1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507
|
/*
* MinIO Go Library for Amazon S3 Compatible Cloud Storage
* Copyright 2015-2017 MinIO, Inc.
*
* 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 minio
import (
"bytes"
"context"
"encoding/base64"
"errors"
"fmt"
"io"
"net/http"
"sort"
"time"
"github.com/minio/minio-go/v7/pkg/encrypt"
"github.com/minio/minio-go/v7/pkg/s3utils"
"golang.org/x/net/http/httpguts"
)
// ReplicationStatus represents replication status of object
type ReplicationStatus string
const (
// ReplicationStatusPending indicates replication is pending
ReplicationStatusPending ReplicationStatus = "PENDING"
// ReplicationStatusComplete indicates replication completed ok
ReplicationStatusComplete ReplicationStatus = "COMPLETED"
// ReplicationStatusFailed indicates replication failed
ReplicationStatusFailed ReplicationStatus = "FAILED"
// ReplicationStatusReplica indicates object is a replica of a source
ReplicationStatusReplica ReplicationStatus = "REPLICA"
// ReplicationStatusReplicaEdge indicates object is a replica of a edge source
ReplicationStatusReplicaEdge ReplicationStatus = "REPLICA-EDGE"
)
// Empty returns true if no replication status set.
func (r ReplicationStatus) Empty() bool {
return r == ""
}
// AdvancedPutOptions for internal use - to be utilized by replication, ILM transition
// implementation on MinIO server
type AdvancedPutOptions struct {
SourceVersionID string
SourceETag string
ReplicationStatus ReplicationStatus
SourceMTime time.Time
ReplicationRequest bool
RetentionTimestamp time.Time
TaggingTimestamp time.Time
LegalholdTimestamp time.Time
ReplicationValidityCheck bool
}
// PutObjectOptions represents options specified by user for PutObject call
type PutObjectOptions struct {
UserMetadata map[string]string
UserTags map[string]string
Progress io.Reader
ContentType string
ContentEncoding string
ContentDisposition string
ContentLanguage string
CacheControl string
Expires time.Time
Mode RetentionMode
RetainUntilDate time.Time
ServerSideEncryption encrypt.ServerSide
NumThreads uint
StorageClass string
WebsiteRedirectLocation string
PartSize uint64
LegalHold LegalHoldStatus
SendContentMd5 bool
DisableContentSha256 bool
DisableMultipart bool
// AutoChecksum is the type of checksum that will be added if no other checksum is added,
// like MD5 or SHA256 streaming checksum, and it is feasible for the upload type.
// If none is specified CRC32C is used, since it is generally the fastest.
AutoChecksum ChecksumType
// Checksum will force a checksum of the specific type.
// This requires that the client was created with "TrailingHeaders:true" option,
// and that the destination server supports it.
// Unavailable with V2 signatures & Google endpoints.
// This will disable content MD5 checksums if set.
Checksum ChecksumType
// ConcurrentStreamParts will create NumThreads buffers of PartSize bytes,
// fill them serially and upload them in parallel.
// This can be used for faster uploads on non-seekable or slow-to-seek input.
ConcurrentStreamParts bool
Internal AdvancedPutOptions
customHeaders http.Header
}
// SetMatchETag if etag matches while PUT MinIO returns an error
// this is a MinIO specific extension to support optimistic locking
// semantics.
func (opts *PutObjectOptions) SetMatchETag(etag string) {
if opts.customHeaders == nil {
opts.customHeaders = http.Header{}
}
if etag == "*" {
opts.customHeaders.Set("If-Match", "*")
} else {
opts.customHeaders.Set("If-Match", "\""+etag+"\"")
}
}
// SetMatchETagExcept if etag does not match while PUT MinIO returns an
// error this is a MinIO specific extension to support optimistic locking
// semantics.
func (opts *PutObjectOptions) SetMatchETagExcept(etag string) {
if opts.customHeaders == nil {
opts.customHeaders = http.Header{}
}
if etag == "*" {
opts.customHeaders.Set("If-None-Match", "*")
} else {
opts.customHeaders.Set("If-None-Match", "\""+etag+"\"")
}
}
// getNumThreads - gets the number of threads to be used in the multipart
// put object operation
func (opts PutObjectOptions) getNumThreads() (numThreads int) {
if opts.NumThreads > 0 {
numThreads = int(opts.NumThreads)
} else {
numThreads = totalWorkers
}
return
}
// Header - constructs the headers from metadata entered by user in
// PutObjectOptions struct
func (opts PutObjectOptions) Header() (header http.Header) {
header = make(http.Header)
contentType := opts.ContentType
if contentType == "" {
contentType = "application/octet-stream"
}
header.Set("Content-Type", contentType)
if opts.ContentEncoding != "" {
header.Set("Content-Encoding", opts.ContentEncoding)
}
if opts.ContentDisposition != "" {
header.Set("Content-Disposition", opts.ContentDisposition)
}
if opts.ContentLanguage != "" {
header.Set("Content-Language", opts.ContentLanguage)
}
if opts.CacheControl != "" {
header.Set("Cache-Control", opts.CacheControl)
}
if !opts.Expires.IsZero() {
header.Set("Expires", opts.Expires.UTC().Format(http.TimeFormat))
}
if opts.Mode != "" {
header.Set(amzLockMode, opts.Mode.String())
}
if !opts.RetainUntilDate.IsZero() {
header.Set("X-Amz-Object-Lock-Retain-Until-Date", opts.RetainUntilDate.Format(time.RFC3339))
}
if opts.LegalHold != "" {
header.Set(amzLegalHoldHeader, opts.LegalHold.String())
}
if opts.ServerSideEncryption != nil {
opts.ServerSideEncryption.Marshal(header)
}
if opts.StorageClass != "" {
header.Set(amzStorageClass, opts.StorageClass)
}
if opts.WebsiteRedirectLocation != "" {
header.Set(amzWebsiteRedirectLocation, opts.WebsiteRedirectLocation)
}
if !opts.Internal.ReplicationStatus.Empty() {
header.Set(amzBucketReplicationStatus, string(opts.Internal.ReplicationStatus))
}
if !opts.Internal.SourceMTime.IsZero() {
header.Set(minIOBucketSourceMTime, opts.Internal.SourceMTime.Format(time.RFC3339Nano))
}
if opts.Internal.SourceETag != "" {
header.Set(minIOBucketSourceETag, opts.Internal.SourceETag)
}
if opts.Internal.ReplicationRequest {
header.Set(minIOBucketReplicationRequest, "true")
}
if opts.Internal.ReplicationValidityCheck {
header.Set(minIOBucketReplicationCheck, "true")
}
if !opts.Internal.LegalholdTimestamp.IsZero() {
header.Set(minIOBucketReplicationObjectLegalHoldTimestamp, opts.Internal.LegalholdTimestamp.Format(time.RFC3339Nano))
}
if !opts.Internal.RetentionTimestamp.IsZero() {
header.Set(minIOBucketReplicationObjectRetentionTimestamp, opts.Internal.RetentionTimestamp.Format(time.RFC3339Nano))
}
if !opts.Internal.TaggingTimestamp.IsZero() {
header.Set(minIOBucketReplicationTaggingTimestamp, opts.Internal.TaggingTimestamp.Format(time.RFC3339Nano))
}
if len(opts.UserTags) != 0 {
header.Set(amzTaggingHeader, s3utils.TagEncode(opts.UserTags))
}
for k, v := range opts.UserMetadata {
if isAmzHeader(k) || isStandardHeader(k) || isStorageClassHeader(k) || isMinioHeader(k) {
header.Set(k, v)
} else {
header.Set("x-amz-meta-"+k, v)
}
}
// set any other additional custom headers.
for k, v := range opts.customHeaders {
header[k] = v
}
return
}
// validate() checks if the UserMetadata map has standard headers or and raises an error if so.
func (opts PutObjectOptions) validate(c *Client) (err error) {
for k, v := range opts.UserMetadata {
if !httpguts.ValidHeaderFieldName(k) || isStandardHeader(k) || isSSEHeader(k) || isStorageClassHeader(k) || isMinioHeader(k) {
return errInvalidArgument(k + " unsupported user defined metadata name")
}
if !httpguts.ValidHeaderFieldValue(v) {
return errInvalidArgument(v + " unsupported user defined metadata value")
}
}
if opts.Mode != "" && !opts.Mode.IsValid() {
return errInvalidArgument(opts.Mode.String() + " unsupported retention mode")
}
if opts.LegalHold != "" && !opts.LegalHold.IsValid() {
return errInvalidArgument(opts.LegalHold.String() + " unsupported legal-hold status")
}
if opts.Checksum.IsSet() {
switch {
case !c.trailingHeaderSupport:
return errInvalidArgument("Checksum requires Client with TrailingHeaders enabled")
case c.overrideSignerType.IsV2():
return errInvalidArgument("Checksum cannot be used with v2 signatures")
case s3utils.IsGoogleEndpoint(*c.endpointURL):
return errInvalidArgument("Checksum cannot be used with GCS endpoints")
}
}
return nil
}
// completedParts is a collection of parts sortable by their part numbers.
// used for sorting the uploaded parts before completing the multipart request.
type completedParts []CompletePart
func (a completedParts) Len() int { return len(a) }
func (a completedParts) Swap(i, j int) { a[i], a[j] = a[j], a[i] }
func (a completedParts) Less(i, j int) bool { return a[i].PartNumber < a[j].PartNumber }
// PutObject creates an object in a bucket.
//
// You must have WRITE permissions on a bucket to create an object.
//
// - For size smaller than 16MiB PutObject automatically does a
// single atomic PUT operation.
//
// - For size larger than 16MiB PutObject automatically does a
// multipart upload operation.
//
// - For size input as -1 PutObject does a multipart Put operation
// until input stream reaches EOF. Maximum object size that can
// be uploaded through this operation will be 5TiB.
//
// WARNING: Passing down '-1' will use memory and these cannot
// be reused for best outcomes for PutObject(), pass the size always.
//
// NOTE: Upon errors during upload multipart operation is entirely aborted.
func (c *Client) PutObject(ctx context.Context, bucketName, objectName string, reader io.Reader, objectSize int64,
opts PutObjectOptions,
) (info UploadInfo, err error) {
if objectSize < 0 && opts.DisableMultipart {
return UploadInfo{}, errors.New("object size must be provided with disable multipart upload")
}
err = opts.validate(c)
if err != nil {
return UploadInfo{}, err
}
return c.putObjectCommon(ctx, bucketName, objectName, reader, objectSize, opts)
}
func (c *Client) putObjectCommon(ctx context.Context, bucketName, objectName string, reader io.Reader, size int64, opts PutObjectOptions) (info UploadInfo, err error) {
// Check for largest object size allowed.
if size > int64(maxMultipartPutObjectSize) {
return UploadInfo{}, errEntityTooLarge(size, maxMultipartPutObjectSize, bucketName, objectName)
}
opts.AutoChecksum.SetDefault(ChecksumCRC32C)
// NOTE: Streaming signature is not supported by GCS.
if s3utils.IsGoogleEndpoint(*c.endpointURL) {
return c.putObject(ctx, bucketName, objectName, reader, size, opts)
}
partSize := opts.PartSize
if opts.PartSize == 0 {
partSize = minPartSize
}
if c.overrideSignerType.IsV2() {
if size >= 0 && size < int64(partSize) || opts.DisableMultipart {
return c.putObject(ctx, bucketName, objectName, reader, size, opts)
}
return c.putObjectMultipart(ctx, bucketName, objectName, reader, size, opts)
}
if size < 0 {
if opts.DisableMultipart {
return UploadInfo{}, errors.New("no length provided and multipart disabled")
}
if opts.ConcurrentStreamParts && opts.NumThreads > 1 {
return c.putObjectMultipartStreamParallel(ctx, bucketName, objectName, reader, opts)
}
return c.putObjectMultipartStreamNoLength(ctx, bucketName, objectName, reader, opts)
}
if size <= int64(partSize) || opts.DisableMultipart {
return c.putObject(ctx, bucketName, objectName, reader, size, opts)
}
return c.putObjectMultipartStream(ctx, bucketName, objectName, reader, size, opts)
}
func (c *Client) putObjectMultipartStreamNoLength(ctx context.Context, bucketName, objectName string, reader io.Reader, opts PutObjectOptions) (info UploadInfo, err error) {
// Input validation.
if err = s3utils.CheckValidBucketName(bucketName); err != nil {
return UploadInfo{}, err
}
if err = s3utils.CheckValidObjectName(objectName); err != nil {
return UploadInfo{}, err
}
// Total data read and written to server. should be equal to
// 'size' at the end of the call.
var totalUploadedSize int64
// Complete multipart upload.
var complMultipartUpload completeMultipartUpload
// Calculate the optimal parts info for a given size.
totalPartsCount, partSize, _, err := OptimalPartInfo(-1, opts.PartSize)
if err != nil {
return UploadInfo{}, err
}
if opts.Checksum.IsSet() {
opts.SendContentMd5 = false
opts.AutoChecksum = opts.Checksum
}
if !opts.SendContentMd5 {
addAutoChecksumHeaders(&opts)
}
// Initiate a new multipart upload.
uploadID, err := c.newUploadID(ctx, bucketName, objectName, opts)
if err != nil {
return UploadInfo{}, err
}
delete(opts.UserMetadata, "X-Amz-Checksum-Algorithm")
defer func() {
if err != nil {
c.abortMultipartUpload(ctx, bucketName, objectName, uploadID)
}
}()
// Part number always starts with '1'.
partNumber := 1
// Initialize parts uploaded map.
partsInfo := make(map[int]ObjectPart)
// Create a buffer.
buf := make([]byte, partSize)
// Create checksums
// CRC32C is ~50% faster on AMD64 @ 30GB/s
customHeader := make(http.Header)
crc := opts.AutoChecksum.Hasher()
for partNumber <= totalPartsCount {
length, rerr := readFull(reader, buf)
if rerr == io.EOF && partNumber > 1 {
break
}
if rerr != nil && rerr != io.ErrUnexpectedEOF && rerr != io.EOF {
return UploadInfo{}, rerr
}
var md5Base64 string
if opts.SendContentMd5 {
// Calculate md5sum.
hash := c.md5Hasher()
hash.Write(buf[:length])
md5Base64 = base64.StdEncoding.EncodeToString(hash.Sum(nil))
hash.Close()
} else {
crc.Reset()
crc.Write(buf[:length])
cSum := crc.Sum(nil)
customHeader.Set(opts.AutoChecksum.Key(), base64.StdEncoding.EncodeToString(cSum))
}
// Update progress reader appropriately to the latest offset
// as we read from the source.
rd := newHook(bytes.NewReader(buf[:length]), opts.Progress)
// Proceed to upload the part.
p := uploadPartParams{bucketName: bucketName, objectName: objectName, uploadID: uploadID, reader: rd, partNumber: partNumber, md5Base64: md5Base64, size: int64(length), sse: opts.ServerSideEncryption, streamSha256: !opts.DisableContentSha256, customHeader: customHeader}
objPart, uerr := c.uploadPart(ctx, p)
if uerr != nil {
return UploadInfo{}, uerr
}
// Save successfully uploaded part metadata.
partsInfo[partNumber] = objPart
// Save successfully uploaded size.
totalUploadedSize += int64(length)
// Increment part number.
partNumber++
// For unknown size, Read EOF we break away.
// We do not have to upload till totalPartsCount.
if rerr == io.EOF {
break
}
}
// Loop over total uploaded parts to save them in
// Parts array before completing the multipart request.
allParts := make([]ObjectPart, 0, len(partsInfo))
for i := 1; i < partNumber; i++ {
part, ok := partsInfo[i]
if !ok {
return UploadInfo{}, errInvalidArgument(fmt.Sprintf("Missing part number %d", i))
}
allParts = append(allParts, part)
complMultipartUpload.Parts = append(complMultipartUpload.Parts, CompletePart{
ETag: part.ETag,
PartNumber: part.PartNumber,
ChecksumCRC32: part.ChecksumCRC32,
ChecksumCRC32C: part.ChecksumCRC32C,
ChecksumSHA1: part.ChecksumSHA1,
ChecksumSHA256: part.ChecksumSHA256,
ChecksumCRC64NVME: part.ChecksumCRC64NVME,
})
}
// Sort all completed parts.
sort.Sort(completedParts(complMultipartUpload.Parts))
opts = PutObjectOptions{
ServerSideEncryption: opts.ServerSideEncryption,
AutoChecksum: opts.AutoChecksum,
}
applyAutoChecksum(&opts, allParts)
uploadInfo, err := c.completeMultipartUpload(ctx, bucketName, objectName, uploadID, complMultipartUpload, opts)
if err != nil {
return UploadInfo{}, err
}
uploadInfo.Size = totalUploadedSize
return uploadInfo, nil
}
|