1
1
mirror of https://github.com/go-gitea/gitea synced 2024-12-30 12:34:28 +00:00
gitea/vendor/github.com/minio/minio-go/v7/api.go

900 lines
26 KiB
Go
Raw Normal View History

Add a storage layer for attachments (#11387) * Add a storage layer for attachments * Fix some bug * fix test * Fix copyright head and lint * Fix bug * Add setting for minio and flags for migrate-storage * Add documents * fix lint * Add test for minio store type on attachments * fix test * fix test * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Add warning when storage migrated successfully * Fix drone * fix test * rebase * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * remove log on xorm * Fi download bug * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * Add URL function to serve attachments directly from S3/Minio * Add ability to enable/disable redirection in attachment configuration * Fix typo * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * don't change unrelated files * Fix lint * Fix build * update go.mod and go.sum * Use github.com/minio/minio-go/v6 * Remove unused function * Upgrade minio to v7 and some other improvements * fix lint * Fix go mod Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> Co-authored-by: Tyler <tystuyfzand@gmail.com>
2020-08-18 04:23:45 +00:00
/*
* MinIO Go Library for Amazon S3 Compatible Cloud Storage
* Copyright 2015-2018 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"
"errors"
"fmt"
"io"
"io/ioutil"
"math/rand"
"net"
"net/http"
"net/http/cookiejar"
"net/http/httputil"
"net/url"
"os"
"runtime"
"strings"
"sync"
"time"
md5simd "github.com/minio/md5-simd"
"github.com/minio/minio-go/v7/pkg/credentials"
"github.com/minio/minio-go/v7/pkg/s3utils"
"github.com/minio/minio-go/v7/pkg/signer"
"golang.org/x/net/publicsuffix"
)
// Client implements Amazon S3 compatible methods.
type Client struct {
/// Standard options.
// Parsed endpoint url provided by the user.
endpointURL *url.URL
// Holds various credential providers.
credsProvider *credentials.Credentials
// Custom signerType value overrides all credentials.
overrideSignerType credentials.SignatureType
// User supplied.
appInfo struct {
appName string
appVersion string
}
// Indicate whether we are using https or not
secure bool
// Needs allocation.
httpClient *http.Client
bucketLocCache *bucketLocationCache
// Advanced functionality.
isTraceEnabled bool
traceErrorsOnly bool
traceOutput io.Writer
// S3 specific accelerated endpoint.
s3AccelerateEndpoint string
// Region endpoint
region string
// Random seed.
random *rand.Rand
// lookup indicates type of url lookup supported by server. If not specified,
// default to Auto.
lookup BucketLookupType
// Factory for MD5 hash functions.
md5Hasher func() md5simd.Hasher
sha256Hasher func() md5simd.Hasher
}
// Options for New method
type Options struct {
Creds *credentials.Credentials
Secure bool
Transport http.RoundTripper
Region string
BucketLookup BucketLookupType
// Custom hash routines. Leave nil to use standard.
CustomMD5 func() md5simd.Hasher
CustomSHA256 func() md5simd.Hasher
}
// Global constants.
const (
libraryName = "minio-go"
2021-02-28 23:08:33 +00:00
libraryVersion = "v7.0.10"
Add a storage layer for attachments (#11387) * Add a storage layer for attachments * Fix some bug * fix test * Fix copyright head and lint * Fix bug * Add setting for minio and flags for migrate-storage * Add documents * fix lint * Add test for minio store type on attachments * fix test * fix test * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Add warning when storage migrated successfully * Fix drone * fix test * rebase * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * remove log on xorm * Fi download bug * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * Add URL function to serve attachments directly from S3/Minio * Add ability to enable/disable redirection in attachment configuration * Fix typo * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * don't change unrelated files * Fix lint * Fix build * update go.mod and go.sum * Use github.com/minio/minio-go/v6 * Remove unused function * Upgrade minio to v7 and some other improvements * fix lint * Fix go mod Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> Co-authored-by: Tyler <tystuyfzand@gmail.com>
2020-08-18 04:23:45 +00:00
)
// User Agent should always following the below style.
// Please open an issue to discuss any new changes here.
//
// MinIO (OS; ARCH) LIB/VER APP/VER
const (
libraryUserAgentPrefix = "MinIO (" + runtime.GOOS + "; " + runtime.GOARCH + ") "
libraryUserAgent = libraryUserAgentPrefix + libraryName + "/" + libraryVersion
)
// BucketLookupType is type of url lookup supported by server.
type BucketLookupType int
// Different types of url lookup supported by the server.Initialized to BucketLookupAuto
const (
BucketLookupAuto BucketLookupType = iota
BucketLookupDNS
BucketLookupPath
)
// New - instantiate minio client with options
func New(endpoint string, opts *Options) (*Client, error) {
if opts == nil {
return nil, errors.New("no options provided")
}
clnt, err := privateNew(endpoint, opts)
if err != nil {
return nil, err
}
// Google cloud storage should be set to signature V2, force it if not.
if s3utils.IsGoogleEndpoint(*clnt.endpointURL) {
clnt.overrideSignerType = credentials.SignatureV2
}
// If Amazon S3 set to signature v4.
if s3utils.IsAmazonEndpoint(*clnt.endpointURL) {
clnt.overrideSignerType = credentials.SignatureV4
}
return clnt, nil
}
// EndpointURL returns the URL of the S3 endpoint.
func (c *Client) EndpointURL() *url.URL {
endpoint := *c.endpointURL // copy to prevent callers from modifying internal state
return &endpoint
}
// lockedRandSource provides protected rand source, implements rand.Source interface.
type lockedRandSource struct {
lk sync.Mutex
src rand.Source
}
// Int63 returns a non-negative pseudo-random 63-bit integer as an int64.
func (r *lockedRandSource) Int63() (n int64) {
r.lk.Lock()
n = r.src.Int63()
r.lk.Unlock()
return
}
// Seed uses the provided seed value to initialize the generator to a
// deterministic state.
func (r *lockedRandSource) Seed(seed int64) {
r.lk.Lock()
r.src.Seed(seed)
r.lk.Unlock()
}
// Redirect requests by re signing the request.
func (c *Client) redirectHeaders(req *http.Request, via []*http.Request) error {
if len(via) >= 5 {
return errors.New("stopped after 5 redirects")
}
if len(via) == 0 {
return nil
}
lastRequest := via[len(via)-1]
var reAuth bool
for attr, val := range lastRequest.Header {
// if hosts do not match do not copy Authorization header
if attr == "Authorization" && req.Host != lastRequest.Host {
reAuth = true
continue
}
if _, ok := req.Header[attr]; !ok {
req.Header[attr] = val
}
}
*c.endpointURL = *req.URL
value, err := c.credsProvider.Get()
if err != nil {
return err
}
var (
signerType = value.SignerType
accessKeyID = value.AccessKeyID
secretAccessKey = value.SecretAccessKey
sessionToken = value.SessionToken
region = c.region
)
// Custom signer set then override the behavior.
if c.overrideSignerType != credentials.SignatureDefault {
signerType = c.overrideSignerType
}
// If signerType returned by credentials helper is anonymous,
// then do not sign regardless of signerType override.
if value.SignerType == credentials.SignatureAnonymous {
signerType = credentials.SignatureAnonymous
}
if reAuth {
// Check if there is no region override, if not get it from the URL if possible.
if region == "" {
region = s3utils.GetRegionFromURL(*c.endpointURL)
}
switch {
case signerType.IsV2():
return errors.New("signature V2 cannot support redirection")
case signerType.IsV4():
signer.SignV4(*req, accessKeyID, secretAccessKey, sessionToken, getDefaultLocation(*c.endpointURL, region))
}
}
return nil
}
func privateNew(endpoint string, opts *Options) (*Client, error) {
// construct endpoint.
endpointURL, err := getEndpointURL(endpoint, opts.Secure)
if err != nil {
return nil, err
}
// Initialize cookies to preserve server sent cookies if any and replay
// them upon each request.
jar, err := cookiejar.New(&cookiejar.Options{PublicSuffixList: publicsuffix.List})
if err != nil {
return nil, err
}
// instantiate new Client.
clnt := new(Client)
// Save the credentials.
clnt.credsProvider = opts.Creds
// Remember whether we are using https or not
clnt.secure = opts.Secure
// Save endpoint URL, user agent for future uses.
clnt.endpointURL = endpointURL
transport := opts.Transport
if transport == nil {
transport, err = DefaultTransport(opts.Secure)
if err != nil {
return nil, err
}
}
// Instantiate http client and bucket location cache.
clnt.httpClient = &http.Client{
Jar: jar,
Transport: transport,
CheckRedirect: clnt.redirectHeaders,
}
// Sets custom region, if region is empty bucket location cache is used automatically.
if opts.Region == "" {
opts.Region = s3utils.GetRegionFromURL(*clnt.endpointURL)
}
clnt.region = opts.Region
// Instantiate bucket location cache.
clnt.bucketLocCache = newBucketLocationCache()
// Introduce a new locked random seed.
clnt.random = rand.New(&lockedRandSource{src: rand.NewSource(time.Now().UTC().UnixNano())})
// Add default md5 hasher.
clnt.md5Hasher = opts.CustomMD5
clnt.sha256Hasher = opts.CustomSHA256
if clnt.md5Hasher == nil {
clnt.md5Hasher = newMd5Hasher
}
if clnt.sha256Hasher == nil {
clnt.sha256Hasher = newSHA256Hasher
}
// Sets bucket lookup style, whether server accepts DNS or Path lookup. Default is Auto - determined
// by the SDK. When Auto is specified, DNS lookup is used for Amazon/Google cloud endpoints and Path for all other endpoints.
clnt.lookup = opts.BucketLookup
// Return.
return clnt, nil
}
// SetAppInfo - add application details to user agent.
func (c *Client) SetAppInfo(appName string, appVersion string) {
// if app name and version not set, we do not set a new user agent.
if appName != "" && appVersion != "" {
c.appInfo.appName = appName
c.appInfo.appVersion = appVersion
}
}
// TraceOn - enable HTTP tracing.
func (c *Client) TraceOn(outputStream io.Writer) {
// if outputStream is nil then default to os.Stdout.
if outputStream == nil {
outputStream = os.Stdout
}
// Sets a new output stream.
c.traceOutput = outputStream
// Enable tracing.
c.isTraceEnabled = true
}
// TraceErrorsOnlyOn - same as TraceOn, but only errors will be traced.
func (c *Client) TraceErrorsOnlyOn(outputStream io.Writer) {
c.TraceOn(outputStream)
c.traceErrorsOnly = true
}
// TraceErrorsOnlyOff - Turns off the errors only tracing and everything will be traced after this call.
// If all tracing needs to be turned off, call TraceOff().
func (c *Client) TraceErrorsOnlyOff() {
c.traceErrorsOnly = false
}
// TraceOff - disable HTTP tracing.
func (c *Client) TraceOff() {
// Disable tracing.
c.isTraceEnabled = false
c.traceErrorsOnly = false
}
// SetS3TransferAccelerate - turns s3 accelerated endpoint on or off for all your
// requests. This feature is only specific to S3 for all other endpoints this
// function does nothing. To read further details on s3 transfer acceleration
// please vist -
// http://docs.aws.amazon.com/AmazonS3/latest/dev/transfer-acceleration.html
func (c *Client) SetS3TransferAccelerate(accelerateEndpoint string) {
if s3utils.IsAmazonEndpoint(*c.endpointURL) {
c.s3AccelerateEndpoint = accelerateEndpoint
}
}
// Hash materials provides relevant initialized hash algo writers
// based on the expected signature type.
//
// - For signature v4 request if the connection is insecure compute only sha256.
// - For signature v4 request if the connection is secure compute only md5.
// - For anonymous request compute md5.
func (c *Client) hashMaterials(isMd5Requested bool) (hashAlgos map[string]md5simd.Hasher, hashSums map[string][]byte) {
hashSums = make(map[string][]byte)
hashAlgos = make(map[string]md5simd.Hasher)
if c.overrideSignerType.IsV4() {
if c.secure {
hashAlgos["md5"] = c.md5Hasher()
} else {
hashAlgos["sha256"] = c.sha256Hasher()
}
} else {
if c.overrideSignerType.IsAnonymous() {
hashAlgos["md5"] = c.md5Hasher()
}
}
if isMd5Requested {
hashAlgos["md5"] = c.md5Hasher()
}
return hashAlgos, hashSums
}
// requestMetadata - is container for all the values to make a request.
type requestMetadata struct {
// If set newRequest presigns the URL.
presignURL bool
// User supplied.
bucketName string
objectName string
queryValues url.Values
customHeader http.Header
expires int64
// Generated by our internal code.
bucketLocation string
contentBody io.Reader
contentLength int64
contentMD5Base64 string // carries base64 encoded md5sum
contentSHA256Hex string // carries hex encoded sha256sum
}
// dumpHTTP - dump HTTP request and response.
func (c Client) dumpHTTP(req *http.Request, resp *http.Response) error {
// Starts http dump.
_, err := fmt.Fprintln(c.traceOutput, "---------START-HTTP---------")
if err != nil {
return err
}
// Filter out Signature field from Authorization header.
origAuth := req.Header.Get("Authorization")
if origAuth != "" {
req.Header.Set("Authorization", redactSignature(origAuth))
}
// Only display request header.
reqTrace, err := httputil.DumpRequestOut(req, false)
if err != nil {
return err
}
// Write request to trace output.
_, err = fmt.Fprint(c.traceOutput, string(reqTrace))
if err != nil {
return err
}
// Only display response header.
var respTrace []byte
// For errors we make sure to dump response body as well.
if resp.StatusCode != http.StatusOK &&
resp.StatusCode != http.StatusPartialContent &&
resp.StatusCode != http.StatusNoContent {
respTrace, err = httputil.DumpResponse(resp, true)
if err != nil {
return err
}
} else {
respTrace, err = httputil.DumpResponse(resp, false)
if err != nil {
return err
}
}
// Write response to trace output.
_, err = fmt.Fprint(c.traceOutput, strings.TrimSuffix(string(respTrace), "\r\n"))
if err != nil {
return err
}
// Ends the http dump.
_, err = fmt.Fprintln(c.traceOutput, "---------END-HTTP---------")
if err != nil {
return err
}
// Returns success.
return nil
}
// do - execute http request.
func (c Client) do(req *http.Request) (*http.Response, error) {
resp, err := c.httpClient.Do(req)
if err != nil {
// Handle this specifically for now until future Golang versions fix this issue properly.
if urlErr, ok := err.(*url.Error); ok {
if strings.Contains(urlErr.Err.Error(), "EOF") {
return nil, &url.Error{
Op: urlErr.Op,
URL: urlErr.URL,
Err: errors.New("Connection closed by foreign host " + urlErr.URL + ". Retry again."),
}
}
}
return nil, err
}
// Response cannot be non-nil, report error if thats the case.
if resp == nil {
msg := "Response is empty. " + reportIssue
return nil, errInvalidArgument(msg)
}
// If trace is enabled, dump http request and response,
// except when the traceErrorsOnly enabled and the response's status code is ok
if c.isTraceEnabled && !(c.traceErrorsOnly && resp.StatusCode == http.StatusOK) {
err = c.dumpHTTP(req, resp)
if err != nil {
return nil, err
}
}
return resp, nil
}
// List of success status.
var successStatus = []int{
http.StatusOK,
http.StatusNoContent,
http.StatusPartialContent,
}
// executeMethod - instantiates a given method, and retries the
// request upon any error up to maxRetries attempts in a binomially
// delayed manner using a standard back off algorithm.
func (c Client) executeMethod(ctx context.Context, method string, metadata requestMetadata) (res *http.Response, err error) {
var retryable bool // Indicates if request can be retried.
var bodySeeker io.Seeker // Extracted seeker from io.Reader.
var reqRetry = MaxRetry // Indicates how many times we can retry the request
if metadata.contentBody != nil {
// Check if body is seekable then it is retryable.
bodySeeker, retryable = metadata.contentBody.(io.Seeker)
switch bodySeeker {
case os.Stdin, os.Stdout, os.Stderr:
retryable = false
}
// Retry only when reader is seekable
if !retryable {
reqRetry = 1
}
// Figure out if the body can be closed - if yes
// we will definitely close it upon the function
// return.
bodyCloser, ok := metadata.contentBody.(io.Closer)
if ok {
defer bodyCloser.Close()
}
}
// Create cancel context to control 'newRetryTimer' go routine.
retryCtx, cancel := context.WithCancel(ctx)
// Indicate to our routine to exit cleanly upon return.
defer cancel()
// Blank indentifier is kept here on purpose since 'range' without
// blank identifiers is only supported since go1.4
// https://golang.org/doc/go1.4#forrange.
for range c.newRetryTimer(retryCtx, reqRetry, DefaultRetryUnit, DefaultRetryCap, MaxJitter) {
// Retry executes the following function body if request has an
// error until maxRetries have been exhausted, retry attempts are
// performed after waiting for a given period of time in a
// binomial fashion.
if retryable {
// Seek back to beginning for each attempt.
if _, err = bodySeeker.Seek(0, 0); err != nil {
// If seek failed, no need to retry.
return nil, err
}
}
// Instantiate a new request.
var req *http.Request
req, err = c.newRequest(ctx, method, metadata)
if err != nil {
errResponse := ToErrorResponse(err)
if isS3CodeRetryable(errResponse.Code) {
continue // Retry.
}
return nil, err
}
// Initiate the request.
res, err = c.do(req)
if err != nil {
if errors.Is(err, context.Canceled) || errors.Is(err, context.DeadlineExceeded) {
Add a storage layer for attachments (#11387) * Add a storage layer for attachments * Fix some bug * fix test * Fix copyright head and lint * Fix bug * Add setting for minio and flags for migrate-storage * Add documents * fix lint * Add test for minio store type on attachments * fix test * fix test * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Add warning when storage migrated successfully * Fix drone * fix test * rebase * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * remove log on xorm * Fi download bug * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * Add URL function to serve attachments directly from S3/Minio * Add ability to enable/disable redirection in attachment configuration * Fix typo * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * don't change unrelated files * Fix lint * Fix build * update go.mod and go.sum * Use github.com/minio/minio-go/v6 * Remove unused function * Upgrade minio to v7 and some other improvements * fix lint * Fix go mod Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> Co-authored-by: Tyler <tystuyfzand@gmail.com>
2020-08-18 04:23:45 +00:00
return nil, err
}
// Retry the request
Add a storage layer for attachments (#11387) * Add a storage layer for attachments * Fix some bug * fix test * Fix copyright head and lint * Fix bug * Add setting for minio and flags for migrate-storage * Add documents * fix lint * Add test for minio store type on attachments * fix test * fix test * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Add warning when storage migrated successfully * Fix drone * fix test * rebase * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * remove log on xorm * Fi download bug * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * Add URL function to serve attachments directly from S3/Minio * Add ability to enable/disable redirection in attachment configuration * Fix typo * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * don't change unrelated files * Fix lint * Fix build * update go.mod and go.sum * Use github.com/minio/minio-go/v6 * Remove unused function * Upgrade minio to v7 and some other improvements * fix lint * Fix go mod Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> Co-authored-by: Tyler <tystuyfzand@gmail.com>
2020-08-18 04:23:45 +00:00
continue
}
// For any known successful http status, return quickly.
for _, httpStatus := range successStatus {
if httpStatus == res.StatusCode {
return res, nil
}
}
// Read the body to be saved later.
errBodyBytes, err := ioutil.ReadAll(res.Body)
// res.Body should be closed
closeResponse(res)
if err != nil {
return nil, err
}
// Save the body.
errBodySeeker := bytes.NewReader(errBodyBytes)
res.Body = ioutil.NopCloser(errBodySeeker)
// For errors verify if its retryable otherwise fail quickly.
errResponse := ToErrorResponse(httpRespToErrorResponse(res, metadata.bucketName, metadata.objectName))
// Save the body back again.
errBodySeeker.Seek(0, 0) // Seek back to starting point.
res.Body = ioutil.NopCloser(errBodySeeker)
// Bucket region if set in error response and the error
// code dictates invalid region, we can retry the request
// with the new region.
//
// Additionally we should only retry if bucketLocation and custom
// region is empty.
if c.region == "" {
switch errResponse.Code {
case "AuthorizationHeaderMalformed":
fallthrough
case "InvalidRegion":
fallthrough
case "AccessDenied":
if errResponse.Region == "" {
// Region is empty we simply return the error.
return res, err
}
// Region is not empty figure out a way to
// handle this appropriately.
if metadata.bucketName != "" {
// Gather Cached location only if bucketName is present.
if location, cachedOk := c.bucketLocCache.Get(metadata.bucketName); cachedOk && location != errResponse.Region {
c.bucketLocCache.Set(metadata.bucketName, errResponse.Region)
continue // Retry.
}
} else {
// This is for ListBuckets() fallback.
if errResponse.Region != metadata.bucketLocation {
// Retry if the error response has a different region
// than the request we just made.
metadata.bucketLocation = errResponse.Region
continue // Retry
}
}
}
}
// Verify if error response code is retryable.
if isS3CodeRetryable(errResponse.Code) {
continue // Retry.
}
// Verify if http status code is retryable.
if isHTTPStatusRetryable(res.StatusCode) {
continue // Retry.
}
// For all other cases break out of the retry loop.
break
}
// Return an error when retry is canceled or deadlined
if e := retryCtx.Err(); e != nil {
return nil, e
}
return res, err
}
// newRequest - instantiate a new HTTP request for a given method.
func (c Client) newRequest(ctx context.Context, method string, metadata requestMetadata) (req *http.Request, err error) {
// If no method is supplied default to 'POST'.
if method == "" {
method = http.MethodPost
}
location := metadata.bucketLocation
if location == "" {
if metadata.bucketName != "" {
// Gather location only if bucketName is present.
location, err = c.getBucketLocation(ctx, metadata.bucketName)
if err != nil {
return nil, err
}
}
if location == "" {
location = getDefaultLocation(*c.endpointURL, c.region)
}
}
// Look if target url supports virtual host.
// We explicitly disallow MakeBucket calls to not use virtual DNS style,
// since the resolution may fail.
isMakeBucket := (metadata.objectName == "" && method == http.MethodPut && len(metadata.queryValues) == 0)
isVirtualHost := c.isVirtualHostStyleRequest(*c.endpointURL, metadata.bucketName) && !isMakeBucket
// Construct a new target URL.
targetURL, err := c.makeTargetURL(metadata.bucketName, metadata.objectName, location,
isVirtualHost, metadata.queryValues)
if err != nil {
return nil, err
}
// Initialize a new HTTP request for the method.
req, err = http.NewRequestWithContext(ctx, method, targetURL.String(), nil)
Add a storage layer for attachments (#11387) * Add a storage layer for attachments * Fix some bug * fix test * Fix copyright head and lint * Fix bug * Add setting for minio and flags for migrate-storage * Add documents * fix lint * Add test for minio store type on attachments * fix test * fix test * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Add warning when storage migrated successfully * Fix drone * fix test * rebase * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * remove log on xorm * Fi download bug * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * refactor the codes * add trace * Fix test * Add URL function to serve attachments directly from S3/Minio * Add ability to enable/disable redirection in attachment configuration * Fix typo * Add a storage layer for attachments * Add setting for minio and flags for migrate-storage * fix lint * Add test for minio store type on attachments * Apply suggestions from code review Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> * Fix drone * fix test * Fix test * display the error on console * Move minio test to amd64 since minio docker don't support arm64 * don't change unrelated files * Fix lint * Fix build * update go.mod and go.sum * Use github.com/minio/minio-go/v6 * Remove unused function * Upgrade minio to v7 and some other improvements * fix lint * Fix go mod Co-authored-by: guillep2k <18600385+guillep2k@users.noreply.github.com> Co-authored-by: Tyler <tystuyfzand@gmail.com>
2020-08-18 04:23:45 +00:00
if err != nil {
return nil, err
}
// Get credentials from the configured credentials provider.
value, err := c.credsProvider.Get()
if err != nil {
return nil, err
}
var (
signerType = value.SignerType
accessKeyID = value.AccessKeyID
secretAccessKey = value.SecretAccessKey
sessionToken = value.SessionToken
)
// Custom signer set then override the behavior.
if c.overrideSignerType != credentials.SignatureDefault {
signerType = c.overrideSignerType
}
// If signerType returned by credentials helper is anonymous,
// then do not sign regardless of signerType override.
if value.SignerType == credentials.SignatureAnonymous {
signerType = credentials.SignatureAnonymous
}
// Generate presign url if needed, return right here.
if metadata.expires != 0 && metadata.presignURL {
if signerType.IsAnonymous() {
return nil, errInvalidArgument("Presigned URLs cannot be generated with anonymous credentials.")
}
if signerType.IsV2() {
// Presign URL with signature v2.
req = signer.PreSignV2(*req, accessKeyID, secretAccessKey, metadata.expires, isVirtualHost)
} else if signerType.IsV4() {
// Presign URL with signature v4.
req = signer.PreSignV4(*req, accessKeyID, secretAccessKey, sessionToken, location, metadata.expires)
}
return req, nil
}
// Set 'User-Agent' header for the request.
c.setUserAgent(req)
// Set all headers.
for k, v := range metadata.customHeader {
req.Header.Set(k, v[0])
}
// Go net/http notoriously closes the request body.
// - The request Body, if non-nil, will be closed by the underlying Transport, even on errors.
// This can cause underlying *os.File seekers to fail, avoid that
// by making sure to wrap the closer as a nop.
if metadata.contentLength == 0 {
req.Body = nil
} else {
req.Body = ioutil.NopCloser(metadata.contentBody)
}
// Set incoming content-length.
req.ContentLength = metadata.contentLength
if req.ContentLength <= -1 {
// For unknown content length, we upload using transfer-encoding: chunked.
req.TransferEncoding = []string{"chunked"}
}
// set md5Sum for content protection.
if len(metadata.contentMD5Base64) > 0 {
req.Header.Set("Content-Md5", metadata.contentMD5Base64)
}
// For anonymous requests just return.
if signerType.IsAnonymous() {
return req, nil
}
switch {
case signerType.IsV2():
// Add signature version '2' authorization header.
req = signer.SignV2(*req, accessKeyID, secretAccessKey, isVirtualHost)
case metadata.objectName != "" && metadata.queryValues == nil && method == http.MethodPut && metadata.customHeader.Get("X-Amz-Copy-Source") == "" && !c.secure:
// Streaming signature is used by default for a PUT object request. Additionally we also
// look if the initialized client is secure, if yes then we don't need to perform
// streaming signature.
req = signer.StreamingSignV4(req, accessKeyID,
secretAccessKey, sessionToken, location, metadata.contentLength, time.Now().UTC())
default:
// Set sha256 sum for signature calculation only with signature version '4'.
shaHeader := unsignedPayload
if metadata.contentSHA256Hex != "" {
shaHeader = metadata.contentSHA256Hex
}
req.Header.Set("X-Amz-Content-Sha256", shaHeader)
// Add signature version '4' authorization header.
req = signer.SignV4(*req, accessKeyID, secretAccessKey, sessionToken, location)
}
// Return request.
return req, nil
}
// set User agent.
func (c Client) setUserAgent(req *http.Request) {
req.Header.Set("User-Agent", libraryUserAgent)
if c.appInfo.appName != "" && c.appInfo.appVersion != "" {
req.Header.Set("User-Agent", libraryUserAgent+" "+c.appInfo.appName+"/"+c.appInfo.appVersion)
}
}
// makeTargetURL make a new target url.
func (c Client) makeTargetURL(bucketName, objectName, bucketLocation string, isVirtualHostStyle bool, queryValues url.Values) (*url.URL, error) {
host := c.endpointURL.Host
// For Amazon S3 endpoint, try to fetch location based endpoint.
if s3utils.IsAmazonEndpoint(*c.endpointURL) {
if c.s3AccelerateEndpoint != "" && bucketName != "" {
// http://docs.aws.amazon.com/AmazonS3/latest/dev/transfer-acceleration.html
// Disable transfer acceleration for non-compliant bucket names.
if strings.Contains(bucketName, ".") {
return nil, errTransferAccelerationBucket(bucketName)
}
// If transfer acceleration is requested set new host.
// For more details about enabling transfer acceleration read here.
// http://docs.aws.amazon.com/AmazonS3/latest/dev/transfer-acceleration.html
host = c.s3AccelerateEndpoint
} else {
// Do not change the host if the endpoint URL is a FIPS S3 endpoint.
if !s3utils.IsAmazonFIPSEndpoint(*c.endpointURL) {
// Fetch new host based on the bucket location.
host = getS3Endpoint(bucketLocation)
}
}
}
// Save scheme.
scheme := c.endpointURL.Scheme
// Strip port 80 and 443 so we won't send these ports in Host header.
// The reason is that browsers and curl automatically remove :80 and :443
// with the generated presigned urls, then a signature mismatch error.
if h, p, err := net.SplitHostPort(host); err == nil {
if scheme == "http" && p == "80" || scheme == "https" && p == "443" {
host = h
}
}
urlStr := scheme + "://" + host + "/"
// Make URL only if bucketName is available, otherwise use the
// endpoint URL.
if bucketName != "" {
// If endpoint supports virtual host style use that always.
// Currently only S3 and Google Cloud Storage would support
// virtual host style.
if isVirtualHostStyle {
urlStr = scheme + "://" + bucketName + "." + host + "/"
if objectName != "" {
urlStr = urlStr + s3utils.EncodePath(objectName)
}
} else {
// If not fall back to using path style.
urlStr = urlStr + bucketName + "/"
if objectName != "" {
urlStr = urlStr + s3utils.EncodePath(objectName)
}
}
}
// If there are any query values, add them to the end.
if len(queryValues) > 0 {
urlStr = urlStr + "?" + s3utils.QueryEncode(queryValues)
}
return url.Parse(urlStr)
}
// returns true if virtual hosted style requests are to be used.
func (c *Client) isVirtualHostStyleRequest(url url.URL, bucketName string) bool {
if bucketName == "" {
return false
}
if c.lookup == BucketLookupDNS {
return true
}
if c.lookup == BucketLookupPath {
return false
}
// default to virtual only for Amazon/Google storage. In all other cases use
// path style requests
return s3utils.IsVirtualHostSupported(url, bucketName)
}