utils

package
v1.9.0 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Oct 13, 2026 License: Apache-2.0 Imports: 41 Imported by: 0

Documentation

Overview

code modified from golang std library src/internal/filepathlite/path.go to support path separator '/' for all platforms.

Index

Constants

View Source 
const (
	ContextKeyRegion           = httpctx.ContextKeyRegion
	ContextKeyStartTime        = httpctx.ContextKeyStartTime
	ContextKeyIsRoot           = httpctx.ContextKeyIsRoot
	ContextKeyRootAccessKey    = httpctx.ContextKeyRootAccessKey
	ContextKeyAccount          = httpctx.ContextKeyAccount
	ContextKeyAuthenticated    = httpctx.ContextKeyAuthenticated
	ContextKeyPublicBucket     = httpctx.ContextKeyPublicBucket
	ContextKeyParsedAcl        = httpctx.ContextKeyParsedAcl
	ContextKeySkipResBodyLog   = httpctx.ContextKeySkipResBodyLog
	ContextKeyBodyReader       = httpctx.ContextKeyBodyReader
	ContextKeyBodyStream       = httpctx.ContextKeyBodyStream
	ContextKeySkip             = httpctx.ContextKeySkip
	ContextKeyStack            = httpctx.ContextKeyStack
	ContextKeyBucketOwner      = httpctx.ContextKeyBucketOwner
	ContextKeyObjectPostResult = httpctx.ContextKeyObjectPostResult
	ContextKeyRequestID        = httpctx.ContextKeyRequestID
	ContextKeyHostID           = httpctx.ContextKeyHostID
	ContextKeyWebsiteConfig    = httpctx.ContextKeyWebsiteConfig
)
View Source 
const (
	HeaderAmzRequestID = "x-amz-request-id"
	HeaderAmzID2       = "x-amz-id-2"
)
View Source 
const (
	// Standard S3 limiters
	LimiterTypeMaxKeys          = "max-keys"           // ListObjects, ListObjectVersions
	LimiterTypeMaxParts         = "max-parts"          // ListParts
	LimiterTypeMaxUploads       = "max-uploads"        // ListMultipartUploads
	LimiterTypePartNumberMarker = "part-number-marker" // ListParts(partNumberMarker)
	LimiterTypeVersionsMaxKeys  = "versions_max_keys"  // ListObjectVersions (internal alias mapped to max-keys)
	LimiterTypeMaxBuckets       = "max-buckets"        // ListBuckets
)
View Source 
const AwsChunkedEncoding = "aws-chunked"

AwsChunkedEncoding is the coding a client announces when it frames a body in aws-chunked, as the SDKs do to carry a trailing checksum.

Variables

This section is empty.

Functions

func AddCRCChecksum added in v1.0.15 

func AddCRCChecksum(algo types.ChecksumAlgorithm, crc, partCrc string, partLen int64) (string, error)

AddCRCChecksum calculates the composite CRC checksum after adding the part crc. Only CRC32, CRC32C, and CRC64NVME are supported. The input checksums must be base64-encoded strings.

func ApplyOverride added in v1.0.18 

func ApplyOverride(original, override *string) *string

ApplyOverride returns the override value if it exists and status is 200, otherwise returns original

func Base64SumString added in v1.0.11 

func Base64SumString(b []byte) string

Base64SumString converts the hash bytes to the b64 encoded string checksum value

func CheckPresignedSignature 

func CheckPresignedSignature(ctx fiber.Ctx, auth AuthData, derivedKey []byte) error

CheckPresignedSignature validates a presigned request's signature against derivedKey

func CheckValidSignature 

func CheckValidSignature(ctx fiber.Ctx, auth AuthData, derivedKey []byte, checksum string, tdate time.Time, contentLen int64) (string, error)

CheckValidSignature validates the ctx v4 auth signature against derivedKey — the request's kSigning value, either derived locally from a known secret or obtained from a standalone IAM service that never reveals the secret itself.

func ConvertPtrToStringPtr added in v1.0.17 

func ConvertPtrToStringPtr[T any](val *T) *string

Converts any pointer to a string pointer

func ConvertTaggingXMLToQueryString added in v1.4.0 

func ConvertTaggingXMLToQueryString(data []byte) (string, error)

ConvertTaggingXMLToQueryString parses object tagging XML and converts it into the x-amz-tagging query-string form expected by PutObject requests.

func ConvertToStringPtr added in v1.0.17 

func ConvertToStringPtr[T any](val T) *string

Converts any type to a string pointer

func DetectResourceType added in v1.5.0 

func DetectResourceType(ctx fiber.Ctx) s3err.ResourceType

func EnsureRequestIDs added in v1.5.0 

func EnsureRequestIDs(ctx fiber.Ctx) (requestID, hostID string)

EnsureRequestIDs makes sure the request-local IDs exist and are present in the response headers. Existing local values are reused so headers and XML bodies stay consistent throughout the request.

func ExtractChecksumType added in v1.0.15 

func ExtractChecksumType(ctx fiber.Ctx) (checksumType, error)

Extracts and validates the checksum type from the 'X-Amz-Trailer' header

func ExtractMetadataFromFields added in v1.4.0 

func ExtractMetadataFromFields(fields map[string]string) (map[string]string, error)

func FilterObjectAttributes 

func FilterObjectAttributes(attrs map[s3response.ObjectAttributes]struct{}, output s3response.GetObjectAttributesResponse) s3response.GetObjectAttributesResponse

func FormatDatePtrToString added in v1.0.17 

func FormatDatePtrToString(date *time.Time, format string) *string

Formats the date with the given formatting and returns a string pointer

func FormatRetainUntilDate added in v1.9.0 

func FormatRetainUntilDate(date *time.Time) *string

FormatRetainUntilDate formats an Object Lock retain until date for a response header: in UTC to the millisecond, with the milliseconds left out when they are zero.

func FromSlash added in v1.0.15 

func FromSlash(path string) string

func GenerateObjectLocation added in v1.1.0 

func GenerateObjectLocation(ctx fiber.Ctx, virtualDomain, bucket, object string) string

GenerateObjectLocation generates the object location path-styled or host-styled depending on the gateway configuration

func GetInt64 added in v1.0.17 

func GetInt64(n *int64) int64

GetInt64 returns the value of int64 pointer

func GetQueryParam added in v1.0.18 

func GetQueryParam(ctx fiber.Ctx, key string) *string

GetQueryParam returns a pointer to the query parameter value if it exists

func GetStringPtr added in v1.0.17 

func GetStringPtr(str string) *string

Returns the provided string pointer

func GetUserMetaData 

func GetUserMetaData(headers *fasthttp.RequestHeader) (map[string]string, error)

GetUserMetaData extracts user metadata from headers with the "x-amz-meta-" prefix. Keys are normalized to lowercase and duplicate headers are merged as comma-separated values. The total metadata size is validated against the 2 KB S3 limit.

func HasAwsChunkedEncoding added in v1.9.0 

func HasAwsChunkedEncoding(contentEncoding string) bool

HasAwsChunkedEncoding reports whether a Content-Encoding value carries the aws-chunked token.

func HasPayloadIntegrityCheck added in v1.9.0 

func HasPayloadIntegrityCheck(ctx fiber.Ctx) bool

HasPayloadIntegrityCheck reports whether an upload carries an integrity check of its body: a Content-MD5 header, an x-amz-checksum-* header, or an x-amz-checksum-* trailer. A payload hash, signed or not, and chunk signatures don't count as one.

func HexBytes added in v1.5.0 

func HexBytes(s string) string

func HostID added in v1.5.0 

func HostID(ctx fiber.Ctx) string

func IsAnonymousPayloadHashSupported added in v1.0.19 

func IsAnonymousPayloadHashSupported(hash string) error

IsAnonymousPayloadHashSupported returns error if payload hash is streaming signed. e.g. "STREAMING-AWS4-HMAC-SHA256-PAYLOAD", "STREAMING-AWS4-ECDSA-P256-SHA256-PAYLOAD" ...

func IsChecksumAlgorithmValid added in v1.0.11 

func IsChecksumAlgorithmValid(alg types.ChecksumAlgorithm) error

func IsChecksumComposable added in v1.0.15 

func IsChecksumComposable(algo types.ChecksumAlgorithm) bool

IsChecksumComposable tests if the final full-object crc can be calculated from the part crc values.

func IsChecksumTypeValid added in v1.0.11 

func IsChecksumTypeValid(t types.ChecksumType) error

Validates the provided checksum type

func IsObjectNameValid added in v1.0.15 

func IsObjectNameValid(name string) bool

func IsPresignedURLAuth added in v1.0.19 

func IsPresignedURLAuth(ctx fiber.Ctx) bool

IsPresignedURLAuth determines if the request is presigned: which is authorization with query params

func IsPresignedURLAuthV2 added in v1.5.0 

func IsPresignedURLAuthV2(ctx fiber.Ctx) bool

IsPresignedURLAuthV2 determines if the request is query-string signed with aws v2 signer

func IsSpecialPayload 

func IsSpecialPayload(str string) bool

IsSpecialPayload checks for special authorization types

func IsStreamingPayload added in v1.0.11 

func IsStreamingPayload(str string) bool

IsStreamingPayload checks for streaming/unsigned authorization types

func IsUnsignedPaylod added in v1.0.19 

func IsUnsignedPaylod(hash string) bool

IsUnsignedPaylod checks if the provided payload hash type is "UNSIGNED-PAYLOAD"

func IsUnsignedStreamingPayload added in v1.0.15 

func IsUnsignedStreamingPayload(str string) bool

Checks if the provided string is unsigned payload trailer type

func IsValidBucketName 

func IsValidBucketName(bucket string) bool

func IsValidChecksum added in v1.0.11 

func IsValidChecksum(checksum string, algorithm types.ChecksumAlgorithm) bool

func IsValidOwnership added in v1.0.3 

func IsValidOwnership(val types.ObjectOwnership) bool

func IsValidSha256PayloadHeader added in v1.5.0 

func IsValidSha256PayloadHeader(value string) bool

IsValidSha256PayloadHeader checks if the provided x-amz-content-sha256 payload header is valid special payload type or a valid sh256 hash

func MergeDeleteObjectsResult added in v1.8.0 

func MergeDeleteObjectsResult(objects []types.ObjectIdentifier, checkErrs []error, backendResult s3response.DeleteResult) s3response.DeleteResult

MergeDeleteObjectsResult builds the final DeleteObjects response, preserving the order objects were requested in across both the Deleted and Error lists. objects is the full request; checkErrs is VerifyObjectsAccess's per-object result for it (nil entries were sent to the backend); backendResult is the backend's response for just those.

The backend's own Deleted/Error order is not assumed to match the order its objects were sent in, so objects are matched back to their backend result by identity (key + version) rather than by position, with a FIFO queue per identity to keep duplicate keys in the same request each paired with their own result. versionID is "" for a keyed (unversioned) delete, matching how both the request and every backend's response represent "no version specified" — as a nil pointer.

func NewChunkReader 

func NewChunkReader(ctx fiber.Ctx, r io.Reader, authdata AuthData, canonicalString string, derivedKey []byte, date time.Time) (io.Reader, error)

func NewHash added in v1.9.0 

func NewHash(ht HashType) (hash.Hash, error)

NewHash returns the hash.Hash implementing the given checksum algorithm.

func NewS3HostID added in v1.5.0 

func NewS3HostID() string

NewS3HostID generates a new s3-style host ID

func NewS3RequestID added in v1.5.0 

func NewS3RequestID() string

NewS3RequestID returns a request ID, for example "5MRQJ97RHWJ4FMX9".

func NewSignedChunkReader added in v1.0.10 

func NewSignedChunkReader(r io.Reader, authdata AuthData, canonicalString string, derivedKey []byte, date time.Time, chType checksumType, requireTrailer bool, cLength int64) (io.Reader, error)

NewChunkReader reads from request body io.Reader and parses out the chunk metadata in stream. The headers are validated for proper signatures. Reading from the chunk reader will read only the object data stream without the chunk headers/trailers. derivedKey is the same SigV4 kSigning value the seed request's Authorization header was already checked against, reused here rather than re-derived or re-fetched.

func NewTLSListener added in v1.2.0 

func NewTLSListener(network string, address string, getCertificateFunc func(*tls.ClientHelloInfo) (*tls.Certificate, error)) (net.Listener, error)

func ParsObjectLockHdrs 

func ParsObjectLockHdrs(ctx fiber.Ctx) (*objLockCfg, error)

ParsObjectLockHdrs parses and validates the Object Lock headers of an object write. A header sent with an empty value is validated, not ignored. Whether a header is sent is told by its name: fasthttp may hold an empty value as a nil slice, so Peek returning nil doesn't mean it is absent.

func ParseContentEncoding added in v1.9.0 

func ParseContentEncoding(ctx fiber.Ctx) string

ParseContentEncoding returns the Content-Encoding to store for a request, dropping the aws-chunked token unless the request declares a non streaming payload type in x-amz-content-sha256.

S3 keys this on the declaration alone. A streaming payload type frames the body in aws-chunked, and a request that declares no payload type - a presigned URL, signed as UNSIGNED-PAYLOAD and carrying no header - may frame it too, so both drop the token. A declared hex digest or UNSIGNED-PAYLOAD says the body is sent as-is, so there the token is a coding the client chose and is stored as sent.

S3 applies this to the Content-Encoding header itself, before any API sees it, so every controller that reads the header inherits it - including CopyObject and CreateMultipartUpload, which carry no body to frame. POSTObject takes its value from a form field and ignores the header, so it stores the token as sent and must not call this.

func ParseCreateMpChecksumHeaders added in v1.0.11 

func ParseCreateMpChecksumHeaders(ctx fiber.Ctx) (types.ChecksumAlgorithm, types.ChecksumType, error)

Parses and validates the x-amz-checksum-algorithm and x-amz-checksum-type headers

func ParseDecodedContentLength added in v1.0.20 

func ParseDecodedContentLength(ctx fiber.Ctx) (int64, error)

ParseDecodedContentLength extracts and validates the 'x-amz-decoded-content-length' from fiber context

func ParseDeleteObjects 

func ParseDeleteObjects(objs []types.ObjectIdentifier) (result []string)

func ParseIfMatchSize added in v1.0.18 

func ParseIfMatchSize(ctx fiber.Ctx) *int64

ParseIfMatchSize parses the 'x-amz-if-match-size' to *int64 if parsing fails, returns nil

func ParseMaxLimiter added in v1.2.0 

func ParseMaxLimiter(limiter string, lt LimiterType) (int32, error)

ParseMaxLimiter parses and validates the pagination limiter

  • Empty value → return default depending on limiter type.
  • Non-numeric or out-of-range → return the corresponding InvalidArgument error.
  • Negative values → return specific negative-value errors per limiter type.
  • Values above defaultMaxLimiter are clamped.
  • The versions_max_keys limiter is normalized to max-keys for validation.

func ParseObjectAttributes 

func ParseObjectAttributes(ctx fiber.Ctx) (map[s3response.ObjectAttributes]struct{}, error)

func ParseObjectLockFields added in v1.9.0 

func ParseObjectLockFields(fields map[string]string) (*objLockCfg, error)

ParseObjectLockFields parses and validates the Object Lock fields of a browser-based POST upload form, whose field names are lowercase. A field sent with an empty value is validated, not ignored.

func ParsePreconditionDateHeader added in v1.0.18 

func ParsePreconditionDateHeader(date string) *time.Time

ParsePreconditionDateHeader tries to parse the given date string as - RFC1123 - RFC3339 both are valid

func ParsePreconditionDateHeaders added in v1.0.18 

func ParsePreconditionDateHeaders(ctx fiber.Ctx, opts ...preconditionOpt) (*time.Time, *time.Time)

ParsePreconditionDateHeaders parses the "If-Modified-Since" and "If-Unmodified-Since" headers from fiber context to *time.Time

func ParsePreconditionMatchHeaders added in v1.0.18 

func ParsePreconditionMatchHeaders(ctx fiber.Ctx, opts ...preconditionOpt) (*string, *string)

ParsePreconditionMatchHeaders extracts "If-Match" and "If-None-Match" headers from fiber Ctx

func ParseTagging added in v1.0.13 

func ParseTagging(data []byte, limit TagLimit) (map[string]string, error)

Parses and validates tagging

func RequestID added in v1.5.0 

func RequestID(ctx fiber.Ctx) string

func SetBucketNameValidationStrict added in v1.0.19 

func SetBucketNameValidationStrict(strict bool)

func SetMetaHeaders 

func SetMetaHeaders(ctx fiber.Ctx, meta map[string]string)

func SetRegionMismatchHeader added in v1.9.0 

func SetRegionMismatchHeader(ctx fiber.Ctx, err error)

SetRegionMismatchHeader reports the region the request should have been signed for in the x-amz-bucket-region response header, when err is a signing region mismatch. HEAD requests carry no response body, so the header is the only channel through which a client can discover the gateway region and retry against it, which is how the aws sdks recover from the mismatch.

func SetResponseHeaders 

func SetResponseHeaders(ctx fiber.Ctx, headers []CustomHeader)

func SignPostPolicy added in v1.4.0 

func SignPostPolicy(base64Policy string, derivedKey []byte) (string, error)

SignPostPolicy signs a POST-policy document with derivedKey

func StreamResponseBody added in v1.0.10 

func StreamResponseBody(ctx fiber.Ctx, rdr io.ReadCloser, bodysize int)

Streams the response body by chunks

func TrimQuotes added in v1.3.0 

func TrimQuotes(str string) string

func ValidateBucketRegion added in v1.9.0 

func ValidateBucketRegion(bucketRegion, region string) error

ValidateBucketRegion checks a ListBuckets bucket-region value. Every bucket lives in the gateway region, so like an S3 regional endpoint the gateway rejects any other region, matching it case-insensitively.

func ValidateCopySource added in v1.0.17 

func ValidateCopySource(input string) error

ValidateCopySource parses and validates the copy-source

func ValidateDate 

func ValidateDate(date time.Time) error

func ValidateLocationConstraint added in v1.9.0 

func ValidateLocationConstraint(constraint *string, region string) error

ValidateLocationConstraint checks a CreateBucket location constraint. The global endpoint serves us-east-1 and takes no constraint; any other region is a region specific endpoint and requires the constraint to name it.

func ValidateSSEHeaders added in v1.9.0 

func ValidateSSEHeaders(ssec SSECHeaders, serverSideEncryption string) error

ValidateSSEHeaders rejects SSE-C combined with another encryption method and encryption methods that this gateway cannot honor.

func ValidateVersionId added in v1.0.19 

func ValidateVersionId(ctx fiber.Ctx) error

ValidateVersionId ensures the versionId query parameter, if specified, isn't empty. S3 rejects both "?versionId=" and the valueless "?versionId" form, and rejects the request if any of the repeated values is empty.

func ValidateWebsiteRedirectLocation added in v1.6.0 

func ValidateWebsiteRedirectLocation(location string) error

func WithCopySource added in v1.0.18 

func WithCopySource() preconditionOpt

Types

type AuthData 

type AuthData = sigv4auth.AuthData

AuthData is the parsed authorization data from the header.

func ParseAuthorization 

func ParseAuthorization(authorization string) (AuthData, error)

ParseAuthorization returns the parsed fields for the aws v4 auth header example authorization string from aws docs: Authorization: AWS4-HMAC-SHA256 Credential=AKIAIOSFODNN7EXAMPLE/20130524/us-east-1/s3/aws4_request, SignedHeaders=host;range;x-amz-date, Signature=fe5f80f77d5fa3beca038a248ff027d0445342fe2855ddc963176630326f1024

func ParsePresignedURIParts 

func ParsePresignedURIParts(ctx fiber.Ctx, region string) (AuthData, error)

https://docs.aws.amazon.com/AmazonS3/latest/API/sigv4-query-string-auth.html

ParsePresignedURIParts parses and validates request URL query parameters

?X-Amz-Algorithm=AWS4-HMAC-SHA256 &X-Amz-Credential=access-key-id/20130721/us-east-1/s3/aws4_request &X-Amz-Date=20130721T201207Z &X-Amz-Expires=86400 &X-Amz-SignedHeaders=host &X-Amz-Signature=1e68ad45c1db540284a4a1eca3884c293ba1a0ff63ab9db9a15b5b29dfa02cd8

type ChecksumValues added in v1.0.13 

type ChecksumValues map[types.ChecksumAlgorithm]string

func ParseCalculatedChecksumFields added in v1.4.0 

func ParseCalculatedChecksumFields(fields map[string]string) (ChecksumValues, error)

ParseCalculatedChecksumFields parses and validates object POST checksum fields

func ParseCalculatedChecksumHeaders added in v1.0.17 

func ParseCalculatedChecksumHeaders(ctx fiber.Ctx) (ChecksumValues, error)

ParseCalculatedChecksumHeaders parses and validates x-amz-checksum-x header keys e.g x-amz-checksum-crc32, x-amz-checksum-sha256 ...

func ParseChecksumHeadersAndSdkAlgo added in v1.0.17 

func ParseChecksumHeadersAndSdkAlgo(ctx fiber.Ctx) (types.ChecksumAlgorithm, ChecksumValues, error)

ParseChecksumHeadersAndSdkAlgo parses/validates 'x-amz-sdk-checksum-algorithm' and 'x-amz-checksum-x' precalculated request headers

func ParseCompleteMpChecksumHeaders added in v1.0.19 

func ParseCompleteMpChecksumHeaders(ctx fiber.Ctx) (ChecksumValues, error)

ParseCompleteMpChecksumHeaders parses and validates the 'CompleteMultipartUpload' x-amz-checksum-x headers by supporting both 'checksum' and 'checksum-<part_length>' formats

func (ChecksumValues) Headers added in v1.0.13 

func (cv ChecksumValues) Headers() string

Headers concatinates checksum algorithm by prefixing each with 'x-amz-checksum-' e.g. "x-amz-checksum-crc64nvme, x-amz-checksum-sha1"

type ChunkReader 

type ChunkReader struct {
	// contains filtered or unexported fields
}

ChunkReader reads from chunked upload request body, and returns object data stream

func (*ChunkReader) Algorithm added in v1.1.0 

func (cr *ChunkReader) Algorithm() string

Algorithm returns the checksum algorithm

func (*ChunkReader) Checksum added in v1.1.0 

func (cr *ChunkReader) Checksum() string

Checksum returns the parsed trailing checksum

func (*ChunkReader) Read 

func (cr *ChunkReader) Read(p []byte) (int, error)

Read satisfies the io.Reader for this type

type CompositeChecksumReader added in v1.0.11 

type CompositeChecksumReader struct {
	// contains filtered or unexported fields
}

func NewCompositeChecksumReader added in v1.0.11 

func NewCompositeChecksumReader(ht HashType) (*CompositeChecksumReader, error)

NewCompositeChecksumReader initializes a composite checksum processor, which decodes and validates the provided checksums and returns the final checksum based on the previous processings.

The supported checksum types are: - CRC32 - CRC32C - SHA1 - SHA256 - SHA512 - MD5 - XXHASH64 - XXHASH3 - XXHASH128

func (*CompositeChecksumReader) Process added in v1.0.11 

func (ccr *CompositeChecksumReader) Process(checksum string) error

Decodes and writes the checksum in the hasher

func (*CompositeChecksumReader) Sum added in v1.0.11 

func (ccr *CompositeChecksumReader) Sum() string

Returns the base64 encoded composite checksum

type ConditionalHeaders added in v1.0.18 

type ConditionalHeaders struct {
	IfMatch       *string
	IfNoneMatch   *string
	IfModSince    *time.Time
	IfUnmodeSince *time.Time
}

ConditionalHeaders holds the conditional header values

func ParsePreconditionHeaders added in v1.0.18 

func ParsePreconditionHeaders(ctx fiber.Ctx, opts ...preconditionOpt) ConditionalHeaders

ParsePreconditionHeaders parses the precondition headers: - If-Match - If-None-Match - If-Modified-Since - If-Unmodified-Since

type ContentLengthReader added in v1.9.0 

type ContentLengthReader struct {
	// contains filtered or unexported fields
}

ContentLengthReader turns an EOF that arrives before Content-Length bytes were read into an IncompleteBody error.

fasthttp's request stream reports a client that closed the connection in the middle of a fixed-length body as a plain io.EOF (only the chunked transfer-encoding path is converted to io.ErrUnexpectedEOF), and io.Copy treats io.EOF as a normal end of stream. Without this check an aborted upload is committed as a complete, shorter object. aws-chunked bodies do not need this: their chunk readers already validate the framing.

func NewContentLengthReader added in v1.9.0 

func NewContentLengthReader(r io.Reader, contentLength int64) *ContentLengthReader

func (*ContentLengthReader) Read added in v1.9.0 

func (cr *ContentLengthReader) Read(p []byte) (int, error)

type ContextKey added in v1.0.15 

type ContextKey = httpctx.ContextKey

Region, StartTime, IsRoot, Account, AccessKey context locals are set to default values in middlewares.SetDefaultValues to avoid the nil interface conversions

type CredentialsScope added in v1.4.0 

type CredentialsScope = sigv4auth.CredentialsScope

func ParseCredentials added in v1.4.0 

func ParseCredentials(input string, errHandler CredsError) (*CredentialsScope, error)

type CredsError added in v1.4.0 

type CredsError interface {
	MalformedCredential(string) s3err.S3Error
	IncorrectService(string, string) s3err.S3Error
	IncorrectTerminal(string, string) s3err.S3Error
	InvalidDateFormat(string, string) s3err.S3Error
}

type CustomHeader 

type CustomHeader struct {
	Key   string
	Value string
}

type HashReader 

type HashReader struct {
	// contains filtered or unexported fields
}

HashReader is an io.Reader that calculates the checksum as the data is read

func NewHashReader 

func NewHashReader(r io.Reader, expectedSum string, ht HashType) (*HashReader, error)

NewHashReader intializes an io.Reader from an underlying io.Reader that calculates the checksum while the reader is being read from. If the sum provided is not "", the reader will return an error when the underlying reader returns io.EOF if the checksum does not match the provided expected checksum. If the provided sum is "", then the Sum() method can still be used to get the current checksum for the data read so far.

func (*HashReader) Read 

func (hr *HashReader) Read(p []byte) (int, error)

Read allows *HashReader to be used as an io.Reader

func (*HashReader) SetReader added in v1.0.11 

func (hr *HashReader) SetReader(r io.Reader)

func (*HashReader) Sum 

func (hr *HashReader) Sum() string

Sum returns the checksum hash of the data read so far

func (*HashReader) Type added in v1.0.11 

func (hr *HashReader) Type() HashType

type HashType 

type HashType string

HashType identifies the checksum algorithm to be used 

const (
	// HashTypeContentMD5 generates MD5 checksum for the Content-MD5 header.
	HashTypeContentMD5 HashType = "content-md5"
	// HashTypeMd5 generates MD5 Base64-Encoded checksum for the data stream
	HashTypeMd5 HashType = "md5"
	// HashTypeSha256 generates SHA256 Base64-Encoded checksum for the data stream
	HashTypeSha256 HashType = "sha256"
	// HashTypeSha256Hex generates SHA256 hex encoded checksum for the data stream
	HashTypeSha256Hex HashType = "sha256-hex"
	// HashTypeSha1 generates SHA1 Base64-Encoded checksum for the data stream
	HashTypeSha1 HashType = "sha1"
	// HashTypeSha512 generates SHA512 Base64-Encoded checksum for the data stream
	HashTypeSha512 HashType = "sha512"
	// HashTypeCRC32 generates CRC32 Base64-Encoded checksum for the data stream
	HashTypeCRC32 HashType = "crc32"
	// HashTypeCRC32C generates CRC32C Base64-Encoded checksum for the data stream
	HashTypeCRC32C HashType = "crc32c"
	// HashTypeCRC64NVME generates CRC64NVME Base64-Encoded checksum for the data stream
	HashTypeCRC64NVME HashType = "crc64nvme"
	// HashTypeXXHASH64 generates XXHASH64 Base64-Encoded checksum for the data stream
	HashTypeXXHASH64 HashType = "xxhash64"
	// HashTypeXXHASH3 generates XXHASH3 Base64-Encoded checksum for the data stream
	HashTypeXXHASH3 HashType = "xxhash3"
	// HashTypeXXHASH128 generates XXHASH128 Base64-Encoded checksum for the data stream
	HashTypeXXHASH128 HashType = "xxhash128"
	// HashTypeNone is a no-op checksum for the data stream
	HashTypeNone HashType = "none"
)

type LimiterType added in v1.2.0 

type LimiterType string

LimiterType represents the name of a pagination limiter parameter.

type MpFileReader added in v1.4.0 

type MpFileReader interface {
	io.Reader
	Length() int64
}

MpFileReader is the streaming interface for the file part of a multipart POST. It extends io.Reader with a Length method that returns the number of file-content bytes actually delivered to callers (boundary and delimiter bytes are not counted).

type MpParseResult added in v1.4.0 

type MpParseResult struct {
	// Fields contains all non-file form fields collected before the file part.
	Fields map[string]string
	// FileRdr streams the file payload without buffering the entire part in memory.
	FileRdr MpFileReader
	// ContentLength is the expected byte length of the file payload only.
	ContentLength int64
}

type MultipartParser added in v1.4.0 

type MultipartParser struct {
	// contains filtered or unexported fields
}

MultipartParser parses S3 browser-based POST multipart/form-data in a streaming way. It buffers regular form fields, but it does not buffer the file part.

func NewMultipartParser added in v1.4.0 

func NewMultipartParser(body io.Reader, boundary string, requestContentLength int64) (*MultipartParser, error)

NewMultipartParser creates a new streaming multipart parser. boundary should be the raw boundary value from Content-Type,

func (*MultipartParser) Parse added in v1.4.0 

func (mp *MultipartParser) Parse() (*MpParseResult, error)

Parse parses all non-file fields and returns:

  • form values
  • a streaming file reader
  • file content length

The returned file reader MUST be read until EOF, otherwise final-boundary validation is not triggered.

type SSECHeaders added in v1.9.0 

type SSECHeaders struct {
	Algorithm *string
	Key       *string
	KeyMD5    *string
}

SSECHeaders holds the SSE-C (server-side encryption with customer-provided keys) request values. Unset values are nil.

func ExtractSSECHeaders added in v1.9.0 

func ExtractSSECHeaders(ctx fiber.Ctx) SSECHeaders

ExtractSSECHeaders reads SSE-C request headers without validating them.

func ParseCopySourceSSECHeaders added in v1.9.0 

func ParseCopySourceSSECHeaders(ctx fiber.Ctx) (SSECHeaders, error)

ParseCopySourceSSECHeaders extracts and validates the x-amz-copy-source-server-side-encryption-customer-* request headers.

func ParseSSECFields added in v1.9.0 

func ParseSSECFields(fields map[string]string) (SSECHeaders, error)

ParseSSECFields extracts and validates the SSE-C values from POST object form fields, whose names are expected to be lowercased.

func ParseSSECHeaders added in v1.9.0 

func ParseSSECHeaders(ctx fiber.Ctx) (SSECHeaders, error)

ParseSSECHeaders extracts and validates the x-amz-server-side-encryption-customer-* request headers.

type TagLimit added in v1.0.13 

type TagLimit int

TagLimit specifies the allowed tag count in a tag set 

const (
	// Tag limit for bucket tagging
	TagLimitBucket TagLimit = 50
	// Tag limit for object tagging
	TagLimitObject TagLimit = 10
)

type UnsignedChunkReader added in v1.0.10 

type UnsignedChunkReader struct {
	// contains filtered or unexported fields
}

UnsignedChunkReader decodes AWS aws-chunked unsigned streaming request bodies. It strips chunk headers/trailers, validates chunk framing and trailing checksums, and exposes only object payload bytes through Read.

The reader is intentionally streaming: chunk payload bytes are read directly into the caller's buffer. It only keeps bounded header/trailer parsing state and a few counters, so a large client-declared chunk does not become a large allocation in the gateway.

func NewUnsignedChunkReader added in v1.0.10 

func NewUnsignedChunkReader(r io.Reader, ct checksumType, decContentLength int64) (*UnsignedChunkReader, error)

func (*UnsignedChunkReader) Algorithm added in v1.0.20 

func (ucr *UnsignedChunkReader) Algorithm() string

Algorithm returns the checksum algorithm

func (*UnsignedChunkReader) Checksum added in v1.0.20 

func (ucr *UnsignedChunkReader) Checksum() string

Checksum returns the parsed trailing checksum

func (*UnsignedChunkReader) Read added in v1.0.10 

func (ucr *UnsignedChunkReader) Read(p []byte) (int, error)

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.