f..eeeee9a43a0c 100644 --- a/rlp/doc.go +++ b/rlp/doc.go @@ -17,17 +17,142 @@ /* Package rlp implements the RLP serialization format. -The purpose of RLP (Recursive Linear Prefix) is to encode arbitrarily -nested arrays of binary data, and RLP is the main encoding method used -to serialize objects in Ethereum. The only purpose of RLP is to encode -structure; encoding specific atomic data types (eg. strings, ints, -floats) is left up to higher-order protocols; in Ethereum integers -must be represented in big endian binary form with no leading zeroes -(thus making the integer value zero equivalent to the empty byte -array). - -RLP values are distinguished by a type tag. The type tag precedes the -value in the input stream and defines the size and kind of the bytes -that follow. +The purpose of RLP (Recursive Linear Prefix) is to encode arbitrarily nested arrays of +binary data, and RLP is the main encoding method used to serialize objects in Ethereum. +The only purpose of RLP is to encode structure; encoding specific atomic data types (eg. +strings, ints, floats) is left up to higher-order protocols. In Ethereum integers must be +represented in big endian binary form with no leading zeroes (thus making the integer +value zero equivalent to the empty string). + +RLP values are distinguished by a type tag. The type tag precedes the value in the input +stream and defines the size and kind of the bytes that follow. + +# Encoding Rules + +Package rlp uses reflection and encodes RLP based on the Go type of the value. + +If the type implements the Encoder interface, Encode calls EncodeRLP. It does not +call EncodeRLP on nil pointer values. + +To encode a pointer, the value being pointed to is encoded. A nil pointer to a struct +type, slice or array always encodes as an empty RLP list unless the slice or array has +element type byte. A nil pointer to any other value encodes as the empty string. + +Struct values are encoded as an RLP list of all their encoded public fields. Recursive +struct types are supported. + +To encode slices and arrays, the elements are encoded as an RLP list of the value's +elements. Note that arrays and slices with element type uint8 or byte are always encoded +as an RLP string. + +A Go string is encoded as an RLP string. + +An unsigned integer value is encoded as an RLP string. Zero always encodes as an empty RLP +string. big.Int values are treated as integers. Signed integers (int, int8, int16, ...) +are not supported and will return an error when encoding. + +Boolean values are encoded as the unsigned integers zero (false) and one (true). + +An interface value encodes as the value contained in the interface. + +Floating point numbers, maps, channels and functions are not supported. + +# Decoding Rules + +Decoding uses the following type-dependent rules: + +If the type implements the Decoder interface, DecodeRLP is called. + +To decode into a pointer, the value will be decoded as the element type of the pointer. If +the pointer is nil, a new value of the pointer's element type is allocated. If the pointer +is non-nil, the existing value will be reused. Note that package rlp never leaves a +pointer-type struct field as nil unless one of the "nil" struct tags is present. + +To decode into a struct, decoding expects the input to be an RLP list. The decoded +elements of the list are assigned to each public field in the order given by the struct's +definition. The input list must contain an element for each decoded field. Decoding +returns an error if there are too few or too many elements for the struct. + +To decode into a slice, the input must be a list and the resulting slice will contain the +input elements in order. For byte slices, the input must be an RLP string. Array types +decode similarly, with the additional restriction that the number of input elements (or +bytes) must match the array's defined length. + +To decode into a Go string, the input must be an RLP string. The input bytes are taken +as-is and will not necessarily be valid UTF-8. + +To decode into an unsigned integer type, the input must also be an RLP string. The bytes +are interpreted as a big endian representation of the integer. If the RLP string is larger +than the bit size of the type, decoding will return an error. Decode also supports +*big.Int. There is no size limit for big integers. + +To decode into a boolean, the input must contain an unsigned integer of value zero (false) +or one (true). + +To decode into an interface value, one of these types is stored in the value: + + []interface{}, for RLP lists + []byte, for RLP strings + +Non-empty interface types are not supported when decoding. +Signed integers, floating point numbers, maps, channels and functions cannot be decoded into. + +# Struct Tags + +As with other encoding packages, the "-" tag ignores fields. + + type StructWithIgnoredField struct{ + Ignored uint `rlp:"-"` + Field uint + } + +Go struct values encode/decode as RLP lists. There are two ways of influencing the mapping +of fields to list elements. The "tail" tag, which may only be used on the last exported +struct field, allows slurping up any excess list elements into a slice. + + type StructWithTail struct{ + Field uint + Tail []string `rlp:"tail"` + } + +The "optional" tag says that the field may be omitted if it is zero-valued. If this tag is +used on a struct field, all subsequent public fields must also be declared optional. + +When encoding a struct with optional fields, the output RLP list contains all values up to +the last non-zero optional field. + +When decoding into a struct, optional fields may be omitted from the end of the input +list. For the example below, this means input lists of one, two, or three elements are +accepted. + + type StructWithOptionalFields struct{ + Required uint + Optional1 uint `rlp:"optional"` + Optional2 uint `rlp:"optional"` + } + +The "nil", "nilList" and "nilString" tags apply to pointer-typed fields only, and change +the decoding rules for the field type. For regular pointer fields without the "nil" tag, +input values must always match the required input length exactly and the decoder does not +produce nil values. When the "nil" tag is set, input values of size zero decode as a nil +pointer. This is especially useful for recursive types. + + type StructWithNilField struct { + Field *[3]byte `rlp:"nil"` + } + +In the example above, Field allows two possible input sizes. For input 0xC180 (a list +containing an empty string) Field is set to nil after decoding. For input 0xC483000000 (a +list containing a 3-byte string), Field is set to a non-nil array pointer. + +RLP supports two kinds of empty values: empty lists and empty strings. When using the +"nil" tag, the kind of empty value allowed for a type is chosen automatically. A field +whose Go type is a pointer to an unsigned integer, string, boolean or byte array/slice +expects an empty RLP string. Any other pointer field type encodes/decodes as an empty RLP +list. + +The choice of null value can be made explicit with the "nilList" and "nilString" struct +tags. Using these tags encodes/decodes a Go nil pointer value as the empty RLP value kind +defined by the tag. */ package rlp diff --git a/rlp/encbuffer.go b/rlp/encbuffer.go new file mode 100644 index 000000000000..8d3a3b2293a5 --- /dev/null +++ b/rlp/encbuffer.go @@ -0,0 +1,423 @@ +// Copyright 2022 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +package rlp + +import ( + "encoding/binary" + "io" + "math/big" + "reflect" + "sync" + + "github.com/holiman/uint256" +) + +type encBuffer struct { + str []byte // string data, contains everything except list headers + lheads []listhead // all list headers + lhsize int // sum of sizes of all encoded list headers + sizebuf [9]byte // auxiliary buffer for uint encoding +} + +// The global encBuffer pool. +var encBufferPool = sync.Pool{ + New: func() interface{} { return new(encBuffer) }, +} + +func getEncBuffer() *encBuffer { + buf := encBufferPool.Get().(*encBuffer) + buf.reset() + return buf +} + +func (buf *encBuffer) reset() { + buf.lhsize = 0 + buf.str = buf.str[:0] + buf.lheads = buf.lheads[:0] +} + +// size returns the length of the encoded data. +func (buf *encBuffer) size() int { + return len(buf.str) + buf.lhsize +} + +// makeBytes creates the encoder output. +func (buf *encBuffer) makeBytes() []byte { + out := make([]byte, buf.size()) + buf.copyTo(out) + return out +} + +func (buf *encBuffer) copyTo(dst []byte) { + strpos := 0 + pos := 0 + for _, head := range buf.lheads { + // write string data before header + n := copy(dst[pos:], buf.str[strpos:head.offset]) + pos += n + strpos += n + // write the header + enc := head.encode(dst[pos:]) + pos += len(enc) + } + // copy string data after the last list header + copy(dst[pos:], buf.str[strpos:]) +} + +// writeTo writes the encoder output to w. +func (buf *encBuffer) writeTo(w io.Writer) (err error) { + strpos := 0 + for _, head := range buf.lheads { + // write string data before header + if head.offset-strpos > 0 { + n, err := w.Write(buf.str[strpos:head.offset]) + strpos += n + if err != nil { + return err + } + } + // write the header + enc := head.encode(buf.sizebuf[:]) + if _, err = w.Write(enc); err != nil { + return err + } + } + if strpos < len(buf.str) { + // write string data after the last list header + _, err = w.Write(buf.str[strpos:]) + } + return err +} + +// Write implements io.Writer and appends b directly to the output. +func (buf *encBuffer) Write(b []byte) (int, error) { + buf.str = append(buf.str, b...) + return len(b), nil +} + +// writeBool writes b as the integer 0 (false) or 1 (true). +func (buf *encBuffer) writeBool(b bool) { + if b { + buf.str = append(buf.str, 0x01) + } else { + buf.str = append(buf.str, 0x80) + } +} + +func (buf *encBuffer) writeUint64(i uint64) { + if i == 0 { + buf.str = append(buf.str, 0x80) + } else if i < 128 { + // fits single byte + buf.str = append(buf.str, byte(i)) + } else { + s := putint(buf.sizebuf[1:], i) + buf.sizebuf[0] = 0x80 + byte(s) + buf.str = append(buf.str, buf.sizebuf[:s+1]...) + } +} + +func (buf *encBuffer) writeBytes(b []byte) { + if len(b) == 1 && b[0] <= 0x7F { + // fits single byte, no string header + buf.str = append(buf.str, b[0]) + } else { + buf.encodeStringHeader(len(b)) + buf.str = append(buf.str, b...) + } +} + +func (buf *encBuffer) writeString(s string) { + buf.writeBytes([]byte(s)) +} + +// wordBytes is the number of bytes in a big.Word +const wordBytes = (32 << (uint64(^big.Word(0)) >> 63)) / 8 + +// writeBigInt writes i as an integer. +func (buf *encBuffer) writeBigInt(i *big.Int) { + bitlen := i.BitLen() + if bitlen <= 64 { + buf.writeUint64(i.Uint64()) + return + } + // Integer is larger than 64 bits, encode from i.Bits(). + // The minimal byte length is bitlen rounded up to the next + // multiple of 8, divided by 8. + length := ((bitlen + 7) & -8) >> 3 + buf.encodeStringHeader(length) + buf.str = append(buf.str, make([]byte, length)...) + index := length + bytesBuf := buf.str[len(buf.str)-length:] + for _, d := range i.Bits() { + for j := 0; j < wordBytes && index > 0; j++ { + index-- + bytesBuf[index] = byte(d) + d >>= 8 + } + } +} + +// writeUint256 writes z as an integer. +func (buf *encBuffer) writeUint256(z *uint256.Int) { + bitlen := z.BitLen() + if bitlen <= 64 { + buf.writeUint64(z.Uint64()) + return + } + nBytes := byte((bitlen + 7) / 8) + var b [33]byte + binary.BigEndian.PutUint64(b[1:9], z[3]) + binary.BigEndian.PutUint64(b[9:17], z[2]) + binary.BigEndian.PutUint64(b[17:25], z[1]) + binary.BigEndian.PutUint64(b[25:33], z[0]) + b[32-nBytes] = 0x80 + nBytes + buf.str = append(buf.str, b[32-nBytes:]...) +} + +// list adds a new list header to the header stack. It returns the index of the header. +// Call listEnd with this index after encoding the content of the list. +func (buf *encBuffer) list() int { + buf.lheads = append(buf.lheads, listhead{offset: len(buf.str), size: buf.lhsize}) + return len(buf.lheads) - 1 +} + +func (buf *encBuffer) listEnd(index int) { + lh := &buf.lheads[index] + lh.size = buf.size() - lh.offset - lh.size + if lh.size < 56 { + buf.lhsize++ // length encoded into kind tag + } else { + buf.lhsize += 1 + intsize(uint64(lh.size)) + } +} + +func (buf *encBuffer) encode(val interface{}) error { + rval := reflect.ValueOf(val) + writer, err := cachedWriter(rval.Type()) + if err != nil { + return err + } + return writer(rval, buf) +} + +func (buf *encBuffer) encodeStringHeader(size int) { + if size < 56 { + buf.str = append(buf.str, 0x80+byte(size)) + } else { + sizesize := putint(buf.sizebuf[1:], uint64(size)) + buf.sizebuf[0] = 0xB7 + byte(sizesize) + buf.str = append(buf.str, buf.sizebuf[:sizesize+1]...) + } +} + +// encReader is the io.Reader returned by EncodeToReader. +// It releases its encbuf at EOF. +type encReader struct { + buf *encBuffer // the buffer we're reading from. this is nil when we're at EOF. + lhpos int // index of list header that we're reading + strpos int // current position in string buffer + piece []byte // next piece to be read +} + +func (r *encReader) Read(b []byte) (n int, err error) { + for { + if r.piece = r.next(); r.piece == nil { + // Put the encode buffer back into the pool at EOF when it + // is first encountered. Subsequent calls still return EOF + // as the error but the buffer is no longer valid. + if r.buf != nil { + encBufferPool.Put(r.buf) + r.buf = nil + } + return n, io.EOF + } + nn := copy(b[n:], r.piece) + n += nn + if nn < len(r.piece) { + // piece didn't fit, see you next time. + r.piece = r.piece[nn:] + return n, nil + } + r.piece = nil + } +} + +// next returns the next piece of data to be read. +// it returns nil at EOF. +func (r *encReader) next() []byte { + switch { + case r.buf == nil: + return nil + + case r.piece != nil: + // There is still data available for reading. + return r.piece + + case r.lhpos < len(r.buf.lheads): + // We're before the last list header. + head := r.buf.lheads[r.lhpos] + sizebefore := head.offset - r.strpos + if sizebefore > 0 { + // String data before header. + p := r.buf.str[r.strpos:head.offset] + r.strpos += sizebefore + return p + } + r.lhpos++ + return head.encode(r.buf.sizebuf[:]) + + case r.strpos < len(r.buf.str): + // String data at the end, after all list headers. + p := r.buf.str[r.strpos:] + r.strpos = len(r.buf.str) + return p + + default: + return nil + } +} + +func encBufferFromWriter(w io.Writer) *encBuffer { + switch w := w.(type) { + case EncoderBuffer: + return w.buf + case *EncoderBuffer: + return w.buf + case *encBuffer: + return w + default: + return nil + } +} + +// EncoderBuffer is a buffer for incremental encoding. +// +// The zero value is NOT ready for use. To get a usable buffer, +// create it using NewEncoderBuffer or call Reset. +type EncoderBuffer struct { + buf *encBuffer + dst io.Writer + + ownBuffer bool +} + +// NewEncoderBuffer creates an encoder buffer. +func NewEncoderBuffer(dst io.Writer) EncoderBuffer { + var w EncoderBuffer + w.Reset(dst) + return w +} + +// Reset truncates the buffer and sets the output destination. +func (w *EncoderBuffer) Reset(dst io.Writer) { + if w.buf != nil && !w.ownBuffer { + panic("can't Reset derived EncoderBuffer") + } + + // If the destination writer has an *encBuffer, use it. + // Note that w.ownBuffer is left false here. + if dst != nil { + if outer := encBufferFromWriter(dst); outer != nil { + *w = EncoderBuffer{outer, nil, false} + return + } + } + + // Get a fresh buffer. + if w.buf == nil { + w.buf = encBufferPool.Get().(*encBuffer) + w.ownBuffer = true + } + w.buf.reset() + w.dst = dst +} + +// Flush writes encoded RLP data to the output writer. This can only be called once. +// If you want to re-use the buffer after Flush, you must call Reset. +func (w *EncoderBuffer) Flush() error { + var err error + if w.dst != nil { + err = w.buf.writeTo(w.dst) + } + // Release the internal buffer. + if w.ownBuffer { + encBufferPool.Put(w.buf) + } + *w = EncoderBuffer{} + return err +} + +// ToBytes returns the encoded bytes. +func (w *EncoderBuffer) ToBytes() []byte { + return w.buf.makeBytes() +} + +// AppendToBytes appends the encoded bytes to dst. +func (w *EncoderBuffer) AppendToBytes(dst []byte) []byte { + size := w.buf.size() + out := append(dst, make([]byte, size)...) + w.buf.copyTo(out[len(dst):]) + return out +} + +// Write appends b directly to the encoder output. +func (w EncoderBuffer) Write(b []byte) (int, error) { + return w.buf.Write(b) +} + +// WriteBool writes b as the integer 0 (false) or 1 (true). +func (w EncoderBuffer) WriteBool(b bool) { + w.buf.writeBool(b) +} + +// WriteUint64 encodes an unsigned integer. +func (w EncoderBuffer) WriteUint64(i uint64) { + w.buf.writeUint64(i) +} + +// WriteBigInt encodes a big.Int as an RLP string. +// Note: Unlike with Encode, the sign of i is ignored. +func (w EncoderBuffer) WriteBigInt(i *big.Int) { + w.buf.writeBigInt(i) +} + +// WriteUint256 encodes uint256.Int as an RLP string. +func (w EncoderBuffer) WriteUint256(i *uint256.Int) { + w.buf.writeUint256(i) +} + +// WriteBytes encodes b as an RLP string. +func (w EncoderBuffer) WriteBytes(b []byte) { + w.buf.writeBytes(b) +} + +// WriteString encodes s as an RLP string. +func (w EncoderBuffer) WriteString(s string) { + w.buf.writeString(s) +} + +// List starts a list. It returns an internal index. Call EndList with +// this index after encoding the content to finish the list. +func (w EncoderBuffer) List() int { + return w.buf.list() +} + +// ListEnd finishes the given list. +func (w EncoderBuffer) ListEnd(index int) { + w.buf.listEnd(index) +} diff --git a/rlp/encbuffer_example_test.go b/rlp/encbuffer_example_test.go new file mode 100644 index 000000000000..f737f3e40c72 --- /dev/null +++ b/rlp/encbuffer_example_test.go @@ -0,0 +1,45 @@ +// Copyright 2022 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +package rlp_test + +import ( + "bytes" + "fmt" + + "github.com/XinFinOrg/XDPoSChain/rlp" +) + +func ExampleEncoderBuffer() { + var w bytes.Buffer + + // Encode [4, [5, 6]] to w. + buf := rlp.NewEncoderBuffer(&w) + l1 := buf.List() + buf.WriteUint64(4) + l2 := buf.List() + buf.WriteUint64(5) + buf.WriteUint64(6) + buf.ListEnd(l2) + buf.ListEnd(l1) + + if err := buf.Flush(); err != nil { + panic(err) + } + fmt.Printf("%X\n", w.Bytes()) + // Output: + // C404C20506 +} diff --git a/rlp/encode.go b/rlp/encode.go index 44592c2f53ed..9435cfc22c6b 100644 --- a/rlp/encode.go +++ b/rlp/encode.go @@ -17,20 +17,28 @@ package rlp import ( + "errors" "fmt" "io" "math/big" "reflect" - "sync" + + "github.com/XinFinOrg/XDPoSChain/rlp/internal/rlpstruct" + "github.com/holiman/uint256" ) var ( // Common encoded values. // These are useful when implementing EncodeRLP. + + // EmptyString is the encoding of an empty string. EmptyString = []byte{0x80} - EmptyList = []byte{0xC0} + // EmptyList is the encoding of an empty list. + EmptyList = []byte{0xC0} ) +var ErrNegativeBigInt = errors.New("rlp: cannot encode negative big.Int") + // Encoder is implemented by types that require custom // encoding rules or want to encode private fields. type Encoder interface { @@ -49,80 +57,48 @@ type Encoder interface { // perform many small writes in some cases. Consider making w // buffered. // -// Encode uses the following type-dependent encoding rules: -// -// If the type implements the Encoder interface, Encode calls -// EncodeRLP. This is true even for nil pointers, please see the -// documentation for Encoder. -// -// To encode a pointer, the value being pointed to is encoded. For nil -// pointers, Encode will encode the zero value of the type. A nil -// pointer to a struct type always encodes as an empty RLP list. -// A nil pointer to an array encodes as an empty list (or empty string -// if the array has element type byte). -// -// Struct values are encoded as an RLP list of all their encoded -// public fields. Recursive struct types are supported. -// -// To encode slices and arrays, the elements are encoded as an RLP -// list of the value's elements. Note that arrays and slices with -// element type uint8 or byte are always encoded as an RLP string. -// -// A Go string is encoded as an RLP string. -// -// An unsigned integer value is encoded as an RLP string. Zero always -// encodes as an empty RLP string. Encode also supports *big.Int. -// -// An interface value encodes as the value contained in the interface. -// -// Boolean values are not supported, nor are signed integers, floating -// point numbers, maps, channels and functions. +// Please see package-level documentation of encoding rules. func Encode(w io.Writer, val interface{}) error { - if outer, ok := w.(*encbuf); ok { - // Encode was called by some type's EncodeRLP. - // Avoid copying by writing to the outer encbuf directly. - return outer.encode(val) + // Optimization: reuse *encBuffer when called by EncodeRLP. + if buf := encBufferFromWriter(w); buf != nil { + return buf.encode(val) } - eb := encbufPool.Get().(*encbuf) - defer encbufPool.Put(eb) - eb.reset() - if err := eb.encode(val); err != nil { + + buf := getEncBuffer() + defer encBufferPool.Put(buf) + if err := buf.encode(val); err != nil { return err } - return eb.toWriter(w) + return buf.writeTo(w) } -// EncodeBytes returns the RLP encoding of val. -// Please see the documentation of Encode for the encoding rules. +// EncodeToBytes returns the RLP encoding of val. +// Please see package-level documentation for the encoding rules. func EncodeToBytes(val interface{}) ([]byte, error) { - eb := encbufPool.Get().(*encbuf) - defer encbufPool.Put(eb) - eb.reset() - if err := eb.encode(val); err != nil { + buf := getEncBuffer() + defer encBufferPool.Put(buf) + + if err := buf.encode(val); err != nil { return nil, err } - return eb.toBytes(), nil + return buf.makeBytes(), nil } -// EncodeReader returns a reader from which the RLP encoding of val +// EncodeToReader returns a reader from which the RLP encoding of val // can be read. The returned size is the total size of the encoded // data. // // Please see the documentation of Encode for the encoding rules. func EncodeToReader(val interface{}) (size int, r io.Reader, err error) { - eb := encbufPool.Get().(*encbuf) - eb.reset() - if err := eb.encode(val); err != nil { + buf := getEncBuffer() + if err := buf.encode(val); err != nil { + encBufferPool.Put(buf) return 0, nil, err } - return eb.size(), &encReader{buf: eb}, nil -} - -type encbuf struct { - str []byte // string data, contains everything except list headers - lheads []*listhead // all list headers - lhsize int // sum of sizes of all encoded list headers - sizebuf []byte // 9-byte auxiliary buffer for uint encoding + // Note: can't put the reader back into the pool here + // because it is held by encReader. The reader puts it + // back when it has been fully consumed. + return buf.size(), &encReader{buf: buf}, nil } type listhead struct { @@ -151,214 +127,32 @@ func puthead(buf []byte, smalltag, largetag byte, size uint64) int { if size < 56 { buf[0] = smalltag + byte(size) return 1 - } else { - sizesize := putint(buf[1:], size) - buf[0] = largetag + byte(sizesize) - return sizesize + 1 - } -} - -// encbufs are pooled. -var encbufPool = sync.Pool{ - New: func() interface{} { return &encbuf{sizebuf: make([]byte, 9)} }, -} - -func (w *encbuf) reset() { - w.lhsize = 0 - if w.str != nil { - w.str = w.str[:0] - } - if w.lheads != nil { - w.lheads = w.lheads[:0] - } -} - -// encbuf implements io.Writer so it can be passed it into EncodeRLP. -func (w *encbuf) Write(b []byte) (int, error) { - w.str = append(w.str, b...) - return len(b), nil -} - -func (w *encbuf) encode(val interface{}) error { - rval := reflect.ValueOf(val) - ti, err := cachedTypeInfo(rval.Type(), tags{}) - if err != nil { - return err - } - return ti.writer(rval, w) -} - -func (w *encbuf) encodeStringHeader(size int) { - if size < 56 { - w.str = append(w.str, 0x80+byte(size)) - } else { - // TODO: encode to w.str directly - sizesize := putint(w.sizebuf[1:], uint64(size)) - w.sizebuf[0] = 0xB7 + byte(sizesize) - w.str = append(w.str, w.sizebuf[:sizesize+1]...) - } -} - -func (w *encbuf) encodeString(b []byte) { - if len(b) == 1 && b[0] <= 0x7F { - // fits single byte, no string header - w.str = append(w.str, b[0]) - } else { - w.encodeStringHeader(len(b)) - w.str = append(w.str, b...) - } -} - -func (w *encbuf) list() *listhead { - lh := &listhead{offset: len(w.str), size: w.lhsize} - w.lheads = append(w.lheads, lh) - return lh -} - -func (w *encbuf) listEnd(lh *listhead) { - lh.size = w.size() - lh.offset - lh.size - if lh.size < 56 { - w.lhsize += 1 // length encoded into kind tag - } else { - w.lhsize += 1 + intsize(uint64(lh.size)) - } -} - -func (w *encbuf) size() int { - return len(w.str) + w.lhsize -} - -func (w *encbuf) toBytes() []byte { - out := make([]byte, w.size()) - strpos := 0 - pos := 0 - for _, head := range w.lheads { - // write string data before header - n := copy(out[pos:], w.str[strpos:head.offset]) - pos += n - strpos += n - // write the header - enc := head.encode(out[pos:]) - pos += len(enc) - } - // copy string data after the last list header - copy(out[pos:], w.str[strpos:]) - return out -} - -func (w *encbuf) toWriter(out io.Writer) (err error) { - strpos := 0 - for _, head := range w.lheads { - // write string data before header - if head.offset-strpos > 0 { - n, err := out.Write(w.str[strpos:head.offset]) - strpos += n - if err != nil { - return err - } - } - // write the header - enc := head.encode(w.sizebuf) - if _, err = out.Write(enc); err != nil { - return err - } - } - if strpos < len(w.str) { - // write string data after the last list header - _, err = out.Write(w.str[strpos:]) - } - return err -} - -// encReader is the io.Reader returned by EncodeToReader. -// It releases its encbuf at EOF. -type encReader struct { - buf *encbuf // the buffer we're reading from. this is nil when we're at EOF. - lhpos int // index of list header that we're reading - strpos int // current position in string buffer - piece []byte // next piece to be read -} - -func (r *encReader) Read(b []byte) (n int, err error) { - for { - if r.piece = r.next(); r.piece == nil { - // Put the encode buffer back into the pool at EOF when it - // is first encountered. Subsequent calls still return EOF - // as the error but the buffer is no longer valid. - if r.buf != nil { - encbufPool.Put(r.buf) - r.buf = nil - } - return n, io.EOF - } - nn := copy(b[n:], r.piece) - n += nn - if nn < len(r.piece) { - // piece didn't fit, see you next time. - r.piece = r.piece[nn:] - return n, nil - } - r.piece = nil - } -} - -// next returns the next piece of data to be read. -// it returns nil at EOF. -func (r *encReader) next() []byte { - switch { - case r.buf == nil: - return nil - - case r.piece != nil: - // There is still data available for reading. - return r.piece - - case r.lhpos < len(r.buf.lheads): - // We're before the last list header. - head := r.buf.lheads[r.lhpos] - sizebefore := head.offset - r.strpos - if sizebefore > 0 { - // String data before header. - p := r.buf.str[r.strpos:head.offset] - r.strpos += sizebefore - return p - } else { - r.lhpos++ - return head.encode(r.buf.sizebuf) - } - - case r.strpos < len(r.buf.str): - // String data at the end, after all list headers. - p := r.buf.str[r.strpos:] - r.strpos = len(r.buf.str) - return p - - default: - return nil } + sizesize := putint(buf[1:], size) + buf[0] = largetag + byte(sizesize) + return sizesize + 1 } -var ( - encoderInterface = reflect.TypeOf(new(Encoder)).Elem() - big0 = big.NewInt(0) -) +var encoderInterface = reflect.TypeOf(new(Encoder)).Elem() // makeWriter creates a writer function for the given type. -func makeWriter(typ reflect.Type, ts tags) (writer, error) { +func makeWriter(typ reflect.Type, ts rlpstruct.Tags) (writer, error) { kind := typ.Kind() switch { case typ == rawValueType: return writeRawValue, nil - case typ.Implements(encoderInterface): - return writeEncoder, nil - case kind != reflect.Ptr && reflect.PtrTo(typ).Implements(encoderInterface): - return writeEncoderNoPtr, nil - case kind == reflect.Interface: - return writeInterface, nil - case typ.AssignableTo(reflect.PtrTo(bigInt)): + case typ.AssignableTo(reflect.PointerTo(bigInt)): return writeBigIntPtr, nil case typ.AssignableTo(bigInt): return writeBigIntNoPtr, nil + case typ == reflect.PointerTo(u256Int): + return writeU256IntPtr, nil + case typ == u256Int: + return writeU256IntNoPtr, nil + case kind == reflect.Ptr: + return makePtrWriter(typ, ts) + case reflect.PointerTo(typ).Implements(encoderInterface): + return makeEncoderWriter(typ), nil case isUint(kind): return writeUint, nil case kind == reflect.Bool: @@ -368,97 +162,116 @@ func makeWriter(typ reflect.Type, ts tags) (writer, error) { case kind == reflect.Slice && isByte(typ.Elem()): return writeBytes, nil case kind == reflect.Array && isByte(typ.Elem()): - return writeByteArray, nil + return makeByteArrayWriter(typ), nil case kind == reflect.Slice || kind == reflect.Array: return makeSliceWriter(typ, ts) case kind == reflect.Struct: return makeStructWriter(typ) - case kind == reflect.Ptr: - return makePtrWriter(typ) + case kind == reflect.Interface: + return writeInterface, nil default: return nil, fmt.Errorf("rlp: type %v is not RLP-serializable", typ) } } -func isByte(typ reflect.Type) bool { - return typ.Kind() == reflect.Uint8 && !typ.Implements(encoderInterface) -} - -func writeRawValue(val reflect.Value, w *encbuf) error { +func writeRawValue(val reflect.Value, w *encBuffer) error { w.str = append(w.str, val.Bytes()...) return nil } -func writeUint(val reflect.Value, w *encbuf) error { - i := val.Uint() - if i == 0 { - w.str = append(w.str, 0x80) - } else if i < 128 { - // fits single byte - w.str = append(w.str, byte(i)) - } else { - // TODO: encode int to w.str directly - s := putint(w.sizebuf[1:], i) - w.sizebuf[0] = 0x80 + byte(s) - w.str = append(w.str, w.sizebuf[:s+1]...) - } +func writeUint(val reflect.Value, w *encBuffer) error { + w.writeUint64(val.Uint()) return nil } -func writeBool(val reflect.Value, w *encbuf) error { - if val.Bool() { - w.str = append(w.str, 0x01) - } else { - w.str = append(w.str, 0x80) - } +func writeBool(val reflect.Value, w *encBuffer) error { + w.writeBool(val.Bool()) return nil } -func writeBigIntPtr(val reflect.Value, w *encbuf) error { +func writeBigIntPtr(val reflect.Value, w *encBuffer) error { ptr := val.Interface().(*big.Int) if ptr == nil { w.str = append(w.str, 0x80) return nil } - return writeBigInt(ptr, w) + if ptr.Sign() == -1 { + return ErrNegativeBigInt + } + w.writeBigInt(ptr) + return nil } -func writeBigIntNoPtr(val reflect.Value, w *encbuf) error { +func writeBigIntNoPtr(val reflect.Value, w *encBuffer) error { i := val.Interface().(big.Int) - return writeBigInt(&i, w) + if i.Sign() == -1 { + return ErrNegativeBigInt + } + w.writeBigInt(&i) + return nil } -func writeBigInt(i *big.Int, w *encbuf) error { - if cmp := i.Cmp(big0); cmp == -1 { - return fmt.Errorf("rlp: cannot encode negative *big.Int") - } else if cmp == 0 { +func writeU256IntPtr(val reflect.Value, w *encBuffer) error { + ptr := val.Interface().(*uint256.Int) + if ptr == nil { w.str = append(w.str, 0x80) - } else { - w.encodeString(i.Bytes()) + return nil } + w.writeUint256(ptr) + return nil +} + +func writeU256IntNoPtr(val reflect.Value, w *encBuffer) error { + i := val.Interface().(uint256.Int) + w.writeUint256(&i) return nil } -func writeBytes(val reflect.Value, w *encbuf) error { - w.encodeString(val.Bytes()) +func writeBytes(val reflect.Value, w *encBuffer) error { + w.writeBytes(val.Bytes()) return nil } -func writeByteArray(val reflect.Value, w *encbuf) error { - if !val.CanAddr() { - // Slice requires the value to be addressable. - // Make it addressable by copying. - copy := reflect.New(val.Type()).Elem() - copy.Set(val) - val = copy +func makeByteArrayWriter(typ reflect.Type) writer { + switch typ.Len() { + case 0: + return writeLengthZeroByteArray + case 1: + return writeLengthOneByteArray + default: + length := typ.Len() + return func(val reflect.Value, w *encBuffer) error { + if !val.CanAddr() { + // Getting the byte slice of val requires it to be addressable. Make it + // addressable by copying. + copy := reflect.New(val.Type()).Elem() + copy.Set(val) + val = copy + } + slice := byteArrayBytes(val, length) + w.encodeStringHeader(len(slice)) + w.str = append(w.str, slice...) + return nil + } } - size := val.Len() - slice := val.Slice(0, size).Bytes() - w.encodeString(slice) +} + +func writeLengthZeroByteArray(val reflect.Value, w *encBuffer) error { + w.str = append(w.str, 0x80) return nil } -func writeString(val reflect.Value, w *encbuf) error { +func writeLengthOneByteArray(val reflect.Value, w *encBuffer) error { + b := byte(val.Index(0).Uint()) + if b <= 0x7f { + w.str = append(w.str, b) + } else { + w.str = append(w.str, 0x81, b) + } + return nil +} + +func writeString(val reflect.Value, w *encBuffer) error { s := val.String() if len(s) == 1 && s[0] <= 0x7f { // fits single byte, no string header @@ -470,27 +283,7 @@ func writeString(val reflect.Value, w *encbuf) error { return nil } -func writeEncoder(val reflect.Value, w *encbuf) error { - return val.Interface().(Encoder).EncodeRLP(w) -} - -// writeEncoderNoPtr handles non-pointer values that implement Encoder -// with a pointer receiver. -func writeEncoderNoPtr(val reflect.Value, w *encbuf) error { - if !val.CanAddr() { - // We can't get the address. It would be possible to make the - // value addressable by creating a shallow copy, but this - // creates other problems so we're not doing it (yet). - // - // package json simply doesn't call MarshalJSON for cases like - // this, but encodes the value as if it didn't implement the - // interface. We don't want to handle it that way. - return fmt.Errorf("rlp: game over: unadressable value of type %v, EncodeRLP is pointer method", val.Type()) - } - return val.Addr().Interface().(Encoder).EncodeRLP(w) -} - -func writeInterface(val reflect.Value, w *encbuf) error { +func writeInterface(val reflect.Value, w *encBuffer) error { if val.IsNil() { // Write empty list. This is consistent with the previous RLP // encoder that we had and should therefore avoid any @@ -499,31 +292,51 @@ func writeInterface(val reflect.Value, w *encbuf) error { return nil } eval := val.Elem() - ti, err := cachedTypeInfo(eval.Type(), tags{}) + writer, err := cachedWriter(eval.Type()) if err != nil { return err } - return ti.writer(eval, w) + return writer(eval, w) } -func makeSliceWriter(typ reflect.Type, ts tags) (writer, error) { - etypeinfo, err := cachedTypeInfo1(typ.Elem(), tags{}) - if err != nil { - return nil, err +func makeSliceWriter(typ reflect.Type, ts rlpstruct.Tags) (writer, error) { + etypeinfo := theTC.infoWhileGenerating(typ.Elem(), rlpstruct.Tags{}) + if etypeinfo.writerErr != nil { + return nil, etypeinfo.writerErr } - writer := func(val reflect.Value, w *encbuf) error { - if !ts.tail { - defer w.listEnd(w.list()) + + var wfn writer + if ts.Tail { + // This is for struct tail slices. + // w.list is not called for them. + wfn = func(val reflect.Value, w *encBuffer) error { + vlen := val.Len() + for i := 0; i < vlen; i++ { + if err := etypeinfo.writer(val.Index(i), w); err != nil { + return err + } + } + return nil } - vlen := val.Len() - for i := 0; i < vlen; i++ { - if err := etypeinfo.writer(val.Index(i), w); err != nil { - return err + } else { + // This is for regular slices and arrays. + wfn = func(val reflect.Value, w *encBuffer) error { + vlen := val.Len() + if vlen == 0 { + w.str = append(w.str, 0xC0) + return nil + } + listOffset := w.list() + for i := 0; i < vlen; i++ { + if err := etypeinfo.writer(val.Index(i), w); err != nil { + return err + } } + w.listEnd(listOffset) + return nil } - return nil } - return writer, nil + return wfn, nil } func makeStructWriter(typ reflect.Type) (writer, error) { @@ -531,56 +344,86 @@ func makeStructWriter(typ reflect.Type) (writer, error) { if err != nil { return nil, err } - writer := func(val reflect.Value, w *encbuf) error { - lh := w.list() - for _, f := range fields { - if err := f.info.writer(val.Field(f.index), w); err != nil { - return err + for _, f := range fields { + if f.info.writerErr != nil { + return nil, structFieldError{typ, f.index, f.info.writerErr} + } + } + + var writer writer + firstOptionalField := firstOptionalField(fields) + if firstOptionalField == len(fields) { + // This is the writer function for structs without any optional fields. + writer = func(val reflect.Value, w *encBuffer) error { + lh := w.list() + for _, f := range fields { + if err := f.info.writer(val.Field(f.index), w); err != nil { + return err + } } + w.listEnd(lh) + return nil + } + } else { + // If there are any "optional" fields, the writer needs to perform additional + // checks to determine the output list length. + writer = func(val reflect.Value, w *encBuffer) error { + lastField := len(fields) - 1 + for ; lastField >= firstOptionalField; lastField-- { + if !val.Field(fields[lastField].index).IsZero() { + break + } + } + lh := w.list() + for i := 0; i <= lastField; i++ { + if err := fields[i].info.writer(val.Field(fields[i].index), w); err != nil { + return err + } + } + w.listEnd(lh) + return nil } - w.listEnd(lh) - return nil } return writer, nil } -func makePtrWriter(typ reflect.Type) (writer, error) { - etypeinfo, err := cachedTypeInfo1(typ.Elem(), tags{}) - if err != nil { - return nil, err +func makePtrWriter(typ reflect.Type, ts rlpstruct.Tags) (writer, error) { + nilEncoding := byte(0xC0) + if typeNilKind(typ.Elem(), ts) == String { + nilEncoding = 0x80 } - // determine nil pointer handler - var nilfunc func(*encbuf) error - kind := typ.Elem().Kind() - switch { - case kind == reflect.Array && isByte(typ.Elem().Elem()): - nilfunc = func(w *encbuf) error { - w.str = append(w.str, 0x80) - return nil - } - case kind == reflect.Struct || kind == reflect.Array: - nilfunc = func(w *encbuf) error { - // encoding the zero value of a struct/array could trigger - // infinite recursion, avoid that. - w.listEnd(w.list()) - return nil - } - default: - zero := reflect.Zero(typ.Elem()) - nilfunc = func(w *encbuf) error { - return etypeinfo.writer(zero, w) + etypeinfo := theTC.infoWhileGenerating(typ.Elem(), rlpstruct.Tags{}) + if etypeinfo.writerErr != nil { + return nil, etypeinfo.writerErr + } + + writer := func(val reflect.Value, w *encBuffer) error { + if ev := val.Elem(); ev.IsValid() { + return etypeinfo.writer(ev, w) } + w.str = append(w.str, nilEncoding) + return nil } + return writer, nil +} - writer := func(val reflect.Value, w *encbuf) error { - if val.IsNil() { - return nilfunc(w) - } else { - return etypeinfo.writer(val.Elem(), w) +func makeEncoderWriter(typ reflect.Type) writer { + if typ.Implements(encoderInterface) { + return func(val reflect.Value, w *encBuffer) error { + return val.Interface().(Encoder).EncodeRLP(w) + } + } + w := func(val reflect.Value, w *encBuffer) error { + if !val.CanAddr() { + // package json simply doesn't call MarshalJSON for this case, but encodes the + // value as if it didn't implement the interface. We don't want to handle it that + // way. + return fmt.Errorf("rlp: unaddressable value of type %v, EncodeRLP is pointer method", val.Type()) } + return val.Addr().Interface().(Encoder).EncodeRLP(w) } - return writer, err + return w } // putint writes i to the beginning of b in big endian byte diff --git a/rlp/encode_test.go b/rlp/encode_test.go index 12e8f275551b..5fc2d116efda 100644 --- a/rlp/encode_test.go +++ b/rlp/encode_test.go @@ -22,8 +22,12 @@ import ( "fmt" "io" "math/big" + "runtime" "sync" "testing" + + "github.com/XinFinOrg/XDPoSChain/common/math" + "github.com/holiman/uint256" ) type testEncoder struct { @@ -32,12 +36,19 @@ type testEncoder struct { func (e *testEncoder) EncodeRLP(w io.Writer) error { if e == nil { - w.Write([]byte{0, 0, 0, 0}) - } else if e.err != nil { + panic("EncodeRLP called on nil value") + } + if e.err != nil { return e.err - } else { - w.Write([]byte{0, 1, 0, 1, 0, 1, 0, 1, 0, 1}) } + w.Write([]byte{0, 1, 0, 1, 0, 1, 0, 1, 0, 1}) + return nil +} + +type testEncoderValueMethod struct{} + +func (e testEncoderValueMethod) EncodeRLP(w io.Writer) error { + w.Write([]byte{0xFA, 0xFE, 0xF0}) return nil } @@ -48,6 +59,13 @@ func (e byteEncoder) EncodeRLP(w io.Writer) error { return nil } +type undecodableEncoder func() + +func (f undecodableEncoder) EncodeRLP(w io.Writer) error { + w.Write([]byte{0xF5, 0xF5, 0xF5}) + return nil +} + type encodableReader struct { A, B uint } @@ -102,35 +120,95 @@ var encTests = []encTest{ {val: big.NewInt(0xFFFFFFFFFFFF), output: "86FFFFFFFFFFFF"}, {val: big.NewInt(0xFFFFFFFFFFFFFF), output: "87FFFFFFFFFFFFFF"}, { - val: big.NewInt(0).SetBytes(unhex("102030405060708090A0B0C0D0E0F2")), + val: new(big.Int).SetBytes(unhex("102030405060708090A0B0C0D0E0F2")), output: "8F102030405060708090A0B0C0D0E0F2", }, { - val: big.NewInt(0).SetBytes(unhex("0100020003000400050006000700080009000A000B000C000D000E01")), + val: new(big.Int).SetBytes(unhex("0100020003000400050006000700080009000A000B000C000D000E01")), output: "9C0100020003000400050006000700080009000A000B000C000D000E01", }, { - val: big.NewInt(0).SetBytes(unhex("010000000000000000000000000000000000000000000000000000000000000000")), + val: new(big.Int).SetBytes(unhex("010000000000000000000000000000000000000000000000000000000000000000")), output: "A1010000000000000000000000000000000000000000000000000000000000000000", }, + { + val: veryBigInt, + output: "89FFFFFFFFFFFFFFFFFF", + }, + { + val: veryVeryBigInt, + output: "B848FFFFFFFFFFFFFFFFF800000000000000001BFFFFFFFFFFFFFFFFC8000000000000000045FFFFFFFFFFFFFFFFC800000000000000001BFFFFFFFFFFFFFFFFF8000000000000000001", + }, // non-pointer big.Int {val: *big.NewInt(0), output: "80"}, {val: *big.NewInt(0xFFFFFF), output: "83FFFFFF"}, // negative ints are not supported - {val: big.NewInt(-1), error: "rlp: cannot encode negative *big.Int"}, - - // byte slices, strings + {val: big.NewInt(-1), error: "rlp: cannot encode negative big.Int"}, + {val: *big.NewInt(-1), error: "rlp: cannot encode negative big.Int"}, + + // uint256 + {val: uint256.NewInt(0), output: "80"}, + {val: uint256.NewInt(1), output: "01"}, + {val: uint256.NewInt(127), output: "7F"}, + {val: uint256.NewInt(128), output: "8180"}, + {val: uint256.NewInt(256), output: "820100"}, + {val: uint256.NewInt(1024), output: "820400"}, + {val: uint256.NewInt(0xFFFFFF), output: "83FFFFFF"}, + {val: uint256.NewInt(0xFFFFFFFF), output: "84FFFFFFFF"}, + {val: uint256.NewInt(0xFFFFFFFFFF), output: "85FFFFFFFFFF"}, + {val: uint256.NewInt(0xFFFFFFFFFFFF), output: "86FFFFFFFFFFFF"}, + {val: uint256.NewInt(0xFFFFFFFFFFFFFF), output: "87FFFFFFFFFFFFFF"}, + { + val: new(uint256.Int).SetBytes(unhex("102030405060708090A0B0C0D0E0F2")), + output: "8F102030405060708090A0B0C0D0E0F2", + }, + { + val: new(uint256.Int).SetBytes(unhex("0100020003000400050006000700080009000A000B000C000D000E01")), + output: "9C0100020003000400050006000700080009000A000B000C000D000E01", + }, + // non-pointer uint256.Int + {val: *uint256.NewInt(0), output: "80"}, + {val: *uint256.NewInt(0xFFFFFF), output: "83FFFFFF"}, + + // byte arrays + {val: [0]byte{}, output: "80"}, + {val: [1]byte{0}, output: "00"}, + {val: [1]byte{1}, output: "01"}, + {val: [1]byte{0x7F}, output: "7F"}, + {val: [1]byte{0x80}, output: "8180"}, + {val: [1]byte{0xFF}, output: "81FF"}, + {val: [3]byte{1, 2, 3}, output: "83010203"}, + {val: [57]byte{1, 2, 3}, output: "B839010203000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"}, + + // named byte type arrays + {val: [0]namedByteType{}, output: "80"}, + {val: [1]namedByteType{0}, output: "00"}, + {val: [1]namedByteType{1}, output: "01"}, + {val: [1]namedByteType{0x7F}, output: "7F"}, + {val: [1]namedByteType{0x80}, output: "8180"}, + {val: [1]namedByteType{0xFF}, output: "81FF"}, + {val: [3]namedByteType{1, 2, 3}, output: "83010203"}, + {val: [57]namedByteType{1, 2, 3}, output: "B839010203000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"}, + + // byte slices {val: []byte{}, output: "80"}, + {val: []byte{0}, output: "00"}, {val: []byte{0x7E}, output: "7E"}, {val: []byte{0x7F}, output: "7F"}, {val: []byte{0x80}, output: "8180"}, {val: []byte{1, 2, 3}, output: "83010203"}, + // named byte type slices + {val: []namedByteType{}, output: "80"}, + {val: []namedByteType{0}, output: "00"}, + {val: []namedByteType{0x7E}, output: "7E"}, + {val: []namedByteType{0x7F}, output: "7F"}, + {val: []namedByteType{0x80}, output: "8180"}, {val: []namedByteType{1, 2, 3}, output: "83010203"}, - {val: [...]namedByteType{1, 2, 3}, output: "83010203"}, + // strings {val: "", output: "80"}, {val: "\x7E", output: "7E"}, {val: "\x7F", output: "7F"}, @@ -203,6 +281,12 @@ var encTests = []encTest{ output: "F90200CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376CF84617364668471776572847A786376", }, + // Non-byte arrays are encoded as lists. + // Note that it is important to test [4]uint64 specifically, + // because that's the underlying type of uint256.Int. + {val: [4]uint32{1, 2, 3, 4}, output: "C401020304"}, + {val: [4]uint64{1, 2, 3, 4}, output: "C401020304"}, + // RawValue {val: RawValue(unhex("01")), output: "01"}, {val: RawValue(unhex("82FFFF")), output: "82FFFF"}, @@ -213,11 +297,34 @@ var encTests = []encTest{ {val: simplestruct{A: 3, B: "foo"}, output: "C50383666F6F"}, {val: &recstruct{5, nil}, output: "C205C0"}, {val: &recstruct{5, &recstruct{4, &recstruct{3, nil}}}, output: "C605C404C203C0"}, + {val: &intField{X: 3}, error: "rlp: type int is not RLP-serializable (struct field rlp.intField.X)"}, + + // struct tag "-" + {val: &ignoredField{A: 1, B: 2, C: 3}, output: "C20103"}, + + // struct tag "tail" {val: &tailRaw{A: 1, Tail: []RawValue{unhex("02"), unhex("03")}}, output: "C3010203"}, {val: &tailRaw{A: 1, Tail: []RawValue{unhex("02")}}, output: "C20102"}, {val: &tailRaw{A: 1, Tail: []RawValue{}}, output: "C101"}, {val: &tailRaw{A: 1, Tail: nil}, output: "C101"}, - {val: &hasIgnoredField{A: 1, B: 2, C: 3}, output: "C20103"}, + + // struct tag "optional" + {val: &optionalFields{}, output: "C180"}, + {val: &optionalFields{A: 1}, output: "C101"}, + {val: &optionalFields{A: 1, B: 2}, output: "C20102"}, + {val: &optionalFields{A: 1, B: 2, C: 3}, output: "C3010203"}, + {val: &optionalFields{A: 1, B: 0, C: 3}, output: "C3018003"}, + {val: &optionalAndTailField{A: 1}, output: "C101"}, + {val: &optionalAndTailField{A: 1, B: 2}, output: "C20102"}, + {val: &optionalAndTailField{A: 1, Tail: []uint{5, 6}}, output: "C401800506"}, + {val: &optionalAndTailField{A: 1, Tail: []uint{5, 6}}, output: "C401800506"}, + {val: &optionalBigIntField{A: 1}, output: "C101"}, + {val: &optionalPtrField{A: 1}, output: "C101"}, + {val: &optionalPtrFieldNil{A: 1}, output: "C101"}, + {val: &multipleOptionalFields{A: nil, B: nil}, output: "C0"}, + {val: &multipleOptionalFields{A: &[3]byte{1, 2, 3}, B: &[3]byte{1, 2, 3}}, output: "C88301020383010203"}, + {val: &multipleOptionalFields{A: nil, B: &[3]byte{1, 2, 3}}, output: "C58083010203"}, // encodes without error but decode will fail + {val: &nonOptionalPtrField{A: 1}, output: "C20180"}, // encodes without error but decode will fail // nil {val: (*uint)(nil), output: "80"}, @@ -225,26 +332,73 @@ var encTests = []encTest{ {val: (*[]byte)(nil), output: "80"}, {val: (*[10]byte)(nil), output: "80"}, {val: (*big.Int)(nil), output: "80"}, + {val: (*uint256.Int)(nil), output: "80"}, {val: (*[]string)(nil), output: "C0"}, {val: (*[10]string)(nil), output: "C0"}, {val: (*[]interface{})(nil), output: "C0"}, {val: (*[]struct{ uint })(nil), output: "C0"}, {val: (*interface{})(nil), output: "C0"}, + // nil struct fields + { + val: struct { + X *[]byte + }{}, + output: "C180", + }, + { + val: struct { + X *[2]byte + }{}, + output: "C180", + }, + { + val: struct { + X *uint64 + }{}, + output: "C180", + }, + { + val: struct { + X *uint64 `rlp:"nilList"` + }{}, + output: "C1C0", + }, + { + val: struct { + X *[]uint64 + }{}, + output: "C1C0", + }, + { + val: struct { + X *[]uint64 `rlp:"nilString"` + }{}, + output: "C180", + }, + // interfaces {val: []io.Reader{reader}, output: "C3C20102"}, // the contained value is a struct // Encoder - {val: (*testEncoder)(nil), output: "00000000"}, + {val: (*testEncoder)(nil), output: "C0"}, {val: &testEncoder{}, output: "00010001000100010001"}, {val: &testEncoder{errors.New("test error")}, error: "test error"}, - // verify that pointer method testEncoder.EncodeRLP is called for + {val: struct{ E testEncoderValueMethod }{}, output: "C3FAFEF0"}, + {val: struct{ E *testEncoderValueMethod }{}, output: "C1C0"}, + + // Verify that the Encoder interface works for unsupported types like func(). + {val: undecodableEncoder(func() {}), output: "F5F5F5"}, + + // Verify that pointer method testEncoder.EncodeRLP is called for // addressable non-pointer values. {val: &struct{ TE testEncoder }{testEncoder{}}, output: "CA00010001000100010001"}, {val: &struct{ TE testEncoder }{testEncoder{errors.New("test error")}}, error: "test error"}, - // verify the error for non-addressable non-pointer Encoder - {val: testEncoder{}, error: "rlp: game over: unadressable value of type rlp.testEncoder, EncodeRLP is pointer method"}, - // verify the special case for []byte + + // Verify the error for non-addressable non-pointer Encoder. + {val: testEncoder{}, error: "rlp: unaddressable value of type rlp.testEncoder, EncodeRLP is pointer method"}, + + // Verify Encoder takes precedence over []byte. {val: []byteEncoder{0, 1, 2, 3, 4}, output: "C5C0C0C0C0C0"}, } @@ -280,6 +434,21 @@ func TestEncodeToBytes(t *testing.T) { runEncTests(t, EncodeToBytes) } +func TestEncodeAppendToBytes(t *testing.T) { + buffer := make([]byte, 20) + runEncTests(t, func(val interface{}) ([]byte, error) { + w := NewEncoderBuffer(nil) + defer w.Flush() + + err := Encode(w, val) + if err != nil { + return nil, err + } + output := w.AppendToBytes(buffer[:0]) + return output, nil + }) +} + func TestEncodeToReader(t *testing.T) { runEncTests(t, func(val interface{}) ([]byte, error) { _, r, err := EncodeToReader(val) @@ -338,3 +507,132 @@ func TestEncodeToReaderReturnToPool(t *testing.T) { } wg.Wait() } + +var sink interface{} + +func BenchmarkIntsize(b *testing.B) { + for i := 0; i < b.N; i++ { + sink = intsize(0x12345678) + } +} + +func BenchmarkPutint(b *testing.B) { + buf := make([]byte, 8) + for i := 0; i < b.N; i++ { + putint(buf, 0x12345678) + sink = buf + } +} + +func BenchmarkEncodeBigInts(b *testing.B) { + ints := make([]*big.Int, 200) + for i := range ints { + ints[i] = math.BigPow(2, int64(i)) + } + out := bytes.NewBuffer(make([]byte, 0, 4096)) + b.ResetTimer() + b.ReportAllocs() + + for i := 0; i < b.N; i++ { + out.Reset() + if err := Encode(out, ints); err != nil { + b.Fatal(err) + } + } +} + +func BenchmarkEncodeU256Ints(b *testing.B) { + ints := make([]*uint256.Int, 200) + for i := range ints { + ints[i], _ = uint256.FromBig(math.BigPow(2, int64(i))) + } + out := bytes.NewBuffer(make([]byte, 0, 4096)) + b.ResetTimer() + b.ReportAllocs() + + for i := 0; i < b.N; i++ { + out.Reset() + if err := Encode(out, ints); err != nil { + b.Fatal(err) + } + } +} + +func BenchmarkEncodeConcurrentInterface(b *testing.B) { + type struct1 struct { + A string + B *big.Int + C [20]byte + } + value := []interface{}{ + uint(999), + &struct1{A: "hello", B: big.NewInt(0xFFFFFFFF)}, + [10]byte{1, 2, 3, 4, 5, 6}, + []string{"yeah", "yeah", "yeah"}, + } + + var wg sync.WaitGroup + for cpu := 0; cpu < runtime.NumCPU(); cpu++ { + wg.Add(1) + go func() { + defer wg.Done() + + var buffer bytes.Buffer + for i := 0; i < b.N; i++ { + buffer.Reset() + err := Encode(&buffer, value) + if err != nil { + panic(err) + } + } + }() + } + wg.Wait() +} + +type byteArrayStruct struct { + A [20]byte + B [32]byte + C [32]byte +} + +func BenchmarkEncodeByteArrayStruct(b *testing.B) { + var out bytes.Buffer + var value byteArrayStruct + + b.ReportAllocs() + for i := 0; i < b.N; i++ { + out.Reset() + if err := Encode(&out, &value); err != nil { + b.Fatal(err) + } + } +} + +type structSliceElem struct { + X uint64 + Y uint64 + Z uint64 +} + +type structPtrSlice []*structSliceElem + +func BenchmarkEncodeStructPtrSlice(b *testing.B) { + var out bytes.Buffer + var value = structPtrSlice{ + &structSliceElem{1, 1, 1}, + &structSliceElem{2, 2, 2}, + &structSliceElem{3, 3, 3}, + &structSliceElem{5, 5, 5}, + &structSliceElem{6, 6, 6}, + &structSliceElem{7, 7, 7}, + } + + b.ReportAllocs() + for i := 0; i < b.N; i++ { + out.Reset() + if err := Encode(&out, &value); err != nil { + b.Fatal(err) + } + } +} diff --git a/rlp/encoder_example_test.go b/rlp/encoder_example_test.go index 1cffa241c259..da0465a53312 100644 --- a/rlp/encoder_example_test.go +++ b/rlp/encoder_example_test.go @@ -14,11 +14,13 @@ // You should have received a copy of the GNU Lesser General Public License // along with the go-ethereum library. If not, see . -package rlp +package rlp_test import ( "fmt" "io" + + "github.com/XinFinOrg/XDPoSChain/rlp" ) type MyCoolType struct { @@ -28,27 +30,19 @@ type MyCoolType struct { // EncodeRLP writes x as RLP list [a, b] that omits the Name field. func (x *MyCoolType) EncodeRLP(w io.Writer) (err error) { - // Note: the receiver can be a nil pointer. This allows you to - // control the encoding of nil, but it also means that you have to - // check for a nil receiver. - if x == nil { - err = Encode(w, []uint{0, 0}) - } else { - err = Encode(w, []uint{x.a, x.b}) - } - return err + return rlp.Encode(w, []uint{x.a, x.b}) } func ExampleEncoder() { var t *MyCoolType // t is nil pointer to MyCoolType - bytes, _ := EncodeToBytes(t) + bytes, _ := rlp.EncodeToBytes(t) fmt.Printf("%v → %X\n", t, bytes) t = &MyCoolType{Name: "foobar", a: 5, b: 6} - bytes, _ = EncodeToBytes(t) + bytes, _ = rlp.EncodeToBytes(t) fmt.Printf("%v → %X\n", t, bytes) // Output: - // → C28080 + // → C0 // &{foobar 5 6} → C20506 } diff --git a/rlp/internal/rlpstruct/rlpstruct.go b/rlp/internal/rlpstruct/rlpstruct.go new file mode 100644 index 000000000000..2e3eeb688193 --- /dev/null +++ b/rlp/internal/rlpstruct/rlpstruct.go @@ -0,0 +1,213 @@ +// Copyright 2022 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +// Package rlpstruct implements struct processing for RLP encoding/decoding. +// +// In particular, this package handles all rules around field filtering, +// struct tags and nil value determination. +package rlpstruct + +import ( + "fmt" + "reflect" + "strings" +) + +// Field represents a struct field. +type Field struct { + Name string + Index int + Exported bool + Type Type + Tag string +} + +// Type represents the attributes of a Go type. +type Type struct { + Name string + Kind reflect.Kind + IsEncoder bool // whether type implements rlp.Encoder + IsDecoder bool // whether type implements rlp.Decoder + Elem *Type // non-nil for Kind values of Ptr, Slice, Array +} + +// DefaultNilValue determines whether a nil pointer to t encodes/decodes +// as an empty string or empty list. +func (t Type) DefaultNilValue() NilKind { + k := t.Kind + if isUint(k) || k == reflect.String || k == reflect.Bool || isByteArray(t) { + return NilKindString + } + return NilKindList +} + +// NilKind is the RLP value encoded in place of nil pointers. +type NilKind uint8 + +const ( + NilKindString NilKind = 0x80 + NilKindList NilKind = 0xC0 +) + +// Tags represents struct tags. +type Tags struct { + // rlp:"nil" controls whether empty input results in a nil pointer. + // nilKind is the kind of empty value allowed for the field. + NilKind NilKind + NilOK bool + + // rlp:"optional" allows for a field to be missing in the input list. + // If this is set, all subsequent fields must also be optional. + Optional bool + + // rlp:"tail" controls whether this field swallows additional list elements. It can + // only be set for the last field, which must be of slice type. + Tail bool + + // rlp:"-" ignores fields. + Ignored bool +} + +// TagError is raised for invalid struct tags. +type TagError struct { + StructType string + + // These are set by this package. + Field string + Tag string + Err string +} + +func (e TagError) Error() string { + field := "field " + e.Field + if e.StructType != "" { + field = e.StructType + "." + e.Field + } + return fmt.Sprintf("rlp: invalid struct tag %q for %s (%s)", e.Tag, field, e.Err) +} + +// ProcessFields filters the given struct fields, returning only fields +// that should be considered for encoding/decoding. +func ProcessFields(allFields []Field) ([]Field, []Tags, error) { + lastPublic := lastPublicField(allFields) + + // Gather all exported fields and their tags. + var fields []Field + var tags []Tags + for _, field := range allFields { + if !field.Exported { + continue + } + ts, err := parseTag(field, lastPublic) + if err != nil { + return nil, nil, err + } + if ts.Ignored { + continue + } + fields = append(fields, field) + tags = append(tags, ts) + } + + // Verify optional field consistency. If any optional field exists, + // all fields after it must also be optional. Note: optional + tail + // is supported. + var anyOptional bool + var firstOptionalName string + for i, ts := range tags { + name := fields[i].Name + if ts.Optional || ts.Tail { + if !anyOptional { + firstOptionalName = name + } + anyOptional = true + } else { + if anyOptional { + msg := fmt.Sprintf("must be optional because preceding field %q is optional", firstOptionalName) + return nil, nil, TagError{Field: name, Err: msg} + } + } + } + return fields, tags, nil +} + +func parseTag(field Field, lastPublic int) (Tags, error) { + name := field.Name + tag := reflect.StructTag(field.Tag) + var ts Tags + for _, t := range strings.Split(tag.Get("rlp"), ",") { + switch t = strings.TrimSpace(t); t { + case "": + // empty tag is allowed for some reason + case "-": + ts.Ignored = true + case "nil", "nilString", "nilList": + ts.NilOK = true + if field.Type.Kind != reflect.Ptr { + return ts, TagError{Field: name, Tag: t, Err: "field is not a pointer"} + } + switch t { + case "nil": + ts.NilKind = field.Type.Elem.DefaultNilValue() + case "nilString": + ts.NilKind = NilKindString + case "nilList": + ts.NilKind = NilKindList + } + case "optional": + ts.Optional = true + if ts.Tail { + return ts, TagError{Field: name, Tag: t, Err: `also has "tail" tag`} + } + case "tail": + ts.Tail = true + if field.Index != lastPublic { + return ts, TagError{Field: name, Tag: t, Err: "must be on last field"} + } + if ts.Optional { + return ts, TagError{Field: name, Tag: t, Err: `also has "optional" tag`} + } + if field.Type.Kind != reflect.Slice { + return ts, TagError{Field: name, Tag: t, Err: "field type is not slice"} + } + default: + return ts, TagError{Field: name, Tag: t, Err: "unknown tag"} + } + } + return ts, nil +} + +func lastPublicField(fields []Field) int { + last := 0 + for _, f := range fields { + if f.Exported { + last = f.Index + } + } + return last +} + +func isUint(k reflect.Kind) bool { + return k >= reflect.Uint && k <= reflect.Uintptr +} + +func isByte(typ Type) bool { + return typ.Kind == reflect.Uint8 && !typ.IsEncoder +} + +func isByteArray(typ Type) bool { + return (typ.Kind == reflect.Slice || typ.Kind == reflect.Array) && isByte(*typ.Elem) +} diff --git a/rlp/iterator.go b/rlp/iterator.go new file mode 100644 index 000000000000..95bd3f258208 --- /dev/null +++ b/rlp/iterator.go @@ -0,0 +1,59 @@ +// Copyright 2020 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +package rlp + +type listIterator struct { + data []byte + next []byte + err error +} + +// NewListIterator creates an iterator for the (list) represented by data +func NewListIterator(data RawValue) (*listIterator, error) { + k, t, c, err := readKind(data) + if err != nil { + return nil, err + } + if k != List { + return nil, ErrExpectedList + } + it := &listIterator{ + data: data[t : t+c], + } + return it, nil +} + +// Next forwards the iterator one step, returns true if it was not at end yet +func (it *listIterator) Next() bool { + if len(it.data) == 0 { + return false + } + _, t, c, err := readKind(it.data) + it.next = it.data[:t+c] + it.data = it.data[t+c:] + it.err = err + return true +} + +// Value returns the current value +func (it *listIterator) Value() []byte { + return it.next +} + +func (it *listIterator) Err() error { + return it.err +} diff --git a/rlp/iterator_test.go b/rlp/iterator_test.go new file mode 100644 index 000000000000..82ac7bfa6eb7 --- /dev/null +++ b/rlp/iterator_test.go @@ -0,0 +1,59 @@ +// Copyright 2020 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +package rlp + +import ( + "testing" + + "github.com/XinFinOrg/XDPoSChain/common/hexutil" +) + +// TestIterator tests some basic things about the ListIterator. A more +// comprehensive test can be found in core/rlp_test.go, where we can +// use both types and rlp without dependency cycles +func TestIterator(t *testing.T) { + bodyRlpHex := "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" + bodyRlp := hexutil.MustDecode(bodyRlpHex) + + it, err := NewListIterator(bodyRlp) + if err != nil { + t.Fatal(err) + } + // Check that txs exist + if !it.Next() { + t.Fatal("expected two elems, got zero") + } + txs := it.Value() + // Check that uncles exist + if !it.Next() { + t.Fatal("expected two elems, got one") + } + txit, err := NewListIterator(txs) + if err != nil { + t.Fatal(err) + } + var i = 0 + for txit.Next() { + if txit.err != nil { + t.Fatal(txit.err) + } + i++ + } + if exp := 2; i != exp { + t.Errorf("count wrong, expected %d got %d", i, exp) + } +} diff --git a/rlp/raw.go b/rlp/raw.go index 74f2cb86328f..773aa7e614e8 100644 --- a/rlp/raw.go +++ b/rlp/raw.go @@ -28,12 +28,53 @@ type RawValue []byte var rawValueType = reflect.TypeOf(RawValue{}) +// StringSize returns the encoded size of a string. +func StringSize(s string) uint64 { + switch { + case len(s) == 0: + return 1 + case len(s) == 1: + if s[0] <= 0x7f { + return 1 + } else { + return 2 + } + default: + return uint64(headsize(uint64(len(s))) + len(s)) + } +} + +// BytesSize returns the encoded size of a byte slice. +func BytesSize(b []byte) uint64 { + switch { + case len(b) == 0: + return 1 + case len(b) == 1: + if b[0] <= 0x7f { + return 1 + } else { + return 2 + } + default: + return uint64(headsize(uint64(len(b))) + len(b)) + } +} + // ListSize returns the encoded size of an RLP list with the given // content size. func ListSize(contentSize uint64) uint64 { return uint64(headsize(contentSize)) + contentSize } +// IntSize returns the encoded size of the integer x. Note: The return type of this +// function is 'int' for backwards-compatibility reasons. The result is always positive. +func IntSize(x uint64) int { + if x < 0x80 { + return 1 + } + return 1 + intsize(x) +} + // Split returns the content of first RLP value and any // bytes after the value as subslices of b. func Split(b []byte) (k Kind, content, rest []byte, err error) { @@ -57,6 +98,32 @@ func SplitString(b []byte) (content, rest []byte, err error) { return content, rest, nil } +// SplitUint64 decodes an integer at the beginning of b. +// It also returns the remaining data after the integer in 'rest'. +func SplitUint64(b []byte) (x uint64, rest []byte, err error) { + content, rest, err := SplitString(b) + if err != nil { + return 0, b, err + } + switch { + case len(content) == 0: + return 0, rest, nil + case len(content) == 1: + if content[0] == 0 { + return 0, b, ErrCanonInt + } + return uint64(content[0]), rest, nil + case len(content) > 8: + return 0, b, errUintOverflow + default: + x, err = readSize(content, byte(len(content))) + if err != nil { + return 0, b, ErrCanonInt + } + return x, rest, nil + } +} + // SplitList splits b into the content of a list and any remaining // bytes after the list. func SplitList(b []byte) (content, rest []byte, err error) { diff --git a/rlp/raw_test.go b/rlp/raw_test.go index 2aad042100ea..7b3255eca36b 100644 --- a/rlp/raw_test.go +++ b/rlp/raw_test.go @@ -18,9 +18,10 @@ package rlp import ( "bytes" + "errors" "io" - "reflect" "testing" + "testing/quick" ) func TestCountValues(t *testing.T) { @@ -53,21 +54,84 @@ func TestCountValues(t *testing.T) { if count != test.count { t.Errorf("test %d: count mismatch, got %d want %d\ninput: %s", i, count, test.count, test.input) } - if !reflect.DeepEqual(err, test.err) { + if !errors.Is(err, test.err) { t.Errorf("test %d: err mismatch, got %q want %q\ninput: %s", i, err, test.err, test.input) } } } -func TestSplitTypes(t *testing.T) { - if _, _, err := SplitString(unhex("C100")); err != ErrExpectedString { - t.Errorf("SplitString returned %q, want %q", err, ErrExpectedString) +func TestSplitString(t *testing.T) { + for i, test := range []string{ + "C0", + "C100", + "C3010203", + "C88363617483646F67", + "F8384C6F72656D20697073756D20646F6C6F722073697420616D65742C20636F6E7365637465747572206164697069736963696E6720656C6974", + } { + if _, _, err := SplitString(unhex(test)); !errors.Is(err, ErrExpectedString) { + t.Errorf("test %d: error mismatch: have %q, want %q", i, err, ErrExpectedString) + } + } +} + +func TestSplitList(t *testing.T) { + for i, test := range []string{ + "80", + "00", + "01", + "8180", + "81FF", + "820400", + "83636174", + "83646F67", + "B8384C6F72656D20697073756D20646F6C6F722073697420616D65742C20636F6E7365637465747572206164697069736963696E6720656C6974", + } { + if _, _, err := SplitList(unhex(test)); !errors.Is(err, ErrExpectedList) { + t.Errorf("test %d: error mismatch: have %q, want %q", i, err, ErrExpectedList) + } } - if _, _, err := SplitList(unhex("01")); err != ErrExpectedList { - t.Errorf("SplitString returned %q, want %q", err, ErrExpectedList) +} + +func TestSplitUint64(t *testing.T) { + tests := []struct { + input string + val uint64 + rest string + err error + }{ + {"01", 1, "", nil}, + {"7FFF", 0x7F, "FF", nil}, + {"80FF", 0, "FF", nil}, + {"81FAFF", 0xFA, "FF", nil}, + {"82FAFAFF", 0xFAFA, "FF", nil}, + {"83FAFAFAFF", 0xFAFAFA, "FF", nil}, + {"84FAFAFAFAFF", 0xFAFAFAFA, "FF", nil}, + {"85FAFAFAFAFAFF", 0xFAFAFAFAFA, "FF", nil}, + {"86FAFAFAFAFAFAFF", 0xFAFAFAFAFAFA, "FF", nil}, + {"87FAFAFAFAFAFAFAFF", 0xFAFAFAFAFAFAFA, "FF", nil}, + {"88FAFAFAFAFAFAFAFAFF", 0xFAFAFAFAFAFAFAFA, "FF", nil}, + + // errors + {"", 0, "", io.ErrUnexpectedEOF}, + {"00", 0, "00", ErrCanonInt}, + {"81", 0, "81", ErrValueTooLarge}, + {"8100", 0, "8100", ErrCanonSize}, + {"8200FF", 0, "8200FF", ErrCanonInt}, + {"8103FF", 0, "8103FF", ErrCanonSize}, + {"89FAFAFAFAFAFAFAFAFAFF", 0, "89FAFAFAFAFAFAFAFAFAFF", errUintOverflow}, } - if _, _, err := SplitList(unhex("81FF")); err != ErrExpectedList { - t.Errorf("SplitString returned %q, want %q", err, ErrExpectedList) + + for i, test := range tests { + val, rest, err := SplitUint64(unhex(test.input)) + if val != test.val { + t.Errorf("test %d: val mismatch: got %x, want %x (input %q)", i, val, test.val, test.input) + } + if !bytes.Equal(rest, unhex(test.rest)) { + t.Errorf("test %d: rest mismatch: got %x, want %s (input %q)", i, rest, test.rest, test.input) + } + if err != test.err { + t.Errorf("test %d: error mismatch: got %q, want %q", i, err, test.err) + } } } @@ -78,7 +142,9 @@ func TestSplit(t *testing.T) { val, rest string err error }{ + {input: "00FFFF", kind: Byte, val: "00", rest: "FFFF"}, {input: "01FFFF", kind: Byte, val: "01", rest: "FFFF"}, + {input: "7FFFFF", kind: Byte, val: "7F", rest: "FFFF"}, {input: "80FFFF", kind: String, val: "", rest: "FFFF"}, {input: "C3010203", kind: List, val: "010203"}, @@ -194,3 +260,79 @@ func TestReadSize(t *testing.T) { } } } + +func TestAppendUint64(t *testing.T) { + tests := []struct { + input uint64 + slice []byte + output string + }{ + {0, nil, "80"}, + {1, nil, "01"}, + {2, nil, "02"}, + {127, nil, "7F"}, + {128, nil, "8180"}, + {129, nil, "8181"}, + {0xFFFFFF, nil, "83FFFFFF"}, + {127, []byte{1, 2, 3}, "0102037F"}, + {0xFFFFFF, []byte{1, 2, 3}, "01020383FFFFFF"}, + } + + for _, test := range tests { + x := AppendUint64(test.slice, test.input) + if !bytes.Equal(x, unhex(test.output)) { + t.Errorf("AppendUint64(%v, %d): got %x, want %s", test.slice, test.input, x, test.output) + } + + // Check that IntSize returns the appended size. + length := len(x) - len(test.slice) + if s := IntSize(test.input); s != length { + t.Errorf("IntSize(%d): got %d, want %d", test.input, s, length) + } + } +} + +func TestAppendUint64Random(t *testing.T) { + fn := func(i uint64) bool { + enc, _ := EncodeToBytes(i) + encAppend := AppendUint64(nil, i) + return bytes.Equal(enc, encAppend) + } + config := quick.Config{MaxCountScale: 50} + if err := quick.Check(fn, &config); err != nil { + t.Fatal(err) + } +} + +func TestBytesSize(t *testing.T) { + tests := []struct { + v []byte + size uint64 + }{ + {v: []byte{}, size: 1}, + {v: []byte{0x1}, size: 1}, + {v: []byte{0x7E}, size: 1}, + {v: []byte{0x7F}, size: 1}, + {v: []byte{0x80}, size: 2}, + {v: []byte{0xFF}, size: 2}, + {v: []byte{0xFF, 0xF0}, size: 3}, + {v: make([]byte, 55), size: 56}, + {v: make([]byte, 56), size: 58}, + } + + for _, test := range tests { + s := BytesSize(test.v) + if s != test.size { + t.Errorf("BytesSize(%#x) -> %d, want %d", test.v, s, test.size) + } + s = StringSize(string(test.v)) + if s != test.size { + t.Errorf("StringSize(%#x) -> %d, want %d", test.v, s, test.size) + } + // Sanity check: + enc, _ := EncodeToBytes(test.v) + if uint64(len(enc)) != test.size { + t.Errorf("len(EncodeToBytes(%#x)) -> %d, test says %d", test.v, len(enc), test.size) + } + } +} diff --git a/rlp/rlpgen/gen.go b/rlp/rlpgen/gen.go new file mode 100644 index 000000000000..ed502c09a7e3 --- /dev/null +++ b/rlp/rlpgen/gen.go @@ -0,0 +1,800 @@ +// Copyright 2022 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +package main + +import ( + "bytes" + "fmt" + "go/format" + "go/types" + "sort" + + "github.com/XinFinOrg/XDPoSChain/rlp/internal/rlpstruct" +) + +// buildContext keeps the data needed for make*Op. +type buildContext struct { + topType *types.Named // the type we're creating methods for + + encoderIface *types.Interface + decoderIface *types.Interface + rawValueType *types.Named + + typeToStructCache map[types.Type]*rlpstruct.Type +} + +func newBuildContext(packageRLP *types.Package) *buildContext { + enc := packageRLP.Scope().Lookup("Encoder").Type().Underlying() + dec := packageRLP.Scope().Lookup("Decoder").Type().Underlying() + rawv := packageRLP.Scope().Lookup("RawValue").Type() + return &buildContext{ + typeToStructCache: make(map[types.Type]*rlpstruct.Type), + encoderIface: enc.(*types.Interface), + decoderIface: dec.(*types.Interface), + rawValueType: rawv.(*types.Named), + } +} + +func (bctx *buildContext) isEncoder(typ types.Type) bool { + return types.Implements(typ, bctx.encoderIface) +} + +func (bctx *buildContext) isDecoder(typ types.Type) bool { + return types.Implements(typ, bctx.decoderIface) +} + +// typeToStructType converts typ to rlpstruct.Type. +func (bctx *buildContext) typeToStructType(typ types.Type) *rlpstruct.Type { + if prev := bctx.typeToStructCache[typ]; prev != nil { + return prev // short-circuit for recursive types. + } + + // Resolve named types to their underlying type, but keep the name. + name := types.TypeString(typ, nil) + for { + utype := typ.Underlying() + if utype == typ { + break + } + typ = utype + } + + // Create the type and store it in cache. + t := &rlpstruct.Type{ + Name: name, + Kind: typeReflectKind(typ), + IsEncoder: bctx.isEncoder(typ), + IsDecoder: bctx.isDecoder(typ), + } + bctx.typeToStructCache[typ] = t + + // Assign element type. + switch typ.(type) { + case *types.Array, *types.Slice, *types.Pointer: + etype := typ.(interface{ Elem() types.Type }).Elem() + t.Elem = bctx.typeToStructType(etype) + } + return t +} + +// genContext is passed to the gen* methods of op when generating +// the output code. It tracks packages to be imported by the output +// file and assigns unique names of temporary variables. +type genContext struct { + inPackage *types.Package + imports map[string]struct{} + tempCounter int +} + +func newGenContext(inPackage *types.Package) *genContext { + return &genContext{ + inPackage: inPackage, + imports: make(map[string]struct{}), + } +} + +func (ctx *genContext) temp() string { + v := fmt.Sprintf("_tmp%d", ctx.tempCounter) + ctx.tempCounter++ + return v +} + +func (ctx *genContext) resetTemp() { + ctx.tempCounter = 0 +} + +func (ctx *genContext) addImport(path string) { + if path == ctx.inPackage.Path() { + return // avoid importing the package that we're generating in. + } + // TODO: renaming? + ctx.imports[path] = struct{}{} +} + +// importsList returns all packages that need to be imported. +func (ctx *genContext) importsList() []string { + imp := make([]string, 0, len(ctx.imports)) + for k := range ctx.imports { + imp = append(imp, k) + } + sort.Strings(imp) + return imp +} + +// qualify is the types.Qualifier used for printing types. +func (ctx *genContext) qualify(pkg *types.Package) string { + if pkg.Path() == ctx.inPackage.Path() { + return "" + } + ctx.addImport(pkg.Path()) + // TODO: renaming? + return pkg.Name() +} + +type op interface { + // genWrite creates the encoder. The generated code should write v, + // which is any Go expression, to the rlp.EncoderBuffer 'w'. + genWrite(ctx *genContext, v string) string + + // genDecode creates the decoder. The generated code should read + // a value from the rlp.Stream 'dec' and store it to dst. + genDecode(ctx *genContext) (string, string) +} + +// basicOp handles basic types bool, uint*, string. +type basicOp struct { + typ types.Type + writeMethod string // EncoderBuffer writer method name + writeArgType types.Type // parameter type of writeMethod + decMethod string + decResultType types.Type // return type of decMethod + decUseBitSize bool // if true, result bit size is appended to decMethod +} + +func (*buildContext) makeBasicOp(typ *types.Basic) (op, error) { + op := basicOp{typ: typ} + kind := typ.Kind() + switch { + case kind == types.Bool: + op.writeMethod = "WriteBool" + op.writeArgType = types.Typ[types.Bool] + op.decMethod = "Bool" + op.decResultType = types.Typ[types.Bool] + case kind >= types.Uint8 && kind <= types.Uint64: + op.writeMethod = "WriteUint64" + op.writeArgType = types.Typ[types.Uint64] + op.decMethod = "Uint" + op.decResultType = typ + op.decUseBitSize = true + case kind == types.String: + op.writeMethod = "WriteString" + op.writeArgType = types.Typ[types.String] + op.decMethod = "String" + op.decResultType = types.Typ[types.String] + default: + return nil, fmt.Errorf("unhandled basic type: %v", typ) + } + return op, nil +} + +func (*buildContext) makeByteSliceOp(typ *types.Slice) op { + if !isByte(typ.Elem()) { + panic("non-byte slice type in makeByteSliceOp") + } + bslice := types.NewSlice(types.Typ[types.Uint8]) + return basicOp{ + typ: typ, + writeMethod: "WriteBytes", + writeArgType: bslice, + decMethod: "Bytes", + decResultType: bslice, + } +} + +func (bctx *buildContext) makeRawValueOp() op { + bslice := types.NewSlice(types.Typ[types.Uint8]) + return basicOp{ + typ: bctx.rawValueType, + writeMethod: "Write", + writeArgType: bslice, + decMethod: "Raw", + decResultType: bslice, + } +} + +func (op basicOp) writeNeedsConversion() bool { + return !types.AssignableTo(op.typ, op.writeArgType) +} + +func (op basicOp) decodeNeedsConversion() bool { + return !types.AssignableTo(op.decResultType, op.typ) +} + +func (op basicOp) genWrite(ctx *genContext, v string) string { + if op.writeNeedsConversion() { + v = fmt.Sprintf("%s(%s)", op.writeArgType, v) + } + return fmt.Sprintf("w.%s(%s)\n", op.writeMethod, v) +} + +func (op basicOp) genDecode(ctx *genContext) (string, string) { + var ( + resultV = ctx.temp() + result = resultV + method = op.decMethod + ) + if op.decUseBitSize { + // Note: For now, this only works for platform-independent integer + // sizes. makeBasicOp forbids the platform-dependent types. + var sizes types.StdSizes + method = fmt.Sprintf("%s%d", op.decMethod, sizes.Sizeof(op.typ)*8) + } + + // Call the decoder method. + var b bytes.Buffer + fmt.Fprintf(&b, "%s, err := dec.%s()\n", resultV, method) + fmt.Fprintf(&b, "if err != nil { return err }\n") + if op.decodeNeedsConversion() { + conv := ctx.temp() + fmt.Fprintf(&b, "%s := %s(%s)\n", conv, types.TypeString(op.typ, ctx.qualify), resultV) + result = conv + } + return result, b.String() +} + +// byteArrayOp handles [...]byte. +type byteArrayOp struct { + typ types.Type + name types.Type // name != typ for named byte array types (e.g. common.Address) +} + +func (bctx *buildContext) makeByteArrayOp(name *types.Named, typ *types.Array) byteArrayOp { + nt := types.Type(name) + if name == nil { + nt = typ + } + return byteArrayOp{typ, nt} +} + +func (op byteArrayOp) genWrite(ctx *genContext, v string) string { + return fmt.Sprintf("w.WriteBytes(%s[:])\n", v) +} + +func (op byteArrayOp) genDecode(ctx *genContext) (string, string) { + var resultV = ctx.temp() + + var b bytes.Buffer + fmt.Fprintf(&b, "var %s %s\n", resultV, types.TypeString(op.name, ctx.qualify)) + fmt.Fprintf(&b, "if err := dec.ReadBytes(%s[:]); err != nil { return err }\n", resultV) + return resultV, b.String() +} + +// bigIntOp handles big.Int. +// This exists because big.Int has it's own decoder operation on rlp.Stream, +// but the decode method returns *big.Int, so it needs to be dereferenced. +type bigIntOp struct { + pointer bool +} + +func (op bigIntOp) genWrite(ctx *genContext, v string) string { + var b bytes.Buffer + + fmt.Fprintf(&b, "if %s.Sign() == -1 {\n", v) + fmt.Fprintf(&b, " return rlp.ErrNegativeBigInt\n") + fmt.Fprintf(&b, "}\n") + dst := v + if !op.pointer { + dst = "&" + v + } + fmt.Fprintf(&b, "w.WriteBigInt(%s)\n", dst) + + // Wrap with nil check. + if op.pointer { + code := b.String() + b.Reset() + fmt.Fprintf(&b, "if %s == nil {\n", v) + fmt.Fprintf(&b, " w.Write(rlp.EmptyString)") + fmt.Fprintf(&b, "} else {\n") + fmt.Fprint(&b, code) + fmt.Fprintf(&b, "}\n") + } + + return b.String() +} + +func (op bigIntOp) genDecode(ctx *genContext) (string, string) { + var resultV = ctx.temp() + + var b bytes.Buffer + fmt.Fprintf(&b, "%s, err := dec.BigInt()\n", resultV) + fmt.Fprintf(&b, "if err != nil { return err }\n") + + result := resultV + if !op.pointer { + result = "(*" + resultV + ")" + } + return result, b.String() +} + +// uint256Op handles "github.com/holiman/uint256".Int +type uint256Op struct { + pointer bool +} + +func (op uint256Op) genWrite(ctx *genContext, v string) string { + var b bytes.Buffer + + dst := v + if !op.pointer { + dst = "&" + v + } + fmt.Fprintf(&b, "w.WriteUint256(%s)\n", dst) + + // Wrap with nil check. + if op.pointer { + code := b.String() + b.Reset() + fmt.Fprintf(&b, "if %s == nil {\n", v) + fmt.Fprintf(&b, " w.Write(rlp.EmptyString)") + fmt.Fprintf(&b, "} else {\n") + fmt.Fprint(&b, code) + fmt.Fprintf(&b, "}\n") + } + + return b.String() +} + +func (op uint256Op) genDecode(ctx *genContext) (string, string) { + ctx.addImport("github.com/holiman/uint256") + + var b bytes.Buffer + resultV := ctx.temp() + fmt.Fprintf(&b, "var %s uint256.Int\n", resultV) + fmt.Fprintf(&b, "if err := dec.ReadUint256(&%s); err != nil { return err }\n", resultV) + + result := resultV + if op.pointer { + result = "&" + resultV + } + return result, b.String() +} + +// encoderDecoderOp handles rlp.Encoder and rlp.Decoder. +// In order to be used with this, the type must implement both interfaces. +// This restriction may be lifted in the future by creating separate ops for +// encoding and decoding. +type encoderDecoderOp struct { + typ types.Type +} + +func (op encoderDecoderOp) genWrite(ctx *genContext, v string) string { + return fmt.Sprintf("if err := %s.EncodeRLP(w); err != nil { return err }\n", v) +} + +func (op encoderDecoderOp) genDecode(ctx *genContext) (string, string) { + // DecodeRLP must have pointer receiver, and this is verified in makeOp. + etyp := op.typ.(*types.Pointer).Elem() + var resultV = ctx.temp() + + var b bytes.Buffer + fmt.Fprintf(&b, "%s := new(%s)\n", resultV, types.TypeString(etyp, ctx.qualify)) + fmt.Fprintf(&b, "if err := %s.DecodeRLP(dec); err != nil { return err }\n", resultV) + return resultV, b.String() +} + +// ptrOp handles pointer types. +type ptrOp struct { + elemTyp types.Type + elem op + nilOK bool + nilValue rlpstruct.NilKind +} + +func (bctx *buildContext) makePtrOp(elemTyp types.Type, tags rlpstruct.Tags) (op, error) { + elemOp, err := bctx.makeOp(nil, elemTyp, rlpstruct.Tags{}) + if err != nil { + return nil, err + } + op := ptrOp{elemTyp: elemTyp, elem: elemOp} + + // Determine nil value. + if tags.NilOK { + op.nilOK = true + op.nilValue = tags.NilKind + } else { + styp := bctx.typeToStructType(elemTyp) + op.nilValue = styp.DefaultNilValue() + } + return op, nil +} + +func (op ptrOp) genWrite(ctx *genContext, v string) string { + // Note: in writer functions, accesses to v are read-only, i.e. v is any Go + // expression. To make all accesses work through the pointer, we substitute + // v with (*v). This is required for most accesses including `v`, `call(v)`, + // and `v[index]` on slices. + // + // For `v.field` and `v[:]` on arrays, the dereference operation is not required. + var vv string + _, isStruct := op.elem.(structOp) + _, isByteArray := op.elem.(byteArrayOp) + if isStruct || isByteArray { + vv = v + } else { + vv = fmt.Sprintf("(*%s)", v) + } + + var b bytes.Buffer + fmt.Fprintf(&b, "if %s == nil {\n", v) + fmt.Fprintf(&b, " w.Write([]byte{0x%X})\n", op.nilValue) + fmt.Fprintf(&b, "} else {\n") + fmt.Fprintf(&b, " %s", op.elem.genWrite(ctx, vv)) + fmt.Fprintf(&b, "}\n") + return b.String() +} + +func (op ptrOp) genDecode(ctx *genContext) (string, string) { + result, code := op.elem.genDecode(ctx) + if !op.nilOK { + // If nil pointers are not allowed, we can just decode the element. + return "&" + result, code + } + + // nil is allowed, so check the kind and size first. + // If size is zero and kind matches the nilKind of the type, + // the value decodes as a nil pointer. + var ( + resultV = ctx.temp() + kindV = ctx.temp() + sizeV = ctx.temp() + wantKind string + ) + if op.nilValue == rlpstruct.NilKindList { + wantKind = "rlp.List" + } else { + wantKind = "rlp.String" + } + var b bytes.Buffer + fmt.Fprintf(&b, "var %s %s\n", resultV, types.TypeString(types.NewPointer(op.elemTyp), ctx.qualify)) + fmt.Fprintf(&b, "if %s, %s, err := dec.Kind(); err != nil {\n", kindV, sizeV) + fmt.Fprintf(&b, " return err\n") + fmt.Fprintf(&b, "} else if %s != 0 || %s != %s {\n", sizeV, kindV, wantKind) + fmt.Fprint(&b, code) + fmt.Fprintf(&b, " %s = &%s\n", resultV, result) + fmt.Fprintf(&b, "}\n") + return resultV, b.String() +} + +// structOp handles struct types. +type structOp struct { + named *types.Named + typ *types.Struct + fields []*structField + optionalFields []*structField +} + +type structField struct { + name string + typ types.Type + elem op +} + +func (bctx *buildContext) makeStructOp(named *types.Named, typ *types.Struct) (op, error) { + // Convert fields to []rlpstruct.Field. + var allStructFields []rlpstruct.Field + for i := 0; i < typ.NumFields(); i++ { + f := typ.Field(i) + allStructFields = append(allStructFields, rlpstruct.Field{ + Name: f.Name(), + Exported: f.Exported(), + Index: i, + Tag: typ.Tag(i), + Type: *bctx.typeToStructType(f.Type()), + }) + } + + // Filter/validate fields. + fields, tags, err := rlpstruct.ProcessFields(allStructFields) + if err != nil { + return nil, err + } + + // Create field ops. + var op = structOp{named: named, typ: typ} + for i, field := range fields { + // Advanced struct tags are not supported yet. + tag := tags[i] + if err := checkUnsupportedTags(field.Name, tag); err != nil { + return nil, err + } + typ := typ.Field(field.Index).Type() + elem, err := bctx.makeOp(nil, typ, tags[i]) + if err != nil { + return nil, fmt.Errorf("field %s: %v", field.Name, err) + } + f := &structField{name: field.Name, typ: typ, elem: elem} + if tag.Optional { + op.optionalFields = append(op.optionalFields, f) + } else { + op.fields = append(op.fields, f) + } + } + return op, nil +} + +func checkUnsupportedTags(field string, tag rlpstruct.Tags) error { + if tag.Tail { + return fmt.Errorf(`field %s has unsupported struct tag "tail"`, field) + } + return nil +} + +func (op structOp) genWrite(ctx *genContext, v string) string { + var b bytes.Buffer + var listMarker = ctx.temp() + fmt.Fprintf(&b, "%s := w.List()\n", listMarker) + for _, field := range op.fields { + selector := v + "." + field.name + fmt.Fprint(&b, field.elem.genWrite(ctx, selector)) + } + op.writeOptionalFields(&b, ctx, v) + fmt.Fprintf(&b, "w.ListEnd(%s)\n", listMarker) + return b.String() +} + +func (op structOp) writeOptionalFields(b *bytes.Buffer, ctx *genContext, v string) { + if len(op.optionalFields) == 0 { + return + } + // First check zero-ness of all optional fields. + var zeroV = make([]string, len(op.optionalFields)) + for i, field := range op.optionalFields { + selector := v + "." + field.name + zeroV[i] = ctx.temp() + fmt.Fprintf(b, "%s := %s\n", zeroV[i], nonZeroCheck(selector, field.typ, ctx.qualify)) + } + // Now write the fields. + for i, field := range op.optionalFields { + selector := v + "." + field.name + cond := "" + for j := i; j < len(op.optionalFields); j++ { + if j > i { + cond += " || " + } + cond += zeroV[j] + } + fmt.Fprintf(b, "if %s {\n", cond) + fmt.Fprint(b, field.elem.genWrite(ctx, selector)) + fmt.Fprintf(b, "}\n") + } +} + +func (op structOp) genDecode(ctx *genContext) (string, string) { + // Get the string representation of the type. + // Here, named types are handled separately because the output + // would contain a copy of the struct definition otherwise. + var typeName string + if op.named != nil { + typeName = types.TypeString(op.named, ctx.qualify) + } else { + typeName = types.TypeString(op.typ, ctx.qualify) + } + + // Create struct object. + var resultV = ctx.temp() + var b bytes.Buffer + fmt.Fprintf(&b, "var %s %s\n", resultV, typeName) + + // Decode fields. + fmt.Fprintf(&b, "{\n") + fmt.Fprintf(&b, "if _, err := dec.List(); err != nil { return err }\n") + for _, field := range op.fields { + result, code := field.elem.genDecode(ctx) + fmt.Fprintf(&b, "// %s:\n", field.name) + fmt.Fprint(&b, code) + fmt.Fprintf(&b, "%s.%s = %s\n", resultV, field.name, result) + } + op.decodeOptionalFields(&b, ctx, resultV) + fmt.Fprintf(&b, "if err := dec.ListEnd(); err != nil { return err }\n") + fmt.Fprintf(&b, "}\n") + return resultV, b.String() +} + +func (op structOp) decodeOptionalFields(b *bytes.Buffer, ctx *genContext, resultV string) { + var suffix bytes.Buffer + for _, field := range op.optionalFields { + result, code := field.elem.genDecode(ctx) + fmt.Fprintf(b, "// %s:\n", field.name) + fmt.Fprintf(b, "if dec.MoreDataInList() {\n") + fmt.Fprint(b, code) + fmt.Fprintf(b, "%s.%s = %s\n", resultV, field.name, result) + fmt.Fprintf(&suffix, "}\n") + } + suffix.WriteTo(b) +} + +// sliceOp handles slice types. +type sliceOp struct { + typ *types.Slice + elemOp op +} + +func (bctx *buildContext) makeSliceOp(typ *types.Slice) (op, error) { + elemOp, err := bctx.makeOp(nil, typ.Elem(), rlpstruct.Tags{}) + if err != nil { + return nil, err + } + return sliceOp{typ: typ, elemOp: elemOp}, nil +} + +func (op sliceOp) genWrite(ctx *genContext, v string) string { + var ( + listMarker = ctx.temp() // holds return value of w.List() + iterElemV = ctx.temp() // iteration variable + elemCode = op.elemOp.genWrite(ctx, iterElemV) + ) + + var b bytes.Buffer + fmt.Fprintf(&b, "%s := w.List()\n", listMarker) + fmt.Fprintf(&b, "for _, %s := range %s {\n", iterElemV, v) + fmt.Fprint(&b, elemCode) + fmt.Fprintf(&b, "}\n") + fmt.Fprintf(&b, "w.ListEnd(%s)\n", listMarker) + return b.String() +} + +func (op sliceOp) genDecode(ctx *genContext) (string, string) { + var sliceV = ctx.temp() // holds the output slice + elemResult, elemCode := op.elemOp.genDecode(ctx) + + var b bytes.Buffer + fmt.Fprintf(&b, "var %s %s\n", sliceV, types.TypeString(op.typ, ctx.qualify)) + fmt.Fprintf(&b, "if _, err := dec.List(); err != nil { return err }\n") + fmt.Fprintf(&b, "for dec.MoreDataInList() {\n") + fmt.Fprintf(&b, " %s", elemCode) + fmt.Fprintf(&b, " %s = append(%s, %s)\n", sliceV, sliceV, elemResult) + fmt.Fprintf(&b, "}\n") + fmt.Fprintf(&b, "if err := dec.ListEnd(); err != nil { return err }\n") + return sliceV, b.String() +} + +func (bctx *buildContext) makeOp(name *types.Named, typ types.Type, tags rlpstruct.Tags) (op, error) { + switch typ := typ.(type) { + case *types.Named: + if isBigInt(typ) { + return bigIntOp{}, nil + } + if isUint256(typ) { + return uint256Op{}, nil + } + if typ == bctx.rawValueType { + return bctx.makeRawValueOp(), nil + } + if bctx.isDecoder(typ) { + return nil, fmt.Errorf("type %v implements rlp.Decoder with non-pointer receiver", typ) + } + // TODO: same check for encoder? + return bctx.makeOp(typ, typ.Underlying(), tags) + case *types.Pointer: + if isBigInt(typ.Elem()) { + return bigIntOp{pointer: true}, nil + } + if isUint256(typ.Elem()) { + return uint256Op{pointer: true}, nil + } + // Encoder/Decoder interfaces. + if bctx.isEncoder(typ) { + if bctx.isDecoder(typ) { + return encoderDecoderOp{typ}, nil + } + return nil, fmt.Errorf("type %v implements rlp.Encoder but not rlp.Decoder", typ) + } + if bctx.isDecoder(typ) { + return nil, fmt.Errorf("type %v implements rlp.Decoder but not rlp.Encoder", typ) + } + // Default pointer handling. + return bctx.makePtrOp(typ.Elem(), tags) + case *types.Basic: + return bctx.makeBasicOp(typ) + case *types.Struct: + return bctx.makeStructOp(name, typ) + case *types.Slice: + etyp := typ.Elem() + if isByte(etyp) && !bctx.isEncoder(etyp) { + return bctx.makeByteSliceOp(typ), nil + } + return bctx.makeSliceOp(typ) + case *types.Array: + etyp := typ.Elem() + if isByte(etyp) && !bctx.isEncoder(etyp) { + return bctx.makeByteArrayOp(name, typ), nil + } + return nil, fmt.Errorf("unhandled array type: %v", typ) + default: + return nil, fmt.Errorf("unhandled type: %v", typ) + } +} + +// generateDecoder generates the DecodeRLP method on 'typ'. +func generateDecoder(ctx *genContext, typ string, op op) []byte { + ctx.resetTemp() + ctx.addImport(pathOfPackageRLP) + + result, code := op.genDecode(ctx) + var b bytes.Buffer + fmt.Fprintf(&b, "func (obj *%s) DecodeRLP(dec *rlp.Stream) error {\n", typ) + fmt.Fprint(&b, code) + fmt.Fprintf(&b, " *obj = %s\n", result) + fmt.Fprintf(&b, " return nil\n") + fmt.Fprintf(&b, "}\n") + return b.Bytes() +} + +// generateEncoder generates the EncodeRLP method on 'typ'. +func generateEncoder(ctx *genContext, typ string, op op) []byte { + ctx.resetTemp() + ctx.addImport("io") + ctx.addImport(pathOfPackageRLP) + + var b bytes.Buffer + fmt.Fprintf(&b, "func (obj *%s) EncodeRLP(_w io.Writer) error {\n", typ) + fmt.Fprintf(&b, " w := rlp.NewEncoderBuffer(_w)\n") + fmt.Fprint(&b, op.genWrite(ctx, "obj")) + fmt.Fprintf(&b, " return w.Flush()\n") + fmt.Fprintf(&b, "}\n") + return b.Bytes() +} + +func (bctx *buildContext) generate(typ *types.Named, encoder, decoder bool) ([]byte, error) { + bctx.topType = typ + + pkg := typ.Obj().Pkg() + op, err := bctx.makeOp(nil, typ, rlpstruct.Tags{}) + if err != nil { + return nil, err + } + + var ( + ctx = newGenContext(pkg) + encSource []byte + decSource []byte + ) + if encoder { + encSource = generateEncoder(ctx, typ.Obj().Name(), op) + } + if decoder { + decSource = generateDecoder(ctx, typ.Obj().Name(), op) + } + + var b bytes.Buffer + fmt.Fprintf(&b, "package %s\n\n", pkg.Name()) + for _, imp := range ctx.importsList() { + fmt.Fprintf(&b, "import %q\n", imp) + } + if encoder { + fmt.Fprintln(&b) + b.Write(encSource) + } + if decoder { + fmt.Fprintln(&b) + b.Write(decSource) + } + + source := b.Bytes() + // fmt.Println(string(source)) + return format.Source(source) +} diff --git a/rlp/rlpgen/gen_test.go b/rlp/rlpgen/gen_test.go new file mode 100644 index 000000000000..3b4f5df28765 --- /dev/null +++ b/rlp/rlpgen/gen_test.go @@ -0,0 +1,107 @@ +// Copyright 2022 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +package main + +import ( + "bytes" + "fmt" + "go/ast" + "go/importer" + "go/parser" + "go/token" + "go/types" + "os" + "path/filepath" + "testing" +) + +// Package RLP is loaded only once and reused for all tests. +var ( + testFset = token.NewFileSet() + testImporter = importer.ForCompiler(testFset, "source", nil).(types.ImporterFrom) + testPackageRLP *types.Package +) + +func init() { + cwd, err := os.Getwd() + if err != nil { + panic(err) + } + testPackageRLP, err = testImporter.ImportFrom(pathOfPackageRLP, cwd, 0) + if err != nil { + panic(fmt.Errorf("can't load package RLP: %v", err)) + } +} + +var tests = []string{"uints", "nil", "rawvalue", "optional", "bigint", "uint256"} + +func TestOutput(t *testing.T) { + for _, test := range tests { + test := test + t.Run(test, func(t *testing.T) { + inputFile := filepath.Join("testdata", test+".in.txt") + outputFile := filepath.Join("testdata", test+".out.txt") + bctx, typ, err := loadTestSource(inputFile, "Test") + if err != nil { + t.Fatal("error loading test source:", err) + } + output, err := bctx.generate(typ, true, true) + if err != nil { + t.Fatal("error in generate:", err) + } + + // Set this environment variable to regenerate the test outputs. + if os.Getenv("WRITE_TEST_FILES") != "" { + os.WriteFile(outputFile, output, 0644) + } + + // Check if output matches. + wantOutput, err := os.ReadFile(outputFile) + if err != nil { + t.Fatal("error loading expected test output:", err) + } + if !bytes.Equal(output, wantOutput) { + t.Fatalf("output mismatch, want: %v got %v", string(wantOutput), string(output)) + } + }) + } +} + +func loadTestSource(file string, typeName string) (*buildContext, *types.Named, error) { + // Load the test input. + content, err := os.ReadFile(file) + if err != nil { + return nil, nil, err + } + f, err := parser.ParseFile(testFset, file, content, 0) + if err != nil { + return nil, nil, err + } + conf := types.Config{Importer: testImporter} + pkg, err := conf.Check("test", testFset, []*ast.File{f}, nil) + if err != nil { + return nil, nil, err + } + + // Find the test struct. + bctx := newBuildContext(testPackageRLP) + typ, err := lookupStructType(pkg.Scope(), typeName) + if err != nil { + return nil, nil, fmt.Errorf("can't find type %s: %v", typeName, err) + } + return bctx, typ, nil +} diff --git a/rlp/rlpgen/main.go b/rlp/rlpgen/main.go new file mode 100644 index 000000000000..727188230606 --- /dev/null +++ b/rlp/rlpgen/main.go @@ -0,0 +1,144 @@ +// Copyright 2022 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +package main + +import ( + "bytes" + "errors" + "flag" + "fmt" + "go/types" + "os" + + "golang.org/x/tools/go/packages" +) + +const pathOfPackageRLP = "github.com/XinFinOrg/XDPoSChain/rlp" + +func main() { + var ( + pkgdir = flag.String("dir", ".", "input package") + output = flag.String("out", "-", "output file (default is stdout)") + genEncoder = flag.Bool("encoder", true, "generate EncodeRLP?") + genDecoder = flag.Bool("decoder", false, "generate DecodeRLP?") + typename = flag.String("type", "", "type to generate methods for") + ) + flag.Parse() + + cfg := Config{ + Dir: *pkgdir, + Type: *typename, + GenerateEncoder: *genEncoder, + GenerateDecoder: *genDecoder, + } + code, err := cfg.process() + if err != nil { + fatal(err) + } + if *output == "-" { + os.Stdout.Write(code) + } else if err := os.WriteFile(*output, code, 0600); err != nil { + fatal(err) + } +} + +func fatal(args ...interface{}) { + fmt.Fprintln(os.Stderr, args...) + os.Exit(1) +} + +type Config struct { + Dir string // input package directory + Type string + + GenerateEncoder bool + GenerateDecoder bool +} + +// process generates the Go code. +func (cfg *Config) process() (code []byte, err error) { + // Load packages. + pcfg := &packages.Config{ + Mode: packages.NeedName | packages.NeedTypes, + Dir: cfg.Dir, + } + ps, err := packages.Load(pcfg, pathOfPackageRLP, ".") + if err != nil { + return nil, err + } + if len(ps) == 0 { + return nil, fmt.Errorf("no Go package found in %s", cfg.Dir) + } + packages.PrintErrors(ps) + + // Find the packages that were loaded. + var ( + pkg *types.Package + packageRLP *types.Package + ) + for _, p := range ps { + if len(p.Errors) > 0 { + return nil, fmt.Errorf("package %s has errors", p.PkgPath) + } + if p.PkgPath == pathOfPackageRLP { + packageRLP = p.Types + } else { + pkg = p.Types + } + } + bctx := newBuildContext(packageRLP) + + // Find the type and generate. + typ, err := lookupStructType(pkg.Scope(), cfg.Type) + if err != nil { + return nil, fmt.Errorf("can't find %s in %s: %v", cfg.Type, pkg, err) + } + code, err = bctx.generate(typ, cfg.GenerateEncoder, cfg.GenerateDecoder) + if err != nil { + return nil, err + } + + // Add build comments. + // This is done here to avoid processing these lines with gofmt. + var header bytes.Buffer + fmt.Fprint(&header, "// Code generated by rlpgen. DO NOT EDIT.\n\n") + return append(header.Bytes(), code...), nil +} + +func lookupStructType(scope *types.Scope, name string) (*types.Named, error) { + typ, err := lookupType(scope, name) + if err != nil { + return nil, err + } + _, ok := typ.Underlying().(*types.Struct) + if !ok { + return nil, errors.New("not a struct type") + } + return typ, nil +} + +func lookupType(scope *types.Scope, name string) (*types.Named, error) { + obj := scope.Lookup(name) + if obj == nil { + return nil, errors.New("no such identifier") + } + typ, ok := obj.(*types.TypeName) + if !ok { + return nil, errors.New("not a type") + } + return typ.Type().(*types.Named), nil +} diff --git a/rlp/rlpgen/testdata/bigint.in.txt b/rlp/rlpgen/testdata/bigint.in.txt new file mode 100644 index 000000000000..d23d84a28763 --- /dev/null +++ b/rlp/rlpgen/testdata/bigint.in.txt @@ -0,0 +1,10 @@ +// -*- mode: go -*- + +package test + +import "math/big" + +type Test struct { + Int *big.Int + IntNoPtr big.Int +} diff --git a/rlp/rlpgen/testdata/bigint.out.txt b/rlp/rlpgen/testdata/bigint.out.txt new file mode 100644 index 000000000000..faab1bed461c --- /dev/null +++ b/rlp/rlpgen/testdata/bigint.out.txt @@ -0,0 +1,49 @@ +package test + +import "github.com/XinFinOrg/XDPoSChain/rlp" +import "io" + +func (obj *Test) EncodeRLP(_w io.Writer) error { + w := rlp.NewEncoderBuffer(_w) + _tmp0 := w.List() + if obj.Int == nil { + w.Write(rlp.EmptyString) + } else { + if obj.Int.Sign() == -1 { + return rlp.ErrNegativeBigInt + } + w.WriteBigInt(obj.Int) + } + if obj.IntNoPtr.Sign() == -1 { + return rlp.ErrNegativeBigInt + } + w.WriteBigInt(&obj.IntNoPtr) + w.ListEnd(_tmp0) + return w.Flush() +} + +func (obj *Test) DecodeRLP(dec *rlp.Stream) error { + var _tmp0 Test + { + if _, err := dec.List(); err != nil { + return err + } + // Int: + _tmp1, err := dec.BigInt() + if err != nil { + return err + } + _tmp0.Int = _tmp1 + // IntNoPtr: + _tmp2, err := dec.BigInt() + if err != nil { + return err + } + _tmp0.IntNoPtr = (*_tmp2) + if err := dec.ListEnd(); err != nil { + return err + } + } + *obj = _tmp0 + return nil +} diff --git a/rlp/rlpgen/testdata/nil.in.txt b/rlp/rlpgen/testdata/nil.in.txt new file mode 100644 index 000000000000..a28ff344874d --- /dev/null +++ b/rlp/rlpgen/testdata/nil.in.txt @@ -0,0 +1,30 @@ +// -*- mode: go -*- + +package test + +type Aux struct{ + A uint32 +} + +type Test struct{ + Uint8 *byte `rlp:"nil"` + Uint8List *byte `rlp:"nilList"` + + Uint32 *uint32 `rlp:"nil"` + Uint32List *uint32 `rlp:"nilList"` + + Uint64 *uint64 `rlp:"nil"` + Uint64List *uint64 `rlp:"nilList"` + + String *string `rlp:"nil"` + StringList *string `rlp:"nilList"` + + ByteArray *[3]byte `rlp:"nil"` + ByteArrayList *[3]byte `rlp:"nilList"` + + ByteSlice *[]byte `rlp:"nil"` + ByteSliceList *[]byte `rlp:"nilList"` + + Struct *Aux `rlp:"nil"` + StructString *Aux `rlp:"nilString"` +} diff --git a/rlp/rlpgen/testdata/nil.out.txt b/rlp/rlpgen/testdata/nil.out.txt new file mode 100644 index 000000000000..7f3459682b4b --- /dev/null +++ b/rlp/rlpgen/testdata/nil.out.txt @@ -0,0 +1,289 @@ +package test + +import "github.com/XinFinOrg/XDPoSChain/rlp" +import "io" + +func (obj *Test) EncodeRLP(_w io.Writer) error { + w := rlp.NewEncoderBuffer(_w) + _tmp0 := w.List() + if obj.Uint8 == nil { + w.Write([]byte{0x80}) + } else { + w.WriteUint64(uint64((*obj.Uint8))) + } + if obj.Uint8List == nil { + w.Write([]byte{0xC0}) + } else { + w.WriteUint64(uint64((*obj.Uint8List))) + } + if obj.Uint32 == nil { + w.Write([]byte{0x80}) + } else { + w.WriteUint64(uint64((*obj.Uint32))) + } + if obj.Uint32List == nil { + w.Write([]byte{0xC0}) + } else { + w.WriteUint64(uint64((*obj.Uint32List))) + } + if obj.Uint64 == nil { + w.Write([]byte{0x80}) + } else { + w.WriteUint64((*obj.Uint64)) + } + if obj.Uint64List == nil { + w.Write([]byte{0xC0}) + } else { + w.WriteUint64((*obj.Uint64List)) + } + if obj.String == nil { + w.Write([]byte{0x80}) + } else { + w.WriteString((*obj.String)) + } + if obj.StringList == nil { + w.Write([]byte{0xC0}) + } else { + w.WriteString((*obj.StringList)) + } + if obj.ByteArray == nil { + w.Write([]byte{0x80}) + } else { + w.WriteBytes(obj.ByteArray[:]) + } + if obj.ByteArrayList == nil { + w.Write([]byte{0xC0}) + } else { + w.WriteBytes(obj.ByteArrayList[:]) + } + if obj.ByteSlice == nil { + w.Write([]byte{0x80}) + } else { + w.WriteBytes((*obj.ByteSlice)) + } + if obj.ByteSliceList == nil { + w.Write([]byte{0xC0}) + } else { + w.WriteBytes((*obj.ByteSliceList)) + } + if obj.Struct == nil { + w.Write([]byte{0xC0}) + } else { + _tmp1 := w.List() + w.WriteUint64(uint64(obj.Struct.A)) + w.ListEnd(_tmp1) + } + if obj.StructString == nil { + w.Write([]byte{0x80}) + } else { + _tmp2 := w.List() + w.WriteUint64(uint64(obj.StructString.A)) + w.ListEnd(_tmp2) + } + w.ListEnd(_tmp0) + return w.Flush() +} + +func (obj *Test) DecodeRLP(dec *rlp.Stream) error { + var _tmp0 Test + { + if _, err := dec.List(); err != nil { + return err + } + // Uint8: + var _tmp2 *byte + if _tmp3, _tmp4, err := dec.Kind(); err != nil { + return err + } else if _tmp4 != 0 || _tmp3 != rlp.String { + _tmp1, err := dec.Uint8() + if err != nil { + return err + } + _tmp2 = &_tmp1 + } + _tmp0.Uint8 = _tmp2 + // Uint8List: + var _tmp6 *byte + if _tmp7, _tmp8, err := dec.Kind(); err != nil { + return err + } else if _tmp8 != 0 || _tmp7 != rlp.List { + _tmp5, err := dec.Uint8() + if err != nil { + return err + } + _tmp6 = &_tmp5 + } + _tmp0.Uint8List = _tmp6 + // Uint32: + var _tmp10 *uint32 + if _tmp11, _tmp12, err := dec.Kind(); err != nil { + return err + } else if _tmp12 != 0 || _tmp11 != rlp.String { + _tmp9, err := dec.Uint32() + if err != nil { + return err + } + _tmp10 = &_tmp9 + } + _tmp0.Uint32 = _tmp10 + // Uint32List: + var _tmp14 *uint32 + if _tmp15, _tmp16, err := dec.Kind(); err != nil { + return err + } else if _tmp16 != 0 || _tmp15 != rlp.List { + _tmp13, err := dec.Uint32() + if err != nil { + return err + } + _tmp14 = &_tmp13 + } + _tmp0.Uint32List = _tmp14 + // Uint64: + var _tmp18 *uint64 + if _tmp19, _tmp20, err := dec.Kind(); err != nil { + return err + } else if _tmp20 != 0 || _tmp19 != rlp.String { + _tmp17, err := dec.Uint64() + if err != nil { + return err + } + _tmp18 = &_tmp17 + } + _tmp0.Uint64 = _tmp18 + // Uint64List: + var _tmp22 *uint64 + if _tmp23, _tmp24, err := dec.Kind(); err != nil { + return err + } else if _tmp24 != 0 || _tmp23 != rlp.List { + _tmp21, err := dec.Uint64() + if err != nil { + return err + } + _tmp22 = &_tmp21 + } + _tmp0.Uint64List = _tmp22 + // String: + var _tmp26 *string + if _tmp27, _tmp28, err := dec.Kind(); err != nil { + return err + } else if _tmp28 != 0 || _tmp27 != rlp.String { + _tmp25, err := dec.String() + if err != nil { + return err + } + _tmp26 = &_tmp25 + } + _tmp0.String = _tmp26 + // StringList: + var _tmp30 *string + if _tmp31, _tmp32, err := dec.Kind(); err != nil { + return err + } else if _tmp32 != 0 || _tmp31 != rlp.List { + _tmp29, err := dec.String() + if err != nil { + return err + } + _tmp30 = &_tmp29 + } + _tmp0.StringList = _tmp30 + // ByteArray: + var _tmp34 *[3]byte + if _tmp35, _tmp36, err := dec.Kind(); err != nil { + return err + } else if _tmp36 != 0 || _tmp35 != rlp.String { + var _tmp33 [3]byte + if err := dec.ReadBytes(_tmp33[:]); err != nil { + return err + } + _tmp34 = &_tmp33 + } + _tmp0.ByteArray = _tmp34 + // ByteArrayList: + var _tmp38 *[3]byte + if _tmp39, _tmp40, err := dec.Kind(); err != nil { + return err + } else if _tmp40 != 0 || _tmp39 != rlp.List { + var _tmp37 [3]byte + if err := dec.ReadBytes(_tmp37[:]); err != nil { + return err + } + _tmp38 = &_tmp37 + } + _tmp0.ByteArrayList = _tmp38 + // ByteSlice: + var _tmp42 *[]byte + if _tmp43, _tmp44, err := dec.Kind(); err != nil { + return err + } else if _tmp44 != 0 || _tmp43 != rlp.String { + _tmp41, err := dec.Bytes() + if err != nil { + return err + } + _tmp42 = &_tmp41 + } + _tmp0.ByteSlice = _tmp42 + // ByteSliceList: + var _tmp46 *[]byte + if _tmp47, _tmp48, err := dec.Kind(); err != nil { + return err + } else if _tmp48 != 0 || _tmp47 != rlp.List { + _tmp45, err := dec.Bytes() + if err != nil { + return err + } + _tmp46 = &_tmp45 + } + _tmp0.ByteSliceList = _tmp46 + // Struct: + var _tmp51 *Aux + if _tmp52, _tmp53, err := dec.Kind(); err != nil { + return err + } else if _tmp53 != 0 || _tmp52 != rlp.List { + var _tmp49 Aux + { + if _, err := dec.List(); err != nil { + return err + } + // A: + _tmp50, err := dec.Uint32() + if err != nil { + return err + } + _tmp49.A = _tmp50 + if err := dec.ListEnd(); err != nil { + return err + } + } + _tmp51 = &_tmp49 + } + _tmp0.Struct = _tmp51 + // StructString: + var _tmp56 *Aux + if _tmp57, _tmp58, err := dec.Kind(); err != nil { + return err + } else if _tmp58 != 0 || _tmp57 != rlp.String { + var _tmp54 Aux + { + if _, err := dec.List(); err != nil { + return err + } + // A: + _tmp55, err := dec.Uint32() + if err != nil { + return err + } + _tmp54.A = _tmp55 + if err := dec.ListEnd(); err != nil { + return err + } + } + _tmp56 = &_tmp54 + } + _tmp0.StructString = _tmp56 + if err := dec.ListEnd(); err != nil { + return err + } + } + *obj = _tmp0 + return nil +} diff --git a/rlp/rlpgen/testdata/optional.in.txt b/rlp/rlpgen/testdata/optional.in.txt new file mode 100644 index 000000000000..f1ac9f7899d1 --- /dev/null +++ b/rlp/rlpgen/testdata/optional.in.txt @@ -0,0 +1,17 @@ +// -*- mode: go -*- + +package test + +type Aux struct { + A uint64 +} + +type Test struct { + Uint64 uint64 `rlp:"optional"` + Pointer *uint64 `rlp:"optional"` + String string `rlp:"optional"` + Slice []uint64 `rlp:"optional"` + Array [3]byte `rlp:"optional"` + NamedStruct Aux `rlp:"optional"` + AnonStruct struct{ A string } `rlp:"optional"` +} diff --git a/rlp/rlpgen/testdata/optional.out.txt b/rlp/rlpgen/testdata/optional.out.txt new file mode 100644 index 000000000000..8b4cfa18171c --- /dev/null +++ b/rlp/rlpgen/testdata/optional.out.txt @@ -0,0 +1,153 @@ +package test + +import "github.com/XinFinOrg/XDPoSChain/rlp" +import "io" + +func (obj *Test) EncodeRLP(_w io.Writer) error { + w := rlp.NewEncoderBuffer(_w) + _tmp0 := w.List() + _tmp1 := obj.Uint64 != 0 + _tmp2 := obj.Pointer != nil + _tmp3 := obj.String != "" + _tmp4 := len(obj.Slice) > 0 + _tmp5 := obj.Array != ([3]byte{}) + _tmp6 := obj.NamedStruct != (Aux{}) + _tmp7 := obj.AnonStruct != (struct{ A string }{}) + if _tmp1 || _tmp2 || _tmp3 || _tmp4 || _tmp5 || _tmp6 || _tmp7 { + w.WriteUint64(obj.Uint64) + } + if _tmp2 || _tmp3 || _tmp4 || _tmp5 || _tmp6 || _tmp7 { + if obj.Pointer == nil { + w.Write([]byte{0x80}) + } else { + w.WriteUint64((*obj.Pointer)) + } + } + if _tmp3 || _tmp4 || _tmp5 || _tmp6 || _tmp7 { + w.WriteString(obj.String) + } + if _tmp4 || _tmp5 || _tmp6 || _tmp7 { + _tmp8 := w.List() + for _, _tmp9 := range obj.Slice { + w.WriteUint64(_tmp9) + } + w.ListEnd(_tmp8) + } + if _tmp5 || _tmp6 || _tmp7 { + w.WriteBytes(obj.Array[:]) + } + if _tmp6 || _tmp7 { + _tmp10 := w.List() + w.WriteUint64(obj.NamedStruct.A) + w.ListEnd(_tmp10) + } + if _tmp7 { + _tmp11 := w.List() + w.WriteString(obj.AnonStruct.A) + w.ListEnd(_tmp11) + } + w.ListEnd(_tmp0) + return w.Flush() +} + +func (obj *Test) DecodeRLP(dec *rlp.Stream) error { + var _tmp0 Test + { + if _, err := dec.List(); err != nil { + return err + } + // Uint64: + if dec.MoreDataInList() { + _tmp1, err := dec.Uint64() + if err != nil { + return err + } + _tmp0.Uint64 = _tmp1 + // Pointer: + if dec.MoreDataInList() { + _tmp2, err := dec.Uint64() + if err != nil { + return err + } + _tmp0.Pointer = &_tmp2 + // String: + if dec.MoreDataInList() { + _tmp3, err := dec.String() + if err != nil { + return err + } + _tmp0.String = _tmp3 + // Slice: + if dec.MoreDataInList() { + var _tmp4 []uint64 + if _, err := dec.List(); err != nil { + return err + } + for dec.MoreDataInList() { + _tmp5, err := dec.Uint64() + if err != nil { + return err + } + _tmp4 = append(_tmp4, _tmp5) + } + if err := dec.ListEnd(); err != nil { + return err + } + _tmp0.Slice = _tmp4 + // Array: + if dec.MoreDataInList() { + var _tmp6 [3]byte + if err := dec.ReadBytes(_tmp6[:]); err != nil { + return err + } + _tmp0.Array = _tmp6 + // NamedStruct: + if dec.MoreDataInList() { + var _tmp7 Aux + { + if _, err := dec.List(); err != nil { + return err + } + // A: + _tmp8, err := dec.Uint64() + if err != nil { + return err + } + _tmp7.A = _tmp8 + if err := dec.ListEnd(); err != nil { + return err + } + } + _tmp0.NamedStruct = _tmp7 + // AnonStruct: + if dec.MoreDataInList() { + var _tmp9 struct{ A string } + { + if _, err := dec.List(); err != nil { + return err + } + // A: + _tmp10, err := dec.String() + if err != nil { + return err + } + _tmp9.A = _tmp10 + if err := dec.ListEnd(); err != nil { + return err + } + } + _tmp0.AnonStruct = _tmp9 + } + } + } + } + } + } + } + if err := dec.ListEnd(); err != nil { + return err + } + } + *obj = _tmp0 + return nil +} diff --git a/rlp/rlpgen/testdata/rawvalue.in.txt b/rlp/rlpgen/testdata/rawvalue.in.txt new file mode 100644 index 000000000000..daa050c3f65e --- /dev/null +++ b/rlp/rlpgen/testdata/rawvalue.in.txt @@ -0,0 +1,11 @@ +// -*- mode: go -*- + +package test + +import "github.com/XinFinOrg/XDPoSChain/rlp" + +type Test struct { + RawValue rlp.RawValue + PointerToRawValue *rlp.RawValue + SliceOfRawValue []rlp.RawValue +} diff --git a/rlp/rlpgen/testdata/rawvalue.out.txt b/rlp/rlpgen/testdata/rawvalue.out.txt new file mode 100644 index 000000000000..35bf145dcc71 --- /dev/null +++ b/rlp/rlpgen/testdata/rawvalue.out.txt @@ -0,0 +1,64 @@ +package test + +import "github.com/XinFinOrg/XDPoSChain/rlp" +import "io" + +func (obj *Test) EncodeRLP(_w io.Writer) error { + w := rlp.NewEncoderBuffer(_w) + _tmp0 := w.List() + w.Write(obj.RawValue) + if obj.PointerToRawValue == nil { + w.Write([]byte{0x80}) + } else { + w.Write((*obj.PointerToRawValue)) + } + _tmp1 := w.List() + for _, _tmp2 := range obj.SliceOfRawValue { + w.Write(_tmp2) + } + w.ListEnd(_tmp1) + w.ListEnd(_tmp0) + return w.Flush() +} + +func (obj *Test) DecodeRLP(dec *rlp.Stream) error { + var _tmp0 Test + { + if _, err := dec.List(); err != nil { + return err + } + // RawValue: + _tmp1, err := dec.Raw() + if err != nil { + return err + } + _tmp0.RawValue = _tmp1 + // PointerToRawValue: + _tmp2, err := dec.Raw() + if err != nil { + return err + } + _tmp0.PointerToRawValue = &_tmp2 + // SliceOfRawValue: + var _tmp3 []rlp.RawValue + if _, err := dec.List(); err != nil { + return err + } + for dec.MoreDataInList() { + _tmp4, err := dec.Raw() + if err != nil { + return err + } + _tmp3 = append(_tmp3, _tmp4) + } + if err := dec.ListEnd(); err != nil { + return err + } + _tmp0.SliceOfRawValue = _tmp3 + if err := dec.ListEnd(); err != nil { + return err + } + } + *obj = _tmp0 + return nil +} diff --git a/rlp/rlpgen/testdata/uint256.in.txt b/rlp/rlpgen/testdata/uint256.in.txt new file mode 100644 index 000000000000..ed16e0a7882f --- /dev/null +++ b/rlp/rlpgen/testdata/uint256.in.txt @@ -0,0 +1,10 @@ +// -*- mode: go -*- + +package test + +import "github.com/holiman/uint256" + +type Test struct { + Int *uint256.Int + IntNoPtr uint256.Int +} diff --git a/rlp/rlpgen/testdata/uint256.out.txt b/rlp/rlpgen/testdata/uint256.out.txt new file mode 100644 index 000000000000..b560aa0e42fe --- /dev/null +++ b/rlp/rlpgen/testdata/uint256.out.txt @@ -0,0 +1,44 @@ +package test + +import "github.com/XinFinOrg/XDPoSChain/rlp" +import "github.com/holiman/uint256" +import "io" + +func (obj *Test) EncodeRLP(_w io.Writer) error { + w := rlp.NewEncoderBuffer(_w) + _tmp0 := w.List() + if obj.Int == nil { + w.Write(rlp.EmptyString) + } else { + w.WriteUint256(obj.Int) + } + w.WriteUint256(&obj.IntNoPtr) + w.ListEnd(_tmp0) + return w.Flush() +} + +func (obj *Test) DecodeRLP(dec *rlp.Stream) error { + var _tmp0 Test + { + if _, err := dec.List(); err != nil { + return err + } + // Int: + var _tmp1 uint256.Int + if err := dec.ReadUint256(&_tmp1); err != nil { + return err + } + _tmp0.Int = &_tmp1 + // IntNoPtr: + var _tmp2 uint256.Int + if err := dec.ReadUint256(&_tmp2); err != nil { + return err + } + _tmp0.IntNoPtr = _tmp2 + if err := dec.ListEnd(); err != nil { + return err + } + } + *obj = _tmp0 + return nil +} diff --git a/rlp/rlpgen/testdata/uints.in.txt b/rlp/rlpgen/testdata/uints.in.txt new file mode 100644 index 000000000000..8095da997d96 --- /dev/null +++ b/rlp/rlpgen/testdata/uints.in.txt @@ -0,0 +1,10 @@ +// -*- mode: go -*- + +package test + +type Test struct{ + A uint8 + B uint16 + C uint32 + D uint64 +} diff --git a/rlp/rlpgen/testdata/uints.out.txt b/rlp/rlpgen/testdata/uints.out.txt new file mode 100644 index 000000000000..cf973ec9a43b --- /dev/null +++ b/rlp/rlpgen/testdata/uints.out.txt @@ -0,0 +1,53 @@ +package test + +import "github.com/XinFinOrg/XDPoSChain/rlp" +import "io" + +func (obj *Test) EncodeRLP(_w io.Writer) error { + w := rlp.NewEncoderBuffer(_w) + _tmp0 := w.List() + w.WriteUint64(uint64(obj.A)) + w.WriteUint64(uint64(obj.B)) + w.WriteUint64(uint64(obj.C)) + w.WriteUint64(obj.D) + w.ListEnd(_tmp0) + return w.Flush() +} + +func (obj *Test) DecodeRLP(dec *rlp.Stream) error { + var _tmp0 Test + { + if _, err := dec.List(); err != nil { + return err + } + // A: + _tmp1, err := dec.Uint8() + if err != nil { + return err + } + _tmp0.A = _tmp1 + // B: + _tmp2, err := dec.Uint16() + if err != nil { + return err + } + _tmp0.B = _tmp2 + // C: + _tmp3, err := dec.Uint32() + if err != nil { + return err + } + _tmp0.C = _tmp3 + // D: + _tmp4, err := dec.Uint64() + if err != nil { + return err + } + _tmp0.D = _tmp4 + if err := dec.ListEnd(); err != nil { + return err + } + } + *obj = _tmp0 + return nil +} diff --git a/rlp/rlpgen/types.go b/rlp/rlpgen/types.go new file mode 100644 index 000000000000..ea7dc96d8813 --- /dev/null +++ b/rlp/rlpgen/types.go @@ -0,0 +1,124 @@ +// Copyright 2022 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +package main + +import ( + "fmt" + "go/types" + "reflect" +) + +// typeReflectKind gives the reflect.Kind that represents typ. +func typeReflectKind(typ types.Type) reflect.Kind { + switch typ := typ.(type) { + case *types.Basic: + k := typ.Kind() + if k >= types.Bool && k <= types.Complex128 { + // value order matches for Bool..Complex128 + return reflect.Bool + reflect.Kind(k-types.Bool) + } + if k == types.String { + return reflect.String + } + if k == types.UnsafePointer { + return reflect.UnsafePointer + } + panic(fmt.Errorf("unhandled BasicKind %v", k)) + case *types.Array: + return reflect.Array + case *types.Chan: + return reflect.Chan + case *types.Interface: + return reflect.Interface + case *types.Map: + return reflect.Map + case *types.Pointer: + return reflect.Ptr + case *types.Signature: + return reflect.Func + case *types.Slice: + return reflect.Slice + case *types.Struct: + return reflect.Struct + default: + panic(fmt.Errorf("unhandled type %T", typ)) + } +} + +// nonZeroCheck returns the expression that checks whether 'v' is a non-zero value of type 'vtyp'. +func nonZeroCheck(v string, vtyp types.Type, qualify types.Qualifier) string { + // Resolve type name. + typ := resolveUnderlying(vtyp) + switch typ := typ.(type) { + case *types.Basic: + k := typ.Kind() + switch { + case k == types.Bool: + return v + case k >= types.Uint && k <= types.Complex128: + return fmt.Sprintf("%s != 0", v) + case k == types.String: + return fmt.Sprintf(`%s != ""`, v) + default: + panic(fmt.Errorf("unhandled BasicKind %v", k)) + } + case *types.Array, *types.Struct: + return fmt.Sprintf("%s != (%s{})", v, types.TypeString(vtyp, qualify)) + case *types.Interface, *types.Pointer, *types.Signature: + return fmt.Sprintf("%s != nil", v) + case *types.Slice, *types.Map: + return fmt.Sprintf("len(%s) > 0", v) + default: + panic(fmt.Errorf("unhandled type %T", typ)) + } +} + +// isBigInt checks whether 'typ' is "math/big".Int. +func isBigInt(typ types.Type) bool { + named, ok := typ.(*types.Named) + if !ok { + return false + } + name := named.Obj() + return name.Pkg().Path() == "math/big" && name.Name() == "Int" +} + +// isUint256 checks whether 'typ' is "github.com/holiman/uint256".Int. +func isUint256(typ types.Type) bool { + named, ok := typ.(*types.Named) + if !ok { + return false + } + name := named.Obj() + return name.Pkg().Path() == "github.com/holiman/uint256" && name.Name() == "Int" +} + +// isByte checks whether the underlying type of 'typ' is uint8. +func isByte(typ types.Type) bool { + basic, ok := resolveUnderlying(typ).(*types.Basic) + return ok && basic.Kind() == types.Uint8 +} + +func resolveUnderlying(typ types.Type) types.Type { + for { + t := typ.Underlying() + if t == typ { + return t + } + typ = t + } +} diff --git a/rlp/safe.go b/rlp/safe.go new file mode 100644 index 000000000000..3c910337b6a2 --- /dev/null +++ b/rlp/safe.go @@ -0,0 +1,27 @@ +// Copyright 2021 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +//go:build nacl || js || !cgo +// +build nacl js !cgo + +package rlp + +import "reflect" + +// byteArrayBytes returns a slice of the byte array v. +func byteArrayBytes(v reflect.Value, length int) []byte { + return v.Slice(0, length).Bytes() +} diff --git a/rlp/typecache.go b/rlp/typecache.go index 3df799e1ecd5..57e4f2e46a06 100644 --- a/rlp/typecache.go +++ b/rlp/typecache.go @@ -18,139 +18,221 @@ package rlp import ( "fmt" + "maps" "reflect" - "strings" "sync" -) + "sync/atomic" -var ( - typeCacheMutex sync.RWMutex - typeCache = make(map[typekey]*typeinfo) + "github.com/XinFinOrg/XDPoSChain/rlp/internal/rlpstruct" ) +// typeinfo is an entry in the type cache. type typeinfo struct { - decoder - writer -} - -// represents struct tags -type tags struct { - // rlp:"nil" controls whether empty input results in a nil pointer. - nilOK bool - // rlp:"tail" controls whether this field swallows additional list - // elements. It can only be set for the last field, which must be - // of slice type. - tail bool - // rlp:"-" ignores fields. - ignored bool + decoder decoder + decoderErr error // error from makeDecoder + writer writer + writerErr error // error from makeWriter } +// typekey is the key of a type in typeCache. It includes the struct tags because +// they might generate a different decoder. type typekey struct { reflect.Type - // the key must include the struct tags because they - // might generate a different decoder. - tags + rlpstruct.Tags } type decoder func(*Stream, reflect.Value) error -type writer func(reflect.Value, *encbuf) error +type writer func(reflect.Value, *encBuffer) error + +var theTC = newTypeCache() + +type typeCache struct { + cur atomic.Value + + // This lock synchronizes writers. + mu sync.Mutex + next map[typekey]*typeinfo +} + +func newTypeCache() *typeCache { + c := new(typeCache) + c.cur.Store(make(map[typekey]*typeinfo)) + return c +} + +func cachedDecoder(typ reflect.Type) (decoder, error) { + info := theTC.info(typ) + return info.decoder, info.decoderErr +} + +func cachedWriter(typ reflect.Type) (writer, error) { + info := theTC.info(typ) + return info.writer, info.writerErr +} -func cachedTypeInfo(typ reflect.Type, tags tags) (*typeinfo, error) { - typeCacheMutex.RLock() - info := typeCache[typekey{typ, tags}] - typeCacheMutex.RUnlock() - if info != nil { - return info, nil +func (c *typeCache) info(typ reflect.Type) *typeinfo { + key := typekey{Type: typ} + if info := c.cur.Load().(map[typekey]*typeinfo)[key]; info != nil { + return info } - // not in the cache, need to generate info for this type. - typeCacheMutex.Lock() - defer typeCacheMutex.Unlock() - return cachedTypeInfo1(typ, tags) + + // Not in the cache, need to generate info for this type. + return c.generate(typ, rlpstruct.Tags{}) +} + +func (c *typeCache) generate(typ reflect.Type, tags rlpstruct.Tags) *typeinfo { + c.mu.Lock() + defer c.mu.Unlock() + + cur := c.cur.Load().(map[typekey]*typeinfo) + if info := cur[typekey{typ, tags}]; info != nil { + return info + } + + // Copy cur to next. + c.next = maps.Clone(cur) + + // Generate. + info := c.infoWhileGenerating(typ, tags) + + // next -> cur + c.cur.Store(c.next) + c.next = nil + return info } -func cachedTypeInfo1(typ reflect.Type, tags tags) (*typeinfo, error) { +func (c *typeCache) infoWhileGenerating(typ reflect.Type, tags rlpstruct.Tags) *typeinfo { key := typekey{typ, tags} - info := typeCache[key] - if info != nil { - // another goroutine got the write lock first - return info, nil + if info := c.next[key]; info != nil { + return info } - // put a dummmy value into the cache before generating. - // if the generator tries to lookup itself, it will get + // Put a dummy value into the cache before generating. + // If the generator tries to lookup itself, it will get // the dummy value and won't call itself recursively. - typeCache[key] = new(typeinfo) - info, err := genTypeInfo(typ, tags) - if err != nil { - // remove the dummy value if the generator fails - delete(typeCache, key) - return nil, err - } - *typeCache[key] = *info - return typeCache[key], err + info := new(typeinfo) + c.next[key] = info + info.generate(typ, tags) + return info } type field struct { - index int - info *typeinfo + index int + info *typeinfo + optional bool } +// structFields resolves the typeinfo of all public fields in a struct type. func structFields(typ reflect.Type) (fields []field, err error) { + // Convert fields to rlpstruct.Field. + var allStructFields []rlpstruct.Field for i := 0; i < typ.NumField(); i++ { - if f := typ.Field(i); f.PkgPath == "" { // exported - tags, err := parseStructTag(typ, i) - if err != nil { - return nil, err - } - if tags.ignored { - continue - } - info, err := cachedTypeInfo1(f.Type, tags) - if err != nil { - return nil, err - } - fields = append(fields, field{i, info}) + rf := typ.Field(i) + allStructFields = append(allStructFields, rlpstruct.Field{ + Name: rf.Name, + Index: i, + Exported: rf.PkgPath == "", + Tag: string(rf.Tag), + Type: *rtypeToStructType(rf.Type, nil), + }) + } + + // Filter/validate fields. + structFields, structTags, err := rlpstruct.ProcessFields(allStructFields) + if err != nil { + if tagErr, ok := err.(rlpstruct.TagError); ok { + tagErr.StructType = typ.String() + return nil, tagErr } + return nil, err + } + + // Resolve typeinfo. + for i, sf := range structFields { + typ := typ.Field(sf.Index).Type + tags := structTags[i] + info := theTC.infoWhileGenerating(typ, tags) + fields = append(fields, field{sf.Index, info, tags.Optional}) } return fields, nil } -func parseStructTag(typ reflect.Type, fi int) (tags, error) { - f := typ.Field(fi) - var ts tags - for _, t := range strings.Split(f.Tag.Get("rlp"), ",") { - switch t = strings.TrimSpace(t); t { - case "": - case "-": - ts.ignored = true - case "nil": - ts.nilOK = true - case "tail": - ts.tail = true - if fi != typ.NumField()-1 { - return ts, fmt.Errorf(`rlp: invalid struct tag "tail" for %v.%s (must be on last field)`, typ, f.Name) - } - if f.Type.Kind() != reflect.Slice { - return ts, fmt.Errorf(`rlp: invalid struct tag "tail" for %v.%s (field type is not slice)`, typ, f.Name) - } - default: - return ts, fmt.Errorf("rlp: unknown struct tag %q on %v.%s", t, typ, f.Name) +// firstOptionalField returns the index of the first field with "optional" tag. +func firstOptionalField(fields []field) int { + for i, f := range fields { + if f.optional { + return i } } - return ts, nil + return len(fields) } -func genTypeInfo(typ reflect.Type, tags tags) (info *typeinfo, err error) { - info = new(typeinfo) - if info.decoder, err = makeDecoder(typ, tags); err != nil { - return nil, err +type structFieldError struct { + typ reflect.Type + field int + err error +} + +func (e structFieldError) Error() string { + return fmt.Sprintf("%v (struct field %v.%s)", e.err, e.typ, e.typ.Field(e.field).Name) +} + +func (i *typeinfo) generate(typ reflect.Type, tags rlpstruct.Tags) { + i.decoder, i.decoderErr = makeDecoder(typ, tags) + i.writer, i.writerErr = makeWriter(typ, tags) +} + +// rtypeToStructType converts typ to rlpstruct.Type. +func rtypeToStructType(typ reflect.Type, rec map[reflect.Type]*rlpstruct.Type) *rlpstruct.Type { + k := typ.Kind() + if k == reflect.Invalid { + panic("invalid kind") } - if info.writer, err = makeWriter(typ, tags); err != nil { - return nil, err + + if prev := rec[typ]; prev != nil { + return prev // short-circuit for recursive types + } + if rec == nil { + rec = make(map[reflect.Type]*rlpstruct.Type) + } + + t := &rlpstruct.Type{ + Name: typ.String(), + Kind: k, + IsEncoder: typ.Implements(encoderInterface), + IsDecoder: typ.Implements(decoderInterface), + } + rec[typ] = t + if k == reflect.Array || k == reflect.Slice || k == reflect.Ptr { + t.Elem = rtypeToStructType(typ.Elem(), rec) + } + return t +} + +// typeNilKind gives the RLP value kind for nil pointers to 'typ'. +func typeNilKind(typ reflect.Type, tags rlpstruct.Tags) Kind { + styp := rtypeToStructType(typ, nil) + + var nk rlpstruct.NilKind + if tags.NilOK { + nk = tags.NilKind + } else { + nk = styp.DefaultNilValue() + } + switch nk { + case rlpstruct.NilKindString: + return String + case rlpstruct.NilKindList: + return List + default: + panic("invalid nil kind value") } - return info, nil } func isUint(k reflect.Kind) bool { return k >= reflect.Uint && k <= reflect.Uintptr } + +func isByte(typ reflect.Type) bool { + return typ.Kind() == reflect.Uint8 && !typ.Implements(encoderInterface) +} diff --git a/rlp/unsafe.go b/rlp/unsafe.go new file mode 100644 index 000000000000..10868caaf287 --- /dev/null +++ b/rlp/unsafe.go @@ -0,0 +1,30 @@ +// Copyright 2021 The go-ethereum Authors +// This file is part of the go-ethereum library. +// +// The go-ethereum library is free software: you can redistribute it and/or modify +// it under the terms of the GNU Lesser General Public License as published by +// the Free Software Foundation, either version 3 of the License, or +// (at your option) any later version. +// +// The go-ethereum library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Lesser General Public License for more details. +// +// You should have received a copy of the GNU Lesser General Public License +// along with the go-ethereum library. If not, see . + +//go:build !nacl && !js && cgo +// +build !nacl,!js,cgo + +package rlp + +import ( + "reflect" + "unsafe" +) + +// byteArrayBytes returns a slice of the byte array v. +func byteArrayBytes(v reflect.Value, length int) []byte { + return unsafe.Slice((*byte)(unsafe.Pointer(v.UnsafeAddr())), length) +}