shared

package
v0.4.3 Latest Latest
Warning

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

 Go to latest Published: Apr 29, 2016 License: Apache-2.0 Imports: 23 Imported by: 0

Documentation

Index

Constants

View Source 
const (
	// BufferedReaderFlagDelimiter enables reading for a delimiter. This flag is
	// ignored if an MLE flag is set.
	BufferedReaderFlagDelimiter = BufferedReaderFlags(0)

	// BufferedReaderFlagMLE enables reading if length encoded messages.
	// Runlength is read as ASCII (to uint64) until the first byte (ASCII char)
	// of the delimiter string.
	// Only one MLE flag is supported at a time.
	BufferedReaderFlagMLE = BufferedReaderFlags(1)

	// BufferedReaderFlagMLE8 enables reading if length encoded messages.
	// Runlength is read as binary (to uint8).
	// Only one MLE flag is supported at a time.
	BufferedReaderFlagMLE8 = BufferedReaderFlags(2)

	// BufferedReaderFlagMLE16 enables reading if length encoded messages.
	// Runlength is read as binary (to uint16).
	// Only one MLE flag is supported at a time.
	BufferedReaderFlagMLE16 = BufferedReaderFlags(3)

	// BufferedReaderFlagMLE32 enables reading if length encoded messages.
	// Runlength is read as binary (to uint32).
	// Only one MLE flag is supported at a time.
	BufferedReaderFlagMLE32 = BufferedReaderFlags(4)

	// BufferedReaderFlagMLE64 enables reading if length encoded messages.
	// Runlength is read as binary (to uint64).
	// Only one MLE flag is supported at a time.
	BufferedReaderFlagMLE64 = BufferedReaderFlags(5)

	// BufferedReaderFlagMLEFixed enables reading messages with a fixed length.
	// Only one MLE flag is supported at a time.
	BufferedReaderFlagMLEFixed = BufferedReaderFlags(6)

	// BufferedReaderFlagMaskMLE is a bitmask to mask out everything but MLE flags
	BufferedReaderFlagMaskMLE = BufferedReaderFlags(7)

	// BufferedReaderFlagBigEndian sets binary reading to big endian encoding.
	BufferedReaderFlagBigEndian = BufferedReaderFlags(8)

	// BufferedReaderFlagEverything will keep MLE and/or delimiters when
	// building a message.
	BufferedReaderFlagEverything = BufferedReaderFlags(16)
)
View Source 
const (
	// MetricProcessStart is the metric name storing the time when this process
	// has been started.
	MetricProcessStart = "ProcessStart"
	// MetricGoRoutines is the metric name storing the number of active go
	// routines.
	MetricGoRoutines = "GoRoutines"
	// MetricGoVersion holds the go version as Major*10000+Minor*100+Patch
	MetricGoVersion = "GoVersion"
	// MetricMemoryAllocated holds the currently active memory in bytes
	MetricMemoryAllocated = "GoMemoryAllocated"
	// MetricMemoryNumObjects holds the total number of allocated heap objects
	MetricMemoryNumObjects = "GoMemoryNumObjects"
	// MetricMemoryGCEnabled holds 1 or 0 depending on the state of garbage collection
	MetricMemoryGCEnabled = "GoMemoryGCEnabled"
)
View Source 
const (
	// ParserFlagContinue continues parsing at the position of the match.
	// By default the matched token will be skipped. This flag prevents the
	// default behavior. In addition to that the parser will add the parsed
	// token to the value of the next match.
	ParserFlagContinue = ParserFlag(1 << iota)

	// ParserFlagAppend causes the parser to keep the current value for the next
	// match. By default a value will be restarted after each match. This flag
	// prevents the default behavior.
	ParserFlagAppend = ParserFlag(1 << iota)

	// ParserFlagInclude includes the matched token in the read value.
	// By default the value does not contain the matched token.
	ParserFlagInclude = ParserFlag(1 << iota)

	// ParserFlagPush pushes the current state on the stack before making the
	// transition.
	ParserFlagPush = ParserFlag(1 << iota)

	// ParserFlagPop pops the stack and uses the returned state as the next
	// state. If no state is on the stack the regular next state is used.
	ParserFlagPop = ParserFlag(1 << iota)

	// ParserStateStop defines a state that causes the parsing to stop when
	// transitioned into.
	ParserStateStop = ParserStateID(0xFFFFFFFF)
)
View Source 
const (
	// SpinPrioritySuspend should be used for spinning loops that are expected
	// to wait for very long periods of time. The loop will sleep for 1 second
	// after each iteration.
	SpinPrioritySuspend = SpinPriority(1)

	// SpinPriorityLow should be used for spinning loops that are expected to
	// spin for a rather long time before being able to exit.
	// After 100 loops the caller waits for 200 milliseconds.
	SpinPriorityLow = SpinPriority(100)

	// SpinPriorityMedium should be used for spinning loops that are expected to
	// spin for a short amount of time before being able to exit.
	// After 500 loops the caller waits for 100 milliseconds.
	SpinPriorityMedium = SpinPriority(500)

	// SpinPriorityHigh should be used for spinning loops that are expected to
	// almost never spin.
	// After 1000 loops the caller waits for 10 milliseconds.
	SpinPriorityHigh = SpinPriority(1000)

	// SpinPriorityRealtime should be used for loops that should never wait and
	// always spin. This priority will increase CPU load
	SpinPriorityRealtime = SpinPriority(0xFFFFFFFF)

	// SpinTimeSuspend holds the sleep time for SpinPrioritySuspend
	SpinTimeSuspend = time.Second
	// SpinTimeLow holds the sleep time for SpinPriorityLow
	SpinTimeLow = 200 * time.Millisecond
	// SpinTimeMedium holds the sleep time for SpinPriorityMedium
	SpinTimeMedium = 100 * time.Millisecond
	// SpinTimeHigh holds the sleep time for SpinPriorityHigh
	SpinTimeHigh = 10 * time.Millisecond
)

Variables

View Source 
var BufferDataInvalid = bufferError("Invalid data")

BufferDataInvalid is returned when a parsing encounters an error

View Source 
var Metric = metrics{new(sync.RWMutex), make(map[string]*int64)}

Metric allows any part of gollum to store and/or modify metric values by name.

View Source 
var ProcessStartTime time.Time

ProcessStartTime stores the time this process has started. This value is also stored in the metric MetricProcessStart

View Source 
var TypeRegistry = typeRegistry{
	// contains filtered or unexported fields
}

TypeRegistry is the global typeRegistry instance. Use this instance to register plugins.

Functions

func Btoi 

func Btoi(buffer []byte) (uint64, int)

Btoi is a fast byte buffer to unsigned int parser that reads until the first non-number character is found. It returns the number value as well as the length of the number string encountered. If a number could not be parsed the returned length will be 0

func DontPanic added in v0.4.0 

func DontPanic(callback func())

DontPanic can be used instead of RecoverShutdown when using a function without any parameters. E.g. go DontPanic(myFunction)

func EscapeJSON added in v0.4.2 

func EscapeJSON(text string) string

EscapeJSON replaces occurrences of \ and " with escaped versions.

func GetMissingMethods 

func GetMissingMethods(objType reflect.Type, ifaceType reflect.Type) (float32, []interface{})

GetMissingMethods checks if a given object implements all methods of a given interface. It returns the interface coverage [0..1] as well as an array of error messages. If the interface is correctly implemented the coverage is 1 and the error message array is empty.

func IndexN added in v0.4.0 

func IndexN(s, sep string, n int) int

IndexN returns the nth occurrence of sep in s or -1 if there is no nth occurrence of sep.

func IsDisconnectedError added in v0.4.0 

func IsDisconnectedError(err error) bool

IsDisconnectedError returns true if the given error is related to a disconnected socket.

func ItoLen 

func ItoLen(number uint64) int

ItoLen returns the length of an unsingned integer when converted to a string

func Itob 

func Itob(number uint64, buffer []byte) error

Itob writes an unsigned integer to the start of a given byte buffer.

func Itobe 

func Itobe(number uint64, buffer []byte) error

Itobe writes an unsigned integer to the end of a given byte buffer.

func LastIndexN added in v0.4.0 

func LastIndexN(s, sep string, n int) int

LastIndexN returns the nth occurrence of sep in s or -1 if there is no nth occurrence of sep. Searching is going from the end of the string to the start.

func ListFilesByDateMatching added in v0.4.0 

func ListFilesByDateMatching(directory string, pattern string) ([]os.FileInfo, error)

ListFilesByDateMatching gets all files from a directory that match a given regular expression pattern and orders them by modification date (ascending). Directories and symlinks are excluded from the returned list.

func Max3I added in v0.4.0 

func Max3I(a, b, c int) int

Max3I returns the maximum out of three integers

func MaxI added in v0.4.0 

func MaxI(a, b int) int

MaxI returns the maximum out of two integers

func Min3I added in v0.4.0 

func Min3I(a, b, c int) int

Min3I returns the minimum out of three integers

func MinI added in v0.4.0 

func MinI(a, b int) int

MinI returns the minimum out of two integers

func ParseAddress 

func ParseAddress(addressString string) (address, protocol string)

ParseAddress takes an address and tries to extract the protocol from it. Protocols may be prepended by the "protocol://" notation. If no protocol is given, "tcp" is assumed. The first parameter returned is the address, the second denotes the protocol. The protocol is allways returned as lowercase string.

func RecoverShutdown 

func RecoverShutdown()

RecoverShutdown will trigger a shutdown via interrupt if a panic was issued. Typically used as "defer RecoverShutdown()".

func SplitAddress added in v0.4.1 

func SplitAddress(addressString string, defaultProtocol string) (protocol, host, port string, err error)

SplitAddress splits an address of the form "protocol://host:port" into its components. If no protocol is given, the default protocol is used. This function uses net.SplitHostPort.

func SplitPath added in v0.4.0 

func SplitPath(filePath string) (dir string, base string, ext string)

SplitPath separates a file path into directory, filename (without extension) and file extension (with dot). If no directory could be derived "." is returned as a directory. If no file extension could be derived "" is returned as a file extension.

func Unescape 

func Unescape(text string) string

Unescape replaces occurrences of \\n, \\r and \\t with real escape codes.

Types

type BufferReadCallback 

type BufferReadCallback func(msg []byte, sequence uint64)

BufferReadCallback defines the function signature for callbacks passed to ReadAll.

type BufferedReader 

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

BufferedReader is a helper struct to read from any io.Reader into a byte slice. The data can arrive "in pieces" and will be assembled. A data "piece" is considered complete if a delimiter or a certain runlength has been reached. The latter has to be enabled by flag and will disable the default behavior, which is looking for a delimiter string. In addition to that every data "piece" will receive an incrementing sequence number.

func NewBufferedReader 

func NewBufferedReader(bufferSize int, flags BufferedReaderFlags, offsetOrLength int, delimiter string) *BufferedReader

NewBufferedReader creates a new buffered reader that reads messages from a continuous stream of bytes. Messages can be separated from the stream by using common methods such as fixed size, encoded message length or delimiter string. The internal buffer is grown statically (by its original size) if necessary. bufferSize defines the initial size / grow size of the buffer flags configures the parsing method offsetOrLength sets either the runlength offset or fixed message size delimiter defines the delimiter used for textual message parsing

func (*BufferedReader) ReadAll 

func (buffer *BufferedReader) ReadAll(reader io.Reader, callback BufferReadCallback) error

ReadAll calls ReadOne as long as there are messages in the stream. Messages will be send to the given write callback. If callback is nil, data will be read and discarded.

func (*BufferedReader) ReadOne 

func (buffer *BufferedReader) ReadOne(reader io.Reader) (data []byte, seq uint64, more bool, err error)

ReadOne reads the next message from the given stream (if possible) and generates a sequence number for this message. The more return parameter is set to true if there are still messages or parts of messages in the stream. Data and seq is only set if a complete message could be parsed. Errors are returned if reading from the stream failed or the parser encountered an error.

func (*BufferedReader) Reset 

func (buffer *BufferedReader) Reset(sequence uint64)

Reset clears the buffer by resetting its internal state

type BufferedReaderFlags 

type BufferedReaderFlags byte

BufferedReaderFlags is an enum to configure a buffered reader

type ByteStream 

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

ByteStream is a more lightweight variant of bytes.Buffer. The managed byte array is increased to the exact required size and never shrinks. Writing moves an internal offset (for appends) but reading always starts at offset 0.

func NewByteStream 

func NewByteStream(capacity int) ByteStream

NewByteStream creates a new byte stream of the desired capacity

func NewByteStreamFrom 

func NewByteStreamFrom(data []byte) ByteStream

NewByteStreamFrom creates a new byte stream that starts with the given byte array.

func (ByteStream) Bytes 

func (stream ByteStream) Bytes() []byte

Bytes returns a slice of the underlying byte array containing all written data up to this point.

func (ByteStream) Cap 

func (stream ByteStream) Cap() int

Cap returns the capacity of the underlying array. This is equal to cap(stream.Bytes()).

func (ByteStream) Len 

func (stream ByteStream) Len() int

Len returns the length of the underlying array. This is equal to len(stream.Bytes()).

func (*ByteStream) Read 

func (stream *ByteStream) Read(target []byte) (int, error)

Read implements the io.Reader interface. The underlying array is copied to target and io.EOF is returned once no data is left to be read. Please note that ResetRead can rewind the internal read index for this operation.

func (*ByteStream) Reset 

func (stream *ByteStream) Reset()

Reset sets the internal write offset to 0 and calls ResetRead

func (*ByteStream) ResetRead added in v0.4.1 

func (stream *ByteStream) ResetRead()

ResetRead sets the internal read offset to 0. The read offset is only used within the Read function to assure io.Reader compatibility

func (*ByteStream) SetCapacity 

func (stream *ByteStream) SetCapacity(capacity int)

SetCapacity assures that capacity bytes are available in the buffer, growing the managed byte array if needed. The buffer will grow by at least 64 bytes if growing is required.

func (ByteStream) String 

func (stream ByteStream) String() string

String returns a string of the underlying byte array containing all written data up to this point.

func (*ByteStream) Write 

func (stream *ByteStream) Write(source []byte) (int, error)

Write implements the io.Writer interface. This function assures that the capacity of the underlying byte array is enough to store the incoming amount of data. Subsequent writes will allways append to the end of the stream until Reset() is called.

func (*ByteStream) WriteByte 

func (stream *ByteStream) WriteByte(source byte) error

WriteByte writes a single byte to the stream. Capacity will be ensured.

func (*ByteStream) WriteString 

func (stream *ByteStream) WriteString(source string) (int, error)

WriteString is a convenience wrapper for Write([]byte(source))

type Expect 

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

Expect is a helper construct for unittesting

func NewExpect 

func NewExpect(t *testing.T) Expect

NewExpect creates an Expect helper struct with scope set to the name of the calling function.

func (Expect) Equal 

func (e Expect) Equal(val1, val2 interface{}) bool

Equal does a deep equality check on both values and returns true if that test yielded true (val1 == val2)

func (Expect) False 

func (e Expect) False(val bool) bool

False tests if the given value is false

func (Expect) Geq 

func (e Expect) Geq(val1, val2 interface{}) bool

Geq does a numeric greater or equal check on both values and returns true if that test yielded true (val1 >= val2)

func (Expect) Greater 

func (e Expect) Greater(val1, val2 interface{}) bool

Greater does a numeric greater than check on both values and returns true if that test yielded true (val1 > val2)

func (Expect) Leq 

func (e Expect) Leq(val1, val2 interface{}) bool

Leq does a numeric less or euqal check on both values and returns true if that test yielded true (val1 <= val2)

func (Expect) Less 

func (e Expect) Less(val1, val2 interface{}) bool

Less does a numeric less than check on both values and returns true if that test yielded true (val1 < val2)

func (Expect) MapEqual 

func (e Expect) MapEqual(data interface{}, key interface{}, value interface{}) bool

MapEqual returns true if MapSet equals true for the given key and Equal returns true for the returned and given value.

func (Expect) MapGeq 

func (e Expect) MapGeq(data interface{}, key interface{}, value interface{}) bool

MapGeq returns true if MapSet equals true for the given key and Geq returns true for the returned and given value.

func (Expect) MapGreater 

func (e Expect) MapGreater(data interface{}, key interface{}, value interface{}) bool

MapGreater returns true if MapSet equals true for the given key and Greater returns true for the returned and given value.

func (Expect) MapLeq 

func (e Expect) MapLeq(data interface{}, key interface{}, value interface{}) bool

MapLeq returns true if MapSet equals true for the given key and Leq returns true for the returned and given value.

func (Expect) MapLess 

func (e Expect) MapLess(data interface{}, key interface{}, value interface{}) bool

MapLess returns true if MapSet equals true for the given key and Less returns true for the returned and given value.

func (Expect) MapNeq 

func (e Expect) MapNeq(data interface{}, key interface{}, value interface{}) bool

MapNeq returns true if MapSet equals true for the given key and Neq returns true for the returned and given value.

func (Expect) MapNotSet 

func (e Expect) MapNotSet(data interface{}, key interface{}) bool

MapNotSet returns true if the key is not set in the given array

func (Expect) MapSet 

func (e Expect) MapSet(data interface{}, key interface{}) bool

MapSet returns true if the key is set in the given array

func (Expect) Neq 

func (e Expect) Neq(val1, val2 interface{}) bool

Neq does a deep equality check on both values and returns true if that test yielded false (val1 != val2)

func (Expect) Nil 

func (e Expect) Nil(val interface{}) bool

Nil tests if the given value is nil

func (Expect) NoError 

func (e Expect) NoError(err error) bool

NoError tests if the given error is nil. If it is not the error will be logged.

func (Expect) NonBlocking added in v0.4.1 

func (e Expect) NonBlocking(t time.Duration, routine func()) bool

NonBlocking waits for the given timeout for the routine to return. If timed out it is an error.

func (Expect) NotExecuted added in v0.4.0 

func (e Expect) NotExecuted()

NotExecuted always reports an error when called

func (Expect) NotNil 

func (e Expect) NotNil(val interface{}) bool

NotNil tests if the given value is not nil

func (Expect) True 

func (e Expect) True(val bool) bool

True tests if the given value is true

type FilesByDate added in v0.4.0 

type FilesByDate []os.FileInfo

FilesByDate implements the Sort interface by Date for os.FileInfo arrays

func (FilesByDate) Len added in v0.4.0 

func (files FilesByDate) Len() int

Len returns the number of files in the array

func (FilesByDate) Less added in v0.4.0 

func (files FilesByDate) Less(a, b int) bool

Less compares the date of the files stored at a and b as in "[a] modified < [b] modified". If both files were created in the same second the file names are compared by using a lexicographic string compare.

func (FilesByDate) Swap added in v0.4.0 

func (files FilesByDate) Swap(a, b int)

Swap exchanges the values stored at indexes a and b

type Fuse added in v0.4.1 

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

Fuse is a local circuit breaker implementation that is ment to be used to manage the state of a given resource between different threads of execution (consumer/producer). If the resource is not available the fuse is "burned". Components may now wait on that fuse and are woken as soon as the resource becomes available again (the fuse is "activated" again).

func NewFuse added in v0.4.1 

func NewFuse() *Fuse

NewFuse creates a new Fuse and returns it. A new fuse is always active.

func (*Fuse) Activate added in v0.4.1 

func (fuse *Fuse) Activate()

Activate sets the fuse back to the "running" state. An already active fuse cannot be activated again (call is ignored).

func (*Fuse) Burn added in v0.4.1 

func (fuse *Fuse) Burn()

Burn sets the fuse back to the "inactive" state. An already burned fuse cannot be burned again (call is ignored).

func (Fuse) IsBurned added in v0.4.1 

func (fuse Fuse) IsBurned() bool

IsBurned returns true if the fuse in the "inactive" state

func (Fuse) Wait added in v0.4.1 

func (fuse Fuse) Wait()

Wait blocks until the fuse enters active state. Multiple go routines may wait on the same fuse.

type MarshalMap 

type MarshalMap map[string]interface{}

MarshalMap is a wrapper type to attach converter methods to maps normally returned by marshalling methods, i.e. key/value parsers. All methods that do a conversion will return an error if the value stored behind key is not of the expected type or if the key is not existing in the map.

func NewMarshalMap 

func NewMarshalMap() MarshalMap

NewMarshalMap creates a new marshal map (string -> interface{})

func (MarshalMap) Array 

func (mmap MarshalMap) Array(key string) ([]interface{}, error)

Array returns a value at key that is expected to be a []interface{}

func (MarshalMap) Bool 

func (mmap MarshalMap) Bool(key string) (bool, error)

Bool returns a value at key that is expected to be a boolean

func (MarshalMap) Float64 

func (mmap MarshalMap) Float64(key string) (float64, error)

Float64 returns a value at key that is expected to be a float64

func (MarshalMap) Int 

func (mmap MarshalMap) Int(key string) (int, error)

Int returns a value at key that is expected to be an int

func (MarshalMap) Int64 

func (mmap MarshalMap) Int64(key string) (int64, error)

Int64 returns a value at key that is expected to be an int64

func (MarshalMap) Map 

func (mmap MarshalMap) Map(key string) (map[interface{}]interface{}, error)

Map returns a value at key that is expected to be a map[interface{}]interface{}.

func (MarshalMap) MarshalMap 

func (mmap MarshalMap) MarshalMap(key string) (MarshalMap, error)

MarshalMap returns a value at key that is expected to be another MarshalMap This function supports conversion (by copy) from

  • map[interface{}]interface{}

func (MarshalMap) Path 

func (mmap MarshalMap) Path(key string) (interface{}, bool)

Path returns a value from a given value path. Fields can be accessed by their name. Nested fields can be accessed by using "/" as a separator. Arrays can be addressed using the standard array notation "[<index>]". Examples: "key" -> mmap["key"] single value "key1/key2" -> mmap["key1"]["key2"] nested map "key1[0]" -> mmap["key1"][0] nested array "key1[0]key2" -> mmap["key1"][0]["key2"] nested array, nested map

func (MarshalMap) String 

func (mmap MarshalMap) String(key string) (string, error)

Array returns a value at key that is expected to be a string

func (MarshalMap) StringArray 

func (mmap MarshalMap) StringArray(key string) ([]string, error)

StringArray returns a value at key that is expected to be a []string This function supports conversion (by copy) from

  • []interface{}

func (MarshalMap) StringArrayMap 

func (mmap MarshalMap) StringArrayMap(key string) (map[string][]string, error)

StringArrayMap returns a value at key that is expected to be a map[string][]string. This function supports conversion (by copy) from

  • map[interface{}][]interface{}
  • map[interface{}]interface{}
  • map[string]interface{}

func (MarshalMap) StringMap 

func (mmap MarshalMap) StringMap(key string) (map[string]string, error)

StringMap returns a value at key that is expected to be a map[string]string. This function supports conversion (by copy) from

  • map[interface{}]interface{}
  • map[string]interface{}

func (MarshalMap) Uint64 

func (mmap MarshalMap) Uint64(key string) (uint64, error)

Uint64 returns a value at key that is expected to be an uint64

type MetricServer 

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

MetricServer contains state information about the metric server process

func NewMetricServer 

func NewMetricServer() *MetricServer

NewMetricServer creates a new server state for a metric server

func (*MetricServer) Start 

func (server *MetricServer) Start(port int)

Start causes a metric server to listen for a specific port. If this port is accessed a JSON containing all metrics will be returned and the connection is closed.

func (*MetricServer) Stop 

func (server *MetricServer) Stop()

Stop notifies the metric server to halt.

type Mutex added in v0.4.1 

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

Mutex is a lightweight, spinner based mutex implementation, extending the standard go mutex by the possibility to query the mutexe's state. This state is very volatile but can come in handy sometimes.

func NewMutex added in v0.4.1 

func NewMutex(priority SpinPriority) *Mutex

NewMutex creates a new mutex with the given spin priority used during Lock.

func (*Mutex) IsLocked added in v0.4.1 

func (m *Mutex) IsLocked() bool

IsLocked returns the state of this mutex. The result of this function might change directly after call so it should only be used in situations where this fact is not considered problematic.

func (*Mutex) Lock added in v0.4.1 

func (m *Mutex) Lock()

Lock blocks (spins) until the lock becomes available

func (*Mutex) Unlock added in v0.4.1 

func (m *Mutex) Unlock()

Unlock unblocks one routine waiting on lock.

type ParsedFunc 

type ParsedFunc func([]byte, ParserStateID)

ParsedFunc defines a function that a token has been matched

type ParserFlag 

type ParserFlag int

ParserFlag is an enum type for flags used in parser transitions.

type ParserStateID 

type ParserStateID uint32

ParserStateID is used as an integer-based reference to a specific parser state. You can use any number (e.g. a hash) as a parser state representative.

type SpinPriority added in v0.4.0 

type SpinPriority uint32

SpinPriority is used for Spinner priority enum values

type Spinner added in v0.4.0 

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

Spinner is a helper struct for spinning loops.

func NewSpinner added in v0.4.0 

func NewSpinner(priority SpinPriority) Spinner

NewSpinner creates a new helper for spinning loops

func (*Spinner) Reset added in v0.4.0 

func (spin *Spinner) Reset()

Reset sets the internal counter back to 0

func (*Spinner) Yield added in v0.4.0 

func (spin *Spinner) Yield()

Yield should be called in spinning loops and will assure correct spin/wait/schedule behavior according to the set priority.

type StopListener 

type StopListener struct {
	*net.TCPListener
	// contains filtered or unexported fields
}

StopListener is a wrapper around net.TCPListener that allows to gracefully shut down a connection loop.

func NewStopListener 

func NewStopListener(address string) (*StopListener, error)

NewStopListener creates a new, stoppable TCP server connection. Address needs to be cmpliant to net.Listen.

func (*StopListener) Accept 

func (listen *StopListener) Accept() (net.Conn, error)

Accept is analogous to net.TCPListener.Accept but may return a connection closed error if the connection was requested to shut down.

func (*StopListener) Close 

func (listen *StopListener) Close() error

Close requests a connection close on this listener

type StopRequestError 

type StopRequestError struct{}

StopRequestError is not an error but a note that the connection was requested to be closed.

func (StopRequestError) Error 

func (err StopRequestError) Error() string

Error implements the standard error interface

type Transition 

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

Transition defines a token based state change

func NewTransition 

func NewTransition(nextState ParserStateID, flags ParserFlag, callback ParsedFunc) Transition

NewTransition creates a new transition to a given state.

type TransitionDirective 

type TransitionDirective struct {
	State     string
	Token     string
	NextState string
	Flags     ParserFlag
	Callback  ParsedFunc
}

TransitionDirective contains a transition description that can be passed to the AddDirectives functions.

func ParseTransitionDirective 

func ParseTransitionDirective(config string, callbacks map[string]ParsedFunc) (TransitionDirective, error)

ParseTransitionDirective creates a transition directive from a string. This string must be of the following format:

State:Token:NextState:Flag,Flag,...:Function

Spaces will be stripped from all fields but Token. If a fields requires a colon it has to be escaped with a backslash. Other escape characters supported are \n, \r and \t. Flag can be a set of the following flags (input will be converted to lowercase):

  • continue -> ParserFlagContinue
  • append -> ParserFlagAppend
  • include -> ParserFlagInclude
  • push -> ParserFlagPush
  • pop -> ParserFlagPop

The name passed to the function token must be in the callbacks map. If it is not the data of the token will not be written. I.e. in empty string equals "do not write" here. An empty NextState will be translated to the "Stop" state as ususal.

type TransitionParser 

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

TransitionParser defines the behavior of a parser by storing transitions from one state to another.

func NewTransitionParser 

func NewTransitionParser() TransitionParser

NewTransitionParser creates a new transition based parser

func (*TransitionParser) Add 

func (parser *TransitionParser) Add(stateName string, token string, nextStateName string, flags ParserFlag, callback ParsedFunc)

Add adds a new transition to a given parser state.

func (*TransitionParser) AddDirectives 

func (parser *TransitionParser) AddDirectives(directives []TransitionDirective)

AddDirectives is a convenience function to add multiple transitions in as a batch.

func (*TransitionParser) AddTransition 

func (parser *TransitionParser) AddTransition(stateName string, newTrans Transition, token string)

AddTransition adds a transition from a given state to the map

func (*TransitionParser) GetStateID 

func (parser *TransitionParser) GetStateID(stateName string) ParserStateID

GetStateID creates a hash from the given state name. Empty state names will be translated to ParserStateStop.

func (*TransitionParser) GetStateName 

func (parser *TransitionParser) GetStateName(id ParserStateID) string

GetStateName returns the name for the given state id or an empty string if the id could not be found.

func (*TransitionParser) Parse 

func (parser *TransitionParser) Parse(data []byte, state string) ([]byte, ParserStateID)

Parse starts parsing at a given stateID. This function returns the remaining parts of data that did not match a transition as well as the last state the parser has been set to.

func (*TransitionParser) Stop 

func (parser *TransitionParser) Stop(stateName string, token string, flags ParserFlag, callback ParsedFunc)

Stop adds a stop transition to a given parser state.

type TrieNode 

type TrieNode struct {
	PathLen int
	Payload interface{}
	// contains filtered or unexported fields
}

TrieNode represents a single node inside a trie. Each node can contain a payload which can be retrieved after a successfull match. In addition to that PathLen will contain the length of the match.

func NewTrie 

func NewTrie(data []byte, payload interface{}) *TrieNode

NewTrie creates a new root TrieNode

func (*TrieNode) Add 

func (node *TrieNode) Add(data []byte, payload interface{}) *TrieNode

Add adds a new data path to the trie. The TrieNode returned is the (new) root node so you should always reassign the root with the return value of Add.

func (*TrieNode) ForEach 

func (node *TrieNode) ForEach(callback func(*TrieNode))

ForEach applies a function to each node in the tree including and below the passed node.

func (*TrieNode) Match 

func (node *TrieNode) Match(data []byte) *TrieNode

Match compares the trie to the given data stream. Match returns true if data can be completely matched to the trie.

func (*TrieNode) MatchStart 

func (node *TrieNode) MatchStart(data []byte) *TrieNode

MatchStart compares the trie to the beginning of the given data stream. MatchStart returns true if the beginning of data can be matched to the trie.

type WaitGroup added in v0.4.0 

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

WaitGroup is a replacement for sync/waitgroup that allows the internal counter to go negative. This version allows a missed Done to be recovered but will make it a lot harder to detect missing Done or Add calls. Use only where needed.

func (*WaitGroup) Active added in v0.4.0 

func (wg *WaitGroup) Active() bool

Active returns true if the counter is > 0

func (*WaitGroup) Add added in v0.4.0 

func (wg *WaitGroup) Add(delta int)

Add increments the waitgroup counter by the given value. Delta may be negative.

func (*WaitGroup) Done added in v0.4.0 

func (wg *WaitGroup) Done()

Done is the shorthand version for Add(-1)

func (*WaitGroup) Inc added in v0.4.0 

func (wg *WaitGroup) Inc()

Inc is the shorthand version for Add(1)

func (*WaitGroup) IncWhenDone added in v0.4.0 

func (wg *WaitGroup) IncWhenDone()

IncWhenDone wait until the counter is exactly 0 and triggeres an increment if this is found to be true

func (*WaitGroup) Reset added in v0.4.0 

func (wg *WaitGroup) Reset()

Reset sets the counter to 0

func (*WaitGroup) Wait added in v0.4.0 

func (wg *WaitGroup) Wait()

Wait blocks until the counter is 0 or less.

func (*WaitGroup) WaitFor added in v0.4.0 

func (wg *WaitGroup) WaitFor(timeout time.Duration) bool

WaitFor blocks until the counter is 0 or less. If the block takes longer than the given timeout, WaitFor will return false. If duration is 0, Wait is called.

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.