commit 8c9518167d6eabd69407dc7311e9404b9abffb16 Author: Dave Friedel Date: Sun Dec 21 05:54:43 2025 -0500 Initial commit - Tutus Ordered JSON diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml new file mode 100644 index 0000000..c554f77 --- /dev/null +++ b/.github/workflows/run_tests.yml @@ -0,0 +1,119 @@ +name: Tests + +on: + pull_request: + branches: + - master + types: [opened, synchronize] + paths-ignore: + - '*.md' + workflow_dispatch: + +env: + GO111MODULE: "on" + +jobs: + test_cover: + name: Coverage + runs-on: ubuntu-latest + + env: + CGO_ENABLED: 0 + steps: + - uses: actions/checkout@v4 + with: + fetch-depth: 0 + + - name: Set up Go + uses: actions/setup-go@v5 + with: + go-version: 1.25 + + - name: Write coverage profile + run: go test -v ./... -coverprofile=./coverage.txt -covermode=atomic + + - name: Upload coverage results to Codecov + uses: codecov/codecov-action@v2 + with: + fail_ci_if_error: false + path_to_write_report: ./coverage.txt + verbose: true + + tests: + name: Go + runs-on: ${{ matrix.os }} + strategy: + matrix: + go_versions: [ '1.24', '1.25' ] + os: [ubuntu-latest, windows-latest, macos-latest] + exclude: + # Only latest Go version for Windows and MacOS. + - os: windows-latest + go_versions: '1.24' + - os: macos-latest + go_versions: '1.24' + # Exclude latest Go version for Ubuntu as Coverage uses it. + - os: ubuntu-latest + go_versions: '1.25' + fail-fast: false + steps: + - uses: actions/checkout@v4 + with: + fetch-depth: 0 + + - name: Set up Go + uses: actions/setup-go@v5 + with: + go-version: '${{ matrix.go_versions }}' + + - name: Run tests + run: go test -v -race ./... + + lint: + name: Linter + uses: nspcc-dev/.github/.github/workflows/go-linter.yml@master + + codeql: + name: CodeQL + runs-on: ubuntu-latest + + strategy: + fail-fast: false + matrix: + language: [ 'go' ] + # CodeQL supports [ 'cpp', 'csharp', 'go', 'java', 'javascript', 'python' ] + # Learn more: + # https://docs.github.com/en/free-pro-team@latest/github/finding-security-vulnerabilities-and-errors-in-your-code/configuring-code-scanning#changing-the-languages-that-are-analyzed + + steps: + - name: Checkout repository + uses: actions/checkout@v4 + + # Initializes the CodeQL tools for scanning. + - name: Initialize CodeQL + uses: github/codeql-action/init@v3 + with: + languages: ${{ matrix.language }} + # If you wish to specify custom queries, you can do so here or in a config file. + # By default, queries listed here will override any specified in a config file. + # Prefix the list here with "+" to use these queries and those in the config file. + # queries: ./path/to/local/query, your-org/your-repo/queries@main + + # Autobuild attempts to build any compiled languages (C/C++, C#, or Java). + # If this step fails, then you should remove it and run the build manually (see below) + - name: Autobuild + uses: github/codeql-action/autobuild@v3 + + # ℹ️ Command-line programs to run using the OS shell. + # 📚 https://git.io/JvXDl + + # ✏️ If the Autobuild fails above, remove it and uncomment the following three lines + # and modify them (or add more) to build your code if your project + # uses a compiled language + + #- run: | + # make bootstrap + # make release + + - name: Perform CodeQL Analysis + uses: github/codeql-action/analyze@v3 diff --git a/LICENSE b/LICENSE new file mode 100644 index 0000000..6a66aea --- /dev/null +++ b/LICENSE @@ -0,0 +1,27 @@ +Copyright (c) 2009 The Go Authors. All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are +met: + + * Redistributions of source code must retain the above copyright +notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above +copyright notice, this list of conditions and the following disclaimer +in the documentation and/or other materials provided with the +distribution. + * Neither the name of Google Inc. nor the names of its +contributors may be used to endorse or promote products derived from +this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. diff --git a/README.md b/README.md new file mode 100644 index 0000000..c3cb5ef --- /dev/null +++ b/README.md @@ -0,0 +1,21 @@ +# Tutus Ordered JSON + +Ordered JSON serialization library for the Tutus blockchain. + +## Overview + +Provides deterministic JSON marshaling with preserved key ordering. + +## Installation + +```go +import "github.com/tutus-one/tutus-ordered-json" +``` + +## License + +MIT License + +--- + +Part of the [Tutus](https://github.com/tutus-one/tutus-chain) blockchain infrastructure. diff --git a/bench_test.go b/bench_test.go new file mode 100644 index 0000000..8d51375 --- /dev/null +++ b/bench_test.go @@ -0,0 +1,252 @@ +// Copyright 2011 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Large data benchmark. +// The JSON data is a summary of agl's changes in the +// go, webkit, and chromium open source projects. +// We benchmark converting between the JSON form +// and in-memory data structures. + +package json + +import ( + "bytes" + "compress/gzip" + "io" + "os" + "strings" + "testing" +) + +type codeResponse struct { + Tree *codeNode `json:"tree"` + Username string `json:"username"` +} + +type codeNode struct { + Name string `json:"name"` + Kids []*codeNode `json:"kids"` + CLWeight float64 `json:"cl_weight"` + Touches int `json:"touches"` + MinT int64 `json:"min_t"` + MaxT int64 `json:"max_t"` + MeanT int64 `json:"mean_t"` +} + +var codeJSON []byte +var codeStruct codeResponse + +func codeInit() { + f, err := os.Open("testdata/code.json.gz") + if err != nil { + panic(err) + } + defer f.Close() + gz, err := gzip.NewReader(f) + if err != nil { + panic(err) + } + data, err := io.ReadAll(gz) + if err != nil { + panic(err) + } + + codeJSON = data + + if err := Unmarshal(codeJSON, &codeStruct); err != nil { + panic("unmarshal code.json: " + err.Error()) + } + + if data, err = Marshal(&codeStruct); err != nil { + panic("marshal code.json: " + err.Error()) + } + + if !bytes.Equal(data, codeJSON) { + println("different lengths", len(data), len(codeJSON)) + for i := 0; i < len(data) && i < len(codeJSON); i++ { + if data[i] != codeJSON[i] { + println("re-marshal: changed at byte", i) + println("orig: ", string(codeJSON[i-10:i+10])) + println("new: ", string(data[i-10:i+10])) + break + } + } + panic("re-marshal code.json: different result") + } +} + +func BenchmarkCodeEncoder(b *testing.B) { + if codeJSON == nil { + b.StopTimer() + codeInit() + b.StartTimer() + } + b.RunParallel(func(pb *testing.PB) { + enc := NewEncoder(io.Discard) + for pb.Next() { + if err := enc.Encode(&codeStruct); err != nil { + b.Fatal("Encode:", err) + } + } + }) + b.SetBytes(int64(len(codeJSON))) +} + +func BenchmarkCodeMarshal(b *testing.B) { + if codeJSON == nil { + b.StopTimer() + codeInit() + b.StartTimer() + } + b.RunParallel(func(pb *testing.PB) { + for pb.Next() { + if _, err := Marshal(&codeStruct); err != nil { + b.Fatal("Marshal:", err) + } + } + }) + b.SetBytes(int64(len(codeJSON))) +} + +func BenchmarkCodeDecoder(b *testing.B) { + if codeJSON == nil { + b.StopTimer() + codeInit() + b.StartTimer() + } + b.RunParallel(func(pb *testing.PB) { + var buf bytes.Buffer + dec := NewDecoder(&buf) + var r codeResponse + for pb.Next() { + buf.Write(codeJSON) + // hide EOF + buf.WriteByte('\n') + buf.WriteByte('\n') + buf.WriteByte('\n') + if err := dec.Decode(&r); err != nil { + b.Fatal("Decode:", err) + } + } + }) + b.SetBytes(int64(len(codeJSON))) +} + +func BenchmarkDecoderStream(b *testing.B) { + b.StopTimer() + var buf bytes.Buffer + dec := NewDecoder(&buf) + buf.WriteString(`"` + strings.Repeat("x", 1000000) + `"` + "\n\n\n") + var x any + if err := dec.Decode(&x); err != nil { + b.Fatal("Decode:", err) + } + ones := strings.Repeat(" 1\n", 300000) + "\n\n\n" + b.StartTimer() + for i := range b.N { + if i%300000 == 0 { + buf.WriteString(ones) + } + x = nil + if err := dec.Decode(&x); err != nil || x != 1.0 { + b.Fatalf("Decode: %v after %d", err, i) + } + } +} + +func BenchmarkCodeUnmarshal(b *testing.B) { + if codeJSON == nil { + b.StopTimer() + codeInit() + b.StartTimer() + } + b.RunParallel(func(pb *testing.PB) { + for pb.Next() { + var r codeResponse + if err := Unmarshal(codeJSON, &r); err != nil { + b.Fatal("Unmarshal:", err) + } + } + }) + b.SetBytes(int64(len(codeJSON))) +} + +func BenchmarkCodeUnmarshalReuse(b *testing.B) { + if codeJSON == nil { + b.StopTimer() + codeInit() + b.StartTimer() + } + b.RunParallel(func(pb *testing.PB) { + var r codeResponse + for pb.Next() { + if err := Unmarshal(codeJSON, &r); err != nil { + b.Fatal("Unmarshal:", err) + } + } + }) + // TODO(bcmills): Is there a missing b.SetBytes here? +} + +func BenchmarkUnmarshalString(b *testing.B) { + data := []byte(`"hello, world"`) + b.RunParallel(func(pb *testing.PB) { + var s string + for pb.Next() { + if err := Unmarshal(data, &s); err != nil { + b.Fatal("Unmarshal:", err) + } + } + }) +} + +func BenchmarkUnmarshalFloat64(b *testing.B) { + data := []byte(`3.14`) + b.RunParallel(func(pb *testing.PB) { + var f float64 + for pb.Next() { + if err := Unmarshal(data, &f); err != nil { + b.Fatal("Unmarshal:", err) + } + } + }) +} + +func BenchmarkUnmarshalInt64(b *testing.B) { + data := []byte(`3`) + b.RunParallel(func(pb *testing.PB) { + var x int64 + for pb.Next() { + if err := Unmarshal(data, &x); err != nil { + b.Fatal("Unmarshal:", err) + } + } + }) +} + +func BenchmarkIssue10335(b *testing.B) { + b.ReportAllocs() + j := []byte(`{"a":{ }}`) + b.RunParallel(func(pb *testing.PB) { + var s struct{} + for pb.Next() { + if err := Unmarshal(j, &s); err != nil { + b.Fatal(err) + } + } + }) +} + +func BenchmarkUnmapped(b *testing.B) { + b.ReportAllocs() + j := []byte(`{"s": "hello", "y": 2, "o": {"x": 0}, "a": [1, 99, {"x": 1}]}`) + b.RunParallel(func(pb *testing.PB) { + var s struct{} + for pb.Next() { + if err := Unmarshal(j, &s); err != nil { + b.Fatal(err) + } + } + }) +} diff --git a/decode.go b/decode.go new file mode 100644 index 0000000..3b4b375 --- /dev/null +++ b/decode.go @@ -0,0 +1,1281 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Represents JSON data structure using native Go types: booleans, floats, +// strings, arrays, and maps. + +package json + +import ( + "bytes" + "encoding" + "encoding/base64" + "errors" + "fmt" + "reflect" + "runtime" + "strconv" + "unicode" + "unicode/utf16" + "unicode/utf8" +) + +// Unmarshal parses the JSON-encoded data and stores the result +// in the value pointed to by v. If v is nil or not a pointer, +// Unmarshal returns an InvalidUnmarshalError. +// +// Unmarshal uses the inverse of the encodings that +// Marshal uses, allocating maps, slices, and pointers as necessary, +// with the following additional rules: +// +// To unmarshal JSON into a pointer, Unmarshal first handles the case of +// the JSON being the JSON literal null. In that case, Unmarshal sets +// the pointer to nil. Otherwise, Unmarshal unmarshals the JSON into +// the value pointed at by the pointer. If the pointer is nil, Unmarshal +// allocates a new value for it to point to. +// +// To unmarshal JSON into a value implementing the Unmarshaler interface, +// Unmarshal calls that value's UnmarshalJSON method, including +// when the input is a JSON null. +// Otherwise, if the value implements encoding.TextUnmarshaler +// and the input is a JSON quoted string, Unmarshal calls that value's +// UnmarshalText method with the unquoted form of the string. +// +// To unmarshal JSON into a struct, Unmarshal matches incoming object +// keys to the keys used by Marshal (either the struct field name or its tag), +// preferring an exact match but also accepting a case-insensitive match. +// Unmarshal will only set exported fields of the struct. +// +// To unmarshal JSON into an interface value, +// Unmarshal stores one of these in the interface value: +// +// bool, for JSON booleans +// float64, for JSON numbers +// string, for JSON strings +// []any, for JSON arrays +// map[string]any, for JSON objects +// nil for JSON null +// +// To unmarshal a JSON array into a slice, Unmarshal resets the slice length +// to zero and then appends each element to the slice. +// As a special case, to unmarshal an empty JSON array into a slice, +// Unmarshal replaces the slice with a new empty slice. +// +// To unmarshal a JSON array into a Go array, Unmarshal decodes +// JSON array elements into corresponding Go array elements. +// If the Go array is smaller than the JSON array, +// the additional JSON array elements are discarded. +// If the JSON array is smaller than the Go array, +// the additional Go array elements are set to zero values. +// +// To unmarshal a JSON object into a map, Unmarshal first establishes a map to +// use. If the map is nil, Unmarshal allocates a new map. Otherwise Unmarshal +// reuses the existing map, keeping existing entries. Unmarshal then stores +// key-value pairs from the JSON object into the map. The map's key type must +// either be a string, an integer, or implement encoding.TextUnmarshaler. +// +// If a JSON value is not appropriate for a given target type, +// or if a JSON number overflows the target type, Unmarshal +// skips that field and completes the unmarshaling as best it can. +// If no more serious errors are encountered, Unmarshal returns +// an UnmarshalTypeError describing the earliest such error. In any +// case, it's not guaranteed that all the remaining fields following +// the problematic one will be unmarshaled into the target object. +// +// The JSON null value unmarshals into an interface, map, pointer, or slice +// by setting that Go value to nil. Because null is often used in JSON to mean +// “not present,” unmarshaling a JSON null into any other Go type has no effect +// on the value and produces no error. +// +// When unmarshaling quoted strings, invalid UTF-8 or +// invalid UTF-16 surrogate pairs are not treated as an error. +// Instead, they are replaced by the Unicode replacement +// character U+FFFD. +func Unmarshal(data []byte, v any) error { + // Check for well-formedness. + // Avoids filling out half a data structure + // before discovering a JSON syntax error. + var d decodeState + err := checkValid(data, &d.scan) + if err != nil { + return err + } + + d.init(data) + return d.unmarshal(v) +} + +// Unmarshaler is the interface implemented by types +// that can unmarshal a JSON description of themselves. +// The input can be assumed to be a valid encoding of +// a JSON value. UnmarshalJSON must copy the JSON data +// if it wishes to retain the data after returning. +// +// By convention, to approximate the behavior of Unmarshal itself, +// Unmarshalers implement UnmarshalJSON([]byte("null")) as a no-op. +type Unmarshaler interface { + UnmarshalJSON([]byte) error +} + +// An UnmarshalTypeError describes a JSON value that was +// not appropriate for a value of a specific Go type. +type UnmarshalTypeError struct { + Value string // description of JSON value - "bool", "array", "number -5" + Type reflect.Type // type of Go value it could not be assigned to + Offset int64 // error occurred after reading Offset bytes + Struct string // name of the struct type containing the field + Field string // name of the field holding the Go value +} + +func (e *UnmarshalTypeError) Error() string { + if e.Struct != "" || e.Field != "" { + return "json: cannot unmarshal " + e.Value + " into Go struct field " + e.Struct + "." + e.Field + " of type " + e.Type.String() + } + return "json: cannot unmarshal " + e.Value + " into Go value of type " + e.Type.String() +} + +// An UnmarshalFieldError describes a JSON object key that +// led to an unexported (and therefore unwritable) struct field. +// (No longer used; kept for compatibility.) +type UnmarshalFieldError struct { + Key string + Type reflect.Type + Field reflect.StructField +} + +func (e *UnmarshalFieldError) Error() string { + return "json: cannot unmarshal object key " + strconv.Quote(e.Key) + " into unexported field " + e.Field.Name + " of type " + e.Type.String() +} + +// An InvalidUnmarshalError describes an invalid argument passed to Unmarshal. +// (The argument to Unmarshal must be a non-nil pointer.) +type InvalidUnmarshalError struct { + Type reflect.Type +} + +func (e *InvalidUnmarshalError) Error() string { + if e.Type == nil { + return "json: Unmarshal(nil)" + } + + if e.Type.Kind() != reflect.Ptr { + return "json: Unmarshal(non-pointer " + e.Type.String() + ")" + } + return "json: Unmarshal(nil " + e.Type.String() + ")" +} + +func (d *decodeState) unmarshal(v any) (err error) { + defer func() { + if r := recover(); r != nil { + if _, ok := r.(runtime.Error); ok { + panic(r) + } + err = r.(error) + } + }() + + rv := reflect.ValueOf(v) + if rv.Kind() != reflect.Ptr || rv.IsNil() { + return &InvalidUnmarshalError{reflect.TypeOf(v)} + } + + d.scan.reset() + // We decode rv not rv.Elem because the Unmarshaler interface + // test must be applied at the top level of the value. + d.value(rv) + return d.savedError +} + +// A Number represents a JSON number literal. +type Number string + +// String returns the literal text of the number. +func (n Number) String() string { return string(n) } + +// Float64 returns the number as a float64. +func (n Number) Float64() (float64, error) { + return strconv.ParseFloat(string(n), 64) +} + +// Int64 returns the number as an int64. +func (n Number) Int64() (int64, error) { + return strconv.ParseInt(string(n), 10, 64) +} + +// isValidNumber reports whether s is a valid JSON number literal. +func isValidNumber(s string) bool { + // This function implements the JSON numbers grammar. + // See https://tools.ietf.org/html/rfc7159#section-6 + // and http://json.org/number.gif + + if s == "" { + return false + } + + // Optional - + if s[0] == '-' { + s = s[1:] + if s == "" { + return false + } + } + + // Digits + switch { + default: + return false + + case s[0] == '0': + s = s[1:] + + case '1' <= s[0] && s[0] <= '9': + s = s[1:] + for len(s) > 0 && '0' <= s[0] && s[0] <= '9' { + s = s[1:] + } + } + + // . followed by 1 or more digits. + if len(s) >= 2 && s[0] == '.' && '0' <= s[1] && s[1] <= '9' { + s = s[2:] + for len(s) > 0 && '0' <= s[0] && s[0] <= '9' { + s = s[1:] + } + } + + // e or E followed by an optional - or + and + // 1 or more digits. + if len(s) >= 2 && (s[0] == 'e' || s[0] == 'E') { + s = s[1:] + if s[0] == '+' || s[0] == '-' { + s = s[1:] + if s == "" { + return false + } + } + for len(s) > 0 && '0' <= s[0] && s[0] <= '9' { + s = s[1:] + } + } + + // Make sure we are at the end. + return s == "" +} + +// decodeState represents the state while decoding a JSON value. +type decodeState struct { + data []byte + off int // read offset in data + scan scanner + nextscan scanner // for calls to nextValue + errorContext struct { // provides context for type errors + Struct string + Field string + } + savedError error + useNumber bool + useOrderedObject bool +} + +// errPhase is used for errors that should not happen unless +// there is a bug in the JSON decoder or something is editing +// the data slice while the decoder executes. +var errPhase = errors.New("JSON decoder out of sync - data changing underfoot?") + +func (d *decodeState) init(data []byte) *decodeState { + d.data = data + d.off = 0 + d.savedError = nil + d.errorContext.Struct = "" + d.errorContext.Field = "" + return d +} + +// error aborts the decoding by panicking with err. +func (d *decodeState) error(err error) { + panic(d.addErrorContext(err)) +} + +// saveError saves the first err it is called with, +// for reporting at the end of the unmarshal. +func (d *decodeState) saveError(err error) { + if d.savedError == nil { + d.savedError = d.addErrorContext(err) + } +} + +// addErrorContext returns a new error enhanced with information from d.errorContext. +func (d *decodeState) addErrorContext(err error) error { + if d.errorContext.Struct != "" || d.errorContext.Field != "" { + var e *UnmarshalTypeError + if errors.As(err, &e) { + e.Struct = d.errorContext.Struct + e.Field = d.errorContext.Field + return e + } + } + return err +} + +// next cuts off and returns the next full JSON value in d.data[d.off:]. +// The next value is known to be an object or array, not a literal. +func (d *decodeState) next() []byte { + c := d.data[d.off] + item, rest, err := nextValue(d.data[d.off:], &d.nextscan) + if err != nil { + d.error(err) + } + d.off = len(d.data) - len(rest) + + // Our scanner has seen the opening brace/bracket + // and thinks we're still in the middle of the object. + // invent a closing brace/bracket to get it out. + if c == '{' { + d.scan.step(&d.scan, '}') + } else { + d.scan.step(&d.scan, ']') + } + + return item +} + +// scanWhile processes bytes in d.data[d.off:] until it +// receives a scan code not equal to op. +// It updates d.off and returns the new scan code. +func (d *decodeState) scanWhile(op int) int { + var newOp int + for { + if d.off >= len(d.data) { + newOp = d.scan.eof() + d.off = len(d.data) + 1 // mark processed EOF with len+1 + } else { + c := d.data[d.off] + d.off++ + newOp = d.scan.step(&d.scan, c) + } + if newOp != op { + break + } + } + return newOp +} + +// discardObject and discardArray are dummy data targets +// used by the (*decodeState).value method, which +// accepts a zero reflect.Value to discard a value. +// The (*decodeState).object and (*decodeState).array methods, +// however, require a valid reflect.Value destination. +// These are the target values used when the caller of value +// wants to skip a field. +// +// Because these values refer to zero-sized objects +// and thus can't be mutated, they're safe for concurrent use +// by different goroutines unmarshalling skipped fields. +var ( + discardObject = reflect.ValueOf(struct{}{}) + discardArray = reflect.ValueOf([0]any{}) +) + +// value decodes a JSON value from d.data[d.off:] into the value. +// It updates d.off to point past the decoded value. If v is +// invalid, the JSON value is discarded. +func (d *decodeState) value(v reflect.Value) { + switch op := d.scanWhile(scanSkipSpace); op { + default: + d.error(errPhase) + + case scanBeginArray: + if !v.IsValid() { + v = discardArray + } + d.array(v) + + case scanBeginObject: + if !v.IsValid() { + v = discardObject + } + d.object(v) + + case scanBeginLiteral: + d.literal(v) + } +} + +type unquotedValue struct{} + +// valueQuoted is like value but decodes a +// quoted string literal or literal null into an interface value. +// If it finds anything other than a quoted string literal or null, +// valueQuoted returns unquotedValue{}. +func (d *decodeState) valueQuoted() any { + switch op := d.scanWhile(scanSkipSpace); op { + default: + d.error(errPhase) + + case scanBeginArray: + d.array(reflect.Value{}) + + case scanBeginObject: + d.object(reflect.Value{}) + + case scanBeginLiteral: + switch v := d.literalInterface().(type) { + case nil, string: + return v + } + } + return unquotedValue{} +} + +// indirect walks down v allocating pointers as needed, +// until it gets to a non-pointer. +// if it encounters an Unmarshaler, indirect stops and returns that. +// if decodingNull is true, indirect stops at the last pointer so it can be set to nil. +func (d *decodeState) indirect(v reflect.Value, decodingNull bool) (Unmarshaler, encoding.TextUnmarshaler, reflect.Value) { + // If v is a named type and is addressable, + // start with its address, so that if the type has pointer methods, + // we find them. + if v.Kind() != reflect.Ptr && v.Type().Name() != "" && v.CanAddr() { + v = v.Addr() + } + for { + // Load value from interface, but only if the result will be + // usefully addressable. + if v.Kind() == reflect.Interface && !v.IsNil() { + e := v.Elem() + if e.Kind() == reflect.Ptr && !e.IsNil() && (!decodingNull || e.Elem().Kind() == reflect.Ptr) { + v = e + continue + } + } + + if v.Kind() != reflect.Ptr { + break + } + + if v.Elem().Kind() != reflect.Ptr && decodingNull && v.CanSet() { + break + } + if v.IsNil() { + v.Set(reflect.New(v.Type().Elem())) + } + if v.Type().NumMethod() > 0 { + if u, ok := v.Interface().(Unmarshaler); ok { + return u, nil, reflect.Value{} + } + if !decodingNull { + if u, ok := v.Interface().(encoding.TextUnmarshaler); ok { + return nil, u, reflect.Value{} + } + } + } + v = v.Elem() + } + return nil, nil, v +} + +// array consumes an array from d.data[d.off-1:], decoding into the value v. +// the first byte of the array ('[') has been read already. +func (d *decodeState) array(v reflect.Value) { + // Check for unmarshaler. + u, ut, pv := d.indirect(v, false) + if u != nil { + d.off-- + err := u.UnmarshalJSON(d.next()) + if err != nil { + d.error(err) + } + return + } + if ut != nil { + d.saveError(&UnmarshalTypeError{Value: "array", Type: v.Type(), Offset: int64(d.off)}) + d.off-- + d.next() + return + } + + v = pv + + // Check type of target. + switch v.Kind() { + case reflect.Interface: + if v.NumMethod() == 0 { + // Decoding into nil interface? Switch to non-reflect code. + v.Set(reflect.ValueOf(d.arrayInterface())) + return + } + // Otherwise it's invalid. + fallthrough + default: + d.saveError(&UnmarshalTypeError{Value: "array", Type: v.Type(), Offset: int64(d.off)}) + d.off-- + d.next() + return + case reflect.Array, reflect.Slice: + break + } + + i := 0 + for { + // Look ahead for ] - can only happen on first iteration. + op := d.scanWhile(scanSkipSpace) + if op == scanEndArray { + break + } + + // Back up so d.value can have the byte we just read. + d.off-- + d.scan.undo(op) + + // Get element of array, growing if necessary. + if v.Kind() == reflect.Slice { + // Grow slice if necessary + if i >= v.Cap() { + newcap := v.Cap() + v.Cap()/2 + if newcap < 4 { + newcap = 4 + } + newv := reflect.MakeSlice(v.Type(), v.Len(), newcap) + reflect.Copy(newv, v) + v.Set(newv) + } + if i >= v.Len() { + v.SetLen(i + 1) + } + } + + if i < v.Len() { + // Decode into element. + d.value(v.Index(i)) + } else { + // Ran out of fixed array: skip. + d.value(reflect.Value{}) + } + i++ + + // Next token must be , or ]. + op = d.scanWhile(scanSkipSpace) + if op == scanEndArray { + break + } + if op != scanArrayValue { + d.error(errPhase) + } + } + + if i < v.Len() { + if v.Kind() == reflect.Array { + // Array. Zero the rest. + z := reflect.Zero(v.Type().Elem()) + for ; i < v.Len(); i++ { + v.Index(i).Set(z) + } + } else { + v.SetLen(i) + } + } + if i == 0 && v.Kind() == reflect.Slice { + v.Set(reflect.MakeSlice(v.Type(), 0, 0)) + } +} + +var nullLiteral = []byte("null") +var textUnmarshalerType = reflect.TypeOf(new(encoding.TextUnmarshaler)).Elem() + +// object consumes an object from d.data[d.off-1:], decoding into the value v. +// the first byte ('{') of the object has been read already. +func (d *decodeState) object(v reflect.Value) { + // Check for unmarshaler. + u, ut, pv := d.indirect(v, false) + if u != nil { + d.off-- + err := u.UnmarshalJSON(d.next()) + if err != nil { + d.error(err) + } + return + } + if ut != nil { + d.saveError(&UnmarshalTypeError{Value: "object", Type: v.Type(), Offset: int64(d.off)}) + d.off-- + d.next() // skip over { } in input + return + } + v = pv + + // If we are decoding into a OrderedObject then use an OrderedObject + // just for this particular object even if UseOrderedObject was not called. + forceOrderedObject := v.Type() == orderedObjectType + + // Decoding into nil interface? Switch to non-reflect code. + if (v.Kind() == reflect.Interface && v.NumMethod() == 0) || forceOrderedObject { + v.Set(reflect.ValueOf(d.objectInterface(forceOrderedObject))) + return + } + + // Check type of target: + // struct or + // map[T1]T2 where T1 is string, an integer type, + // or an encoding.TextUnmarshaler + switch v.Kind() { + case reflect.Map: + // Map key must either have string kind, have an integer kind, + // or be an encoding.TextUnmarshaler. + t := v.Type() + switch t.Key().Kind() { + case reflect.String, + reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, + reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + default: + if !reflect.PointerTo(t.Key()).Implements(textUnmarshalerType) { + d.saveError(&UnmarshalTypeError{Value: "object", Type: v.Type(), Offset: int64(d.off)}) + d.off-- + d.next() // skip over { } in input + return + } + } + if v.IsNil() { + v.Set(reflect.MakeMap(t)) + } + case reflect.Struct: + // ok + default: + d.saveError(&UnmarshalTypeError{Value: "object", Type: v.Type(), Offset: int64(d.off)}) + d.off-- + d.next() // skip over { } in input + return + } + + var mapElem reflect.Value + + for { + // Read opening " of string key or closing }. + op := d.scanWhile(scanSkipSpace) + if op == scanEndObject { + // closing } - can only happen on first iteration. + break + } + if op != scanBeginLiteral { + d.error(errPhase) + } + + // Read key. + start := d.off - 1 + op = d.scanWhile(scanContinue) + item := d.data[start : d.off-1] + key, ok := unquoteBytes(item) + if !ok { + d.error(errPhase) + } + + // Figure out field corresponding to key. + var subv reflect.Value + destring := false // whether the value is wrapped in a string to be decoded first + + if v.Kind() == reflect.Map { + elemType := v.Type().Elem() + if !mapElem.IsValid() { + mapElem = reflect.New(elemType).Elem() + } else { + mapElem.Set(reflect.Zero(elemType)) + } + subv = mapElem + } else { + var f *field + fields := cachedTypeFields(v.Type()) + for i := range fields { + ff := &fields[i] + if bytes.Equal(ff.nameBytes, key) { + f = ff + break + } + if f == nil && ff.equalFold(ff.nameBytes, key) { + f = ff + } + } + if f != nil { + subv = v + destring = f.quoted + for _, i := range f.index { + if subv.Kind() == reflect.Ptr { + if subv.IsNil() { + subv.Set(reflect.New(subv.Type().Elem())) + } + subv = subv.Elem() + } + subv = subv.Field(i) + } + d.errorContext.Field = f.name + d.errorContext.Struct = v.Type().Name() + } + } + + // Read : before value. + if op == scanSkipSpace { + op = d.scanWhile(scanSkipSpace) + } + if op != scanObjectKey { + d.error(errPhase) + } + + if destring { + switch qv := d.valueQuoted().(type) { + case nil: + d.literalStore(nullLiteral, subv, false) + case string: + d.literalStore([]byte(qv), subv, true) + default: + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal unquoted value into %v", subv.Type())) + } + } else { + d.value(subv) + } + + // Write value back to map; + // if using struct, subv points into struct already. + if v.Kind() == reflect.Map { + kt := v.Type().Key() + var kv reflect.Value + switch { + case kt.Kind() == reflect.String: + kv = reflect.ValueOf(key).Convert(kt) + case reflect.PointerTo(kt).Implements(textUnmarshalerType): + kv = reflect.New(v.Type().Key()) + d.literalStore(item, kv, true) + kv = kv.Elem() + default: + switch kt.Kind() { + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + s := string(key) + n, err := strconv.ParseInt(s, 10, 64) + if err != nil || reflect.Zero(kt).OverflowInt(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + s, Type: kt, Offset: int64(start + 1)}) + return + } + kv = reflect.ValueOf(n).Convert(kt) + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + s := string(key) + n, err := strconv.ParseUint(s, 10, 64) + if err != nil || reflect.Zero(kt).OverflowUint(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + s, Type: kt, Offset: int64(start + 1)}) + return + } + kv = reflect.ValueOf(n).Convert(kt) + default: + panic("json: Unexpected key type") // should never occur + } + } + v.SetMapIndex(kv, subv) + } + + // Next token must be , or }. + op = d.scanWhile(scanSkipSpace) + if op == scanEndObject { + break + } + if op != scanObjectValue { + d.error(errPhase) + } + + d.errorContext.Struct = "" + d.errorContext.Field = "" + } +} + +// literal consumes a literal from d.data[d.off-1:], decoding into the value v. +// The first byte of the literal has been read already +// (that's how the caller knows it's a literal). +func (d *decodeState) literal(v reflect.Value) { + // All bytes inside literal return scanContinue op code. + start := d.off - 1 + op := d.scanWhile(scanContinue) + + // Scan read one byte too far; back up. + d.off-- + d.scan.undo(op) + + if v.IsValid() { + d.literalStore(d.data[start:d.off], v, false) + } +} + +// convertNumber converts the number literal s to a float64 or a Number +// depending on the setting of d.useNumber. +func (d *decodeState) convertNumber(s string) (any, error) { + if d.useNumber { + return Number(s), nil + } + f, err := strconv.ParseFloat(s, 64) + if err != nil { + return nil, &UnmarshalTypeError{Value: "number " + s, Type: reflect.TypeOf(0.0), Offset: int64(d.off)} + } + return f, nil +} + +var numberType = reflect.TypeOf(Number("")) + +// literalStore decodes a literal stored in item into v. +// +// fromQuoted indicates whether this literal came from unwrapping a +// string from the ",string" struct tag option. this is used only to +// produce more helpful error messages. +func (d *decodeState) literalStore(item []byte, v reflect.Value, fromQuoted bool) { + // Check for unmarshaler. + if len(item) == 0 { + //Empty string given + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + return + } + isNull := item[0] == 'n' // null + u, ut, pv := d.indirect(v, isNull) + if u != nil { + err := u.UnmarshalJSON(item) + if err != nil { + d.error(err) + } + return + } + if ut != nil { + if item[0] != '"' { + if fromQuoted { + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + } else { + var val string + switch item[0] { + case 'n': + val = "null" + case 't', 'f': + val = "bool" + default: + val = "number" + } + d.saveError(&UnmarshalTypeError{Value: val, Type: v.Type(), Offset: int64(d.off)}) + } + return + } + s, ok := unquoteBytes(item) + if !ok { + if fromQuoted { + d.error(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + } else { + d.error(errPhase) + } + } + err := ut.UnmarshalText(s) + if err != nil { + d.error(err) + } + return + } + + v = pv + + switch c := item[0]; c { + case 'n': // null + // The main parser checks that only true and false can reach here, + // but if this was a quoted string input, it could be anything. + if fromQuoted && string(item) != "null" { + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + break + } + switch v.Kind() { + case reflect.Interface, reflect.Ptr, reflect.Map, reflect.Slice: + v.Set(reflect.Zero(v.Type())) + // otherwise, ignore null for primitives/string + default: + } + case 't', 'f': // true, false + value := item[0] == 't' + // The main parser checks that only true and false can reach here, + // but if this was a quoted string input, it could be anything. + if fromQuoted && string(item) != "true" && string(item) != "false" { + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + break + } + switch v.Kind() { + default: + if fromQuoted { + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + } else { + d.saveError(&UnmarshalTypeError{Value: "bool", Type: v.Type(), Offset: int64(d.off)}) + } + case reflect.Bool: + v.SetBool(value) + case reflect.Interface: + if v.NumMethod() == 0 { + v.Set(reflect.ValueOf(value)) + } else { + d.saveError(&UnmarshalTypeError{Value: "bool", Type: v.Type(), Offset: int64(d.off)}) + } + } + + case '"': // string + s, ok := unquoteBytes(item) + if !ok { + if fromQuoted { + d.error(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + } else { + d.error(errPhase) + } + } + switch v.Kind() { + default: + d.saveError(&UnmarshalTypeError{Value: "string", Type: v.Type(), Offset: int64(d.off)}) + case reflect.Slice: + if v.Type().Elem().Kind() != reflect.Uint8 { + d.saveError(&UnmarshalTypeError{Value: "string", Type: v.Type(), Offset: int64(d.off)}) + break + } + b := make([]byte, base64.StdEncoding.DecodedLen(len(s))) + n, err := base64.StdEncoding.Decode(b, s) + if err != nil { + d.saveError(err) + break + } + v.SetBytes(b[:n]) + case reflect.String: + v.SetString(string(s)) + case reflect.Interface: + if v.NumMethod() == 0 { + v.Set(reflect.ValueOf(string(s))) + } else { + d.saveError(&UnmarshalTypeError{Value: "string", Type: v.Type(), Offset: int64(d.off)}) + } + } + + default: // number + if c != '-' && (c < '0' || c > '9') { + if fromQuoted { + d.error(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + } else { + d.error(errPhase) + } + } + s := string(item) + switch v.Kind() { + default: + if v.Kind() == reflect.String && v.Type() == numberType { + v.SetString(s) + if !isValidNumber(s) { + d.error(fmt.Errorf("json: invalid number literal, trying to unmarshal %q into Number", item)) + } + break + } + if fromQuoted { + d.error(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + } else { + d.error(&UnmarshalTypeError{Value: "number", Type: v.Type(), Offset: int64(d.off)}) + } + case reflect.Interface: + n, err := d.convertNumber(s) + if err != nil { + d.saveError(err) + break + } + if v.NumMethod() != 0 { + d.saveError(&UnmarshalTypeError{Value: "number", Type: v.Type(), Offset: int64(d.off)}) + break + } + v.Set(reflect.ValueOf(n)) + + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + n, err := strconv.ParseInt(s, 10, 64) + if err != nil || v.OverflowInt(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + s, Type: v.Type(), Offset: int64(d.off)}) + break + } + v.SetInt(n) + + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + n, err := strconv.ParseUint(s, 10, 64) + if err != nil || v.OverflowUint(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + s, Type: v.Type(), Offset: int64(d.off)}) + break + } + v.SetUint(n) + + case reflect.Float32, reflect.Float64: + n, err := strconv.ParseFloat(s, v.Type().Bits()) + if err != nil || v.OverflowFloat(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + s, Type: v.Type(), Offset: int64(d.off)}) + break + } + v.SetFloat(n) + } + } +} + +// The xxxInterface routines build up a value to be stored +// in an empty interface. They are not strictly necessary, +// but they avoid the weight of reflection in this common case. + +// valueInterface is like value but returns any. +func (d *decodeState) valueInterface() any { + switch d.scanWhile(scanSkipSpace) { + default: + d.error(errPhase) + panic("unreachable") + case scanBeginArray: + return d.arrayInterface() + case scanBeginObject: + return d.objectInterface(false) + case scanBeginLiteral: + return d.literalInterface() + } +} + +// arrayInterface is like array but returns []any. +func (d *decodeState) arrayInterface() []any { + var v = make([]any, 0) + for { + // Look ahead for ] - can only happen on first iteration. + op := d.scanWhile(scanSkipSpace) + if op == scanEndArray { + break + } + + // Back up so d.value can have the byte we just read. + d.off-- + d.scan.undo(op) + + v = append(v, d.valueInterface()) + + // Next token must be , or ]. + op = d.scanWhile(scanSkipSpace) + if op == scanEndArray { + break + } + if op != scanArrayValue { + d.error(errPhase) + } + } + return v +} + +// objectInterface is like object but returns map[string]any or []OrderedObject. +func (d *decodeState) objectInterface(forceOrderedObject bool) any { + m := make(map[string]any) + v := make(OrderedObject, 0) + for { + // Read opening " of string key or closing }. + op := d.scanWhile(scanSkipSpace) + if op == scanEndObject { + // closing } - can only happen on first iteration. + break + } + if op != scanBeginLiteral { + d.error(errPhase) + } + + // Read string key. + start := d.off - 1 + op = d.scanWhile(scanContinue) + item := d.data[start : d.off-1] + key, ok := unquote(item) + if !ok { + d.error(errPhase) + } + + // Read : before value. + if op == scanSkipSpace { + op = d.scanWhile(scanSkipSpace) + } + if op != scanObjectKey { + d.error(errPhase) + } + + // Read value. + if d.useOrderedObject || forceOrderedObject { + v = append(v, Member{Key: key, Value: d.valueInterface()}) + } else { + m[key] = d.valueInterface() + } + + // Next token must be , or }. + op = d.scanWhile(scanSkipSpace) + if op == scanEndObject { + break + } + if op != scanObjectValue { + d.error(errPhase) + } + } + + if d.useOrderedObject || forceOrderedObject { + return v + } + return m +} + +// literalInterface is like literal but returns an interface value. +func (d *decodeState) literalInterface() any { + // All bytes inside literal return scanContinue op code. + start := d.off - 1 + op := d.scanWhile(scanContinue) + + // Scan read one byte too far; back up. + d.off-- + d.scan.undo(op) + item := d.data[start:d.off] + + switch c := item[0]; c { + case 'n': // null + return nil + + case 't', 'f': // true, false + return c == 't' + + case '"': // string + s, ok := unquote(item) + if !ok { + d.error(errPhase) + } + return s + + default: // number + if c != '-' && (c < '0' || c > '9') { + d.error(errPhase) + } + n, err := d.convertNumber(string(item)) + if err != nil { + d.saveError(err) + } + return n + } +} + +// getu4 decodes \uXXXX from the beginning of s, returning the hex value, +// or it returns -1. +func getu4(s []byte) rune { + if len(s) < 6 || s[0] != '\\' || s[1] != 'u' { + return -1 + } + r, err := strconv.ParseUint(string(s[2:6]), 16, 64) + if err != nil { + return -1 + } + return rune(r) +} + +// unquote converts a quoted JSON string literal s into an actual string t. +// The rules are different than for Go, so cannot use strconv.Unquote. +func unquote(s []byte) (t string, ok bool) { + s, ok = unquoteBytes(s) + t = string(s) + return +} + +func unquoteBytes(s []byte) (t []byte, ok bool) { + if len(s) < 2 || s[0] != '"' || s[len(s)-1] != '"' { + return + } + s = s[1 : len(s)-1] + + // Check for unusual characters. If there are none, + // then no unquoting is needed, so return a slice of the + // original bytes. + r := 0 + for r < len(s) { + c := s[r] + if c == '\\' || c == '"' || c < ' ' { + break + } + if c < utf8.RuneSelf { + r++ + continue + } + rr, size := utf8.DecodeRune(s[r:]) + if rr == utf8.RuneError && size == 1 { + break + } + r += size + } + if r == len(s) { + return s, true + } + + b := make([]byte, len(s)+2*utf8.UTFMax) + w := copy(b, s[0:r]) + for r < len(s) { + // Out of room? Can only happen if s is full of + // malformed UTF-8 and we're replacing each + // byte with RuneError. + if w >= len(b)-2*utf8.UTFMax { + nb := make([]byte, (len(b)+utf8.UTFMax)*2) + copy(nb, b[0:w]) + b = nb + } + switch c := s[r]; { + case c == '\\': + r++ + if r >= len(s) { + return + } + switch s[r] { + default: + return + case '"', '\\', '/', '\'': + b[w] = s[r] + r++ + w++ + case 'b': + b[w] = '\b' + r++ + w++ + case 'f': + b[w] = '\f' + r++ + w++ + case 'n': + b[w] = '\n' + r++ + w++ + case 'r': + b[w] = '\r' + r++ + w++ + case 't': + b[w] = '\t' + r++ + w++ + case 'u': + r-- + rr := getu4(s[r:]) + if rr < 0 { + return + } + r += 6 + if utf16.IsSurrogate(rr) { + rr1 := getu4(s[r:]) + if dec := utf16.DecodeRune(rr, rr1); dec != unicode.ReplacementChar { + // A valid pair; consume. + r += 6 + w += utf8.EncodeRune(b[w:], dec) + break + } + // Invalid surrogate; fall back to replacement rune. + rr = unicode.ReplacementChar + } + w += utf8.EncodeRune(b[w:], rr) + } + + // Quote, control characters are invalid. + case c == '"', c < ' ': + return + + // ASCII + case c < utf8.RuneSelf: + b[w] = c + r++ + w++ + + // Coerce to well-formed UTF-8. + default: + rr, size := utf8.DecodeRune(s[r:]) + r += size + w += utf8.EncodeRune(b[w:], rr) + } + } + return b[0:w], true +} diff --git a/decode_test.go b/decode_test.go new file mode 100644 index 0000000..2e5ef37 --- /dev/null +++ b/decode_test.go @@ -0,0 +1,2073 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package json + +import ( + "bytes" + "encoding" + "errors" + "fmt" + "image" + "math" + "math/big" + "net" + "reflect" + "strconv" + "strings" + "testing" + "time" +) + +type T struct { + X string + Y int + Z int `json:"-"` +} + +type U struct { + Alphabet string `json:"alpha"` +} + +type V struct { + F1 any + F2 int32 + F3 Number + F4 *VOuter +} + +type VOuter struct { + V V +} + +// ifaceNumAsFloat64/ifaceNumAsNumber are used to test unmarshaling with and +// without UseNumber. +var ifaceNumAsFloat64 = map[string]any{ + "k1": float64(1), + "k2": "s", + "k3": []any{float64(1), float64(2.0), float64(3e-3)}, + "k4": map[string]any{"kk1": "s", "kk2": float64(2)}, +} + +var ifaceNumAsNumber = map[string]any{ + "k1": Number("1"), + "k2": "s", + "k3": []any{Number("1"), Number("2.0"), Number("3e-3")}, + "k4": map[string]any{"kk1": "s", "kk2": Number("2")}, +} + +type tx struct { + x int //nolint:unused // Used via reflection. +} + +type u8 uint8 + +// A type that can unmarshal itself. + +type unmarshaler struct { + T bool +} + +func (u *unmarshaler) UnmarshalJSON(b []byte) error { + *u = unmarshaler{true} // All we need to see that UnmarshalJSON is called. + return nil +} + +type ustruct struct { + M unmarshaler +} + +type unmarshalerText struct { + A, B string +} + +// MarshalText is needed for re-marshaling tests. +func (u unmarshalerText) MarshalText() ([]byte, error) { + return []byte(u.A + ":" + u.B), nil +} + +func (u *unmarshalerText) UnmarshalText(b []byte) error { + pos := bytes.Index(b, []byte(":")) + if pos == -1 { + return errors.New("missing separator") + } + u.A, u.B = string(b[:pos]), string(b[pos+1:]) + return nil +} + +var _ encoding.TextUnmarshaler = (*unmarshalerText)(nil) + +type ustructText struct { + M unmarshalerText +} + +// u8marshal is an integer type that can marshal/unmarshal itself. +type u8marshal uint8 + +func (u8 u8marshal) MarshalText() ([]byte, error) { + return []byte(fmt.Sprintf("u%d", u8)), nil +} + +var errMissingU8Prefix = errors.New("missing 'u' prefix") + +func (u8 *u8marshal) UnmarshalText(b []byte) error { + if !bytes.HasPrefix(b, []byte{'u'}) { + return errMissingU8Prefix + } + n, err := strconv.Atoi(string(b[1:])) + if err != nil { + return err + } + *u8 = u8marshal(n) + return nil +} + +var _ encoding.TextUnmarshaler = (*u8marshal)(nil) + +var ( + um0, um1 unmarshaler // target2 of unmarshaling + ump = &um1 + umtrue = unmarshaler{true} + umslice = []unmarshaler{{true}} + umslicep = new([]unmarshaler) + umstruct = ustruct{unmarshaler{true}} + + um0T, um1T unmarshalerText // target2 of unmarshaling + umpType = &um1T + umtrueXY = unmarshalerText{"x", "y"} + umsliceXY = []unmarshalerText{{"x", "y"}} + umslicepType = new([]unmarshalerText) + umstructType = new(ustructText) + umstructXY = ustructText{unmarshalerText{"x", "y"}} + + ummapType = map[unmarshalerText]bool{} + ummapXY = map[unmarshalerText]bool{{"x", "y"}: true} +) + +// Test data structures for anonymous fields. + +type Point struct { + Z int +} + +type Top struct { + Level0 int + Embed0 + *Embed0a + *Embed0b `json:"e,omitempty"` // treated as named + Embed0c `json:"-"` // ignored + Loop + Embed0p // has Point with X, Y, used + Embed0q // has Point with Z, used + embed // contains exported field +} + +type Embed0 struct { + Level1a int // overridden by Embed0a's Level1a with json tag + Level1b int // used because Embed0a's Level1b is renamed + Level1c int // used because Embed0a's Level1c is ignored + Level1d int // annihilated by Embed0a's Level1d + Level1e int `json:"x"` // annihilated by Embed0a.Level1e +} + +type Embed0a struct { + Level1a int `json:"Level1a,omitempty"` + Level1b int `json:"LEVEL1B,omitempty"` + Level1c int `json:"-"` + Level1d int // annihilated by Embed0's Level1d + Level1f int `json:"x"` // annihilated by Embed0's Level1e +} + +type Embed0b Embed0 + +type Embed0c Embed0 + +type Embed0p struct { + image.Point +} + +type Embed0q struct { + Point +} + +type embed struct { + Q int +} + +type Loop struct { + Loop1 int `json:",omitempty"` + Loop2 int `json:",omitempty"` + *Loop +} + +// From reflect test: +// The X in S6 and S7 annihilate, but they also block the X in S8.S9. +type S5 struct { + S6 + S7 + S8 +} + +type S6 struct { + X int +} + +type S7 S6 + +type S8 struct { + S9 +} + +type S9 struct { + X int + Y int +} + +// From reflect test: +// The X in S11.S6 and S12.S6 annihilate, but they also block the X in S13.S8.S9. +type S10 struct { + S11 + S12 + S13 +} + +type S11 struct { + S6 +} + +type S12 struct { + S6 +} + +type S13 struct { + S8 +} + +type Ambig struct { + // Given "hello", the first match should win. + First int `json:"HELLO"` + Second int `json:"Hello"` +} + +type XYZ struct { + X any + Y any + Z any +} + +func sliceAddr(x []int) *[]int { return &x } +func mapAddr(x map[string]int) *map[string]int { return &x } + +type byteWithMarshalJSON byte + +func (b byteWithMarshalJSON) MarshalJSON() ([]byte, error) { + return []byte(fmt.Sprintf(`"Z%.2x"`, byte(b))), nil +} + +func (b *byteWithMarshalJSON) UnmarshalJSON(data []byte) error { + if len(data) != 5 || data[0] != '"' || data[1] != 'Z' || data[4] != '"' { + return fmt.Errorf("bad quoted string") + } + i, err := strconv.ParseInt(string(data[2:4]), 16, 8) + if err != nil { + return fmt.Errorf("bad hex") + } + *b = byteWithMarshalJSON(i) + return nil +} + +type byteWithPtrMarshalJSON byte + +func (b *byteWithPtrMarshalJSON) MarshalJSON() ([]byte, error) { + return byteWithMarshalJSON(*b).MarshalJSON() +} + +func (b *byteWithPtrMarshalJSON) UnmarshalJSON(data []byte) error { + return (*byteWithMarshalJSON)(b).UnmarshalJSON(data) +} + +type byteWithMarshalText byte + +func (b byteWithMarshalText) MarshalText() ([]byte, error) { + return []byte(fmt.Sprintf(`Z%.2x`, byte(b))), nil +} + +func (b *byteWithMarshalText) UnmarshalText(data []byte) error { + if len(data) != 3 || data[0] != 'Z' { + return fmt.Errorf("bad quoted string") + } + i, err := strconv.ParseInt(string(data[1:3]), 16, 8) + if err != nil { + return fmt.Errorf("bad hex") + } + *b = byteWithMarshalText(i) + return nil +} + +type byteWithPtrMarshalText byte + +func (b *byteWithPtrMarshalText) MarshalText() ([]byte, error) { + return byteWithMarshalText(*b).MarshalText() +} + +func (b *byteWithPtrMarshalText) UnmarshalText(data []byte) error { + return (*byteWithMarshalText)(b).UnmarshalText(data) +} + +type intWithMarshalJSON int + +func (b intWithMarshalJSON) MarshalJSON() ([]byte, error) { + return []byte(fmt.Sprintf(`"Z%.2x"`, int(b))), nil +} + +func (b *intWithMarshalJSON) UnmarshalJSON(data []byte) error { + if len(data) != 5 || data[0] != '"' || data[1] != 'Z' || data[4] != '"' { + return fmt.Errorf("bad quoted string") + } + i, err := strconv.ParseInt(string(data[2:4]), 16, 8) + if err != nil { + return fmt.Errorf("bad hex") + } + *b = intWithMarshalJSON(i) + return nil +} + +type intWithPtrMarshalJSON int + +func (b *intWithPtrMarshalJSON) MarshalJSON() ([]byte, error) { + return intWithMarshalJSON(*b).MarshalJSON() +} + +func (b *intWithPtrMarshalJSON) UnmarshalJSON(data []byte) error { + return (*intWithMarshalJSON)(b).UnmarshalJSON(data) +} + +type intWithMarshalText int + +func (b intWithMarshalText) MarshalText() ([]byte, error) { + return []byte(fmt.Sprintf(`Z%.2x`, int(b))), nil +} + +func (b *intWithMarshalText) UnmarshalText(data []byte) error { + if len(data) != 3 || data[0] != 'Z' { + return fmt.Errorf("bad quoted string") + } + i, err := strconv.ParseInt(string(data[1:3]), 16, 8) + if err != nil { + return fmt.Errorf("bad hex") + } + *b = intWithMarshalText(i) + return nil +} + +type intWithPtrMarshalText int + +func (b *intWithPtrMarshalText) MarshalText() ([]byte, error) { + return intWithMarshalText(*b).MarshalText() +} + +func (b *intWithPtrMarshalText) UnmarshalText(data []byte) error { + return (*intWithMarshalText)(b).UnmarshalText(data) +} + +type unmarshalTest struct { + in string + ptr any + out any + err error + useNumber bool + golden bool + useOrderedObject bool +} + +type B struct { + B bool `json:",string"` +} + +var unmarshalTests = []unmarshalTest{ + // basic types + {in: `true`, ptr: new(bool), out: true}, + {in: `1`, ptr: new(int), out: 1}, + {in: `1.2`, ptr: new(float64), out: 1.2}, + {in: `-5`, ptr: new(int16), out: int16(-5)}, + {in: `2`, ptr: new(Number), out: Number("2"), useNumber: true}, + {in: `2`, ptr: new(Number), out: Number("2")}, + {in: `2`, ptr: new(any), out: float64(2.0)}, + {in: `2`, ptr: new(any), out: Number("2"), useNumber: true}, + {in: `"a\u1234"`, ptr: new(string), out: "a\u1234"}, + {in: `"http:\/\/"`, ptr: new(string), out: "http://"}, + {in: `"g-clef: \uD834\uDD1E"`, ptr: new(string), out: "g-clef: \U0001D11E"}, + {in: `"invalid: \uD834x\uDD1E"`, ptr: new(string), out: "invalid: \uFFFDx\uFFFD"}, + {in: "null", ptr: new(any), out: nil}, + {in: `{"X": [1,2,3], "Y": 4}`, ptr: new(T), out: T{Y: 4}, err: &UnmarshalTypeError{"array", reflect.TypeOf(""), 7, "T", "X"}}, + {in: `{"x": 1}`, ptr: new(tx), out: tx{}}, + {in: `{"F1":1,"F2":2,"F3":3}`, ptr: new(V), out: V{F1: float64(1), F2: int32(2), F3: Number("3")}}, + {in: `{"F1":1,"F2":2,"F3":3}`, ptr: new(V), out: V{F1: Number("1"), F2: int32(2), F3: Number("3")}, useNumber: true}, + {in: `{"k1":1,"k2":"s","k3":[1,2.0,3e-3],"k4":{"kk1":"s","kk2":2}}`, ptr: new(any), out: ifaceNumAsFloat64}, + {in: `{"k1":1,"k2":"s","k3":[1,2.0,3e-3],"k4":{"kk1":"s","kk2":2}}`, ptr: new(any), out: ifaceNumAsNumber, useNumber: true}, + + // raw values with whitespace + {in: "\n true ", ptr: new(bool), out: true}, + {in: "\t 1 ", ptr: new(int), out: 1}, + {in: "\r 1.2 ", ptr: new(float64), out: 1.2}, + {in: "\t -5 \n", ptr: new(int16), out: int16(-5)}, + {in: "\t \"a\\u1234\" \n", ptr: new(string), out: "a\u1234"}, + + // Z has a "-" tag. + {in: `{"Y": 1, "Z": 2}`, ptr: new(T), out: T{Y: 1}}, + + {in: `{"alpha": "abc", "alphabet": "xyz"}`, ptr: new(U), out: U{Alphabet: "abc"}}, + {in: `{"alpha": "abc"}`, ptr: new(U), out: U{Alphabet: "abc"}}, + {in: `{"alphabet": "xyz"}`, ptr: new(U), out: U{}}, + + // syntax errors + {in: `{"X": "foo", "Y"}`, err: &SyntaxError{"invalid character '}' after object key", 17}}, + {in: `[1, 2, 3+]`, err: &SyntaxError{"invalid character '+' after array element", 9}}, + {in: `{"X":12x}`, err: &SyntaxError{"invalid character 'x' after object key:value pair", 8}, useNumber: true}, + + // raw value errors + {in: "\x01 42", err: &SyntaxError{"invalid character '\\x01' looking for beginning of value", 1}}, + {in: " 42 \x01", err: &SyntaxError{"invalid character '\\x01' after top-level value", 5}}, + {in: "\x01 true", err: &SyntaxError{"invalid character '\\x01' looking for beginning of value", 1}}, + {in: " false \x01", err: &SyntaxError{"invalid character '\\x01' after top-level value", 8}}, + {in: "\x01 1.2", err: &SyntaxError{"invalid character '\\x01' looking for beginning of value", 1}}, + {in: " 3.4 \x01", err: &SyntaxError{"invalid character '\\x01' after top-level value", 6}}, + {in: "\x01 \"string\"", err: &SyntaxError{"invalid character '\\x01' looking for beginning of value", 1}}, + {in: " \"string\" \x01", err: &SyntaxError{"invalid character '\\x01' after top-level value", 11}}, + + // array tests + {in: `[1, 2, 3]`, ptr: new([3]int), out: [3]int{1, 2, 3}}, + {in: `[1, 2, 3]`, ptr: new([1]int), out: [1]int{1}}, + {in: `[1, 2, 3]`, ptr: new([5]int), out: [5]int{1, 2, 3, 0, 0}}, + + // empty array to interface test + {in: `[]`, ptr: new([]any), out: []any{}}, + {in: `null`, ptr: new([]any), out: []any(nil)}, + {in: `{"T":[]}`, ptr: new(map[string]any), out: map[string]any{"T": []any{}}}, + {in: `{"T":null}`, ptr: new(map[string]any), out: map[string]any{"T": any(nil)}}, + + // composite tests + {in: allValueIndent, ptr: new(All), out: allValue}, + {in: allValueCompact, ptr: new(All), out: allValue}, + {in: allValueIndent, ptr: new(*All), out: &allValue}, + {in: allValueCompact, ptr: new(*All), out: &allValue}, + {in: pallValueIndent, ptr: new(All), out: pallValue}, + {in: pallValueCompact, ptr: new(All), out: pallValue}, + {in: pallValueIndent, ptr: new(*All), out: &pallValue}, + {in: pallValueCompact, ptr: new(*All), out: &pallValue}, + + // unmarshal interface test + {in: `{"T":false}`, ptr: &um0, out: umtrue}, // use "false" so test will fail if custom unmarshaler is not called + {in: `{"T":false}`, ptr: &ump, out: &umtrue}, + {in: `[{"T":false}]`, ptr: &umslice, out: umslice}, + {in: `[{"T":false}]`, ptr: &umslicep, out: &umslice}, + {in: `{"M":{"T":"x:y"}}`, ptr: &umstruct, out: umstruct}, + + // UnmarshalText interface test + {in: `"x:y"`, ptr: &um0T, out: umtrueXY}, + {in: `"x:y"`, ptr: &umpType, out: &umtrueXY}, + {in: `["x:y"]`, ptr: &umsliceXY, out: umsliceXY}, + {in: `["x:y"]`, ptr: &umslicepType, out: &umsliceXY}, + {in: `{"M":"x:y"}`, ptr: umstructType, out: umstructXY}, + + // integer-keyed map test + { + in: `{"-1":"a","0":"b","1":"c"}`, + ptr: new(map[int]string), + out: map[int]string{-1: "a", 0: "b", 1: "c"}, + }, + { + in: `{"0":"a","10":"c","9":"b"}`, + ptr: new(map[u8]string), + out: map[u8]string{0: "a", 9: "b", 10: "c"}, + }, + { + in: `{"-9223372036854775808":"min","9223372036854775807":"max"}`, + ptr: new(map[int64]string), + out: map[int64]string{math.MinInt64: "min", math.MaxInt64: "max"}, + }, + { + in: `{"18446744073709551615":"max"}`, + ptr: new(map[uint64]string), + out: map[uint64]string{math.MaxUint64: "max"}, + }, + { + in: `{"0":false,"10":true}`, + ptr: new(map[uintptr]bool), + out: map[uintptr]bool{0: false, 10: true}, + }, + + // Check that MarshalText and UnmarshalText take precedence + // over default integer handling in map keys. + { + in: `{"u2":4}`, + ptr: new(map[u8marshal]int), + out: map[u8marshal]int{2: 4}, + }, + { + in: `{"2":4}`, + ptr: new(map[u8marshal]int), + err: errMissingU8Prefix, + }, + + // integer-keyed map errors + { + in: `{"abc":"abc"}`, + ptr: new(map[int]string), + err: &UnmarshalTypeError{Value: "number abc", Type: reflect.TypeOf(0), Offset: 2}, + }, + { + in: `{"256":"abc"}`, + ptr: new(map[uint8]string), + err: &UnmarshalTypeError{Value: "number 256", Type: reflect.TypeOf(uint8(0)), Offset: 2}, + }, + { + in: `{"128":"abc"}`, + ptr: new(map[int8]string), + err: &UnmarshalTypeError{Value: "number 128", Type: reflect.TypeOf(int8(0)), Offset: 2}, + }, + { + in: `{"-1":"abc"}`, + ptr: new(map[uint8]string), + err: &UnmarshalTypeError{Value: "number -1", Type: reflect.TypeOf(uint8(0)), Offset: 2}, + }, + + // Map keys can be encoding.TextUnmarshalers. + {in: `{"x:y":true}`, ptr: &ummapType, out: ummapXY}, + // If multiple values for the same key exists, only the most recent value is used. + {in: `{"x:y":false,"x:y":true}`, ptr: &ummapType, out: ummapXY}, + + // Overwriting of data. + // This is different from package xml, but it's what we've always done. + // Now documented and tested. + {in: `[2]`, ptr: sliceAddr([]int{1}), out: []int{2}}, + {in: `{"key": 2}`, ptr: mapAddr(map[string]int{"old": 0, "key": 1}), out: map[string]int{"key": 2}}, + + { + in: `{ + "Level0": 1, + "Level1b": 2, + "Level1c": 3, + "x": 4, + "Level1a": 5, + "LEVEL1B": 6, + "e": { + "Level1a": 8, + "Level1b": 9, + "Level1c": 10, + "Level1d": 11, + "x": 12 + }, + "Loop1": 13, + "Loop2": 14, + "X": 15, + "Y": 16, + "Z": 17, + "Q": 18 + }`, + ptr: new(Top), + out: Top{ + Level0: 1, + Embed0: Embed0{ + Level1b: 2, + Level1c: 3, + }, + Embed0a: &Embed0a{ + Level1a: 5, + Level1b: 6, + }, + Embed0b: &Embed0b{ + Level1a: 8, + Level1b: 9, + Level1c: 10, + Level1d: 11, + Level1e: 12, + }, + Loop: Loop{ + Loop1: 13, + Loop2: 14, + }, + Embed0p: Embed0p{ + Point: image.Point{X: 15, Y: 16}, + }, + Embed0q: Embed0q{ + Point: Point{Z: 17}, + }, + embed: embed{ + Q: 18, + }, + }, + }, + { + in: `{"hello": 1}`, + ptr: new(Ambig), + out: Ambig{First: 1}, + }, + + { + in: `{"X": 1,"Y":2}`, + ptr: new(S5), + out: S5{S8: S8{S9: S9{Y: 2}}}, + }, + { + in: `{"X": 1,"Y":2}`, + ptr: new(S10), + out: S10{S13: S13{S8: S8{S9: S9{Y: 2}}}}, + }, + + // invalid UTF-8 is coerced to valid UTF-8. + { + in: "\"hello\xffworld\"", + ptr: new(string), + out: "hello\ufffdworld", + }, + { + in: "\"hello\xc2\xc2world\"", + ptr: new(string), + out: "hello\ufffd\ufffdworld", + }, + { + in: "\"hello\xc2\xffworld\"", + ptr: new(string), + out: "hello\ufffd\ufffdworld", + }, + { + in: "\"hello\\ud800world\"", + ptr: new(string), + out: "hello\ufffdworld", + }, + { + in: "\"hello\\ud800\\ud800world\"", + ptr: new(string), + out: "hello\ufffd\ufffdworld", + }, + { + in: "\"hello\\ud800\\ud800world\"", + ptr: new(string), + out: "hello\ufffd\ufffdworld", + }, + { + in: "\"hello\xed\xa0\x80\xed\xb0\x80world\"", + ptr: new(string), + out: "hello\ufffd\ufffd\ufffd\ufffd\ufffd\ufffdworld", + }, + + // Used to be issue 8305, but time.Time implements encoding.TextUnmarshaler so this works now. + { + in: `{"2009-11-10T23:00:00Z": "hello world"}`, + ptr: &map[time.Time]string{}, + out: map[time.Time]string{time.Date(2009, 11, 10, 23, 0, 0, 0, time.UTC): "hello world"}, + }, + + // issue 8305 + { + in: `{"2009-11-10T23:00:00Z": "hello world"}`, + ptr: &map[Point]string{}, + err: &UnmarshalTypeError{Value: "object", Type: reflect.TypeOf(map[Point]string{}), Offset: 1}, + }, + { + in: `{"asdf": "hello world"}`, + ptr: &map[unmarshaler]string{}, + err: &UnmarshalTypeError{Value: "object", Type: reflect.TypeOf(map[unmarshaler]string{}), Offset: 1}, + }, + + // related to issue 13783. + // Go 1.7 changed marshaling a slice of typed byte to use the methods on the byte type, + // similar to marshaling a slice of typed int. + // These tests check that, assuming the byte type also has valid decoding methods, + // either the old base64 string encoding or the new per-element encoding can be + // successfully unmarshaled. The custom unmarshalers were accessible in earlier + // versions of Go, even though the custom marshaler was not. + { + in: `"AQID"`, + ptr: new([]byteWithMarshalJSON), + out: []byteWithMarshalJSON{1, 2, 3}, + }, + { + in: `["Z01","Z02","Z03"]`, + ptr: new([]byteWithMarshalJSON), + out: []byteWithMarshalJSON{1, 2, 3}, + golden: true, + }, + { + in: `"AQID"`, + ptr: new([]byteWithMarshalText), + out: []byteWithMarshalText{1, 2, 3}, + }, + { + in: `["Z01","Z02","Z03"]`, + ptr: new([]byteWithMarshalText), + out: []byteWithMarshalText{1, 2, 3}, + golden: true, + }, + { + in: `"AQID"`, + ptr: new([]byteWithPtrMarshalJSON), + out: []byteWithPtrMarshalJSON{1, 2, 3}, + }, + { + in: `["Z01","Z02","Z03"]`, + ptr: new([]byteWithPtrMarshalJSON), + out: []byteWithPtrMarshalJSON{1, 2, 3}, + golden: true, + }, + { + in: `"AQID"`, + ptr: new([]byteWithPtrMarshalText), + out: []byteWithPtrMarshalText{1, 2, 3}, + }, + { + in: `["Z01","Z02","Z03"]`, + ptr: new([]byteWithPtrMarshalText), + out: []byteWithPtrMarshalText{1, 2, 3}, + golden: true, + }, + + // ints work with the marshaler but not the base64 []byte case + { + in: `["Z01","Z02","Z03"]`, + ptr: new([]intWithMarshalJSON), + out: []intWithMarshalJSON{1, 2, 3}, + golden: true, + }, + { + in: `["Z01","Z02","Z03"]`, + ptr: new([]intWithMarshalText), + out: []intWithMarshalText{1, 2, 3}, + golden: true, + }, + { + in: `["Z01","Z02","Z03"]`, + ptr: new([]intWithPtrMarshalJSON), + out: []intWithPtrMarshalJSON{1, 2, 3}, + golden: true, + }, + { + in: `["Z01","Z02","Z03"]`, + ptr: new([]intWithPtrMarshalText), + out: []intWithPtrMarshalText{1, 2, 3}, + golden: true, + }, + + {in: `0.000001`, ptr: new(float64), out: 0.000001, golden: true}, + {in: `1e-7`, ptr: new(float64), out: 1e-7, golden: true}, + {in: `100000000000000000000`, ptr: new(float64), out: 100000000000000000000.0, golden: true}, + {in: `1e+21`, ptr: new(float64), out: 1e21, golden: true}, + {in: `-0.000001`, ptr: new(float64), out: -0.000001, golden: true}, + {in: `-1e-7`, ptr: new(float64), out: -1e-7, golden: true}, + {in: `-100000000000000000000`, ptr: new(float64), out: -100000000000000000000.0, golden: true}, + {in: `-1e+21`, ptr: new(float64), out: -1e21, golden: true}, + {in: `999999999999999900000`, ptr: new(float64), out: 999999999999999900000.0, golden: true}, + {in: `9007199254740992`, ptr: new(float64), out: 9007199254740992.0, golden: true}, + {in: `9007199254740993`, ptr: new(float64), out: 9007199254740992.0, golden: false}, + + { + in: `{"V": {"F2": "hello"}}`, + ptr: new(VOuter), + err: &UnmarshalTypeError{ + Value: "string", + Struct: "V", + Field: "F2", + Type: reflect.TypeOf(int32(0)), + Offset: 20, + }, + }, + { + in: `{"V": {"F4": {}, "F2": "hello"}}`, + ptr: new(VOuter), + err: &UnmarshalTypeError{ + Value: "string", + Struct: "V", + Field: "F2", + Type: reflect.TypeOf(int32(0)), + Offset: 30, + }, + }, + + // issue 15146. + // invalid inputs in wrongStringTests below. + {in: `{"B":"true"}`, ptr: new(B), out: B{true}, golden: true}, + {in: `{"B":"false"}`, ptr: new(B), out: B{false}, golden: true}, + {in: `{"B": "maybe"}`, ptr: new(B), err: errors.New(`json: invalid use of ,string struct tag, trying to unmarshal "maybe" into bool`)}, + {in: `{"B": "tru"}`, ptr: new(B), err: errors.New(`json: invalid use of ,string struct tag, trying to unmarshal "tru" into bool`)}, + {in: `{"B": "False"}`, ptr: new(B), err: errors.New(`json: invalid use of ,string struct tag, trying to unmarshal "False" into bool`)}, + {in: `{"B": "null"}`, ptr: new(B), out: B{false}}, + {in: `{"B": "nul"}`, ptr: new(B), err: errors.New(`json: invalid use of ,string struct tag, trying to unmarshal "nul" into bool`)}, + + // OrderedObject tests - into any + {in: `{"a":"1","b":2,"c":3}`, ptr: new(any), useOrderedObject: true, + out: OrderedObject{{"a", "1"}, {"b", float64(2)}, {"c", float64(3)}}}, + {in: `[{"a":"1","b":2}]`, ptr: new(any), useOrderedObject: true, + out: []any{OrderedObject{{"a", "1"}, {"b", float64(2)}}}}, + {in: `{"a":null,"b": {"c":true} }`, ptr: new(any), useOrderedObject: true, + out: OrderedObject{{"a", nil}, {"b", OrderedObject{{"c", true}}}}}, + {in: `{"T":[]}`, ptr: new(any), useOrderedObject: true, + out: OrderedObject{{Key: "T", Value: []any{}}}}, + {in: `{"T":null}`, ptr: new(any), useOrderedObject: true, + out: OrderedObject{{Key: "T", Value: nil}}}, + + // OrderedObject tests - into map[string]any + {in: `{"a":"1","b":2,"c":3}`, ptr: new(map[string]any), useOrderedObject: true, + out: map[string]any{"a": "1", "b": float64(2), "c": float64(3)}}, + {in: `{"a":null,"b": {"c":true} }`, ptr: new(map[string]any), useOrderedObject: true, + out: map[string]any{"a": nil, "b": OrderedObject{{"c", true}}}}, + {in: `{"T":[]}`, ptr: new(map[string]any), useOrderedObject: true, + out: map[string]any{"T": []any{}}}, + {in: `{"T":null}`, ptr: new(map[string]any), useOrderedObject: true, + out: map[string]any{"T": nil}}, + + // OrderedObject tests - into OrderedObject + {in: `{"a":"1","b":2,"c":3}`, ptr: new(OrderedObject), useOrderedObject: true, + out: OrderedObject{{"a", "1"}, {"b", float64(2)}, {"c", float64(3)}}}, + {in: `{"a":"1","b":2,"c":3}`, ptr: new(OrderedObject), + out: OrderedObject{{"a", "1"}, {"b", float64(2)}, {"c", float64(3)}}}, + + {in: `{"a":null,"b": {"c":true} }`, ptr: new(OrderedObject), useOrderedObject: true, + out: OrderedObject{{"a", nil}, {"b", OrderedObject{{"c", true}}}}}, + {in: `{"a":null,"b": {"c":true} }`, ptr: new(OrderedObject), + out: OrderedObject{{"a", nil}, {"b", map[string]any{"c": true}}}}, + + // OrderedObject tests -into []OrderedObject + {in: `[{"a":"1","b":2},{"c":3}]`, ptr: &[]OrderedObject{}, + out: []OrderedObject{{{"a", "1"}, {"b", float64(2)}}, {{"c", float64(3)}}}}, +} + +func TestMarshal(t *testing.T) { + b, err := Marshal(allValue) + if err != nil { + t.Fatalf("Marshal allValue: %v", err) + } + if string(b) != allValueCompact { + t.Errorf("Marshal allValueCompact") + diff(t, b, []byte(allValueCompact)) + return + } + + b, err = Marshal(pallValue) + if err != nil { + t.Fatalf("Marshal pallValue: %v", err) + } + if string(b) != pallValueCompact { + t.Errorf("Marshal pallValueCompact") + diff(t, b, []byte(pallValueCompact)) + return + } +} + +var badUTF8 = []struct { + in, out string +}{ + {"hello\xffworld", `"hello\u00FFworld"`}, + {"", `""`}, + {"\xff", `"\u00FF"`}, + {"\xff\xff", `"\u00FF\u00FF"`}, + {"a\xffb", `"a\u00FFb"`}, + {"\xe6\x97\xa5\xe6\x9c\xac\xff\xaa\x9e", `"\u65E5\u672C\u00FF\u00AA\u009E"`}, +} + +func TestMarshalBadUTF8(t *testing.T) { + for _, tt := range badUTF8 { + b, err := Marshal(tt.in) + if string(b) != tt.out || err != nil { + t.Errorf("Marshal(%q) = %#q, %v, want %#q, nil", tt.in, b, err, tt.out) + } + } +} + +func TestMarshalNumberZeroVal(t *testing.T) { + var n Number + out, err := Marshal(n) + if err != nil { + t.Fatal(err) + } + outStr := string(out) + if outStr != "0" { + t.Fatalf("Invalid zero val for Number: %q", outStr) + } +} + +func TestMarshalEmbeds(t *testing.T) { + top := &Top{ + Level0: 1, + Embed0: Embed0{ + Level1b: 2, + Level1c: 3, + }, + Embed0a: &Embed0a{ + Level1a: 5, + Level1b: 6, + }, + Embed0b: &Embed0b{ + Level1a: 8, + Level1b: 9, + Level1c: 10, + Level1d: 11, + Level1e: 12, + }, + Loop: Loop{ + Loop1: 13, + Loop2: 14, + }, + Embed0p: Embed0p{ + Point: image.Point{X: 15, Y: 16}, + }, + Embed0q: Embed0q{ + Point: Point{Z: 17}, + }, + embed: embed{ + Q: 18, + }, + } + b, err := Marshal(top) + if err != nil { + t.Fatal(err) + } + want := "{\"Level0\":1,\"Level1b\":2,\"Level1c\":3,\"Level1a\":5,\"LEVEL1B\":6,\"e\":{\"Level1a\":8,\"Level1b\":9,\"Level1c\":10,\"Level1d\":11,\"x\":12},\"Loop1\":13,\"Loop2\":14,\"X\":15,\"Y\":16,\"Z\":17,\"Q\":18}" + if string(b) != want { + t.Errorf("Wrong marshal result.\n got: %q\nwant: %q", b, want) + } +} + +func TestUnmarshal(t *testing.T) { + for i, tt := range unmarshalTests { + var scan scanner + in := []byte(tt.in) + if err := checkValid(in, &scan); err != nil { + if !reflect.DeepEqual(err, tt.err) { + t.Errorf("#%d: checkValid: %#v", i, err) + continue + } + } + if tt.ptr == nil { + continue + } + + // v = new(right-type) + v := reflect.New(reflect.TypeOf(tt.ptr).Elem()) + dec := NewDecoder(bytes.NewReader(in)) + if tt.useNumber { + dec.UseNumber() + } + if tt.useOrderedObject { + dec.UseOrderedObject() + } + if err := dec.Decode(v.Interface()); !reflect.DeepEqual(err, tt.err) { + t.Errorf("#%d: %v, want %v", i, err, tt.err) + continue + } else if err != nil { + continue + } + if !reflect.DeepEqual(v.Elem().Interface(), tt.out) { + t.Errorf("#%d: mismatch\nhave: %#+v\nwant: %#+v", i, v.Elem().Interface(), tt.out) + data, _ := Marshal(v.Elem().Interface()) + println(string(data)) + data, _ = Marshal(tt.out) + println(string(data)) + continue + } + + // Check round trip also decodes correctly. + if tt.err == nil { + enc, err := Marshal(v.Interface()) + if err != nil { + t.Errorf("#%d: error re-marshaling: %v", i, err) + continue + } + if tt.golden && !bytes.Equal(enc, in) { + t.Errorf("#%d: remarshal mismatch:\nhave: %s\nwant: %s", i, enc, in) + } + vv := reflect.New(reflect.TypeOf(tt.ptr).Elem()) + dec = NewDecoder(bytes.NewReader(enc)) + if tt.useNumber { + dec.UseNumber() + } + if tt.useOrderedObject { + dec.UseOrderedObject() + } + if err := dec.Decode(vv.Interface()); err != nil { + t.Errorf("#%d: error re-unmarshaling %#q: %v", i, enc, err) + continue + } + if !reflect.DeepEqual(v.Elem().Interface(), vv.Elem().Interface()) { + t.Errorf("#%d: mismatch\nhave: %#+v\nwant: %#+v", i, v.Elem().Interface(), vv.Elem().Interface()) + t.Errorf(" In: %q", strings.Map(noSpace, string(in))) + t.Errorf("Marshal: %q", strings.Map(noSpace, string(enc))) + continue + } + } + } +} + +func TestUnmarshalMarshal(t *testing.T) { + initBig() + var v any + if err := Unmarshal(jsonBig, &v); err != nil { + t.Fatalf("Unmarshal: %v", err) + } + b, err := Marshal(v) + if err != nil { + t.Fatalf("Marshal: %v", err) + } + if !bytes.Equal(jsonBig, b) { + t.Errorf("Marshal jsonBig") + diff(t, b, jsonBig) + return + } +} + +var numberTests = []struct { + in string + i int64 + intErr string + f float64 + floatErr string +}{ + {in: "-1.23e1", intErr: "strconv.ParseInt: parsing \"-1.23e1\": invalid syntax", f: -1.23e1}, + {in: "-12", i: -12, f: -12.0}, + {in: "1e1000", intErr: "strconv.ParseInt: parsing \"1e1000\": invalid syntax", floatErr: "strconv.ParseFloat: parsing \"1e1000\": value out of range"}, +} + +// Independent of Decode, basic coverage of the accessors in Number. +func TestNumberAccessors(t *testing.T) { + for _, tt := range numberTests { + n := Number(tt.in) + if s := n.String(); s != tt.in { + t.Errorf("Number(%q).String() is %q", tt.in, s) + } + if i, err := n.Int64(); err == nil && tt.intErr == "" && i != tt.i { + t.Errorf("Number(%q).Int64() is %d", tt.in, i) + } else if (err == nil && tt.intErr != "") || (err != nil && err.Error() != tt.intErr) { + t.Errorf("Number(%q).Int64() wanted error %q but got: %v", tt.in, tt.intErr, err) + } + if f, err := n.Float64(); err == nil && tt.floatErr == "" && f != tt.f { + t.Errorf("Number(%q).Float64() is %g", tt.in, f) + } else if (err == nil && tt.floatErr != "") || (err != nil && err.Error() != tt.floatErr) { + t.Errorf("Number(%q).Float64() wanted error %q but got: %v", tt.in, tt.floatErr, err) + } + } +} + +func TestLargeByteSlice(t *testing.T) { + s0 := make([]byte, 2000) + for i := range s0 { + s0[i] = byte(i) + } + b, err := Marshal(s0) + if err != nil { + t.Fatalf("Marshal: %v", err) + } + var s1 []byte + if err := Unmarshal(b, &s1); err != nil { + t.Fatalf("Unmarshal: %v", err) + } + if !bytes.Equal(s0, s1) { + t.Errorf("Marshal large byte slice") + diff(t, s0, s1) + } +} + +type Xint struct { + X int +} + +func TestUnmarshalInterface(t *testing.T) { + var xint Xint + var i any = &xint + if err := Unmarshal([]byte(`{"X":1}`), &i); err != nil { + t.Fatalf("Unmarshal: %v", err) + } + if xint.X != 1 { + t.Fatalf("Did not write to xint") + } +} + +func TestUnmarshalPtrPtr(t *testing.T) { + var xint Xint + pxint := &xint + if err := Unmarshal([]byte(`{"X":1}`), &pxint); err != nil { + t.Fatalf("Unmarshal: %v", err) + } + if xint.X != 1 { + t.Fatalf("Did not write to xint") + } +} + +func TestEscape(t *testing.T) { + const input = `"foobar"` + " [\u2028 \u2029]" + const expected = `"\u0022foobar\u0022\u003Chtml\u003E [\u2028 \u2029]"` + b, err := Marshal(input) + if err != nil { + t.Fatalf("Marshal error: %v", err) + } + if s := string(b); s != expected { + t.Errorf("Encoding of [%s]:\n got [%s]\nwant [%s]", input, s, expected) + } +} + +// WrongString is a struct that's misusing the ,string modifier. +type WrongString struct { + Message string `json:"result,string"` +} + +type wrongStringTest struct { + in, err string +} + +var wrongStringTests = []wrongStringTest{ + {`{"result":"x"}`, `json: invalid use of ,string struct tag, trying to unmarshal "x" into string`}, + {`{"result":"foo"}`, `json: invalid use of ,string struct tag, trying to unmarshal "foo" into string`}, + {`{"result":"123"}`, `json: invalid use of ,string struct tag, trying to unmarshal "123" into string`}, + {`{"result":123}`, `json: invalid use of ,string struct tag, trying to unmarshal unquoted value into string`}, +} + +// If people misuse the ,string modifier, the error message should be +// helpful, telling the user that they're doing it wrong. +func TestErrorMessageFromMisusedString(t *testing.T) { + for n, tt := range wrongStringTests { + r := strings.NewReader(tt.in) + var s WrongString + err := NewDecoder(r).Decode(&s) + got := fmt.Sprintf("%v", err) + if got != tt.err { + t.Errorf("%d. got err = %q, want %q", n, got, tt.err) + } + } +} + +func noSpace(c rune) rune { + if isSpace(byte(c)) { //only used for ascii + return -1 + } + return c +} + +type All struct { + Bool bool + Int int + Int8 int8 + Int16 int16 + Int32 int32 + Int64 int64 + Uint uint + Uint8 uint8 + Uint16 uint16 + Uint32 uint32 + Uint64 uint64 + Uintptr uintptr + Float32 float32 + Float64 float64 + + Foo string `json:"bar"` + Foo2 string `json:"bar2,dummyopt"` //nolint:staticcheck // It's intentionally wrong. + + IntStr int64 `json:",string"` + + PBool *bool + PInt *int + PInt8 *int8 + PInt16 *int16 + PInt32 *int32 + PInt64 *int64 + PUint *uint + PUint8 *uint8 + PUint16 *uint16 + PUint32 *uint32 + PUint64 *uint64 + PUintptr *uintptr + PFloat32 *float32 + PFloat64 *float64 + + String string + PString *string + + Map map[string]Small + MapP map[string]*Small + PMap *map[string]Small + PMapP *map[string]*Small + + EmptyMap map[string]Small + NilMap map[string]Small + + Slice []Small + SliceP []*Small + PSlice *[]Small + PSliceP *[]*Small + + EmptySlice []Small + NilSlice []Small + + StringSlice []string + ByteSlice []byte + + Small Small + PSmall *Small + PPSmall **Small + + Interface any + PInterface *any + + unexported int //nolint:unused // Not really used, but needed for test. +} + +type Small struct { + Tag string +} + +var allValue = All{ + Bool: true, + Int: 2, + Int8: 3, + Int16: 4, + Int32: 5, + Int64: 6, + Uint: 7, + Uint8: 8, + Uint16: 9, + Uint32: 10, + Uint64: 11, + Uintptr: 12, + Float32: 14.1, + Float64: 15.1, + Foo: "foo", + Foo2: "foo2", + IntStr: 42, + String: "16", + Map: map[string]Small{ + "17": {Tag: "tag17"}, + "18": {Tag: "tag18"}, + }, + MapP: map[string]*Small{ + "19": {Tag: "tag19"}, + "20": nil, + }, + EmptyMap: map[string]Small{}, + Slice: []Small{{Tag: "tag20"}, {Tag: "tag21"}}, + SliceP: []*Small{{Tag: "tag22"}, nil, {Tag: "tag23"}}, + EmptySlice: []Small{}, + StringSlice: []string{"str24", "str25", "str26"}, + ByteSlice: []byte{27, 28, 29}, + Small: Small{Tag: "tag30"}, + PSmall: &Small{Tag: "tag31"}, + Interface: 5.2, +} + +var pallValue = All{ + PBool: &allValue.Bool, + PInt: &allValue.Int, + PInt8: &allValue.Int8, + PInt16: &allValue.Int16, + PInt32: &allValue.Int32, + PInt64: &allValue.Int64, + PUint: &allValue.Uint, + PUint8: &allValue.Uint8, + PUint16: &allValue.Uint16, + PUint32: &allValue.Uint32, + PUint64: &allValue.Uint64, + PUintptr: &allValue.Uintptr, + PFloat32: &allValue.Float32, + PFloat64: &allValue.Float64, + PString: &allValue.String, + PMap: &allValue.Map, + PMapP: &allValue.MapP, + PSlice: &allValue.Slice, + PSliceP: &allValue.SliceP, + PPSmall: &allValue.PSmall, + PInterface: &allValue.Interface, +} + +var allValueIndent = `{ + "Bool": true, + "Int": 2, + "Int8": 3, + "Int16": 4, + "Int32": 5, + "Int64": 6, + "Uint": 7, + "Uint8": 8, + "Uint16": 9, + "Uint32": 10, + "Uint64": 11, + "Uintptr": 12, + "Float32": 14.1, + "Float64": 15.1, + "bar": "foo", + "bar2": "foo2", + "IntStr": "42", + "PBool": null, + "PInt": null, + "PInt8": null, + "PInt16": null, + "PInt32": null, + "PInt64": null, + "PUint": null, + "PUint8": null, + "PUint16": null, + "PUint32": null, + "PUint64": null, + "PUintptr": null, + "PFloat32": null, + "PFloat64": null, + "String": "16", + "PString": null, + "Map": { + "17": { + "Tag": "tag17" + }, + "18": { + "Tag": "tag18" + } + }, + "MapP": { + "19": { + "Tag": "tag19" + }, + "20": null + }, + "PMap": null, + "PMapP": null, + "EmptyMap": {}, + "NilMap": null, + "Slice": [ + { + "Tag": "tag20" + }, + { + "Tag": "tag21" + } + ], + "SliceP": [ + { + "Tag": "tag22" + }, + null, + { + "Tag": "tag23" + } + ], + "PSlice": null, + "PSliceP": null, + "EmptySlice": [], + "NilSlice": null, + "StringSlice": [ + "str24", + "str25", + "str26" + ], + "ByteSlice": "Gxwd", + "Small": { + "Tag": "tag30" + }, + "PSmall": { + "Tag": "tag31" + }, + "PPSmall": null, + "Interface": 5.2, + "PInterface": null +}` + +var allValueCompact = strings.Map(noSpace, allValueIndent) + +var pallValueIndent = `{ + "Bool": false, + "Int": 0, + "Int8": 0, + "Int16": 0, + "Int32": 0, + "Int64": 0, + "Uint": 0, + "Uint8": 0, + "Uint16": 0, + "Uint32": 0, + "Uint64": 0, + "Uintptr": 0, + "Float32": 0, + "Float64": 0, + "bar": "", + "bar2": "", + "IntStr": "0", + "PBool": true, + "PInt": 2, + "PInt8": 3, + "PInt16": 4, + "PInt32": 5, + "PInt64": 6, + "PUint": 7, + "PUint8": 8, + "PUint16": 9, + "PUint32": 10, + "PUint64": 11, + "PUintptr": 12, + "PFloat32": 14.1, + "PFloat64": 15.1, + "String": "", + "PString": "16", + "Map": null, + "MapP": null, + "PMap": { + "17": { + "Tag": "tag17" + }, + "18": { + "Tag": "tag18" + } + }, + "PMapP": { + "19": { + "Tag": "tag19" + }, + "20": null + }, + "EmptyMap": null, + "NilMap": null, + "Slice": null, + "SliceP": null, + "PSlice": [ + { + "Tag": "tag20" + }, + { + "Tag": "tag21" + } + ], + "PSliceP": [ + { + "Tag": "tag22" + }, + null, + { + "Tag": "tag23" + } + ], + "EmptySlice": null, + "NilSlice": null, + "StringSlice": null, + "ByteSlice": null, + "Small": { + "Tag": "" + }, + "PSmall": null, + "PPSmall": { + "Tag": "tag31" + }, + "Interface": null, + "PInterface": 5.2 +}` + +var pallValueCompact = strings.Map(noSpace, pallValueIndent) + +func TestRefUnmarshal(t *testing.T) { + type S struct { + // Ref is defined in encode_test.go. + R0 Ref + R1 *Ref + R2 RefText + R3 *RefText + } + want := S{ + R0: 12, + R1: new(Ref), + R2: 13, + R3: new(RefText), + } + *want.R1 = 12 + *want.R3 = 13 + + var got S + if err := Unmarshal([]byte(`{"R0":"ref","R1":"ref","R2":"ref","R3":"ref"}`), &got); err != nil { + t.Fatalf("Unmarshal: %v", err) + } + if !reflect.DeepEqual(got, want) { + t.Errorf("got %+v, want %+v", got, want) + } +} + +// Test that the empty string doesn't panic decoding when ,string is specified +// Issue 3450. +func TestEmptyString(t *testing.T) { + type T2 struct { + Number1 int `json:",string"` + Number2 int `json:",string"` + } + data := `{"Number1":"1", "Number2":""}` + dec := NewDecoder(strings.NewReader(data)) + var t2 T2 + err := dec.Decode(&t2) + if err == nil { + t.Fatal("Decode: did not return error") + } + if t2.Number1 != 1 { + t.Fatal("Decode: did not set Number1") + } +} + +// Test that a null for ,string is not replaced with the previous quoted string (issue 7046). +// It should also not be an error (issue 2540, issue 8587). +func TestNullString(t *testing.T) { + type T struct { + A int `json:",string"` + B int `json:",string"` + C *int `json:",string"` + } + data := []byte(`{"A": "1", "B": null, "C": null}`) + var s T + s.B = 1 + s.C = new(int) + *s.C = 2 + err := Unmarshal(data, &s) + if err != nil { + t.Fatalf("Unmarshal: %v", err) + } + if s.B != 1 || s.C != nil { + t.Fatalf("after Unmarshal, s.B=%d, s.C=%p, want 1, nil", s.B, s.C) + } +} + +func intp(x int) *int { + p := new(int) + *p = x + return p +} + +func intpp(x *int) **int { + pp := new(*int) + *pp = x + return pp +} + +var interfaceSetTests = []struct { + pre any + json string + post any +}{ + {"foo", `"bar"`, "bar"}, + {"foo", `2`, 2.0}, + {"foo", `true`, true}, + {"foo", `null`, nil}, + + {nil, `null`, nil}, + {new(int), `null`, nil}, + {(*int)(nil), `null`, nil}, + {new(*int), `null`, new(*int)}, + {(**int)(nil), `null`, nil}, + {intp(1), `null`, nil}, + {intpp(nil), `null`, intpp(nil)}, + {intpp(intp(1)), `null`, intpp(nil)}, +} + +func TestInterfaceSet(t *testing.T) { + for _, tt := range interfaceSetTests { + b := struct{ X any }{tt.pre} + blob := `{"X":` + tt.json + `}` + if err := Unmarshal([]byte(blob), &b); err != nil { + t.Errorf("Unmarshal %#q: %v", blob, err) + continue + } + if !reflect.DeepEqual(b.X, tt.post) { + t.Errorf("Unmarshal %#q into %#v: X=%#v, want %#v", blob, tt.pre, b.X, tt.post) + } + } +} + +type NullTest struct { + Bool bool + Int int + Int8 int8 + Int16 int16 + Int32 int32 + Int64 int64 + Uint uint + Uint8 uint8 + Uint16 uint16 + Uint32 uint32 + Uint64 uint64 + Float32 float32 + Float64 float64 + String string + PBool *bool + Map map[string]string + Slice []string + Interface any + + PRaw *RawMessage + PTime *time.Time + PBigInt *big.Int + PText *MustNotUnmarshalText + PBuffer *bytes.Buffer // has methods, just not relevant ones + PStruct *struct{} + + Raw RawMessage + Time time.Time + BigInt big.Int + Text MustNotUnmarshalText + Buffer bytes.Buffer + Struct struct{} +} + +type NullTestStrings struct { + Bool bool `json:",string"` + Int int `json:",string"` + Int8 int8 `json:",string"` + Int16 int16 `json:",string"` + Int32 int32 `json:",string"` + Int64 int64 `json:",string"` + Uint uint `json:",string"` + Uint8 uint8 `json:",string"` + Uint16 uint16 `json:",string"` + Uint32 uint32 `json:",string"` + Uint64 uint64 `json:",string"` + Float32 float32 `json:",string"` + Float64 float64 `json:",string"` + String string `json:",string"` + PBool *bool `json:",string"` + Map map[string]string `json:",string"` //nolint:staticcheck // It's intentionally wrong. + Slice []string `json:",string"` //nolint:staticcheck // It's intentionally wrong. + Interface any `json:",string"` //nolint:staticcheck // It's intentionally wrong. + + PRaw *RawMessage `json:",string"` //nolint:staticcheck // It's intentionally wrong. + PTime *time.Time `json:",string"` //nolint:staticcheck // It's intentionally wrong. + PBigInt *big.Int `json:",string"` //nolint:staticcheck // It's intentionally wrong. + PText *MustNotUnmarshalText `json:",string"` //nolint:staticcheck // It's intentionally wrong. + PBuffer *bytes.Buffer `json:",string"` //nolint:staticcheck // It's intentionally wrong. + PStruct *struct{} `json:",string"` //nolint:staticcheck // It's intentionally wrong. + + Raw RawMessage `json:",string"` //nolint:staticcheck // It's intentionally wrong. + Time time.Time `json:",string"` //nolint:staticcheck // It's intentionally wrong. + BigInt big.Int `json:",string"` //nolint:staticcheck // It's intentionally wrong. + Text MustNotUnmarshalText `json:",string"` //nolint:staticcheck // It's intentionally wrong. + Buffer bytes.Buffer `json:",string"` //nolint:staticcheck // It's intentionally wrong. + Struct struct{} `json:",string"` //nolint:staticcheck // It's intentionally wrong. +} + +// JSON null values should be ignored for primitives and string values instead of resulting in an error. +// Issue 2540. +func TestUnmarshalNulls(t *testing.T) { + // Unmarshal docs: + // The JSON null value unmarshals into an interface, map, pointer, or slice + // by setting that Go value to nil. Because null is often used in JSON to mean + // ``not present,'' unmarshaling a JSON null into any other Go type has no effect + // on the value and produces no error. + + jsonData := []byte(`{ + "Bool" : null, + "Int" : null, + "Int8" : null, + "Int16" : null, + "Int32" : null, + "Int64" : null, + "Uint" : null, + "Uint8" : null, + "Uint16" : null, + "Uint32" : null, + "Uint64" : null, + "Float32" : null, + "Float64" : null, + "String" : null, + "PBool": null, + "Map": null, + "Slice": null, + "Interface": null, + "PRaw": null, + "PTime": null, + "PBigInt": null, + "PText": null, + "PBuffer": null, + "PStruct": null, + "Raw": null, + "Time": null, + "BigInt": null, + "Text": null, + "Buffer": null, + "Struct": null + }`) + nulls := NullTest{ + Bool: true, + Int: 2, + Int8: 3, + Int16: 4, + Int32: 5, + Int64: 6, + Uint: 7, + Uint8: 8, + Uint16: 9, + Uint32: 10, + Uint64: 11, + Float32: 12.1, + Float64: 13.1, + String: "14", + PBool: new(bool), + Map: map[string]string{}, + Slice: []string{}, + Interface: new(MustNotUnmarshalJSON), + PRaw: new(RawMessage), + PTime: new(time.Time), + PBigInt: new(big.Int), + PText: new(MustNotUnmarshalText), + PStruct: new(struct{}), + PBuffer: new(bytes.Buffer), + Raw: RawMessage("123"), + Time: time.Unix(123456789, 0), + BigInt: *big.NewInt(123), + } + + before := nulls.Time.String() + + err := Unmarshal(jsonData, &nulls) + if err != nil { + t.Errorf("Unmarshal of null values failed: %v", err) + } + if !nulls.Bool || nulls.Int != 2 || nulls.Int8 != 3 || nulls.Int16 != 4 || nulls.Int32 != 5 || nulls.Int64 != 6 || + nulls.Uint != 7 || nulls.Uint8 != 8 || nulls.Uint16 != 9 || nulls.Uint32 != 10 || nulls.Uint64 != 11 || + nulls.Float32 != 12.1 || nulls.Float64 != 13.1 || nulls.String != "14" { + t.Errorf("Unmarshal of null values affected primitives") + } + + if nulls.PBool != nil { + t.Errorf("Unmarshal of null did not clear nulls.PBool") + } + if nulls.Map != nil { + t.Errorf("Unmarshal of null did not clear nulls.Map") + } + if nulls.Slice != nil { + t.Errorf("Unmarshal of null did not clear nulls.Slice") + } + if nulls.Interface != nil { + t.Errorf("Unmarshal of null did not clear nulls.Interface") + } + if nulls.PRaw != nil { + t.Errorf("Unmarshal of null did not clear nulls.PRaw") + } + if nulls.PTime != nil { + t.Errorf("Unmarshal of null did not clear nulls.PTime") + } + if nulls.PBigInt != nil { + t.Errorf("Unmarshal of null did not clear nulls.PBigInt") + } + if nulls.PText != nil { + t.Errorf("Unmarshal of null did not clear nulls.PText") + } + if nulls.PBuffer != nil { + t.Errorf("Unmarshal of null did not clear nulls.PBuffer") + } + if nulls.PStruct != nil { + t.Errorf("Unmarshal of null did not clear nulls.PStruct") + } + + if string(nulls.Raw) != "null" { + t.Errorf("Unmarshal of RawMessage null did not record null: %v", string(nulls.Raw)) + } + if nulls.Time.String() != before { + t.Errorf("Unmarshal of time.Time null set time to %v", nulls.Time.String()) + } + if nulls.BigInt.String() != "123" { + t.Errorf("Unmarshal of big.Int null set int to %v", nulls.BigInt.String()) + } +} + +type MustNotUnmarshalJSON struct{} + +func (x MustNotUnmarshalJSON) UnmarshalJSON(data []byte) error { + return errors.New("MustNotUnmarshalJSON was used") +} + +type MustNotUnmarshalText struct{} + +func (x MustNotUnmarshalText) UnmarshalText(text []byte) error { + return errors.New("MustNotUnmarshalText was used") +} + +func TestStringKind(t *testing.T) { + type stringKind string + + var m1, m2 map[stringKind]int + m1 = map[stringKind]int{ + "foo": 42, + } + + data, err := Marshal(m1) + if err != nil { + t.Errorf("Unexpected error marshaling: %v", err) + } + + err = Unmarshal(data, &m2) + if err != nil { + t.Errorf("Unexpected error unmarshaling: %v", err) + } + + if !reflect.DeepEqual(m1, m2) { + t.Error("Items should be equal after encoding and then decoding") + } +} + +// Custom types with []byte as underlying type could not be marshaled +// and then unmarshaled. +// Issue 8962. +func TestByteKind(t *testing.T) { + type byteKind []byte + + a := byteKind("hello") + + data, err := Marshal(a) + if err != nil { + t.Error(err) + } + var b byteKind + err = Unmarshal(data, &b) + if err != nil { + t.Fatal(err) + } + if !reflect.DeepEqual(a, b) { + t.Errorf("expected %v == %v", a, b) + } +} + +// The fix for issue 8962 introduced a regression. +// Issue 12921. +func TestSliceOfCustomByte(t *testing.T) { + type Uint8 uint8 + + a := []Uint8("hello") + + data, err := Marshal(a) + if err != nil { + t.Fatal(err) + } + var b []Uint8 + err = Unmarshal(data, &b) + if err != nil { + t.Fatal(err) + } + if !reflect.DeepEqual(a, b) { + t.Fatalf("expected %v == %v", a, b) + } +} + +var decodeTypeErrorTests = []struct { + dest any + src string +}{ + {new(string), `{"user": "name"}`}, // issue 4628. + {new(error), `{}`}, // issue 4222 + {new(error), `[]`}, + {new(error), `""`}, + {new(error), `123`}, + {new(error), `true`}, +} + +func TestUnmarshalTypeError(t *testing.T) { + for _, item := range decodeTypeErrorTests { + err := Unmarshal([]byte(item.src), item.dest) + if _, ok := err.(*UnmarshalTypeError); !ok { //nolint:errorlint // It must match exactly, it's a test. + t.Errorf("expected type error for Unmarshal(%q, type %T): got %T", + item.src, item.dest, err) + } + } +} + +var unmarshalSyntaxTests = []string{ + "tru", + "fals", + "nul", + "123e", + `"hello`, + `[1,2,3`, + `{"key":1`, + `{"key":1,`, +} + +func TestUnmarshalSyntax(t *testing.T) { + var x any + for _, src := range unmarshalSyntaxTests { + err := Unmarshal([]byte(src), &x) + if _, ok := err.(*SyntaxError); !ok { //nolint:errorlint // It must match exactly, it's a test. + t.Errorf("expected syntax error for Unmarshal(%q): got %T", src, err) + } + } +} + +// Test handling of unexported fields that should be ignored. +// Issue 4660. +type unexportedFields struct { + Name string + m map[string]any `json:"-"` //nolint:unused // Not really used, but important for test. + m2 map[string]any `json:"abcd"` //nolint:unused,govet // Not really used and wrong, but important for test. +} + +func TestUnmarshalUnexported(t *testing.T) { + input := `{"Name": "Bob", "m": {"x": 123}, "m2": {"y": 456}, "abcd": {"z": 789}}` + want := &unexportedFields{Name: "Bob"} + + out := &unexportedFields{} + err := Unmarshal([]byte(input), out) + if err != nil { + t.Errorf("got error %v, expected nil", err) + } + if !reflect.DeepEqual(out, want) { + t.Errorf("got %q, want %q", out, want) + } +} + +// Time3339 is a time.Time which encodes to and from JSON +// as an RFC 3339 time in UTC. +type Time3339 time.Time + +func (t *Time3339) UnmarshalJSON(b []byte) error { + if len(b) < 2 || b[0] != '"' || b[len(b)-1] != '"' { + return fmt.Errorf("types: failed to unmarshal non-string value %q as an RFC 3339 time", b) + } + tm, err := time.Parse(time.RFC3339, string(b[1:len(b)-1])) + if err != nil { + return err + } + *t = Time3339(tm) + return nil +} + +func TestUnmarshalJSONLiteralError(t *testing.T) { + var t3 Time3339 + err := Unmarshal([]byte(`"0000-00-00T00:00:00Z"`), &t3) + if err == nil { + t.Fatalf("expected error; got time %v", time.Time(t3)) + } + if !strings.Contains(err.Error(), "range") { + t.Errorf("got err = %v; want out of range error", err) + } +} + +// Test that extra object elements in an array do not result in a +// "data changing underfoot" error. +// Issue 3717. +func TestSkipArrayObjects(t *testing.T) { + json := `[{}]` + var dest [0]any + + err := Unmarshal([]byte(json), &dest) + if err != nil { + t.Errorf("got error %q, want nil", err) + } +} + +// Test semantics of pre-filled struct fields and pre-filled map fields. +// Issue 4900. +func TestPrefilled(t *testing.T) { + ptrToMap := func(m map[string]any) *map[string]any { return &m } + + // Values here change, cannot reuse table across runs. + var prefillTests = []struct { + in string + ptr any + out any + }{ + { + in: `{"X": 1, "Y": 2}`, + ptr: &XYZ{X: float32(3), Y: int16(4), Z: 1.5}, + out: &XYZ{X: float64(1), Y: float64(2), Z: 1.5}, + }, + { + in: `{"X": 1, "Y": 2}`, + ptr: ptrToMap(map[string]any{"X": float32(3), "Y": int16(4), "Z": 1.5}), + out: ptrToMap(map[string]any{"X": float64(1), "Y": float64(2), "Z": 1.5}), + }, + } + + for _, tt := range prefillTests { + ptrstr := fmt.Sprintf("%v", tt.ptr) + err := Unmarshal([]byte(tt.in), tt.ptr) // tt.ptr edited here + if err != nil { + t.Errorf("Unmarshal: %v", err) + } + if !reflect.DeepEqual(tt.ptr, tt.out) { + t.Errorf("Unmarshal(%#q, %s): have %v, want %v", tt.in, ptrstr, tt.ptr, tt.out) + } + } +} + +var invalidUnmarshalTests = []struct { + v any + want string +}{ + {nil, "json: Unmarshal(nil)"}, + {struct{}{}, "json: Unmarshal(non-pointer struct {})"}, + {(*int)(nil), "json: Unmarshal(nil *int)"}, +} + +func TestInvalidUnmarshal(t *testing.T) { + buf := []byte(`{"a":"1"}`) + for _, tt := range invalidUnmarshalTests { + err := Unmarshal(buf, tt.v) + if err == nil { + t.Errorf("Unmarshal expecting error, got nil") + continue + } + if got := err.Error(); got != tt.want { + t.Errorf("Unmarshal = %q; want %q", got, tt.want) + } + } +} + +var invalidUnmarshalTextTests = []struct { + v any + want string +}{ + {nil, "json: Unmarshal(nil)"}, + {struct{}{}, "json: Unmarshal(non-pointer struct {})"}, + {(*int)(nil), "json: Unmarshal(nil *int)"}, + {new(net.IP), "json: cannot unmarshal number into Go value of type *net.IP"}, +} + +func TestInvalidUnmarshalText(t *testing.T) { + buf := []byte(`123`) + for _, tt := range invalidUnmarshalTextTests { + err := Unmarshal(buf, tt.v) + if err == nil { + t.Errorf("Unmarshal expecting error, got nil") + continue + } + if got := err.Error(); got != tt.want { + t.Errorf("Unmarshal = %q; want %q", got, tt.want) + } + } +} + +// Test that string option is ignored for invalid types. +// Issue 9812. +func TestInvalidStringOption(t *testing.T) { + num := 0 + item := struct { + T time.Time `json:",string"` //nolint:staticcheck // It's intentionally wrong. + M map[string]string `json:",string"` //nolint:staticcheck // It's intentionally wrong. + S []string `json:",string"` //nolint:staticcheck // It's intentionally wrong. + A [1]string `json:",string"` //nolint:staticcheck // It's intentionally wrong. + I any `json:",string"` //nolint:staticcheck // It's intentionally wrong. + P *int `json:",string"` //nolint:staticcheck // It's intentionally wrong. + }{M: make(map[string]string), S: make([]string, 0), I: num, P: &num} + + data, err := Marshal(item) + if err != nil { + t.Fatalf("Marshal: %v", err) + } + + err = Unmarshal(data, &item) + if err != nil { + t.Fatalf("Unmarshal: %v", err) + } +} + +type Obj struct { + A int + B OrderedObject + C string +} + +func TestUnmarshalOrdered(t *testing.T) { + var v Obj + buf := []byte(`{"A": 3, "B": { "A": 5, "B": null}, "C": "C"}`) + exp := Obj{A: 3, B: OrderedObject{{"A", float64(5)}, {"B", nil}}, C: "C"} + err := Unmarshal(buf, &v) + if err != nil { + t.Errorf("Unmarshal unexpected error: %v", err) + } + if !reflect.DeepEqual(exp, v) { + t.Errorf("%v, want %v", v, exp) + } +} diff --git a/encode.go b/encode.go new file mode 100644 index 0000000..a45ffd9 --- /dev/null +++ b/encode.go @@ -0,0 +1,1344 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package json implements encoding and decoding of JSON as defined in +// RFC 4627. The mapping between JSON and Go values is described +// in the documentation for the Marshal and Unmarshal functions. +// +// See "JSON and Go" for an introduction to this package: +// https://golang.org/doc/articles/json_and_go.html +package json + +import ( + "bytes" + "encoding" + "encoding/base64" + "fmt" + "math" + "reflect" + "runtime" + "slices" + "sort" + "strconv" + "strings" + "sync" + "sync/atomic" + "unicode" + "unicode/utf16" + "unicode/utf8" +) + +// Marshal returns the JSON encoding of v. +// +// Marshal traverses the value v recursively. +// If an encountered value implements the Marshaler interface +// and is not a nil pointer, Marshal calls its MarshalJSON method +// to produce JSON. If no MarshalJSON method is present but the +// value implements encoding.TextMarshaler instead, Marshal calls +// its MarshalText method and encodes the result as a JSON string. +// The nil pointer exception is not strictly necessary +// but mimics a similar, necessary exception in the behavior of +// UnmarshalJSON. +// +// Otherwise, Marshal uses the following type-dependent default encodings: +// +// Boolean values encode as JSON booleans. +// +// Floating point, integer, and Number values encode as JSON numbers. +// +// String values encode as JSON strings coerced to valid UTF-8, +// replacing invalid bytes with the Unicode replacement rune. +// The angle brackets "<" and ">" are escaped to "\u003c" and "\u003e" +// to keep some browsers from misinterpreting JSON output as HTML. +// Ampersand "&" is also escaped to "\u0026" for the same reason. +// This escaping can be disabled using an Encoder that had SetEscapeHTML(false) +// called on it. +// +// Array and slice values encode as JSON arrays, except that +// []byte encodes as a base64-encoded string, and a nil slice +// encodes as the null JSON value. +// +// Struct values encode as JSON objects. +// Each exported struct field becomes a member of the object, using the +// field name as the object key, unless the field is omitted for one of the +// reasons given below. +// +// The encoding of each struct field can be customized by the format string +// stored under the "json" key in the struct field's tag. +// The format string gives the name of the field, possibly followed by a +// comma-separated list of options. The name may be empty in order to +// specify options without overriding the default field name. +// +// The "omitempty" option specifies that the field should be omitted +// from the encoding if the field has an empty value, defined as +// false, 0, a nil pointer, a nil interface value, and any empty array, +// slice, map, or string. +// +// As a special case, if the field tag is "-", the field is always omitted. +// Note that a field with name "-" can still be generated using the tag "-,". +// +// Examples of struct field tags and their meanings: +// +// // Field appears in JSON as key "myName". +// Field int `json:"myName"` +// +// // Field appears in JSON as key "myName" and +// // the field is omitted from the object if its value is empty, +// // as defined above. +// Field int `json:"myName,omitempty"` +// +// // Field appears in JSON as key "Field" (the default), but +// // the field is skipped if empty. +// // Note the leading comma. +// Field int `json:",omitempty"` +// +// // Field is ignored by this package. +// Field int `json:"-"` +// +// // Field appears in JSON as key "-". +// Field int `json:"-,"` +// +// The "string" option signals that a field is stored as JSON inside a +// JSON-encoded string. It applies only to fields of string, floating point, +// integer, or boolean types. This extra level of encoding is sometimes used +// when communicating with JavaScript programs: +// +// Int64String int64 `json:",string"` +// +// The key name will be used if it's a non-empty string consisting of +// only Unicode letters, digits, and ASCII punctuation except quotation +// marks, backslash, and comma. +// +// Anonymous struct fields are usually marshaled as if their inner exported fields +// were fields in the outer struct, subject to the usual Go visibility rules amended +// as described in the next paragraph. +// An anonymous struct field with a name given in its JSON tag is treated as +// having that name, rather than being anonymous. +// An anonymous struct field of interface type is treated the same as having +// that type as its name, rather than being anonymous. +// +// The Go visibility rules for struct fields are amended for JSON when +// deciding which field to marshal or unmarshal. If there are +// multiple fields at the same level, and that level is the least +// nested (and would therefore be the nesting level selected by the +// usual Go rules), the following extra rules apply: +// +// 1) Of those fields, if any are JSON-tagged, only tagged fields are considered, +// even if there are multiple untagged fields that would otherwise conflict. +// +// 2) If there is exactly one field (tagged or not according to the first rule), that is selected. +// +// 3) Otherwise there are multiple fields, and all are ignored; no error occurs. +// +// Handling of anonymous struct fields is new in Go 1.1. +// Prior to Go 1.1, anonymous struct fields were ignored. To force ignoring of +// an anonymous struct field in both current and earlier versions, give the field +// a JSON tag of "-". +// +// Map values encode as JSON objects. The map's key type must either be a +// string, an integer type, or implement encoding.TextMarshaler. The map keys +// are sorted and used as JSON object keys by applying the following rules, +// subject to the UTF-8 coercion described for string values above: +// - string keys are used directly +// - encoding.TextMarshalers are marshaled +// - integer keys are converted to strings +// +// OrderedObject values encode as JSON objects. +// The JSON keys are encoded in the order in which they appear in the OrderedObject. +// +// Pointer values encode as the value pointed to. +// A nil pointer encodes as the null JSON value. +// +// Interface values encode as the value contained in the interface. +// A nil interface value encodes as the null JSON value. +// +// Channel, complex, and function values cannot be encoded in JSON. +// Attempting to encode such a value causes Marshal to return +// an UnsupportedTypeError. +// +// JSON cannot represent cyclic data structures and Marshal does not +// handle them. Passing cyclic structures to Marshal will result in +// an infinite recursion. +func Marshal(v any) ([]byte, error) { + e := &encodeState{} + err := e.marshal(v, encOpts{escapeHTML: true}) + if err != nil { + return nil, err + } + return e.Bytes(), nil +} + +// MarshalIndent is like Marshal but applies Indent to format the output. +func MarshalIndent(v any, prefix, indent string) ([]byte, error) { + b, err := Marshal(v) + if err != nil { + return nil, err + } + var buf bytes.Buffer + err = Indent(&buf, b, prefix, indent) + if err != nil { + return nil, err + } + return buf.Bytes(), nil +} + +// HTMLEscape appends to dst the JSON-encoded src with <, >, &, U+2028 and U+2029 +// characters inside string literals changed to \u003c, \u003e, \u0026, \u2028, \u2029 +// so that the JSON will be safe to embed inside HTML