plugeth/swarm/shed/index.go
2019-01-07 13:20:11 +01:00

307 lines
9.0 KiB
Go

// Copyright 2018 The go-ethereum Authors
// This file is part of the go-ethereum library.
//
// The go-ethereum library is free software: you can redistribute it and/or modify
// it under the terms of the GNU Lesser General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// The go-ethereum library is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Lesser General Public License for more details.
//
// You should have received a copy of the GNU Lesser General Public License
// along with the go-ethereum library. If not, see <http://www.gnu.org/licenses/>.
package shed
import (
"bytes"
"github.com/syndtr/goleveldb/leveldb"
)
// Item holds fields relevant to Swarm Chunk data and metadata.
// All information required for swarm storage and operations
// on that storage must be defined here.
// This structure is logically connected to swarm storage,
// the only part of this package that is not generalized,
// mostly for performance reasons.
//
// Item is a type that is used for retrieving, storing and encoding
// chunk data and metadata. It is passed as an argument to Index encoding
// functions, get function and put function.
// But it is also returned with additional data from get function call
// and as the argument in iterator function definition.
type Item struct {
Address []byte
Data []byte
AccessTimestamp int64
StoreTimestamp int64
// UseMockStore is a pointer to identify
// an unset state of the field in Join function.
UseMockStore *bool
}
// Merge is a helper method to construct a new
// Item by filling up fields with default values
// of a particular Item with values from another one.
func (i Item) Merge(i2 Item) (new Item) {
if i.Address == nil {
i.Address = i2.Address
}
if i.Data == nil {
i.Data = i2.Data
}
if i.AccessTimestamp == 0 {
i.AccessTimestamp = i2.AccessTimestamp
}
if i.StoreTimestamp == 0 {
i.StoreTimestamp = i2.StoreTimestamp
}
if i.UseMockStore == nil {
i.UseMockStore = i2.UseMockStore
}
return i
}
// Index represents a set of LevelDB key value pairs that have common
// prefix. It holds functions for encoding and decoding keys and values
// to provide transparent actions on saved data which inclide:
// - getting a particular Item
// - saving a particular Item
// - iterating over a sorted LevelDB keys
// It implements IndexIteratorInterface interface.
type Index struct {
db *DB
prefix []byte
encodeKeyFunc func(fields Item) (key []byte, err error)
decodeKeyFunc func(key []byte) (e Item, err error)
encodeValueFunc func(fields Item) (value []byte, err error)
decodeValueFunc func(keyFields Item, value []byte) (e Item, err error)
}
// IndexFuncs structure defines functions for encoding and decoding
// LevelDB keys and values for a specific index.
type IndexFuncs struct {
EncodeKey func(fields Item) (key []byte, err error)
DecodeKey func(key []byte) (e Item, err error)
EncodeValue func(fields Item) (value []byte, err error)
DecodeValue func(keyFields Item, value []byte) (e Item, err error)
}
// NewIndex returns a new Index instance with defined name and
// encoding functions. The name must be unique and will be validated
// on database schema for a key prefix byte.
func (db *DB) NewIndex(name string, funcs IndexFuncs) (f Index, err error) {
id, err := db.schemaIndexPrefix(name)
if err != nil {
return f, err
}
prefix := []byte{id}
return Index{
db: db,
prefix: prefix,
// This function adjusts Index LevelDB key
// by appending the provided index id byte.
// This is needed to avoid collisions between keys of different
// indexes as all index ids are unique.
encodeKeyFunc: func(e Item) (key []byte, err error) {
key, err = funcs.EncodeKey(e)
if err != nil {
return nil, err
}
return append(append(make([]byte, 0, len(key)+1), prefix...), key...), nil
},
// This function reverses the encodeKeyFunc constructed key
// to transparently work with index keys without their index ids.
// It assumes that index keys are prefixed with only one byte.
decodeKeyFunc: func(key []byte) (e Item, err error) {
return funcs.DecodeKey(key[1:])
},
encodeValueFunc: funcs.EncodeValue,
decodeValueFunc: funcs.DecodeValue,
}, nil
}
// Get accepts key fields represented as Item to retrieve a
// value from the index and return maximum available information
// from the index represented as another Item.
func (f Index) Get(keyFields Item) (out Item, err error) {
key, err := f.encodeKeyFunc(keyFields)
if err != nil {
return out, err
}
value, err := f.db.Get(key)
if err != nil {
return out, err
}
out, err = f.decodeValueFunc(keyFields, value)
if err != nil {
return out, err
}
return out.Merge(keyFields), nil
}
// Put accepts Item to encode information from it
// and save it to the database.
func (f Index) Put(i Item) (err error) {
key, err := f.encodeKeyFunc(i)
if err != nil {
return err
}
value, err := f.encodeValueFunc(i)
if err != nil {
return err
}
return f.db.Put(key, value)
}
// PutInBatch is the same as Put method, but it just
// saves the key/value pair to the batch instead
// directly to the database.
func (f Index) PutInBatch(batch *leveldb.Batch, i Item) (err error) {
key, err := f.encodeKeyFunc(i)
if err != nil {
return err
}
value, err := f.encodeValueFunc(i)
if err != nil {
return err
}
batch.Put(key, value)
return nil
}
// Delete accepts Item to remove a key/value pair
// from the database based on its fields.
func (f Index) Delete(keyFields Item) (err error) {
key, err := f.encodeKeyFunc(keyFields)
if err != nil {
return err
}
return f.db.Delete(key)
}
// DeleteInBatch is the same as Delete just the operation
// is performed on the batch instead on the database.
func (f Index) DeleteInBatch(batch *leveldb.Batch, keyFields Item) (err error) {
key, err := f.encodeKeyFunc(keyFields)
if err != nil {
return err
}
batch.Delete(key)
return nil
}
// IndexIterFunc is a callback on every Item that is decoded
// by iterating on an Index keys.
// By returning a true for stop variable, iteration will
// stop, and by returning the error, that error will be
// propagated to the called iterator method on Index.
type IndexIterFunc func(item Item) (stop bool, err error)
// IterateOptions defines optional parameters for Iterate function.
type IterateOptions struct {
// StartFrom is the Item to start the iteration from.
StartFrom *Item
// If SkipStartFromItem is true, StartFrom item will not
// be iterated on.
SkipStartFromItem bool
// Iterate over items which keys have a common prefix.
Prefix []byte
}
// Iterate function iterates over keys of the Index.
// If IterateOptions is nil, the iterations is over all keys.
func (f Index) Iterate(fn IndexIterFunc, options *IterateOptions) (err error) {
if options == nil {
options = new(IterateOptions)
}
// construct a prefix with Index prefix and optional common key prefix
prefix := append(f.prefix, options.Prefix...)
// start from the prefix
startKey := prefix
if options.StartFrom != nil {
// start from the provided StartFrom Item key value
startKey, err = f.encodeKeyFunc(*options.StartFrom)
if err != nil {
return err
}
}
it := f.db.NewIterator()
defer it.Release()
// move the cursor to the start key
ok := it.Seek(startKey)
if !ok {
// stop iterator if seek has failed
return it.Error()
}
if options.SkipStartFromItem && bytes.Equal(startKey, it.Key()) {
// skip the start from Item if it is the first key
// and it is explicitly configured to skip it
ok = it.Next()
}
for ; ok; ok = it.Next() {
key := it.Key()
if !bytes.HasPrefix(key, prefix) {
break
}
// create a copy of key byte slice not to share leveldb underlaying slice array
keyItem, err := f.decodeKeyFunc(append([]byte(nil), key...))
if err != nil {
return err
}
// create a copy of value byte slice not to share leveldb underlaying slice array
valueItem, err := f.decodeValueFunc(keyItem, append([]byte(nil), it.Value()...))
if err != nil {
return err
}
stop, err := fn(keyItem.Merge(valueItem))
if err != nil {
return err
}
if stop {
break
}
}
return it.Error()
}
// Count returns the number of items in index.
func (f Index) Count() (count int, err error) {
it := f.db.NewIterator()
defer it.Release()
for ok := it.Seek(f.prefix); ok; ok = it.Next() {
key := it.Key()
if key[0] != f.prefix[0] {
break
}
count++
}
return count, it.Error()
}
// CountFrom returns the number of items in index keys
// starting from the key encoded from the provided Item.
func (f Index) CountFrom(start Item) (count int, err error) {
startKey, err := f.encodeKeyFunc(start)
if err != nil {
return 0, err
}
it := f.db.NewIterator()
defer it.Release()
for ok := it.Seek(startKey); ok; ok = it.Next() {
key := it.Key()
if key[0] != f.prefix[0] {
break
}
count++
}
return count, it.Error()
}