Documentation ¶
Overview ¶
Package models implements basic objects used throughout the TICK stack.
Index ¶
- Constants
- Variables
- func AppendMakeKey(dst []byte, name []byte, tags Tags) []byte
- func CheckTime(t time.Time) error
- func CompareTags(a, b Tags) int
- func EnableUintSupport()
- func EscapeMeasurement(in []byte) []byte
- func EscapeStringField(in string) string
- func GetPrecisionMultiplier(precision string) int64
- func MakeKey(name []byte, tags Tags) []byte
- func ParseName(buf []byte) []byte
- func SafeCalcTime(timestamp int64, precision string) (time.Time, error)
- func UnescapeMeasurement(in []byte) []byte
- func ValidKeyToken(s string) bool
- func ValidKeyTokens(name string, tags Tags) bool
- func ValidPrecision(precision string) bool
- type FieldIterator
- type FieldType
- type Fields
- type InlineFNV64a
- type Point
- func MustNewPoint(name string, tags Tags, fields Fields, time time.Time) Point
- func NewPoint(name string, tags Tags, fields Fields, t time.Time) (Point, error)
- func NewPointFromBytes(b []byte) (Point, error)
- func ParsePoints(buf []byte) ([]Point, error)
- func ParsePointsString(buf string) ([]Point, error)
- func ParsePointsWithPrecision(buf []byte, defaultTime time.Time, precision string) ([]Point, error)
- type Points
- type Row
- type Rows
- type Statistic
- type StatisticTags
- type Tag
- type Tags
- func (a Tags) AppendHashKey(dst []byte) []byte
- func (a Tags) Clone() Tags
- func (a *Tags) Delete(key []byte)
- func (a Tags) Equal(other Tags) bool
- func (a Tags) Get(key []byte) []byte
- func (a Tags) GetString(key string) string
- func (a Tags) HashKey() []byte
- func (a Tags) Keys() []string
- func (a Tags) Len() int
- func (a Tags) Less(i, j int) bool
- func (a Tags) Map() map[string]string
- func (a Tags) Merge(other map[string]string) Tags
- func (a *Tags) Set(key, value []byte)
- func (a *Tags) SetString(key, value string)
- func (a Tags) Size() int
- func (a Tags) String() string
- func (a Tags) Swap(i, j int)
- func (a Tags) Values() []string
Constants ¶
const ( // MinNanoTime is the minumum time that can be represented. // // 1677-09-21 00:12:43.145224194 +0000 UTC // // The two lowest minimum integers are used as sentinel values. The // minimum value needs to be used as a value lower than any other value for // comparisons and another separate value is needed to act as a sentinel // default value that is unusable by the user, but usable internally. // Because these two values need to be used for a special purpose, we do // not allow users to write points at these two times. MinNanoTime = int64(math.MinInt64) + 2 // MaxNanoTime is the maximum time that can be represented. // // 2262-04-11 23:47:16.854775806 +0000 UTC // // The highest time represented by a nanosecond needs to be used for an // exclusive range in the shard group, so the maximum time needs to be one // less than the possible maximum number of nanoseconds representable by an // int64 so that we don't lose a point at that one time. MaxNanoTime = int64(math.MaxInt64) - 1 )
const (
// MaxKeyLength is the largest allowed size of the combined measurement and tag keys.
MaxKeyLength = 65535
)
Variables ¶
var ( // ErrPointMustHaveAField is returned when operating on a point that does not have any fields. ErrPointMustHaveAField = errors.New("point without fields is unsupported") // ErrInvalidNumber is returned when a number is expected but not provided. ErrInvalidNumber = errors.New("invalid number") // ErrInvalidPoint is returned when a point cannot be parsed correctly. ErrInvalidPoint = errors.New("point is invalid") )
var ( // ErrTimeOutOfRange gets returned when time is out of the representable range using int64 nanoseconds since the epoch. ErrTimeOutOfRange = fmt.Errorf("time outside range %d - %d", MinNanoTime, MaxNanoTime) )
Functions ¶
func AppendMakeKey ¶
AppendMakeKey appends the key derived from name and tags to dst and returns the extended buffer.
func CompareTags ¶
CompareTags returns -1 if a < b, 1 if a > b, and 0 if a == b.
func EnableUintSupport ¶
func EnableUintSupport()
EnableUintSupport manually enables uint support for the point parser. This function will be removed in the future and only exists for unit tests during the transition.
func EscapeMeasurement ¶
func EscapeStringField ¶
EscapeStringField returns a copy of in with any double quotes or backslashes with escaped values.
func GetPrecisionMultiplier ¶
GetPrecisionMultiplier will return a multiplier for the precision specified.
func SafeCalcTime ¶
SafeCalcTime safely calculates the time given. Will return error if the time is outside the supported range.
func UnescapeMeasurement ¶
func ValidKeyToken ¶
ValidKeyToken returns true if the token used for measurement, tag key, or tag value is a valid unicode string and only contains printable, non-replacement characters.
func ValidKeyTokens ¶
ValidKeyTokens returns true if the measurement name and all tags are valid.
func ValidPrecision ¶
ValidPrecision checks if the precision is known.
Types ¶
type FieldIterator ¶
type FieldIterator interface { // Next indicates whether there any fields remaining. Next() bool // FieldKey returns the key of the current field. FieldKey() []byte // Type returns the FieldType of the current field. Type() FieldType // StringValue returns the string value of the current field. StringValue() string // IntegerValue returns the integer value of the current field. IntegerValue() (int64, error) // UnsignedValue returns the unsigned value of the current field. UnsignedValue() (uint64, error) // BooleanValue returns the boolean value of the current field. BooleanValue() (bool, error) // FloatValue returns the float value of the current field. FloatValue() (float64, error) // Reset resets the iterator to its initial state. Reset() }
FieldIterator provides a low-allocation interface to iterate through a point's fields.
type FieldType ¶
type FieldType int
FieldType represents the type of a field.
const ( // Integer indicates the field's type is integer. Integer FieldType = iota // Float indicates the field's type is float. Float // Boolean indicates the field's type is boolean. Boolean // String indicates the field's type is string. String // Empty is used to indicate that there is no field. Empty // Unsigned indicates the field's type is an unsigned integer. Unsigned )
type Fields ¶
type Fields map[string]interface{}
Fields represents a mapping between a Point's field names and their values.
func (Fields) MarshalBinary ¶
MarshalBinary encodes all the fields to their proper type and returns the binary represenation NOTE: uint64 is specifically not supported due to potential overflow when we decode again later to an int64 NOTE2: uint is accepted, and may be 64 bits, and is for some reason accepted...
type InlineFNV64a ¶
type InlineFNV64a uint64
InlineFNV64a is an alloc-free port of the standard library's fnv64a. See https://en.wikipedia.org/wiki/Fowler%E2%80%93Noll%E2%80%93Vo_hash_function.
func NewInlineFNV64a ¶
func NewInlineFNV64a() InlineFNV64a
NewInlineFNV64a returns a new instance of InlineFNV64a.
func (*InlineFNV64a) Sum64 ¶
func (s *InlineFNV64a) Sum64() uint64
Sum64 returns the uint64 of the current resulting hash.
type Point ¶
type Point interface { // Name return the measurement name for the point. Name() []byte // SetName updates the measurement name for the point. SetName(string) // Tags returns the tag set for the point. Tags() Tags // ForEachTag iterates over each tag invoking fn. If fn return false, iteration stops. ForEachTag(fn func(k, v []byte) bool) // AddTag adds or replaces a tag value for a point. AddTag(key, value string) // SetTags replaces the tags for the point. SetTags(tags Tags) // HasTag returns true if the tag exists for the point. HasTag(tag []byte) bool // Fields returns the fields for the point. Fields() (Fields, error) // Time return the timestamp for the point. Time() time.Time // SetTime updates the timestamp for the point. SetTime(t time.Time) // UnixNano returns the timestamp of the point as nanoseconds since Unix epoch. UnixNano() int64 // HashID returns a non-cryptographic checksum of the point's key. HashID() uint64 // Key returns the key (measurement joined with tags) of the point. Key() []byte // String returns a string representation of the point. If there is a // timestamp associated with the point then it will be specified with the default // precision of nanoseconds. String() string // MarshalBinary returns a binary representation of the point. MarshalBinary() ([]byte, error) // PrecisionString returns a string representation of the point. If there // is a timestamp associated with the point then it will be specified in the // given unit. PrecisionString(precision string) string // RoundedString returns a string representation of the point. If there // is a timestamp associated with the point, then it will be rounded to the // given duration. RoundedString(d time.Duration) string // Split will attempt to return multiple points with the same timestamp whose // string representations are no longer than size. Points with a single field or // a point without a timestamp may exceed the requested size. Split(size int) []Point // Round will round the timestamp of the point to the given duration. Round(d time.Duration) // StringSize returns the length of the string that would be returned by String(). StringSize() int // AppendString appends the result of String() to the provided buffer and returns // the result, potentially reducing string allocations. AppendString(buf []byte) []byte // FieldIterator retuns a FieldIterator that can be used to traverse the // fields of a point without constructing the in-memory map. FieldIterator() FieldIterator }
Point defines the values that will be written to the database.
func MustNewPoint ¶
MustNewPoint returns a new point with the given measurement name, tags, fields and timestamp. If an unsupported field value (NaN) is passed, this function panics.
func NewPoint ¶
NewPoint returns a new point with the given measurement name, tags, fields and timestamp. If an unsupported field value (NaN, or +/-Inf) or out of range time is passed, this function returns an error.
func NewPointFromBytes ¶
NewPointFromBytes returns a new Point from a marshalled Point.
func ParsePoints ¶
ParsePoints returns a slice of Points from a text representation of a point with each point separated by newlines. If any points fail to parse, a non-nil error will be returned in addition to the points that parsed successfully.
func ParsePointsString ¶
ParsePointsString is identical to ParsePoints but accepts a string.
func ParsePointsWithPrecision ¶
ParsePointsWithPrecision is similar to ParsePoints, but allows the caller to provide a precision for time.
NOTE: to minimize heap allocations, the returned Points will refer to subslices of buf. This can have the unintended effect preventing buf from being garbage collected.
type Row ¶
type Row struct { Name string `json:"name,omitempty"` Tags map[string]string `json:"tags,omitempty"` Columns []string `json:"columns,omitempty"` Values [][]interface{} `json:"values,omitempty"` Partial bool `json:"partial,omitempty"` }
Row represents a single row returned from the execution of a statement.
func (*Row) SameSeries ¶
SameSeries returns true if r contains values for the same series as o.
type Statistic ¶
type Statistic struct { Name string `json:"name"` Tags map[string]string `json:"tags"` Values map[string]interface{} `json:"values"` }
Statistic is the representation of a statistic used by the monitoring service.
func NewStatistic ¶
NewStatistic returns an initialized Statistic.
type StatisticTags ¶
StatisticTags is a map that can be merged with others without causing mutations to either map.
func (StatisticTags) Merge ¶
func (t StatisticTags) Merge(tags map[string]string) map[string]string
Merge creates a new map containing the merged contents of tags and t. If both tags and the receiver map contain the same key, the value in tags is used in the resulting map.
Merge always returns a usable map.
type Tag ¶
Tag represents a single key/value tag pair.
type Tags ¶
type Tags []Tag
Tags represents a sorted list of tags.
func ParseKey ¶
ParseKey returns the measurement name and tags from a point.
NOTE: to minimize heap allocations, the returned Tags will refer to subslices of buf. This can have the unintended effect preventing buf from being garbage collected.
func ParseKeyBytes ¶
func (Tags) AppendHashKey ¶
AppendHashKey appends the result of hashing all of a tag's keys and values to dst and returns the extended buffer.
func (Tags) Clone ¶
Clone returns a copy of the slice where the elements are a result of calling `Clone` on the original elements
Tags associated with a Point created by ParsePointsWithPrecision will hold references to the byte slice that was parsed. Use Clone to create Tags with new byte slices that do not refer to the argument to ParsePointsWithPrecision.
func (Tags) Merge ¶
Merge merges the tags combining the two. If both define a tag with the same key, the merged value overwrites the old value. A new map is returned.
func (Tags) Size ¶
Size returns the number of bytes needed to store all tags. Note, this is the number of bytes needed to store all keys and values and does not account for data structures or delimiters for example.