transport

package module
v0.0.0-...-d739d82 Latest Latest
Warning

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

 Go to latest Published: Feb 11, 2025 License: MIT Imports: 17 Imported by: 0

README

Overview

Go Reference Coverage

This library provides a common set of operations for writing apt transport methods in golang. It does not handle the exact logic that covers every possible transport protocol, but does allow for an easy to use interface when communicating with apt or apt-get.

In addition to providing a simple interface for acquiring resources in custom transport methods, this API also exports all pieces necessary to recreate the common interfaces. This is done to provide users with more granular support over Message deserialization, Method acquisition, and introspection.

Usage

To add this as a go module simply do

$ go get occult.work/apt/transport@latest

NOTE: There is no occult.work/apt package.

Testing

To run unit tests, simply run go test ./.... All mocking, testing data, etc. is taken care of. For more "fun" output, users can use gotestfmt

$ go test -v ./... -json -cover ${PWD} 2>&1 | gotestfmt

Documentation

Index

Examples

Constants

View Source 
const (
	StatusCodeCapabilities             = 100
	StatusCodeLog                      = 101
	StatusCodeStatus                   = 102
	StatusCodeRedirect                 = 103
	StatusCodeWarning                  = 104
	StatusCodeURIStart                 = 200
	StatusCodeURIDone                  = 201
	StatusCodeAuxRequest               = 351
	StatusCodeURIFailure               = 400
	StatusCodeGeneralFailure           = 401
	StatusCodeAuthorizationRequired    = 402
	StatusCodeMediaFailure             = 403
	StatusCodeURIAcquire               = 600
	StatusCodeConfiguration            = 601
	StatusCodeAuthorizationCredentials = 602
	StatusCodeMediaChanged             = 603
)

Variables

View Source 
var (
	ErrDestinationNotPointer = errors.New("destination for value is not a pointer")
	ErrDestinationNotStruct  = errors.New("destination for value is not a struct")
	ErrDestinationIsNil      = errors.New("destination for value is nil")

	ErrSourceNotPointer = errors.New("source for value is not a pointer")
	ErrSourceNotStruct  = errors.New("source for value is not a struct")
	ErrSourceIsNil      = errors.New("source for value is nil")

	ErrMessageScannerNoData = errors.New("message scanner has not data")

	ErrMessageHeaderNotFound  = errors.New("message header not found")
	ErrMessageHeaderMalformed = errors.New("message header malformed")

	ErrFieldEntryInvalid = errors.New("header field entry is invalid")
	ErrFieldsEmpty       = errors.New("header fields are empty")

	ErrInvalidConfigurationItem = errors.New("configuration item is invalid")

	ErrEmptyInformationalMessage = errors.New("informational message is empty")

	ErrNotImplemented = errors.New("not implemented")
)

Functions

func CanonicalFieldsKey 

func CanonicalFieldsKey(key string) string

CanonicalFieldsKey returns the canonical format of the field key. The canonicalization conversts the first letter of each word in the string to upper case. All other letters are lowercased. For example, the canonical key for "send-config" is "Send-Config". If the key contains a space or invalid bytes, it is returned without modifications.

func GetFieldName 

func GetFieldName(field reflect.StructField) string

GetFieldName returns either the name of the given struct field or the value of the struct tag "transport"

func ScanMessages 

func ScanMessages(data []byte, atEOF bool) (advance int, token []byte, err error)

ScanMessages is a SplitFunc function for bufio.Scanner that returns the entire set of data starting with a status code and ending in a double newline.

Unlike bufio.ScanLines, this function will error if the last line of input is not a newline.

func StatusText 

func StatusText(code int) string

StatusText returns a text for the APT status code. It returns the empty string if the code is unknown.

func UnmarshalFields 

func UnmarshalFields(fields Fields, destination any) error

UnmarshalFields unmarshals the provided Fields object into the destination provided.

MarshalFields traverses the value of the provided object recursively. If an encountered value implements the FieldUnmarshaler interface (and is not nil or empty), UnmarshalFields calls its UnmarshalFields method to deserialize the Fields object.

Some field representations can be represented with a simple string or map, and thus these two types are permitted without being passed by pointer. Slices are never permitted by this function, unless they implement the FieldMarshaler interface.

The encoding of each field in a struct can be cutomized by the format string stored under the "transport" key in the struct field's tag. The format string gives the name of the field. This is intended to allow for aliasing field names as well as field names that conflict with variable naming requirements in Go.

func UnmarshalMessage 

func UnmarshalMessage(message *Message, destination any) error

Types

type AuthorizationCredentials 

type AuthorizationCredentials struct {
	Password string
	User     string
	Site     string
}

AuthorizationCredentials (status code 602) is sent in response to a 402 Authorization Required.

When received, it will contain the entered username and password.

type AuthorizationRequired 

type AuthorizationRequired struct {
	Site string
}

AuthorizationRequired (status code 402) is sent to APT to request credentials.

The transport method requires a User and Password pair to continue. After sending this message, a Method will expect APT to send a 602 Authorization Credentials message with the required information. It is possible for a transport method to send this message to APT multiple times (both for multiple credential steps as well as retries and timeouts)

type AuxRequest 

type AuxRequest struct {
	MaximumSize int64  `transport:"MaximumSize"`
	ShortDesc   string `transport:"Aux-ShortDesc"`
	Description string `transport:"Aux-Description"`
	URI         string `transport:"Aux-URI"`
}

AuxRequest (status code 351) indicates a request for an auxiliary file to be downloaded by the acquire system (via another method) and made available for the requesting method.

The requester will get a 600 URI Acquire with the URI it requested and the filename will either be an existing file if the request was a success or if the acquire failed for the some reason the file will not exist.

type Capabilities 

type Capabilities struct {
	SingleInstance bool `transport:"Single-Instance"`
	NeedsCleanup   bool `transport:"Needs-Cleanup"`
	Pipeline       bool
	SendURIEncoded bool `transport:"Send-URI-Encoded"`
	SendConfig     bool `transport:"Send-Config"`
	Removable      bool
	AuxRequests    bool
	PreScan        string `transport:"Pre-Scan"`
	Version        string
}

Capabilities (status code 100) messages are used to inform APT of a transport method's feature set.

Displays the capabilities of the transport method. Methods should set the pipeline bit if their underlying protocol supports pipelining. The only known built-in method that does support pipelining is http(s).

type Configuration 

type Configuration map[string]string

Configuration (status code 601) indicates the configuration was sent to the method.

APT will send the configuration 'space' to the transport method. A series of Config-Items fields are sent, each containing an entry from the APT configuration. Each item is placed verbatim into the map directly. No canonicalization occurs, as these values come in directly from the APT configuration space, and not as field headers.

While files found in `/etc/apt.conf.d/*.conf` will sometimes have several settings: 

APT {
  Install-Recommends "false";
  Get {
    Assume-Yes "true";
  };
}

Apt will then convert these into their full namespaced form before the transport receives them: 

APT::Install-Recommends "false";
APT::Get::Assume-Yes "true";

This is the format that the transport method will receive, and is what users should expect to look for. No parsing is done for the user as the content of a setting can be, quite literally, anything.

func (Configuration) Section 

func (cfg Configuration) Section(prefix string) Configuration

Section returns a subsection of the configuration, based on the given prefix. All keys in the returned subsection will have the prefix trimmed. Multiple section lookups can be performed by combining sections together with "::" (e.g., "APT::Get") This allows taking all parts of a configuration and putting it into a smaller lookup: 

confg := cfg.Section("APT::Get")
fmt.Println(confg["Assume-Yes"])

This reduces the amount of work and lookup required when a method has its own configuration section

func (Configuration) UnmarshalFields 

func (cfg Configuration) UnmarshalFields(fields Fields) error

type FieldMarshaler 

type FieldMarshaler interface {
	//MarshalFields encodes the receiver into a Fields instance and returns the
	//result.
	MarshalFields() (Fields, error)
}

type FieldMarshalerError 

type FieldMarshalerError struct {
	Type reflect.Type
	Err  error
	// contains filtered or unexported fields
}

FieldMarshalerError is used when performing automatic reflection-based marshaling into a field.

func (*FieldMarshalerError) Error 

func (err *FieldMarshalerError) Error() string

func (*FieldMarshalerError) Unwrap 

func (err *FieldMarshalerError) Unwrap() error

type FieldType 

type FieldType int
const (
	UnknownFieldType FieldType = iota
	UnsignedFieldType
	IntegerFieldType
	BooleanFieldType
	StringFieldType
	FloatFieldType
	TimeFieldType
	URIFieldType
)

func GetFieldType 

func GetFieldType(value reflect.Value) FieldType

GetFieldType returns the FieldType for the given value.

If the FieldType returned is UnknownFieldType, the value cannot be automatically deserialized from a field value into a Go type.

type FieldUnmarshaler 

type FieldUnmarshaler interface {
	// UnmarshalFields decodes the Fields provided into the receiver.
	UnmarshalFields(Fields) error
}

FieldUnmarshaler is the interface implemented by any object that can unmarshal a Fields instance into itself.

type Fields 

type Fields map[string][]string

Fields represents the key-value pairs in a Message.

The keys should use a canonical form, as returned by transport.CanonicalFieldKey.

This type closely matches the textproto.MIMEfields and net/http.Header types.

func MarshalFields 

func MarshalFields(source any) (Fields, error)

MarshalFields returns the Fields representation of the value provided.

MarshalFields traverses the value of the provided object recursively. If an encountered value implements the FieldMarshaler interface (and is not nil or empty), MarshalFields will use it to produce the Fields object.

Some message representations can be represented with a simple string or map, and thus these two types are permitted without being passed by pointer. Slices are never permitted by this function, unless they implement FieldMarshaler.

The encoding of each field in a struct can be customized by the format string stored under the "transport" key in the struct field's tag. The format string gives the name of the field. This is intended to allow for aliasing field names as well as field names that conflict with variable naming requirements in Go.

func (Fields) Add 

func (fields Fields) Add(key, value string)

Add adds the key, value pair to the fields.

It appends to any existing values associated with key. The key is case-insensitive; it is canonicalized by transport.CanonicalFieldsKey.

func (Fields) Del 

func (fields Fields) Del(key string)

Del deletes the value associated with key. The key is case insensitive. It is canonicalized by transport.CanonicalFieldsKey.

func (Fields) Get 

func (fields Fields) Get(key string) string

Get gets the first value associated with the given key. If there are no values associated with the key, Get returns "".

The key is case insensitive; it is canonicalized by transport.CanonicalFieldsKey. To use non-canonical keys, use the Fields instance directly.

func (Fields) MarshalBinary 

func (fields Fields) MarshalBinary() ([]byte, error)

MarshalBinary turns the Fields into the correct binary representation.

This is primarily called by Fields.Write.

This function will error if the fields is empty or nil.

func (Fields) Set 

func (fields Fields) Set(key, value string)

Set sets the field entries associated with key to the single element value. It replaces any existing values associated with key. The key is case insensitive. It is canonicalized by transport.CanonicalFieldsKey. To use non-canonical keys, assign to the Fields instance directly.

func (Fields) UnmarshalBinary 

func (fields Fields) UnmarshalBinary(data []byte) error

UnmarshalBinary parses the fields fields from the provided byte slice.

This function does not perform validation for the contents of the fields fields. It also does not validate that a fields field is not empty, as this is technically allowed.

BUG(bruxisma): This function does not currently handle multi-line fields.

func (Fields) Values 

func (fields Fields) Values(key string) []string

Values returns all values associated with the given key.

It is case insensitive; it is canonicalized by transport.CanonicalFieldsKey. To use non-canonical keys, access the map directly.

The slice returned is NOT a copy.

func (Fields) Write 

func (fields Fields) Write(writer io.Writer) error

Write writes the Fields as though it were a textproto.MIMEHeader. However, it does not add the trailing newline.

This function performs the write all at once, but does allocate internally.

type GeneralFailure 

type GeneralFailure string

GeneralFailure (status code 401) indicates that some unspecific failure has occurred.

This is used when the transport method is no longer able to continue. Shortly after sending this, the transport method SHOULD terminate. It is intended to for invalid configuration options or other severe conditions.

When using the transport.Method.SendAndReceive method, this is automatically sent if the transport.Method.Handler returns an error.

func (GeneralFailure) Error 

func (failure GeneralFailure) Error() string

func (GeneralFailure) Is 

func (failure GeneralFailure) Is(target error) bool

func (GeneralFailure) MarshalFields 

func (failure GeneralFailure) MarshalFields() (Fields, error)

func (GeneralFailure) MarshalMessage 

func (failure GeneralFailure) MarshalMessage() (*Message, error)

type Handler 

type Handler interface {
	AcquireResource(*MessageWriter, *Request) error
}

A Handler responds to a URI Acquire message.

type HandlerFunc 

type HandlerFunc func(*MessageWriter, *Request) error

func (HandlerFunc) AcquireResource 

func (handler HandlerFunc) AcquireResource(writer *MessageWriter, request *Request) error

type Log 

type Log string

Log (status code 101) messages are used when debugging is enabled.

These messages are ONLY used for debugging.

func (Log) MarshalFields 

func (log Log) MarshalFields() (Fields, error)

func (Log) MarshalMessage 

func (log Log) MarshalMessage() (*Message, error)

type MediaChanged 

type MediaChanged struct {
	Media string
	Fail  string
}

MediaChanged (status code 603) is sent in response to a 403 Media Failure message.

This message is sent in response to a 403 Media Failure message. It indicates the user has changed media and it is safe to proceed.

type MediaFailure 

type MediaFailure struct {
	Media string
	Drive string
}

MediaFailure (status code 403) indicates new media must be inserted.

This method is executed primarily when the transport method deals with multiple media to install packages. This can include resources like disks that mounted, FUSE mounts, or any other transparent "media".

type Message 

type Message struct {
	StatusCode int    // e.g., 100
	Summary    string // e.g., Capabilities
	Fields     Fields // e.g., {"Send-Config": "true"}
}

Message represents the low level data either sent or received via the APT transport method.

Messages are very close to the same messages found in HTTP and other text protocols that are based off of RFC822. However, no attempt is made to enforce a carriage return ("\r").

A raw Message makes no attempt to validate the StatusCode, Summary, or Fields. This is instead handled by higher level APIs within this library, and users are encouraged to use them over raw Messages.

Messages can be marshaled to and from Binary data, as they have a well-formed "wire format".

func MarshalMessage 

func MarshalMessage(value any) (*Message, error)

todo: support passing in an `error` and converting it to a Message

func (*Message) IsFailure 

func (message *Message) IsFailure() bool

func (*Message) IsInformational 

func (message *Message) IsInformational() bool

func (*Message) IsResponse 

func (message *Message) IsResponse() bool

func (*Message) IsSuccessful 

func (message *Message) IsSuccessful() bool

func (*Message) MarshalBinary 

func (message *Message) MarshalBinary() ([]byte, error)

MarshalBinary serializes the receiving Message into a byte slice.

func (*Message) UnmarshalBinary 

func (message *Message) UnmarshalBinary(data []byte) error

UnmarshalBinary deserializes the receiving byte slice into a Message.

This function does not perform any validation on the message's contents, only that it meets the correct layout.

This means it is possible to receive correctly formatted but ultimately invalid messages.

This function is dependent on the behavior of Unmarshalling a Fields object.

type MessageMarshaler 

type MessageMarshaler interface {
	// MarshalMessage encodes the receiver into a Message and returns the result.
	MarshalMessage() (*Message, error)
}

type MessageMarshalerError 

type MessageMarshalerError struct {
	Type reflect.Type
	Err  error
	// contains filtered or unexported fields
}

MessageMarshalerError is used when performing automatic reflection-based marhsalling into a message.

func (*MessageMarshalerError) Error 

func (err *MessageMarshalerError) Error() string

func (*MessageMarshalerError) Unwrap 

func (err *MessageMarshalerError) Unwrap() error

type MessageScanner 

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

MessageScanner provides a convenient interface for reading messages from any io.Reader. Successive calls to Scan() will step through the the reader, skipping the empty newline between messages. The internal function used to split these messages up is ScanMessages. This internal split function cannot be overridden at this time.

Scanning stops unrecoverably at EOF, the first I/O error encountered, or when data is too large to fit into the internal buffer. Much like bufio.Scanner, if more control over scanning messages is required (however unlikely), it is recommended that users utilize a bufio.Reader in conjunction with ScanMessages.

NOTE(bruxisma): It is unlikely that more granular control scanning is required by users, as the input for messages comes from os.Stdin. Additionally, rescanning os.Stdin is out of scope for this library, and indicates that "something wacky" has gone awry with the APT transport method protocol.

func NewMessageScanner 

func NewMessageScanner(reader io.Reader) *MessageScanner

NewMessageScanner will initialize a bufio.Scanner internally, call bufio.Scanner.Split with ScanMessages and then return. This ensures that the order of operations does not result in a panic when scanning.

func (*MessageScanner) Err 

func (scanner *MessageScanner) Err() error

Err returns the first non-EOF error that was encountered by the MessageScanner.

func (*MessageScanner) Message 

func (scanner *MessageScanner) Message() (*Message, error)

Message returns the most recent Message when scanning, or an error. The value returned will NOT be overwritten by subsequent calls to Scan. However, this is done at the cost of an allocation, as the internal bytes buffer is unmarshalled into the returned Message.

If MessageScanner.Scan has not been called, an error is returned.

func (*MessageScanner) Scan 

func (scanner *MessageScanner) Scan() bool

Scan advances the MessageScanner to the next message, which will then be available through the Message method. It returns false when the scan has stopped, either by reaching the end of the input or an error. After Scan returns false, the Err method will return any error that occurred during scanning, unless it was io.EOF.

NOTE(bruxisma): Scan panics if the split function returns empty slices without advancing the input. This is a side effect of uing bufio.Scanner internally.

Example 
// note the extra newline at the end of the message
data := heredoc.Doc(`
    100 Capabilities
    Single-Instance: true
    Needs-Cleanup: true
    Send-URI-Encoded: true
    Removable: false
    Send-Config: true

  `)
scanner := NewMessageScanner(strings.NewReader(data))
for scanner.Scan() {
	_, err := scanner.Message()
	if err != nil {
		log.Fatal(err)
	}
	// do something with the message
}

Output:

type MessageUnmarshaler 

type MessageUnmarshaler interface {
	// UnmarshalMessages decodes a Message into the receiver.
	UnmarshalMessage(*Message) error
}

type MessageWriter 

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

MessageWriter is used to send additional messages back to the consumer.

These messages are sent immediately once called, and can result in a handler being cancelled if an error is sent.

func NewMessageWriter 

func NewMessageWriter(writer io.Writer) *MessageWriter

func (*MessageWriter) Configuration 

func (writer *MessageWriter) Configuration() Configuration

Configuration returns a copy of configuration sent to the Method from APT.

func (*MessageWriter) Debug 

func (writer *MessageWriter) Debug(message string)

Debug is an alias for MessageWriter.Log

func (*MessageWriter) Debugf 

func (writer *MessageWriter) Debugf(format string, args ...any)

Debugf is an alias for MessageWriter.Logf

func (*MessageWriter) Log 

func (writer *MessageWriter) Log(message string)

Writes a transport.Log message to the communication stream.

func (*MessageWriter) Logf 

func (writer *MessageWriter) Logf(format string, args ...any)

Logf writes a transport.Log message to the communication stream, using the provided format specifier.

func (*MessageWriter) Print 

func (writer *MessageWriter) Print(message string)

Print is an alias for MessageWriter.Log

func (*MessageWriter) Printf 

func (writer *MessageWriter) Printf(format string, args ...any)

Printf is an alias for MessageWriter.Logf

func (*MessageWriter) Status 

func (writer *MessageWriter) Status(message string)

Writes a transport.Status message to the communication stream.

func (*MessageWriter) Statusf 

func (writer *MessageWriter) Statusf(format string, args ...any)

Statusf writes a transport.Status message to the communication stream, using the provided format specifier.

func (*MessageWriter) Warning 

func (writer *MessageWriter) Warning(message string)

Writes a transport.Warning message to the communication stream.

func (*MessageWriter) Warningf 

func (writer *MessageWriter) Warningf(format string, args ...any)

Warningf writes a transport.Warning message to the communication stream, using the provided format specifier.

func (*MessageWriter) Write 

func (writer *MessageWriter) Write(message *Message) error

Write attempts to marshal the provided message into a binary wire format, and then write it all at once to the underlying writer.

type Method 

type Method struct {
	Handler Handler
	// contains filtered or unexported fields
}

func NewMethod 

func NewMethod(ctx context.Context, version string, options ...MethodOption) (*Method, error)

func (*Method) SendAndReceive 

func (method *Method) SendAndReceive() error

SendAndReceive is the Method's main loop, and can be considered equivalent to net/http.Server.ListenAndServe.

This function will perform the initial handshake, launch an event queue, and then block on stdin, until it is closed, cannot be read from any longer, or the context is cancelled.

NOTE(bruxisma): The use of context.Context is currently barebones, and will most likely improve over time.

type MethodOption 

type MethodOption func(*Method) error

func WithCapabilities 

func WithCapabilities(capabilities Capabilities) MethodOption

WithCapabilities sets the transport.Capabilities of the Method.

NOTE: transport.Capabilities.SendConfig is always over-written to be true, and transport.Capabilities.Version is always over-written with the version that was passed in to the call to NewMethod.

func WithHandler 

func WithHandler(handler Handler) MethodOption

WithHandler sets the transport.Method.Handler

func WithHandlerFunction 

func WithHandlerFunction(function func(*MessageWriter, *Request) error) MethodOption

WithHandlerFunction sets the transport.Method.Handler to the provided function

func WithStream 

func WithStream(stream *Stream) MethodOption

WithStream allows overwriting the default IO stream for the Method.

type Redirect 

type Redirect struct {
	URI        *url.URL
	NewURI     *url.URL `transport:"New-URI"`
	AltURIs    *url.URL `transport:"Alt-URIs"`
	UsedMirror bool     `transport:"Used-Mirror"`
}

Redirect (status code 103) is currently undocumented by the APT transport method protocol.

type Request 

type Request struct {
	Modified time.Time `transport:"Last-Modified"`
	Source   *url.URL  `transport:"URI"`
	Target   string    `transport:"Filename"`
}

type Status 

type Status string

Status (status code 102) messages are used regarding transportation progress.

Status gives a progress indication for the transportation method. It can be used to show pre-transfer status for internet enabled transport methods.

func (Status) MarshalFields 

func (status Status) MarshalFields() (Fields, error)

func (Status) MarshalMessage 

func (status Status) MarshalMessage() (*Message, error)

type Stream 

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

Stream acts as an io.ReadWriteCloser, that operates on stdin (for reading), and stdout (for writing). There are several ways to open a stream, the simplest of which is to use the default os.Stdin and os.Stdout pointers.

This type is only thread-safe if the underlying io.Writer and io.Reader are thread-safe.

Users must call close on the underlying io.Writer or io.Reader.

func NewStream 

func NewStream() *Stream

NewStream initializes a new Stream with os.Stdin and os.Stdout.

This function does *not* check if the files are actually open and valid.

func NewStreamWith 

func NewStreamWith(input io.Reader, output io.Writer) *Stream

NewStreamWith initializes a new Stream with the given input and output.

This function does *not* check if the provided interfaces are not nil.

func (*Stream) Close 

func (stream *Stream) Close() error

Close is a no-op, as we let the operating system take care of closing os.Stdout and os.Stdin. We also make no effort to know whether we can close the underlying io.Writer and io.Reader.

func (*Stream) Read 

func (stream *Stream) Read(data []byte) (int, error)

func (*Stream) Write 

func (stream *Stream) Write(data []byte) (int, error)

type URIAcquire 

type URIAcquire struct {
	LastModified *time.Time `transport:"Last-Modified"`
	URI          string
	Filename     string
}

URIAcquire (status code 600) indicates that APT is requesting a new URI be added to the acquire list.

The deserialized transport.URIAcquire.LastModified field has the time stamp of the current cache file if applicable. transport.URIAcquire.Filename is the name of the file that the acquired URI should be written to. It is safe for the method to assume it has correct write permissions.

NOTE(bruxisma): This message is effectively "repeated" by the transport.Request type passed to Method's Handler.

type URIDone 

type URIDone struct {
	URI          string
	LastModified string `transport:"Last-Modified"`
	IMSHit       string `transport:"IMS-Hit"`
	Filename     string
	MD5Hash      string `transport:"MD5-Hash"`
	Size         int64
}

URIDone (status code 201) indicates that a URI has completed transferrence.

It is possible to specify a 201 URI Done without a URI Start which would mean no data was transferred, but the file is now available. A transport.URIDone.Filename field is specified when the URI is directly available in the local pathname space. APT will either directly use that file or copy it into another location. It is possible to return fields prefix with Alt- to indicate that another possible for the URI has been found in the local pathname space. This is done if a decompressed version of a gunzip file is found.

BUG(bruxisma): We do not currently support the Alt- prefixed fields.

type URIFailure 

type URIFailure struct {
	URI     string
	Message string
}

URIFailure (status code 400) indicates the URI is not retrievable from this source.

Indicates a fatal URI failure. As with 201 URI Done, 200 URI start is not required to precede this message.

func (*URIFailure) Error 

func (failure *URIFailure) Error() string

func (*URIFailure) Is 

func (failure *URIFailure) Is(target error) bool

type URIStart 

type URIStart struct {
	LastModified string `transport:"Last-Modified"`
	ResumePoint  string `transport:"Resume-Point"`
	URI          string
	Size         int64
}

URIStart (status code 200) indicates that the resource located at the given URI is to begin transferring.

The URI is specified along with stats regarding the file itself.

type Warning 

type Warning string

Warning (status code 104) messages are used to indicate a warning.

Warnings can be used to alert users to possibly problematic conditions.

func (Warning) MarshalFields 

func (warning Warning) MarshalFields() (Fields, error)

func (Warning) MarshalMessage 

func (warning Warning) MarshalMessage() (*Message, error)

Notes

Bugs

  • This function does not currently handle multi-line fields.

  • We do not currently support the Alt- prefixed fields.

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.