From 2ac54e5a7bf864983fa0106ab470ded2c32dbfd5 Mon Sep 17 00:00:00 2001
From: Klaus Post <klauspost@gmail.com>
Date: Tue, 22 Mar 2022 12:39:45 -0700
Subject: [PATCH] ListObjects: Filter lifecycle expired objects (#14606)

For ListObjects and ListObjectsV2 perform lifecycle checks on
all objects before returning. This will filter out objects that are
pending lifecycle expiration.

Bonus: Cheaper server pool conflict resolution by not converting to FileInfo.
---
 cmd/erasure-server-pool.go   | 17 +++++++++++
 cmd/metacache-entries.go     |  2 +-
 cmd/metacache-server-pool.go | 58 ++++++++++++++++++++++++++++++++----
 cmd/metacache-set.go         |  6 ++++
 4 files changed, 76 insertions(+), 7 deletions(-)

diff --git a/cmd/erasure-server-pool.go b/cmd/erasure-server-pool.go
index e427dc860..b13ba2bbb 100644
--- a/cmd/erasure-server-pool.go
+++ b/cmd/erasure-server-pool.go
@@ -33,6 +33,7 @@ import (
 	"github.com/minio/madmin-go"
 	"github.com/minio/minio-go/v7/pkg/set"
 	"github.com/minio/minio-go/v7/pkg/tags"
+	"github.com/minio/minio/internal/bucket/lifecycle"
 	"github.com/minio/minio/internal/config/storageclass"
 	"github.com/minio/minio/internal/logger"
 	"github.com/minio/minio/internal/sync/errgroup"
@@ -1152,6 +1153,14 @@ func maxKeysPlusOne(maxKeys int, addOne bool) int {
 func (z *erasureServerPools) ListObjects(ctx context.Context, bucket, prefix, marker, delimiter string, maxKeys int) (ListObjectsInfo, error) {
 	var loi ListObjectsInfo
 
+	// Automatically remove the object/version is an expiry lifecycle rule can be applied
+	lc, _ := globalLifecycleSys.Get(bucket)
+	if lc != nil {
+		if !lc.HasActiveRules(prefix, true) {
+			lc = nil
+		}
+	}
+
 	if len(prefix) > 0 && maxKeys == 1 && delimiter == "" && marker == "" {
 		// Optimization for certain applications like
 		// - Cohesity
@@ -1162,6 +1171,13 @@ func (z *erasureServerPools) ListObjects(ctx context.Context, bucket, prefix, ma
 		// to avoid the need for ListObjects().
 		objInfo, err := z.GetObjectInfo(ctx, bucket, prefix, ObjectOptions{NoLock: true})
 		if err == nil {
+			if lc != nil {
+				action := evalActionFromLifecycle(ctx, *lc, objInfo, false)
+				switch action {
+				case lifecycle.DeleteVersionAction, lifecycle.DeleteAction:
+					return loi, nil
+				}
+			}
 			loi.Objects = append(loi.Objects, objInfo)
 			return loi, nil
 		}
@@ -1176,6 +1192,7 @@ func (z *erasureServerPools) ListObjects(ctx context.Context, bucket, prefix, ma
 		InclDeleted: false,
 		AskDisks:    globalAPIConfig.getListQuorum(),
 	}
+
 	merged, err := z.listPath(ctx, &opts)
 	if err != nil && err != io.EOF {
 		if !isErrBucketNotFound(err) {
diff --git a/cmd/metacache-entries.go b/cmd/metacache-entries.go
index 30cd0fef0..f0f26534b 100644
--- a/cmd/metacache-entries.go
+++ b/cmd/metacache-entries.go
@@ -498,7 +498,7 @@ func (m *metaCacheEntriesSorted) fileInfoVersions(bucket, prefix, delimiter, aft
 	return versions
 }
 
-// fileInfoVersions converts the metadata to FileInfoVersions where possible.
+// fileInfos converts the metadata to ObjectInfo where possible.
 // Metadata that cannot be decoded is skipped.
 func (m *metaCacheEntriesSorted) fileInfos(bucket, prefix, delimiter string) (objects []ObjectInfo) {
 	objects = make([]ObjectInfo, 0, m.len())
diff --git a/cmd/metacache-server-pool.go b/cmd/metacache-server-pool.go
index 7ff4db7ec..7d350630b 100644
--- a/cmd/metacache-server-pool.go
+++ b/cmd/metacache-server-pool.go
@@ -28,6 +28,7 @@ import (
 	"sync"
 	"time"
 
+	"github.com/minio/minio/internal/bucket/lifecycle"
 	"github.com/minio/minio/internal/logger"
 )
 
@@ -289,6 +290,14 @@ func (z *erasureServerPools) listMerged(ctx context.Context, o listPathOptions,
 	}
 	mu.Unlock()
 
+	// Do lifecycle filtering.
+	if o.lcFilter != nil {
+		filterIn := make(chan metaCacheEntry, 10)
+		go filterLifeCycle(ctx, o.Bucket, o.lcFilter, filterIn, results)
+		// Replace results.
+		results = filterIn
+	}
+
 	// Gather results to a single channel.
 	err := mergeEntryChannels(ctx, inputs, results, func(existing, other *metaCacheEntry) (replace bool) {
 		// Pick object over directory
@@ -298,21 +307,20 @@ func (z *erasureServerPools) listMerged(ctx context.Context, o listPathOptions,
 		if !existing.isDir() && other.isDir() {
 			return false
 		}
-
-		eFIV, err := existing.fileInfo(o.Bucket)
+		eMeta, err := existing.xlmeta()
 		if err != nil {
 			return true
 		}
-		oFIV, err := other.fileInfo(o.Bucket)
+		oMeta, err := other.xlmeta()
 		if err != nil {
 			return false
 		}
 		// Replace if modtime is newer
-		if !oFIV.ModTime.Equal(eFIV.ModTime) {
-			return oFIV.ModTime.After(eFIV.ModTime)
+		if !oMeta.latestModtime().Equal(oMeta.latestModtime()) {
+			return oMeta.latestModtime().After(eMeta.latestModtime())
 		}
 		// Use NumVersions as a final tiebreaker.
-		return oFIV.NumVersions > eFIV.NumVersions
+		return len(oMeta.versions) > len(eMeta.versions)
 	})
 
 	cancelList()
@@ -346,6 +354,44 @@ func (z *erasureServerPools) listMerged(ctx context.Context, o listPathOptions,
 	return nil
 }
 
+// filterLifeCycle will filter out objects if the most recent
+// version should be deleted by lifecycle.
+// out will be closed when there are no more results.
+// When 'in' is closed or the context is canceled the
+// function closes 'out' and exits.
+func filterLifeCycle(ctx context.Context, bucket string, lc *lifecycle.Lifecycle, in <-chan metaCacheEntry, out chan<- metaCacheEntry) {
+	defer close(out)
+	for {
+		var obj metaCacheEntry
+		var ok bool
+		select {
+		case <-ctx.Done():
+			return
+		case obj, ok = <-in:
+			if !ok {
+				return
+			}
+		}
+
+		fi, err := obj.fileInfo(bucket)
+		if err != nil {
+			continue
+		}
+		objInfo := fi.ToObjectInfo(bucket, obj.name)
+		action := evalActionFromLifecycle(ctx, *lc, objInfo, false)
+		switch action {
+		case lifecycle.DeleteVersionAction, lifecycle.DeleteAction:
+			// Skip this entry.
+			continue
+		}
+		select {
+		case <-ctx.Done():
+			return
+		case out <- obj:
+		}
+	}
+}
+
 func (z *erasureServerPools) listAndSave(ctx context.Context, o *listPathOptions) (entries metaCacheEntriesSorted, err error) {
 	// Use ID as the object name...
 	o.pool = z.getAvailablePoolIdx(ctx, minioMetaBucket, o.ID, 10<<20)
diff --git a/cmd/metacache-set.go b/cmd/metacache-set.go
index addaf6460..3639c1888 100644
--- a/cmd/metacache-set.go
+++ b/cmd/metacache-set.go
@@ -32,6 +32,7 @@ import (
 	"time"
 
 	jsoniter "github.com/json-iterator/go"
+	"github.com/minio/minio/internal/bucket/lifecycle"
 	"github.com/minio/minio/internal/color"
 	"github.com/minio/minio/internal/hash"
 	"github.com/minio/minio/internal/logger"
@@ -96,6 +97,11 @@ type listPathOptions struct {
 
 	// pool and set of where the cache is located.
 	pool, set int
+
+	// lcFilter performs filtering based on lifecycle.
+	// This will filter out objects if the most recent version should be deleted by lifecycle.
+	// Is not transferred across request calls.
+	lcFilter *lifecycle.Lifecycle
 }
 
 func init() {