storj/satellite/metabase/list_verify_segments.go
Michal Niewrzal 96d3c41c14 satellite/metabase: convert bucket name to bytes for queries
In case some invalid characters in bucket name we need to cast
bucket name to byte array for query argument. This change is
doing this for some missed cases.

Change-Id: I47d0d8e3c85a69bdf63de1137adcd533dcfe50a8
2023-06-29 10:43:35 +00:00

222 lines
5.9 KiB
Go

// Copyright (C) 2022 Storj Labs, Inc.
// See LICENSE for copying information.
package metabase
import (
"context"
"time"
"storj.io/common/storj"
"storj.io/common/uuid"
"storj.io/private/dbutil/pgutil"
"storj.io/private/tagsql"
)
// ListVerifyLimit is the maximum number of items the client can request for listing.
const ListVerifyLimit = intLimitRange(100000)
// ListVerifySegments contains arguments necessary for listing stream segments.
type ListVerifySegments struct {
CursorStreamID uuid.UUID
CursorPosition SegmentPosition
StreamIDs []uuid.UUID
Limit int
AsOfSystemTime time.Time
AsOfSystemInterval time.Duration
}
// ListVerifySegmentsResult is the result of ListVerifySegments.
type ListVerifySegmentsResult struct {
Segments []VerifySegment
}
// VerifySegment result of listing segments for verifying remote segments.
type VerifySegment struct {
StreamID uuid.UUID
Position SegmentPosition
CreatedAt time.Time
RepairedAt *time.Time
RootPieceID storj.PieceID
Redundancy storj.RedundancyScheme
AliasPieces AliasPieces
}
func (opts *ListVerifySegments) getQueryAndParameters(asof string) (string, []interface{}) {
if len(opts.StreamIDs) == 0 {
return `
SELECT
stream_id, position,
created_at, repaired_at,
root_piece_id, redundancy,
remote_alias_pieces
FROM segments
` + asof + `
WHERE
(stream_id, position) > ($1, $2) AND
inline_data IS NULL AND
remote_alias_pieces IS NOT NULL
ORDER BY stream_id ASC, position ASC
LIMIT $3
`, []interface{}{
opts.CursorStreamID,
opts.CursorPosition,
opts.Limit,
}
}
return `
SELECT
segments.stream_id, segments.position,
segments.created_at, segments.repaired_at,
segments.root_piece_id, segments.redundancy,
segments.remote_alias_pieces
FROM segments
` + asof + `
WHERE
stream_id = ANY($1) AND
(segments.stream_id, segments.position) > ($2, $3) AND
segments.inline_data IS NULL AND
segments.remote_alias_pieces IS NOT NULL
ORDER BY segments.stream_id ASC, segments.position ASC
LIMIT $4
`, []interface{}{
pgutil.UUIDArray(opts.StreamIDs),
opts.CursorStreamID,
opts.CursorPosition,
opts.Limit,
}
}
// ListVerifySegments lists specified stream segments.
func (db *DB) ListVerifySegments(ctx context.Context, opts ListVerifySegments) (result ListVerifySegmentsResult, err error) {
defer mon.Task()(&ctx)(&err)
if opts.Limit <= 0 {
return ListVerifySegmentsResult{}, ErrInvalidRequest.New("invalid limit: %d", opts.Limit)
}
ListVerifyLimit.Ensure(&opts.Limit)
result.Segments = make([]VerifySegment, 0, opts.Limit)
asOfString := db.asOfTime(opts.AsOfSystemTime, opts.AsOfSystemInterval)
query, parameters := opts.getQueryAndParameters(asOfString)
err = withRows(db.db.QueryContext(ctx, query, parameters...))(func(rows tagsql.Rows) error {
for rows.Next() {
var seg VerifySegment
err := rows.Scan(
&seg.StreamID,
&seg.Position,
&seg.CreatedAt,
&seg.RepairedAt,
&seg.RootPieceID,
redundancyScheme{&seg.Redundancy},
&seg.AliasPieces,
)
if err != nil {
return Error.Wrap(err)
}
result.Segments = append(result.Segments, seg)
}
return nil
})
return result, Error.Wrap(err)
}
// ListVerifyBucketList represents a list of buckets.
type ListVerifyBucketList struct {
Buckets []BucketLocation
}
// Add adds a (projectID, bucketName) to the list of buckets to be checked.
func (list *ListVerifyBucketList) Add(projectID uuid.UUID, bucketName string) {
list.Buckets = append(list.Buckets, BucketLocation{
ProjectID: projectID,
BucketName: bucketName,
})
}
// ListBucketsStreamIDsResult is the result of listing segments of a list of buckets.
type ListBucketsStreamIDsResult struct {
StreamIDs []uuid.UUID
Counts []int
LastBucket BucketLocation
}
func (list *ListBucketsStreamIDsResult) addStreamID(streamID uuid.UUID, count int) {
list.StreamIDs = append(list.StreamIDs, streamID)
list.Counts = append(list.Counts, count)
}
// ListBucketsStreamIDs contains arguments necessary for listing stream segments from buckets.
type ListBucketsStreamIDs struct {
BucketList ListVerifyBucketList
CursorBucket BucketLocation
CursorStreamID uuid.UUID
Limit int
AsOfSystemTime time.Time
AsOfSystemInterval time.Duration
}
// ListBucketsStreamIDs lists the streamIDs of a list of buckets.
func (db *DB) ListBucketsStreamIDs(ctx context.Context, opts ListBucketsStreamIDs) (ListBucketsStreamIDsResult, error) {
if opts.Limit <= 0 {
return ListBucketsStreamIDsResult{}, ErrInvalidRequest.New("invalid limit: %d", opts.Limit)
}
ListVerifyLimit.Ensure(&opts.Limit)
result := ListBucketsStreamIDsResult{}
bucketNamesBytes := [][]byte{}
projectIDs := []uuid.UUID{}
for _, bucket := range opts.BucketList.Buckets {
bucketNamesBytes = append(bucketNamesBytes, []byte(bucket.BucketName))
projectIDs = append(projectIDs, bucket.ProjectID)
}
// get the list of stream_ids and segment counts from the objects table
err := withRows(db.db.QueryContext(ctx, `
SELECT DISTINCT project_id, bucket_name, stream_id, segment_count
FROM objects
`+db.asOfTime(opts.AsOfSystemTime, opts.AsOfSystemInterval)+`
WHERE
(project_id, bucket_name, stream_id) > ($4::BYTEA, $5::BYTEA, $6::BYTEA) AND
(project_id, bucket_name) IN (SELECT UNNEST($1::BYTEA[]),UNNEST($2::BYTEA[]))
ORDER BY project_id, bucket_name, stream_id ASC
LIMIT $3
`, pgutil.UUIDArray(projectIDs), pgutil.ByteaArray(bucketNamesBytes),
opts.Limit,
opts.CursorBucket.ProjectID, []byte(opts.CursorBucket.BucketName), opts.CursorStreamID,
))(func(rows tagsql.Rows) error {
for rows.Next() {
var streamID uuid.UUID
var count int
err := rows.Scan(
&result.LastBucket.ProjectID,
&result.LastBucket.BucketName,
&streamID,
&count,
)
if err != nil {
return Error.Wrap(err)
}
result.addStreamID(streamID, count)
}
return nil
})
if err != nil {
return ListBucketsStreamIDsResult{}, err
}
return result, nil
}