decoder: strict mode (#512)

This commit is contained in:
Thomas Pelletier
2021-04-20 21:26:22 -04:00
committed by GitHub
parent dca2103910
commit 9b67e40640
11 changed files with 715 additions and 259 deletions
@@ -7,6 +7,7 @@ package imported_tests
// marked as skipped until we figure out if that's something we want in v2.
import (
"bytes"
"errors"
"fmt"
"reflect"
@@ -1955,66 +1956,80 @@ String2="2"`
assert.Error(t, err)
}
func decoder(doc string) *toml.Decoder {
return toml.NewDecoder(bytes.NewReader([]byte(doc)))
}
func strictDecoder(doc string) *toml.Decoder {
d := decoder(doc)
d.SetStrict(true)
return d
}
func TestDecoderStrict(t *testing.T) {
t.Skip()
// input := `
//[decoded]
// key = ""
//
//[undecoded]
// key = ""
//
// [undecoded.inner]
// key = ""
//
// [[undecoded.array]]
// key = ""
//
// [[undecoded.array]]
// key = ""
//
//`
// var doc struct {
// Decoded struct {
// Key string
// }
// }
//
// expected := `undecoded keys: ["undecoded.array.0.key" "undecoded.array.1.key" "undecoded.inner.key" "undecoded.key"]`
//
// err := NewDecoder(bytes.NewReader([]byte(input))).Strict(true).Decode(&doc)
// if err == nil {
// t.Error("expected error, got none")
// } else if err.Error() != expected {
// t.Errorf("expect err: %s, got: %s", expected, err.Error())
// }
//
// if err := NewDecoder(bytes.NewReader([]byte(input))).Decode(&doc); err != nil {
// t.Errorf("unexpected err: %s", err)
// }
//
// var m map[string]interface{}
// if err := NewDecoder(bytes.NewReader([]byte(input))).Decode(&m); err != nil {
// t.Errorf("unexpected err: %s", err)
// }
input := `
[decoded]
key = ""
[undecoded]
key = ""
[undecoded.inner]
key = ""
[[undecoded.array]]
key = ""
[[undecoded.array]]
key = ""
`
var doc struct {
Decoded struct {
Key string
}
}
err := strictDecoder(input).Decode(&doc)
require.Error(t, err)
require.IsType(t, &toml.StrictMissingError{}, err)
se := err.(*toml.StrictMissingError)
keys := []toml.Key{}
for _, e := range se.Errors {
keys = append(keys, e.Key())
}
expectedKeys := []toml.Key{
{"undecoded"},
{"undecoded", "inner"},
{"undecoded", "array"},
{"undecoded", "array"},
}
require.Equal(t, expectedKeys, keys)
err = decoder(input).Decode(&doc)
require.NoError(t, err)
var m map[string]interface{}
err = decoder(input).Decode(&m)
}
func TestDecoderStrictValid(t *testing.T) {
t.Skip()
// input := `
//[decoded]
// key = ""
//`
// var doc struct {
// Decoded struct {
// Key string
// }
// }
//
// err := NewDecoder(bytes.NewReader([]byte(input))).Strict(true).Decode(&doc)
// if err != nil {
// t.Fatal("unexpected error:", err)
// }
input := `
[decoded]
key = ""
`
var doc struct {
Decoded struct {
Key string
}
}
err := strictDecoder(input).Decode(&doc)
require.NoError(t, err)
}
type docUnmarshalTOML struct {
+50
View File
@@ -0,0 +1,50 @@
package tracker
import (
"github.com/pelletier/go-toml/v2/internal/ast"
)
// KeyTracker is a tracker that keeps track of the current Key as the AST is
// walked.
type KeyTracker struct {
k []string
}
// UpdateTable sets the state of the tracker with the AST table node.
func (t *KeyTracker) UpdateTable(node ast.Node) {
t.reset()
t.Push(node)
}
// UpdateArrayTable sets the state of the tracker with the AST array table node.
func (t *KeyTracker) UpdateArrayTable(node ast.Node) {
t.reset()
t.Push(node)
}
// Push the given key on the stack.
func (t *KeyTracker) Push(node ast.Node) {
it := node.Key()
for it.Next() {
t.k = append(t.k, string(it.Node().Data))
}
}
// Pop key from stack.
func (t *KeyTracker) Pop(node ast.Node) {
it := node.Key()
for it.Next() {
t.k = t.k[:len(t.k)-1]
}
}
// Key returns the current key
func (t *KeyTracker) Key() []string {
k := make([]string, len(t.k))
copy(k, t.k)
return k
}
func (t *KeyTracker) reset() {
t.k = t.k[:0]
}
+200
View File
@@ -0,0 +1,200 @@
package tracker
import (
"fmt"
"github.com/pelletier/go-toml/v2/internal/ast"
)
type keyKind uint8
const (
invalidKind keyKind = iota
valueKind
tableKind
arrayTableKind
)
func (k keyKind) String() string {
switch k {
case invalidKind:
return "invalid"
case valueKind:
return "value"
case tableKind:
return "table"
case arrayTableKind:
return "array table"
}
panic("missing keyKind string mapping")
}
// SeenTracker tracks which keys have been seen with which TOML type to flag duplicates
// and mismatches according to the spec.
type SeenTracker struct {
root *info
current *info
}
type info struct {
parent *info
kind keyKind
children map[string]*info
explicit bool
}
func (i *info) Clear() {
i.children = nil
}
func (i *info) Has(k string) (*info, bool) {
c, ok := i.children[k]
return c, ok
}
func (i *info) SetKind(kind keyKind) {
i.kind = kind
}
func (i *info) CreateTable(k string, explicit bool) *info {
return i.createChild(k, tableKind, explicit)
}
func (i *info) CreateArrayTable(k string, explicit bool) *info {
return i.createChild(k, arrayTableKind, explicit)
}
func (i *info) createChild(k string, kind keyKind, explicit bool) *info {
if i.children == nil {
i.children = make(map[string]*info, 1)
}
x := &info{
parent: i,
kind: kind,
explicit: explicit,
}
i.children[k] = x
return x
}
// CheckExpression takes a top-level node and checks that it does not contain keys
// that have been seen in previous calls, and validates that types are consistent.
func (s *SeenTracker) CheckExpression(node ast.Node) error {
if s.root == nil {
s.root = &info{
kind: tableKind,
}
s.current = s.root
}
switch node.Kind {
case ast.KeyValue:
return s.checkKeyValue(s.current, node)
case ast.Table:
return s.checkTable(node)
case ast.ArrayTable:
return s.checkArrayTable(node)
default:
panic(fmt.Errorf("this should not be a top level node type: %s", node.Kind))
}
}
func (s *SeenTracker) checkTable(node ast.Node) error {
s.current = s.root
it := node.Key()
// handle the first parts of the key, excluding the last one
for it.Next() {
if !it.Node().Next().Valid() {
break
}
k := string(it.Node().Data)
child, found := s.current.Has(k)
if !found {
child = s.current.CreateTable(k, false)
}
s.current = child
}
// handle the last part of the key
k := string(it.Node().Data)
i, found := s.current.Has(k)
if found {
if i.kind != tableKind {
return fmt.Errorf("key %s should be a table", k)
}
if i.explicit {
return fmt.Errorf("table %s already exists", k)
}
i.explicit = true
s.current = i
} else {
s.current = s.current.CreateTable(k, true)
}
return nil
}
func (s *SeenTracker) checkArrayTable(node ast.Node) error {
s.current = s.root
it := node.Key()
// handle the first parts of the key, excluding the last one
for it.Next() {
if !it.Node().Next().Valid() {
break
}
k := string(it.Node().Data)
child, found := s.current.Has(k)
if !found {
child = s.current.CreateTable(k, false)
}
s.current = child
}
// handle the last part of the key
k := string(it.Node().Data)
info, found := s.current.Has(k)
if found {
if info.kind != arrayTableKind {
return fmt.Errorf("key %s already exists but is not an array table", k)
}
info.Clear()
} else {
info = s.current.CreateArrayTable(k, true)
}
s.current = info
return nil
}
func (s *SeenTracker) checkKeyValue(context *info, node ast.Node) error {
it := node.Key()
// handle the first parts of the key, excluding the last one
for it.Next() {
k := string(it.Node().Data)
child, found := context.Has(k)
if found {
if child.kind != tableKind {
return fmt.Errorf("expected %s to be a table, not a %s", k, child.kind)
}
} else {
child = context.CreateTable(k, false)
}
context = child
}
if node.Value().Kind == ast.InlineTable {
context.SetKind(tableKind)
} else {
context.SetKind(valueKind)
}
return nil
}
-199
View File
@@ -1,200 +1 @@
package tracker
import (
"fmt"
"github.com/pelletier/go-toml/v2/internal/ast"
)
type keyKind uint8
const (
invalidKind keyKind = iota
valueKind
tableKind
arrayTableKind
)
func (k keyKind) String() string {
switch k {
case invalidKind:
return "invalid"
case valueKind:
return "value"
case tableKind:
return "table"
case arrayTableKind:
return "array table"
}
panic("missing keyKind string mapping")
}
// Tracks which keys have been seen with which TOML type to flag duplicates
// and mismatches according to the spec.
type Seen struct {
root *info
current *info
}
type info struct {
parent *info
kind keyKind
children map[string]*info
explicit bool
}
func (i *info) Clear() {
i.children = nil
}
func (i *info) Has(k string) (*info, bool) {
c, ok := i.children[k]
return c, ok
}
func (i *info) SetKind(kind keyKind) {
i.kind = kind
}
func (i *info) CreateTable(k string, explicit bool) *info {
return i.createChild(k, tableKind, explicit)
}
func (i *info) CreateArrayTable(k string, explicit bool) *info {
return i.createChild(k, arrayTableKind, explicit)
}
func (i *info) createChild(k string, kind keyKind, explicit bool) *info {
if i.children == nil {
i.children = make(map[string]*info, 1)
}
x := &info{
parent: i,
kind: kind,
explicit: explicit,
}
i.children[k] = x
return x
}
// CheckExpression takes a top-level node and checks that it does not contain keys
// that have been seen in previous calls, and validates that types are consistent.
func (s *Seen) CheckExpression(node ast.Node) error {
if s.root == nil {
s.root = &info{
kind: tableKind,
}
s.current = s.root
}
switch node.Kind {
case ast.KeyValue:
return s.checkKeyValue(s.current, node)
case ast.Table:
return s.checkTable(node)
case ast.ArrayTable:
return s.checkArrayTable(node)
default:
panic(fmt.Errorf("this should not be a top level node type: %s", node.Kind))
}
}
func (s *Seen) checkTable(node ast.Node) error {
s.current = s.root
it := node.Key()
// handle the first parts of the key, excluding the last one
for it.Next() {
if !it.Node().Next().Valid() {
break
}
k := string(it.Node().Data)
child, found := s.current.Has(k)
if !found {
child = s.current.CreateTable(k, false)
}
s.current = child
}
// handle the last part of the key
k := string(it.Node().Data)
i, found := s.current.Has(k)
if found {
if i.kind != tableKind {
return fmt.Errorf("key %s should be a table", k)
}
if i.explicit {
return fmt.Errorf("table %s already exists", k)
}
i.explicit = true
s.current = i
} else {
s.current = s.current.CreateTable(k, true)
}
return nil
}
func (s *Seen) checkArrayTable(node ast.Node) error {
s.current = s.root
it := node.Key()
// handle the first parts of the key, excluding the last one
for it.Next() {
if !it.Node().Next().Valid() {
break
}
k := string(it.Node().Data)
child, found := s.current.Has(k)
if !found {
child = s.current.CreateTable(k, false)
}
s.current = child
}
// handle the last part of the key
k := string(it.Node().Data)
info, found := s.current.Has(k)
if found {
if info.kind != arrayTableKind {
return fmt.Errorf("key %s already exists but is not an array table", k)
}
info.Clear()
} else {
info = s.current.CreateArrayTable(k, true)
}
s.current = info
return nil
}
func (s *Seen) checkKeyValue(context *info, node ast.Node) error {
it := node.Key()
// handle the first parts of the key, excluding the last one
for it.Next() {
k := string(it.Node().Data)
child, found := context.Has(k)
if found {
if child.kind != tableKind {
return fmt.Errorf("expected %s to be a table, not a %s", k, child.kind)
}
} else {
child = context.CreateTable(k, false)
}
context = child
}
if node.Value().Kind == ast.InlineTable {
context.SetKind(tableKind)
} else {
context.SetKind(valueKind)
}
return nil
}
+24
View File
@@ -33,3 +33,27 @@ func SubsliceOffset(data []byte, subslice []byte) int {
return intoffset
}
func BytesRange(start []byte, end []byte) []byte {
if start == nil || end == nil {
panic("cannot call BytesRange with nil")
}
startp := (*reflect.SliceHeader)(unsafe.Pointer(&start))
endp := (*reflect.SliceHeader)(unsafe.Pointer(&end))
if startp.Data > endp.Data {
panic(fmt.Errorf("start pointer address (%d) is after end pointer address (%d)", startp.Data, endp.Data))
}
l := startp.Len
endLen := int(endp.Data-startp.Data) + endp.Len
if endLen > l {
l = endLen
}
if l > startp.Cap {
panic(fmt.Errorf("range length is larger than capacity"))
}
return start[:l]
}
+89
View File
@@ -77,3 +77,92 @@ func TestUnsafeSubsliceOffsetInvalid(t *testing.T) {
})
}
}
func TestUnsafeBytesRange(t *testing.T) {
type fn = func() ([]byte, []byte)
examples := []struct {
desc string
test fn
expected []byte
}{
{
desc: "simple",
test: func() ([]byte, []byte) {
full := []byte("hello world")
return full[1:3], full[6:8]
},
expected: []byte("ello wo"),
},
{
desc: "full",
test: func() ([]byte, []byte) {
full := []byte("hello world")
return full[0:1], full[len(full)-1:]
},
expected: []byte("hello world"),
},
{
desc: "end before start",
test: func() ([]byte, []byte) {
full := []byte("hello world")
return full[len(full)-1:], full[0:1]
},
},
{
desc: "nils",
test: func() ([]byte, []byte) {
return nil, nil
},
},
{
desc: "nils start",
test: func() ([]byte, []byte) {
return nil, []byte("foo")
},
},
{
desc: "nils end",
test: func() ([]byte, []byte) {
return []byte("foo"), nil
},
},
{
desc: "start is end",
test: func() ([]byte, []byte) {
full := []byte("hello world")
return full[1:3], full[1:3]
},
expected: []byte("el"),
},
{
desc: "end contained in start",
test: func() ([]byte, []byte) {
full := []byte("hello world")
return full[1:7], full[2:4]
},
expected: []byte("ello w"),
},
{
desc: "different backing arrays",
test: func() ([]byte, []byte) {
one := []byte("hello world")
two := []byte("hello world")
return one, two
},
},
}
for _, e := range examples {
t.Run(e.desc, func(t *testing.T) {
start, end := e.test()
if e.expected == nil {
require.Panics(t, func() {
unsafe.BytesRange(start, end)
})
} else {
res := unsafe.BytesRange(start, end)
require.Equal(t, e.expected, res)
}
})
}
}