document

package
v0.7.1 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Sep 1, 2020 License: MIT Imports: 13 Imported by: 0

Documentation

Overview

Package document defines types to manipulate and compare documents and values.

Index

Examples

Constants

This section is empty.

Variables

View Source 
var ErrFieldNotFound = errors.New("field not found")

ErrFieldNotFound must be returned by Document implementations, when calling the GetByField method and the field wasn't found in the document.

View Source 
var ErrStreamClosed = errors.New("stream closed")

ErrStreamClosed is used to indicate that a stream must be closed.

View Source 
var ErrValueNotFound = errors.New("value not found")

ErrValueNotFound must be returned by Array implementations, when calling the GetByIndex method and the index wasn't found in the array.

Functions

func ArrayContains 

func ArrayContains(a Array, v Value) (bool, error)

ArrayContains iterates over a and returns whether v is equal to one of its values.

func ArrayLength 

func ArrayLength(a Array) (int, error)

ArrayLength returns the length of an array.

func ArrayToJSON 

func ArrayToJSON(w io.Writer, a Array) error

ArrayToJSON encodes a to w in JSON.

func IteratorToJSON 

func IteratorToJSON(w io.Writer, s Iterator) error

IteratorToJSON encodes all the documents of an iterator to JSON stream.

func IteratorToJSONArray 

func IteratorToJSONArray(w io.Writer, s Iterator) error

IteratorToJSONArray encodes all the documents of an iterator to a JSON array.

func Length 

func Length(d Document) (int, error)

Length returns the length of a document.

func MapScan 

func MapScan(d Document, t interface{}) error

MapScan decodes the document into a map.

func Scan 

func Scan(d Document, targets ...interface{}) error

Scan each field of the document into the given variables.

func ScanValue 

func ScanValue(v Value, t interface{}) error

ScanValue scans v into t.

func SliceScan 

func SliceScan(a Array, t interface{}) error

SliceScan scans a document array into a slice or fixed size array. t must be a pointer to a valid slice or array.

It t is a slice pointer and its capacity is too low, a new slice will be allocated. Otherwise, its length is set to 0 so that its content is overrided.

If t is an array pointer, its capacity must be bigger than the length of a, otherwise an error is returned.

func StructScan 

func StructScan(d Document, t interface{}) error

StructScan scans d into t. t is expected to be a pointer to a struct.

By default, each struct field name is lowercased and the document's GetByField method is called with that name. If there is a match, the value is converted to the struct field type when possible, otherwise an error is returned. The decoding of each struct field can be customized by the format string stored under the "genji" key stored in the struct field's tag. The content of the format string is used instead of the struct field name and passed to the GetByField method.

func ToJSON 

func ToJSON(w io.Writer, d Document) error

ToJSON encodes d to w in JSON.

Types

type Array 

type Array interface {
	// Iterate goes through all the values of the array and calls the given function by passing each one of them.
	// If the given function returns an error, the iteration stops.
	Iterate(fn func(i int, value Value) error) error
	// GetByIndex returns a value by index of the array.
	GetByIndex(i int) (Value, error)
}

An Array contains a set of values.

func SortArray 

func SortArray(a Array) (Array, error)

SortArray creates a new sorted array. Types are sorted in the following ascending order:

  • NULL
  • Booleans
  • Numbers
  • Text / Blob
  • Arrays
  • Documents

It doesn't sort nested arrays.

type Document 

type Document interface {
	// Iterate goes through all the fields of the document and calls the given function by passing each one of them.
	// If the given function returns an error, the iteration stops.
	Iterate(fn func(field string, value Value) error) error
	// GetByField returns a value by field name.
	// Must return ErrFieldNotFound if the field doesnt exist.
	GetByField(field string) (Value, error)
}

A Document represents a group of key value pairs.

func NewFromJSON 

func NewFromJSON(data []byte) (Document, error)

NewFromJSON creates a document from a JSON object.

func NewFromMap 

func NewFromMap(m interface{}) (Document, error)

NewFromMap creates a document from a map. Due to the way maps are designed, iteration order is not guaranteed.

func NewFromStruct 

func NewFromStruct(s interface{}) (Document, error)

NewFromStruct creates a document from a struct using reflection.

type ErrUnsupportedType 

type ErrUnsupportedType struct {
	Value interface{}
	Msg   string
}

ErrUnsupportedType is used to skip struct or array fields that are not supported.

func (*ErrUnsupportedType) Error 

func (e *ErrUnsupportedType) Error() string

type FieldBuffer 

type FieldBuffer struct {
	// contains filtered or unexported fields
}

FieldBuffer stores a group of fields in memory. It implements the Document interface.

func NewFieldBuffer 

func NewFieldBuffer() *FieldBuffer

NewFieldBuffer creates a FieldBuffer.

func (*FieldBuffer) Add 

func (fb *FieldBuffer) Add(field string, v Value) *FieldBuffer

Add a field to the buffer.

func (*FieldBuffer) Copy 

func (fb *FieldBuffer) Copy(d Document) error

Copy deep copies every value of the document to the buffer. If a value is a document or an array, it will be stored as a FieldBuffer or ValueBuffer respectively.

func (*FieldBuffer) Delete 

func (fb *FieldBuffer) Delete(field string) error

Delete a field from the buffer.

func (FieldBuffer) GetByField 

func (fb FieldBuffer) GetByField(field string) (Value, error)

GetByField returns a value by field. Returns an error if the field doesn't exists.

func (FieldBuffer) Iterate 

func (fb FieldBuffer) Iterate(fn func(field string, value Value) error) error

Iterate goes through all the fields of the document and calls the given function by passing each one of them. If the given function returns an error, the iteration stops.

func (*FieldBuffer) Key 

func (fb *FieldBuffer) Key() []byte

Key of the document if any.

func (FieldBuffer) Len 

func (fb FieldBuffer) Len() int

Len of the buffer.

func (*FieldBuffer) MarshalJSON 

func (fb *FieldBuffer) MarshalJSON() ([]byte, error)

MarshalJSON implements the json.Marshaler interface.

func (*FieldBuffer) Replace 

func (fb *FieldBuffer) Replace(field string, v Value) error

Replace the value of the field by v.

func (*FieldBuffer) Reset 

func (fb *FieldBuffer) Reset()

Reset the buffer.

func (*FieldBuffer) ScanDocument 

func (fb *FieldBuffer) ScanDocument(d Document) error

ScanDocument copies all the fields of d to the buffer.

func (*FieldBuffer) Set 

func (fb *FieldBuffer) Set(f string, v Value)

Set replaces a field if it already exists or creates one if not.

func (*FieldBuffer) UnmarshalJSON 

func (fb *FieldBuffer) UnmarshalJSON(data []byte) error

UnmarshalJSON implements the json.Unmarshaler interface.

type Iterator 

type Iterator interface {
	// Iterate goes through all the documents and calls the given function by passing each one of them.
	// If the given function returns an error, the iteration stops.
	Iterate(func(d Document) error) error
}

An Iterator can iterate over documents.

func NewIterator 

func NewIterator(documents ...Document) Iterator

NewIterator creates an iterator that iterates over documents.

type Keyer 

type Keyer interface {
	Key() []byte
}

A Keyer returns the key identifying documents in their storage. This is usually implemented by documents read from storages.

type Scanner 

type Scanner interface {
	ScanDocument(Document) error
}

A Scanner can iterate over a document and scan all the fields.

type Stream 

type Stream struct {
	// contains filtered or unexported fields
}

Stream reads documents of an iterator one by one and passes them through a list of functions for transformation.

func NewStream 

func NewStream(it Iterator) Stream

NewStream creates a stream using the given iterator.

func (Stream) Append 

func (s Stream) Append(it Iterator) Stream

Append adds the given iterator to the stream.

func (Stream) Count 

func (s Stream) Count() (int, error)

Count counts all the documents from the stream.

func (Stream) Filter 

func (s Stream) Filter(fn func(d Document) (bool, error)) Stream

Filter each received document using fn. If fn returns true, the document is kept, otherwise it is skipped. If fn returns an error, the stream is interrupted.

func (Stream) First 

func (s Stream) First() (d Document, err error)

First runs the stream, returns the first document found and closes the stream. If the stream is empty, all return values are nil.

Example 
package main

import (
	"fmt"
	"log"

	"github.com/genjidb/genji"
	"github.com/genjidb/genji/document"
)

func main() {
	db, err := genji.Open(":memory:")
	if err != nil {
		log.Fatal(err)
	}
	defer db.Close()

	err = db.Exec("CREATE TABLE user")
	if err != nil {
		log.Fatal(err)
	}

	err = db.Exec("INSERT INTO user (id, name, age) VALUES (?, ?, ?)", 10, "foo", 15)
	if err != nil {
		log.Fatal(err)
	}

	result, err := db.Query("SELECT id, name, age FROM user WHERE name = ?", "foo")
	if err != nil {
		panic(err)
	}
	defer result.Close()

	d, err := result.First()
	if err != nil {
		panic(err)
	}

	var id uint64
	var name string
	var age uint8

	err = document.Scan(d, &id, &name, &age)
	if err != nil {
		panic(err)
	}

	fmt.Println(id, name, age)

}

Output:

10 foo 15

func (Stream) IsEmpty 

func (s Stream) IsEmpty() bool

IsEmpty returns whether the stream doesn't contain any iterator.

func (Stream) Iterate 

func (s Stream) Iterate(fn func(d Document) error) error

Iterate calls the underlying iterator's iterate method. If this stream was created using the Pipe method, it will apply fn to any document passed by the underlying iterator. If fn returns a document, it will be passed to the next stream. If it returns a nil document, the document will be ignored. If it returns an error, the stream will be interrupted and that error will bubble up and returned by fn, unless that error is ErrStreamClosed, in which case the Iterate method will stop the iteration and return nil. It implements the Iterator interface.

Example 
package main

import (
	"fmt"
	"log"

	"github.com/genjidb/genji"
	"github.com/genjidb/genji/document"
)

func main() {
	type User struct {
		ID      int64
		Name    string
		Age     uint32
		Address struct {
			City    string
			ZipCode string
		}
	}

	db, err := genji.Open(":memory:")
	if err != nil {
		log.Fatal(err)
	}
	defer db.Close()

	err = db.Exec("CREATE TABLE IF NOT EXISTS user")
	if err != nil {
		log.Fatal(err)
	}

	for i := 1; i <= 10; i++ {
		err = db.Exec("INSERT INTO user VALUES ?", &User{
			ID:   int64(i),
			Name: fmt.Sprintf("foo%d", i),
			Age:  uint32(i * 10),
			Address: struct {
				City    string
				ZipCode string
			}{
				City:    "Lyon",
				ZipCode: fmt.Sprintf("69%03d", i),
			},
		})
		if err != nil {
			log.Fatal(err)
		}
	}

	result, err := db.Query(`SELECT id, name, age, address FROM user WHERE age >= 18`)
	if err != nil {
		panic(err)
	}
	defer result.Close()

	err = result.Iterate(func(d document.Document) error {
		// Scan into a struct
		var u User
		err = document.StructScan(d, &u)
		if err != nil {
			return err
		}

		fmt.Println(u)

		// Or scan individual variables
		// Types of variables don't have to exactly match with the types stored
		var id uint64
		var name []byte
		var age uint8
		var address map[string]string

		err = document.Scan(d, &id, &name, &age, &address)
		if err != nil {
			return err
		}

		fmt.Println(id, string(name), age, address)
		return nil
	})
	if err != nil {
		panic(err)
	}

}

Output:

{2 foo2 20 {Lyon 69002}}
2 foo2 20 map[city:Lyon zipcode:69002]
{3 foo3 30 {Lyon 69003}}
3 foo3 30 map[city:Lyon zipcode:69003]
{4 foo4 40 {Lyon 69004}}
4 foo4 40 map[city:Lyon zipcode:69004]
{5 foo5 50 {Lyon 69005}}
5 foo5 50 map[city:Lyon zipcode:69005]
{6 foo6 60 {Lyon 69006}}
6 foo6 60 map[city:Lyon zipcode:69006]
{7 foo7 70 {Lyon 69007}}
7 foo7 70 map[city:Lyon zipcode:69007]
{8 foo8 80 {Lyon 69008}}
8 foo8 80 map[city:Lyon zipcode:69008]
{9 foo9 90 {Lyon 69009}}
9 foo9 90 map[city:Lyon zipcode:69009]
{10 foo10 100 {Lyon 69010}}
10 foo10 100 map[city:Lyon zipcode:69010]

func (Stream) Limit 

func (s Stream) Limit(n int) Stream

Limit interrupts the stream once the number of passed documents have reached n.

func (Stream) Map 

func (s Stream) Map(fn func(d Document) (Document, error)) Stream

Map applies fn to each received document and passes it to the next stream. If fn returns an error, the stream is interrupted.

func (Stream) Offset 

func (s Stream) Offset(n int) Stream

Offset ignores n documents then passes the subsequent ones to the stream.

func (Stream) Pipe 

func (s Stream) Pipe(op StreamOperator) Stream

Pipe creates a new Stream who can read its data from s and apply op to every document passed by its Iterate method.

type StreamOperator 

type StreamOperator func() func(d Document) (Document, error)

An StreamOperator is used to modify a stream. If a stream operator returns a document, it will be passed to the next stream. If it returns a nil document, the document will be ignored. If it returns an error, the stream will be interrupted and that error will bubble up and returned by this function, unless that error is ErrStreamClosed, in which case the Iterate method will stop the iteration and return nil. Stream operators can be reused, and thus, any state or side effect should be kept within the operator closure unless the nature of the operator prevents that.

type Value 

type Value struct {
	Type ValueType
	V    interface{}
}

A Value stores encoded data alongside its type.

func NewArrayValue 

func NewArrayValue(a Array) Value

NewArrayValue returns a value of type Array.

func NewBlobValue 

func NewBlobValue(x []byte) Value

NewBlobValue encodes x and returns a value.

func NewBoolValue 

func NewBoolValue(x bool) Value

NewBoolValue encodes x and returns a value.

func NewDocumentValue 

func NewDocumentValue(d Document) Value

NewDocumentValue returns a value of type Document.

func NewDoubleValue added in v0.7.0 

func NewDoubleValue(x float64) Value

NewDoubleValue encodes x and returns a value.

func NewDurationValue 

func NewDurationValue(d time.Duration) Value

NewDurationValue returns a value of type Duration.

func NewIntegerValue added in v0.7.0 

func NewIntegerValue(x int64) Value

NewIntegerValue encodes x and returns a value whose type depends on the magnitude of x.

func NewNullValue 

func NewNullValue() Value

NewNullValue returns a Null value.

func NewTextValue 

func NewTextValue(x string) Value

NewTextValue encodes x and returns a value.

func NewValue 

func NewValue(x interface{}) (Value, error)

NewValue creates a value whose type is infered from x.

func NewZeroValue 

func NewZeroValue(t ValueType) Value

NewZeroValue returns a value whose value is equal to the Go zero value of the selected type.

func (Value) Add 

func (v Value) Add(u Value) (res Value, err error)

Add u to v and return the result. Only numeric values and booleans can be added together.

func (Value) BitwiseAnd 

func (v Value) BitwiseAnd(u Value) (res Value, err error)

BitwiseAnd calculates v & u and returns the result. Only numeric values and booleans can be calculated together. If both v and u are integers, the result will be an integer.

func (Value) BitwiseOr 

func (v Value) BitwiseOr(u Value) (res Value, err error)

BitwiseOr calculates v | u and returns the result. Only numeric values and booleans can be calculated together. If both v and u are integers, the result will be an integer.

func (Value) BitwiseXor 

func (v Value) BitwiseXor(u Value) (res Value, err error)

BitwiseXor calculates v ^ u and returns the result. Only numeric values and booleans can be calculated together. If both v and u are integers, the result will be an integer.

func (Value) CastAs added in v0.7.0 

func (v Value) CastAs(t ValueType) (Value, error)

CastAs casts v as the selected type when possible.

func (Value) CastAsArray added in v0.7.0 

func (v Value) CastAsArray() (Value, error)

CastAsArray casts according to the following rules: Text: decodes a JSON array, otherwise fails. Any other type is considered an invalid cast.

func (Value) CastAsBlob added in v0.7.0 

func (v Value) CastAsBlob() (Value, error)

CastAsBlob casts according to the following rules: Text: decodes a base64 string, otherwise fails. Any other type is considered an invalid cast.

func (Value) CastAsBool added in v0.7.0 

func (v Value) CastAsBool() (Value, error)

CastAsBool casts according to the following rules: Integer: true if truthy, otherwise false. Text: uses strconv.Parsebool to determine the boolean value, it fails if the text doesn't contain a valid boolean. Any other type is considered an invalid cast.

func (Value) CastAsDocument added in v0.7.0 

func (v Value) CastAsDocument() (Value, error)

CastAsDocument casts according to the following rules: Text: decodes a JSON object, otherwise fails. Any other type is considered an invalid cast.

func (Value) CastAsDouble added in v0.7.0 

func (v Value) CastAsDouble() (Value, error)

CastAsDouble casts according to the following rules: Integer: returns a double version of the integer. Text: uses strconv.ParseFloat to determine the double value, it fails if the text doesn't contain a valid float value. Any other type is considered an invalid cast.

func (Value) CastAsDuration added in v0.7.0 

func (v Value) CastAsDuration() (Value, error)

CastAsDuration casts according to the following rules: Text: decodes using time.ParseDuration, otherwise fails. Any other type is considered an invalid cast.

func (Value) CastAsInteger added in v0.7.0 

func (v Value) CastAsInteger() (Value, error)

CastAsInteger casts according to the following rules: Bool: returns 1 if true, 0 if false. Double: cuts off the decimal and remaining numbers. Duration: returns the number of nanoseconds in the duration. Text: uses strconv.ParseInt to determine the integer value, then casts it to an integer. If it fails uses strconv.ParseFloat to determine the double value, then casts it to an integer It fails if the text doesn't contain a valid float value. Any other type is considered an invalid cast.

func (Value) CastAsText added in v0.7.0 

func (v Value) CastAsText() (Value, error)

CastAsText returns a JSON representation of v. If the representation is a string, it gets unquoted.

func (Value) Div 

func (v Value) Div(u Value) (res Value, err error)

Div calculates v / u and returns the result. Only numeric values and booleans can be calculated together. If both v and u are integers, the result will be an integer.

func (Value) IsEqual 

func (v Value) IsEqual(other Value) (bool, error)

IsEqual returns true if v is equal to the given value.

func (Value) IsGreaterThan 

func (v Value) IsGreaterThan(other Value) (bool, error)

IsGreaterThan returns true if v is greather than the given value.

func (Value) IsGreaterThanOrEqual 

func (v Value) IsGreaterThanOrEqual(other Value) (bool, error)

IsGreaterThanOrEqual returns true if v is greather than or equal to the given value.

func (Value) IsLesserThan 

func (v Value) IsLesserThan(other Value) (bool, error)

IsLesserThan returns true if v is lesser than the given value.

func (Value) IsLesserThanOrEqual 

func (v Value) IsLesserThanOrEqual(other Value) (bool, error)

IsLesserThanOrEqual returns true if v is lesser than or equal to the given value.

func (Value) IsNotEqual 

func (v Value) IsNotEqual(other Value) (bool, error)

IsNotEqual returns true if v is not equal to the given value.

func (Value) IsTruthy 

func (v Value) IsTruthy() (bool, error)

IsTruthy returns whether v is not equal to the zero value of its type.

func (Value) IsZeroValue 

func (v Value) IsZeroValue() (bool, error)

IsZeroValue indicates if the value data is the zero value for the value type. This function doesn't perform any allocation.

func (Value) MarshalJSON 

func (v Value) MarshalJSON() ([]byte, error)

MarshalJSON implements the json.Marshaler interface.

func (Value) Mod 

func (v Value) Mod(u Value) (res Value, err error)

Mod calculates v / u and returns the result. Only numeric values and booleans can be calculated together. If both v and u are integers, the result will be an integer.

func (Value) Mul 

func (v Value) Mul(u Value) (res Value, err error)

Mul calculates v * u and returns the result. Only numeric values and booleans can be calculated together.

func (Value) Scan 

func (v Value) Scan(t interface{}) error

Scan v into t.

func (Value) String 

func (v Value) String() string

String returns a string representation of the value. It implements the fmt.Stringer interface.

func (Value) Sub 

func (v Value) Sub(u Value) (res Value, err error)

Sub calculates v - u and returns the result. Only numeric values and booleans can be calculated together.

type ValueBuffer 

type ValueBuffer []Value

ValueBuffer is an array that holds values in memory.

func NewValueBuffer 

func NewValueBuffer(values ...Value) ValueBuffer

NewValueBuffer creates a buffer of values.

func (ValueBuffer) Append 

func (vb ValueBuffer) Append(v Value) ValueBuffer

Append a value to the buffer and return a new buffer.

func (*ValueBuffer) Copy 

func (vb *ValueBuffer) Copy(a Array) error

Copy deep copies all the values from the given array. If a value is a document or an array, it will be stored as a FieldBuffer or ValueBuffer respectively.

func (ValueBuffer) GetByIndex 

func (vb ValueBuffer) GetByIndex(i int) (Value, error)

GetByIndex returns a value set at the given index. If the index is out of range it returns an error.

func (ValueBuffer) Iterate 

func (vb ValueBuffer) Iterate(fn func(i int, value Value) error) error

Iterate over all the values of the buffer. It implements the Array interface.

func (*ValueBuffer) MarshalJSON 

func (vb *ValueBuffer) MarshalJSON() ([]byte, error)

MarshalJSON implements the json.Marshaler interface.

func (*ValueBuffer) Replace 

func (vb *ValueBuffer) Replace(index int, v Value) error

Replace the value of the index by v.

func (*ValueBuffer) ScanArray 

func (vb *ValueBuffer) ScanArray(a Array) error

ScanArray copies all the values of a to the buffer.

func (*ValueBuffer) UnmarshalJSON 

func (vb *ValueBuffer) UnmarshalJSON(data []byte) error

UnmarshalJSON implements the json.Unmarshaler interface.

type ValuePath 

type ValuePath []ValuePathFragment

A ValuePath represents the path to a particular value within a document.

func (ValuePath) GetValue 

func (p ValuePath) GetValue(d Document) (Value, error)

GetValue from a document.

func (ValuePath) String 

func (p ValuePath) String() string

String representation of all the fragments of the path. It implements the Stringer interface.

type ValuePathFragment added in v0.7.0 

type ValuePathFragment struct {
	FieldName  string
	ArrayIndex int
}

ValuePathFragment is a fragment of a path representing either a field name or the index of an array.

type ValueType 

type ValueType uint8

ValueType represents a value type supported by the database. 

const (
	NullValue ValueType = 0x1

	BoolValue ValueType = 0x5

	// integer family: 0x10 to 0x1F
	IntegerValue ValueType = 0x10

	// double family: 0x20 to 0x2F
	DoubleValue ValueType = 0x20

	// time family: 0x30 to 0x3F
	DurationValue ValueType = 0x30

	// string family: 0x40 to 0x4F
	TextValue ValueType = 0x40

	// blob family: 0x50 to 0x5F
	BlobValue ValueType = 0x50

	// array family: 0x60 to 0x6F
	ArrayValue ValueType = 0x60

	// document family: 0x70 to 0x7F
	DocumentValue ValueType = 0x70
)

List of supported value types. These types are separated by family so that when new types are introduced we don't need to modify them.

func (ValueType) IsNumber 

func (t ValueType) IsNumber() bool

IsNumber returns true if t is either an integer of a float.

func (ValueType) String 

func (t ValueType) String() string

Source Files

View all Source files

Directories

Path Synopsis
encoding
msgpack

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.