featurebase/rrtx.go
Seebs 5ffa4ba803 drop "batched" flag from Add operation
The "batched" flag creates a complexity which is that the return value of Add
might or might not be meaningful, but it doesn't really buy us very much.

If we are concerned about the ops log size of writing single ops as 21-byte
arrays of 1 op rather than as 13-byte ops, we can make the AddN code smarter
about how it writes ops. And probably should.

Along with this, change Remove to use the batched operation form, which
writes a more meaningful ops log, and return a meaningful value for changes
made. Otherwise, it ends up writing potentially thousands of ops to the
ops log without reporting any OpN, because the number of ops written isn't
the same as the number of changes those ops made. This could result in
files growing by megabytes without OpN changing.

There was a comment here about a test failing with RemoveN. I can't prove
it, but I strongly suspect that this was actually a result of that test
case hitting a particular bug that we eventually fixed, and which we might
have fixed sooner if we'd realized why using RemoveN made that test
fail.
2021-04-20 12:03:25 -05:00

759 lines
20 KiB
Go

// Copyright 2020 Pilosa Corp.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package pilosa
import (
"bytes"
"fmt"
"io"
"os"
"path/filepath"
"sort"
"strconv"
"strings"
"sync"
"sync/atomic"
"github.com/pilosa/pilosa/v2/roaring"
txkey "github.com/pilosa/pilosa/v2/short_txkey"
"github.com/pilosa/pilosa/v2/storage"
. "github.com/pilosa/pilosa/v2/vprint" // nolint:staticcheck
"github.com/pkg/errors"
)
// RoaringTx represents a fake transaction object for Roaring storage.
type RoaringTx struct {
write bool
Index *Index
Field *Field
fragment *fragment
o Txo
sn int64 // serial number
done bool
mu sync.Mutex // protect done as it changes state
w *RoaringWrapper
}
func (tx *RoaringTx) IsDone() (done bool) {
tx.mu.Lock()
done = tx.done
tx.mu.Unlock()
return
}
func (tx *RoaringTx) Type() string {
return RoaringTxn
}
func (tx *RoaringTx) Dump(short bool, shard uint64) {
o := tx.o
o.Shard = shard
fmt.Printf("%v\n", tx.Index.StringifiedRoaringKeys(short, false, o))
}
func (tx *RoaringTx) UseRowCache() bool {
return storage.EnableRowCache()
}
// based on view.openFragments()
func roaringMapOfShards(optionalViewPath string) (shardMap map[uint64]bool, err error) {
shardMap = make(map[uint64]bool)
path := filepath.Join(optionalViewPath, "fragments")
file, err := os.Open(path)
if os.IsNotExist(err) {
return
} else if err != nil {
return nil, errors.Wrap(err, "opening fragments directory")
}
defer file.Close()
fis, err := file.Readdir(0)
if err != nil {
return nil, errors.Wrap(err, "reading fragments directory")
}
for _, fi := range fis {
//vv("rrtx next fi = '%v'", fi.Name())
if fi.IsDir() {
continue
}
name := fi.Name()
if strings.HasSuffix(name, ".cache") {
continue
}
// Parse filename into integer.
shard, err := strconv.ParseUint(filepath.Base(name), 10, 64)
if err != nil {
//vv("WARNING: couldn't use non-integer file as shard in index/field/view %s/%s/%s: %s", index, field, view, fi.Name())
//panic(fmt.Sprintf("WARNING: couldn't use non-integer file as shard in index/field/view %s/%s/%s: %s", index, field, view, fi.Name()))
//tx.Index.holder.Logger.Debugf("WARNING: couldn't use non-integer file as shard in index/field/view %s/%s/%s: %s", index, field, view, fi.Name())
continue
}
shardMap[shard] = true
}
return
}
func (tx *RoaringTx) Pointer() string {
return fmt.Sprintf("%p", tx)
}
// NewTxIterator returns a *roaring.Iterator that MUST have Close() called on it BEFORE
// the transaction Commits or Rollsback.
func (tx *RoaringTx) NewTxIterator(index, field, view string, shard uint64) *roaring.Iterator {
b, err := tx.bitmap(index, field, view, shard)
PanicOn(err)
return b.Iterator()
}
// ImportRoaringBits return values changed and rowSet will be inaccurate if
// the data []byte is supplied. This mimics the traditional roaring-per-file
// and should be faster.
func (tx *RoaringTx) ImportRoaringBits(index, field, view string, shard uint64, rit roaring.RoaringIterator, clear bool, log bool, rowSize uint64, data []byte) (changed int, rowSet map[uint64]int, err error) {
f, err := tx.getFragment(index, field, view, shard)
if err != nil {
return 0, nil, err
}
if len(data) > 0 {
// changed and rowSet are ignored anyway when len(data) > 0;
// when we are called from fragment.fillFragmentFromArchive()
// which is the only place the data []byte is supplied.
// blueGreenTx also turns off the checks in this case.
return 0, nil, f.readStorageFromArchive(bytes.NewBuffer(data))
}
changed, rowSet, err = f.storage.ImportRoaringRawIterator(rit, clear, true, rowSize)
return
}
func (tx *RoaringTx) Readonly() bool {
return !tx.write
}
func (tx *RoaringTx) IncrementOpN(index, field, view string, shard uint64, changedN int) {
frag, err := tx.getFragment(index, field, view, shard)
PanicOn(err)
frag.incrementOpN(changedN)
}
func (c *RoaringTx) ApplyFilter(index, field, view string, shard uint64, ckey uint64, filter roaring.BitmapFilter) (err error) {
return GenericApplyFilter(c, index, field, view, shard, ckey, filter)
}
// Rollback
func (tx *RoaringTx) Rollback() {
tx.w.CleanupTx(tx)
}
// Commit
func (tx *RoaringTx) Commit() error {
tx.w.CleanupTx(tx)
return nil
}
func (tx *RoaringTx) RoaringBitmap(index, field, view string, shard uint64) (*roaring.Bitmap, error) {
return tx.bitmap(index, field, view, shard)
}
func (tx *RoaringTx) Container(index, field, view string, shard uint64, key uint64) (*roaring.Container, error) {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return nil, err
}
return b.Containers.Get(key), nil
}
func (tx *RoaringTx) PutContainer(index, field, view string, shard uint64, key uint64, c *roaring.Container) error {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return err
}
b.Containers.Put(key, c)
return nil
}
func (tx *RoaringTx) RemoveContainer(index, field, view string, shard uint64, key uint64) error {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return err
}
b.Containers.Remove(key)
return nil
}
func (tx *RoaringTx) Add(index, field, view string, shard uint64, a ...uint64) (changeCount int, err error) {
//vv("RoaringTx.Add(index='%v', shard='%v') stack=\n%v", index, shard, stack())
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return 0, err
}
// Note: do not replace b.AddN() with b.DirectAddN().
// DirectAddN() does not do op-log operations inside roaring, so the
// on-disk representation no longer matches the in-memory operations.
count, err := b.AddN(a...)
return count, err
}
func (tx *RoaringTx) Remove(index, field, view string, shard uint64, a ...uint64) (changeCount int, err error) {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return 0, err
}
return b.RemoveN(a...)
}
func (tx *RoaringTx) Contains(index, field, view string, shard uint64, v uint64) (exists bool, err error) {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return false, err
}
return b.Contains(v), nil
}
func (tx *RoaringTx) ContainerIterator(index, field, view string, shard uint64, key uint64) (citer roaring.ContainerIterator, found bool, err error) {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return nil, false, errors.Wrap(err, "getting bitmap")
}
//vv("b bitmap back from bitmap(index='%v', field='%v', view='%v', shard='%v')='%#v'", index, field, view, shard, b.Slice())
citer, found = b.Containers.Iterator(key)
return citer, found, nil
}
func (tx *RoaringTx) ForEach(index, field, view string, shard uint64, fn func(i uint64) error) error {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return err
}
return b.ForEach(fn)
}
func (tx *RoaringTx) ForEachRange(index, field, view string, shard uint64, start, end uint64, fn func(uint64) error) error {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return err
}
return b.ForEachRange(start, end, fn)
}
func (tx *RoaringTx) Count(index, field, view string, shard uint64) (uint64, error) {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return 0, err
}
return b.Count(), nil
}
func (tx *RoaringTx) Max(index, field, view string, shard uint64) (uint64, error) {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return 0, err
}
return b.Max(), nil
}
func (tx *RoaringTx) Min(index, field, view string, shard uint64) (uint64, bool, error) {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return 0, false, err
}
v, ok := b.Min()
return v, ok, nil
}
func (tx *RoaringTx) UnionInPlace(index, field, view string, shard uint64, others ...*roaring.Bitmap) error {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return err
}
b.UnionInPlace(others...)
return nil
}
func (tx *RoaringTx) CountRange(index, field, view string, shard uint64, start, end uint64) (uint64, error) {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return 0, err
}
return b.CountRange(start, end), nil
}
func (tx *RoaringTx) OffsetRange(index, field, view string, shard uint64, offset, start, end uint64) (*roaring.Bitmap, error) {
b, err := tx.bitmap(index, field, view, shard)
if err != nil {
return nil, err
}
return b.OffsetRange(offset, start, end), nil
}
// getFragment is used by IncrementOpN() and by bitmap()
func (tx *RoaringTx) getFragment(index, field, view string, shard uint64) (*fragment, error) {
// If a fragment is attached, always use it. Since it was set at Tx creation,
// it is highly likely to be correct.
if tx.fragment != nil {
// but still a basic sanity check.
if tx.fragment.index() != index ||
tx.fragment.field() != field ||
tx.fragment.view() != view ||
tx.fragment.shard != shard {
// still insist that index and shard match, since that is the current scope of all Tx.
if tx.fragment.index() != index ||
tx.fragment.shard != shard {
panic(fmt.Sprintf("different fragment cached vs requested. index='%v', field='%v'; view='%v'; shard='%v'; tx.fragment='%#v'", index, field, view, shard, tx.fragment))
}
// cannot use this fragment.
tx.fragment = nil
} else {
return tx.fragment, nil
}
}
// If a field is attached, start from there.
// Otherwise look up the field from the index.
f := tx.Field
if f == nil {
// we cannot assume that the tx.Index that we "started" on is the same
// as the index we are being queried; it might be foreign: TestExecutor_ForeignIndex
// So go through the holder
idx := tx.Index.holder.Index(index)
if idx == nil {
// only thing we can try is the cached index, and hope we aren't being asked for a foreign index.
f = tx.Index.Field(field)
if f == nil {
return nil, newNotFoundError(ErrFieldNotFound, field)
}
} else {
if f = idx.Field(field); f == nil {
return nil, newNotFoundError(ErrFieldNotFound, field)
}
}
}
// INVAR: f is not nil.
v := f.view(view)
if v == nil {
return nil, errors.Wrapf(ViewNotFound, "getting %s", view)
}
frag := v.Fragment(shard)
if frag == nil {
return nil, errors.Wrapf(FragmentNotFound, "field:%q, view:%q, shard:%d", field, view, shard)
}
// Note: we cannot cache frag into tx.fragment.
// Empirically, it breaks 245 top-level pilosa tests.
// tx.fragment = frag // breaks the world.
return frag, nil
}
const ViewNotFound = Error("view not found")
const FragmentNotFound = Error("fragment not found")
func (tx *RoaringTx) bitmap(index, field, view string, shard uint64) (*roaring.Bitmap, error) {
frag, err := tx.getFragment(index, field, view, shard)
if err != nil {
return nil, errors.Wrap(err, "getFragment")
}
return frag.storage, nil
}
func (tx *RoaringTx) RoaringBitmapReader(index, field, view string, shard uint64, fragmentPathForRoaring string) (r io.ReadCloser, sz int64, err error) {
file, err := os.Open(fragmentPathForRoaring) // open the fragment file
if err != nil {
return nil, -1, err
}
fi, err := file.Stat()
if err != nil {
return nil, -1, errors.Wrap(err, "statting")
}
sz = fi.Size()
r = file
return
}
func (tx *RoaringTx) Group() *TxGroup {
return tx.o.Group
}
func (tx *RoaringTx) Options() Txo {
return tx.o
}
// Sn retreives the serial number of the Tx.
func (tx *RoaringTx) Sn() int64 {
return tx.sn
}
func roaringGetFieldView2Shards(idx *Index) (vs *FieldView2Shards, err error) {
vs = NewFieldView2Shards()
// A) open the index directory
f, err := os.Open(idx.FieldsPath())
if err != nil {
return nil, errors.Wrap(err, "opening directory")
}
defer f.Close()
fieldFIs, err := f.Readdir(0)
if err != nil {
return nil, errors.Wrap(err, "reading directory")
}
//vv("roaringGetFieldView2Shards A) opened index path '%v'", idx.path)
// B) read the name of each field under the index
for _, loopFieldFi := range fieldFIs {
fieldFI := loopFieldFi
if !fieldFI.IsDir() {
continue
}
field := fieldFI.Name()
//vv("roaringGetFieldView2Shards B) on field '%v'", field)
fieldPath := filepath.Join(idx.FieldsPath(), field)
viewsDir := filepath.Join(fieldPath, "views")
file, err := os.Open(viewsDir)
if os.IsNotExist(err) {
//return nil
continue
} else if err != nil {
return nil, errors.Wrapf(err, "opening view directory '%v'", viewsDir)
}
defer file.Close()
// C) read the name of each view under the field
viewFIs, err := file.Readdir(0)
if err != nil {
return nil, errors.Wrapf(err, "reading views directory '%v'", viewsDir)
}
for _, viewFI := range viewFIs {
if !viewFI.IsDir() {
continue
}
view := viewFI.Name()
roaringViewPath := filepath.Join(viewsDir, view)
shardMap, err := roaringMapOfShards(roaringViewPath)
if err != nil {
return nil, errors.Wrapf(err, "reading view path directory '%v'", roaringViewPath)
}
if len(shardMap) == 0 {
//vv("roaringGetFieldView2Shards C) SAVED SPACE! field '%v' view '%v' had no shards", field, view)
continue
}
ss := newShardSetFromMap(shardMap)
fv := txkey.FieldView{Field: field, View: view}
vs.addViewShardSet(fv, ss)
//vv("roaringGetFieldView2Shards C) added field '%v' view '%v' with shards '%#v'", field, view, ss.shards)
}
}
return
}
// inefficient for roaring. Instead use the roaringGetFieldView2Shards() above.
func (tx *RoaringTx) GetSortedFieldViewList(idx *Index, shard uint64) (fvs []txkey.FieldView, err error) {
// A) open the index directory
f, err := os.Open(idx.FieldsPath())
if err != nil {
return nil, errors.Wrap(err, "opening directory")
}
defer f.Close()
fieldFIs, err := f.Readdir(0)
if err != nil {
return nil, errors.Wrap(err, "reading directory")
}
//vv("A) shard %v, opened index path '%v'", shard, idx.path)
// B) read the name of each field under the index
for _, loopFieldFi := range fieldFIs {
fieldFI := loopFieldFi
if !fieldFI.IsDir() {
continue
}
field := fieldFI.Name()
//vv("B) on field '%v'", field)
fieldPath := filepath.Join(idx.FieldsPath(), field)
viewsDir := filepath.Join(fieldPath, "views")
file, err := os.Open(viewsDir)
if os.IsNotExist(err) {
//return nil
continue
} else if err != nil {
return nil, errors.Wrapf(err, "opening view directory '%v'", viewsDir)
}
defer file.Close()
// C) read the name of each view under the field
viewFIs, err := file.Readdir(0)
if err != nil {
return nil, errors.Wrapf(err, "reading views directory '%v'", viewsDir)
}
for _, viewFI := range viewFIs {
if !viewFI.IsDir() {
continue
}
view := viewFI.Name()
roaringViewPath := filepath.Join(viewsDir, view)
shardMap, err := roaringMapOfShards(roaringViewPath)
if err != nil {
return nil, errors.Wrapf(err, "reading view path directory '%v'", roaringViewPath)
}
if len(shardMap) == 0 {
continue
}
// once we know we have data for this shard!
if shardMap[shard] {
fv := txkey.FieldView{Field: field, View: view}
//vv("C) adding fv '%#v'", fv)
fvs = append(fvs, fv)
}
}
}
// directory stuff isn't returned in sorted order, we must sort.
sort.Slice(fvs, func(i, j int) bool {
if fvs[i].Field < fvs[j].Field {
return true
}
if fvs[i].Field > fvs[j].Field {
return false
}
return fvs[i].View < fvs[j].View
})
return
}
func (tx *RoaringTx) GetFieldSizeBytes(index, field string) (uint64, error) {
return 0, nil
}
//////// registrar and wrapper machinery
// roaringRegistrar mirrors the machinery expected
// for all backends for the roaring files approach.
//
type roaringRegistrar struct {
mu sync.Mutex
mp map[*RoaringWrapper]bool
path2db map[string]*RoaringWrapper
}
func (r *roaringRegistrar) Size() int {
r.mu.Lock()
defer r.mu.Unlock()
nmp := len(r.mp)
npa := len(r.path2db)
if nmp != npa {
panic(fmt.Sprintf("nmp=%v, vs npa=%v", nmp, npa))
}
return nmp
}
var globalRoaringReg *roaringRegistrar = newRoaringRegistrar()
func newRoaringRegistrar() *roaringRegistrar {
return &roaringRegistrar{
mp: make(map[*RoaringWrapper]bool),
path2db: make(map[string]*RoaringWrapper),
}
}
func (r *roaringRegistrar) unprotectedRegister(w *RoaringWrapper) {
r.mp[w] = true
r.path2db[w.path] = w
}
// unregister removes w from r
func (r *roaringRegistrar) unregister(w *RoaringWrapper) {
r.mu.Lock()
delete(r.mp, w)
delete(r.path2db, w.path)
r.mu.Unlock()
}
// openRoaringDB will check the registry and make a new instance only
// if one does not exist for its path0. Otherwise it returns
// the existing instance.
func (r *roaringRegistrar) OpenDBWrapper(path string, doAllocZero bool, _ *storage.Config) (DBWrapper, error) {
r.mu.Lock()
defer r.mu.Unlock()
w, ok := r.path2db[path]
if ok {
return w, nil
}
// otherwise, make a new roaring and store it in globalRoaringReg
w = &RoaringWrapper{
reg: r,
path: path,
}
r.unprotectedRegister(w)
return w, nil
}
func (w *RoaringWrapper) SetHolder(h *Holder) {
w.h = h
}
func (w *RoaringWrapper) Path() string {
return w.path
}
func (w *RoaringWrapper) HasData() (has bool, err error) {
return w.h.HasRoaringData()
}
func (w *RoaringWrapper) CleanupTx(tx Tx) {
r := tx.(*RoaringTx)
r.mu.Lock()
defer r.mu.Unlock()
if r.done {
return
}
r.done = true
r.o.dbs.Cleanup(tx) // release the read/write lock.
}
func (w *RoaringWrapper) OpenListString() (r string) {
return "RoaringWrapper.OpenListString() not yet implemented"
}
func (w *RoaringWrapper) OpenSnList() (slc []int64) {
return nil
}
// statically confirm that RoaringTx satisfies the Tx interface.
var _ Tx = (*RoaringTx)(nil)
// RoaringWrapper provides the NewTx() method.
type RoaringWrapper struct {
muDb sync.Mutex
path string
h *Holder
reg *roaringRegistrar
// make RoaringWrapper.Close() idempotent, avoiding panic on double Close()
closed bool
}
var globalNextTxSnRoaring int64
func (w *RoaringWrapper) NewTx(write bool, initialIndexName string, o Txo) (tx Tx, err error) {
sn := atomic.AddInt64(&globalNextTxSnRoaring, 1)
return &RoaringTx{
write: o.Write,
Field: o.Field,
Index: o.Index,
fragment: o.Fragment,
o: o,
sn: sn,
w: w,
}, nil
}
// Close shuts down the Roaring database.
func (w *RoaringWrapper) Close() (err error) {
w.muDb.Lock()
defer w.muDb.Unlock()
if !w.closed {
w.reg.unregister(w)
w.closed = true
}
return nil
}
func (w *RoaringWrapper) IsClosed() (closed bool) {
w.muDb.Lock()
closed = w.closed
w.muDb.Unlock()
return
}
func (w *RoaringWrapper) DeleteDBPath(dbs *DBShard) (err error) {
//vv("RoaringWrapper.DeleteDBPath called on dbs = '%#v'", dbs)
path := dbs.pathForType(roaringTxn)
return os.RemoveAll(path)
}
func (w *RoaringWrapper) DeleteField(index, field, fieldPath string) error {
//vv("RoaringWrapper.DeleteField(index = '%v', field = '%v', fieldPath = '%v'", index, field, fieldPath)
// match txn sn count vs lmdb/etc.
atomic.AddInt64(&globalNextTxSnRoaring, 1)
// under blue-green bolt_roaring, the directory will not be found, b/c bolt will have
// already done the os.RemoveAll(). BUT, RemoveAll returns nil error in this case. Docs:
// "If the path does not exist, RemoveAll returns nil (no error)"
err := os.RemoveAll(fieldPath)
if err != nil {
return errors.Wrap(err, "removing directory")
}
return nil
}
func (w *RoaringWrapper) DeleteFragment(index, field, view string, shard uint64, frag interface{}) error {
// match txn sn count vs lmdb/etc.
atomic.AddInt64(&globalNextTxSnRoaring, 1)
fragment, ok := frag.(*fragment)
if !ok {
return fmt.Errorf("RoaringStore.DeleteFragment must get frag of type *fragment, but got '%T'", frag)
}
// Delete fragment file.
if err := os.Remove(fragment.path()); err != nil {
return errors.Wrap(err, "deleting fragment file")
}
// Delete fragment cache file.
if err := os.Remove(fragment.cachePath()); err != nil {
return errors.Wrap(err, fmt.Sprintf("no cache file to delete for shard %d", fragment.shard))
}
return nil
}