Commit 80153aa5 authored by Kirill Smelkov's avatar Kirill Smelkov

wcfs: xbtree: BTree-diff algorithm

This algorithm will be internally used by ΔBtail in the next patch.

The algorithm would be simple, if we would need to diff two trees
completely. However in ΔBtail only subpart of BTree nodes are tracked(*)
and the diff has to work modulo that tracking set.

No tests now because ΔBtail tests will cover treediff functionality as well.

Some preliminary history:

kirr/wendelin.core@78f2f88b    X wcfs/xbtree: Fix treediff(a, ø)
kirr/wendelin.core@5324547c    X wcfs/xbtree: root(a) must stay in trackSet even after treediff(a,ø)
kirr/wendelin.core@f65f775b    X wcfs/xbtree: treediff(ø, b)
kirr/wendelin.core@c75b1c6f    X wcfs/xbtree: Start killing holeIdx
kirr/wendelin.core@ef5e5183    X treediff ret += δtkeycov
kirr/wendelin.core@9d20f8e8    X treediff: Fix BUG while computing AB coverage
kirr/wendelin.core@ddb28043    X rebuild: Don't return nil for empty ΔPPTreeSubSet - that leads to SIGSEGV
kirr/wendelin.core@f68398c9    X wcfs: Move treediff into its own file

(*) because full BTree scan is needed to discover all of its nodes.

Quoting treediff documentation:

---- 8< ----

treediff provides diff for BTrees

Use δZConnectTracked + treediff to compute BTree-diff caused by δZ:

    δZConnectTracked(δZ, trackSet)                         -> δZTC, δtopsByRoot
    treediff(root, δtops, δZTC, trackSet, zconn{Old,New})  -> δT, δtrack, δtkeycov

δZConnectTracked computes BTree-connected closure of δZ modulo tracked set
and also returns δtopsByRoot to indicate which tree objects were changed and
in which subtree parts. With that information one can call treediff for each
changed root to compute BTree-diff and δ for trackSet itself.

BTree diff algorithm

diffT, diffB and δMerge constitute the diff algorithm implementation.
diff(A,B) works on pair of A and B whole key ranges splitted into regions
covered by tree nodes. The splitting represents current state of recursion
into corresponding tree. If a node in particular key range is Bucket, that
bucket contributes to δ- in case of A, and to δ+ in case of B. If a node in
particular key range is Tree, the algorithm may want to expand that tree
node into its children and to recourse into some of the children.

There are two phases:

- Phase 1 expands A top->down driven by δZTC, adds reached buckets to δ-,
  and queues key regions of those buckets to be processed on B.

- Phase 2 starts processing from queued key regions, expands them on B and
  adds reached buckets to δ+. Then it iterates to reach consistency in between
  A and B because processing buckets on B side may increase δ key coverage,
  and so corresponding key ranges has to be again processed on A. Which in
  turn may increase δ key coverage again, and needs to be processed on B side,
  etc...

The final δ is merge of δ- and δ+.

diffT has more detailed explanation of phase 1 and phase 2 logic.
parent 27df5a3b
This diff is collapsed.
// Copyright (C) 2021 Nexedi SA and Contributors.
// Kirill Smelkov <kirr@nexedi.com>
//
// This program is free software: you can Use, Study, Modify and Redistribute
// it under the terms of the GNU General Public License version 3, or (at your
// option) any later version, as published by the Free Software Foundation.
//
// You can also Link and Combine this program with other software covered by
// the terms of any of the Free Software licenses or any of the Open Source
// Initiative approved licenses and Convey the resulting work. Corresponding
// source of such a combination shall include the source code for all other
// software used.
//
// This program is distributed WITHOUT ANY WARRANTY; without even the implied
// warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
//
// See COPYING file for full licensing terms.
// See https://www.nexedi.com/licensing for rationale and options.
// Package xbtree complements package lab.nexedi.com/kirr/neo/go/zodb/btree.
//
// It provides the following amendments:
//
// - ΔBtail (tail of revisional changes to BTrees).
package xbtree
// this file contains only tree types and utilities.
// main code lives in δbtail.go and treediff.go .
import (
"context"
"fmt"
"lab.nexedi.com/kirr/go123/xerr"
"lab.nexedi.com/kirr/neo/go/zodb"
"lab.nexedi.com/nexedi/wendelin.core/wcfs/internal/set"
"lab.nexedi.com/nexedi/wendelin.core/wcfs/internal/xbtree/blib"
"lab.nexedi.com/nexedi/wendelin.core/wcfs/internal/xzodb"
)
// XXX instead of generics
type Tree = blib.Tree
type Bucket = blib.Bucket
type Node = blib.Node
type TreeEntry = blib.TreeEntry
type BucketEntry = blib.BucketEntry
type Key = blib.Key
type KeyRange = blib.KeyRange
const KeyMax = blib.KeyMax
const KeyMin = blib.KeyMin
// value is assumed to be persistent reference.
// deletion is represented as VDEL.
type Value = zodb.Oid
const VDEL = zodb.InvalidOid
type setOid = set.Oid
// pathEqual returns whether two paths are the same.
func pathEqual(patha, pathb []zodb.Oid) bool {
if len(patha) != len(pathb) {
return false
}
for i, a := range patha {
if pathb[i] != a {
return false
}
}
return true
}
// vnode returns brief human-readable representation of node.
func vnode(node Node) string {
kind := "?"
switch node.(type) {
case *Tree: kind = "T"
case *Bucket: kind = "B"
}
return kind + node.POid().String()
}
// zgetNodeOrNil returns btree node corresponding to zconn.Get(oid) .
// if the node does not exist, (nil, ok) is returned.
func zgetNodeOrNil(ctx context.Context, zconn *zodb.Connection, oid zodb.Oid) (node Node, err error) {
defer xerr.Contextf(&err, "getnode %s@%s", oid, zconn.At())
xnode, err := xzodb.ZGetOrNil(ctx, zconn, oid)
if xnode == nil || err != nil {
return nil, err
}
node, ok := xnode.(Node)
if !ok {
return nil, fmt.Errorf("unexpected type: %s", zodb.ClassOf(xnode))
}
return node, nil
}
func panicf(format string, argv ...interface{}) {
panic(fmt.Sprintf(format, argv...))
}
...@@ -43,6 +43,7 @@ const KeyMin = blib.KeyMin ...@@ -43,6 +43,7 @@ const KeyMin = blib.KeyMin
type setKey = set.I64 type setKey = set.I64
// XXX dup from xbtree (to avoid import cycle)
const VDEL = zodb.InvalidOid const VDEL = zodb.InvalidOid
......
...@@ -22,9 +22,12 @@ package xzodb ...@@ -22,9 +22,12 @@ package xzodb
import ( import (
"context" "context"
"errors"
"fmt" "fmt"
"reflect"
"lab.nexedi.com/kirr/go123/xcontext" "lab.nexedi.com/kirr/go123/xcontext"
"lab.nexedi.com/kirr/go123/xerr"
"lab.nexedi.com/kirr/neo/go/transaction" "lab.nexedi.com/kirr/neo/go/transaction"
"lab.nexedi.com/kirr/neo/go/zodb" "lab.nexedi.com/kirr/neo/go/zodb"
...@@ -80,3 +83,54 @@ func ZOpen(ctx context.Context, zdb *zodb.DB, zopt *zodb.ConnOptions) (_ *ZConn, ...@@ -80,3 +83,54 @@ func ZOpen(ctx context.Context, zdb *zodb.DB, zopt *zodb.ConnOptions) (_ *ZConn,
TxnCtx: txnCtx, TxnCtx: txnCtx,
}, nil }, nil
} }
// ZGetOrNil returns zconn.Get(oid), or (nil,ok) if the object does not exist.
func ZGetOrNil(ctx context.Context, zconn *zodb.Connection, oid zodb.Oid) (_ zodb.IPersistent, err error) {
defer xerr.Contextf(&err, "zget %s@%s", oid, zconn.At())
obj, err := zconn.Get(ctx, oid)
if err != nil {
if IsErrNoData(err) {
err = nil
}
return nil, err
}
// activate the object to find out it really exists
// after removal on storage, the object might have stayed in Connection
// cache due to e.g. PCachePinObject, and it will be PActivate that
// will return "deleted" error.
err = obj.PActivate(ctx)
if err != nil {
if IsErrNoData(err) {
return nil, nil
}
return nil, err
}
obj.PDeactivate()
return obj, nil
}
// IsErrNoData returns whether err is due to NoDataError or NoObjectError.
func IsErrNoData(err error) bool {
var eNoData *zodb.NoDataError
var eNoObject *zodb.NoObjectError
switch {
case errors.As(err, &eNoData):
return true
case errors.As(err, &eNoObject):
return true
default:
return false
}
}
// XidOf returns string representation of object xid.
func XidOf(obj zodb.IPersistent) string {
if obj == nil || reflect.ValueOf(obj).IsNil() {
return "ø"
}
xid := zodb.Xid{At: obj.PJar().At(), Oid: obj.POid()}
return xid.String()
}
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment