diff --git a/vendor/go.opentelemetry.io/otel/api/metric/number.go b/vendor/go.opentelemetry.io/otel/api/metric/number.go
new file mode 100644
index 0000000..c3ca0ed
--- /dev/null
+++ b/vendor/go.opentelemetry.io/otel/api/metric/number.go
@@ -0,0 +1,540 @@
+// Copyright The OpenTelemetry Authors
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+package metric
+
+//go:generate stringer -type=NumberKind
+
+import (
+	"fmt"
+	"math"
+	"sync/atomic"
+
+	"go.opentelemetry.io/otel/internal"
+)
+
+// NumberKind describes the data type of the Number.
+type NumberKind int8
+
+const (
+	// Int64NumberKind means that the Number stores int64.
+	Int64NumberKind NumberKind = iota
+	// Float64NumberKind means that the Number stores float64.
+	Float64NumberKind
+)
+
+// Zero returns a zero value for a given NumberKind
+func (k NumberKind) Zero() Number {
+	switch k {
+	case Int64NumberKind:
+		return NewInt64Number(0)
+	case Float64NumberKind:
+		return NewFloat64Number(0.)
+	default:
+		return Number(0)
+	}
+}
+
+// Minimum returns the minimum representable value
+// for a given NumberKind
+func (k NumberKind) Minimum() Number {
+	switch k {
+	case Int64NumberKind:
+		return NewInt64Number(math.MinInt64)
+	case Float64NumberKind:
+		return NewFloat64Number(-1. * math.MaxFloat64)
+	default:
+		return Number(0)
+	}
+}
+
+// Maximum returns the maximum representable value
+// for a given NumberKind
+func (k NumberKind) Maximum() Number {
+	switch k {
+	case Int64NumberKind:
+		return NewInt64Number(math.MaxInt64)
+	case Float64NumberKind:
+		return NewFloat64Number(math.MaxFloat64)
+	default:
+		return Number(0)
+	}
+}
+
+// Number represents either an integral or a floating point value. It
+// needs to be accompanied with a source of NumberKind that describes
+// the actual type of the value stored within Number.
+type Number uint64
+
+// - constructors
+
+// NewNumberFromRaw creates a new Number from a raw value.
+func NewNumberFromRaw(r uint64) Number {
+	return Number(r)
+}
+
+// NewInt64Number creates an integral Number.
+func NewInt64Number(i int64) Number {
+	return NewNumberFromRaw(internal.Int64ToRaw(i))
+}
+
+// NewFloat64Number creates a floating point Number.
+func NewFloat64Number(f float64) Number {
+	return NewNumberFromRaw(internal.Float64ToRaw(f))
+}
+
+// NewNumberSignChange returns a number with the same magnitude and
+// the opposite sign.  `kind` must describe the kind of number in `nn`.
+//
+// Does not change Uint64NumberKind values.
+func NewNumberSignChange(kind NumberKind, nn Number) Number {
+	switch kind {
+	case Int64NumberKind:
+		return NewInt64Number(-nn.AsInt64())
+	case Float64NumberKind:
+		return NewFloat64Number(-nn.AsFloat64())
+	}
+	return nn
+}
+
+// - as x
+
+// AsNumber gets the Number.
+func (n *Number) AsNumber() Number {
+	return *n
+}
+
+// AsRaw gets the uninterpreted raw value. Might be useful for some
+// atomic operations.
+func (n *Number) AsRaw() uint64 {
+	return uint64(*n)
+}
+
+// AsInt64 assumes that the value contains an int64 and returns it as
+// such.
+func (n *Number) AsInt64() int64 {
+	return internal.RawToInt64(n.AsRaw())
+}
+
+// AsFloat64 assumes that the measurement value contains a float64 and
+// returns it as such.
+func (n *Number) AsFloat64() float64 {
+	return internal.RawToFloat64(n.AsRaw())
+}
+
+// - as x atomic
+
+// AsNumberAtomic gets the Number atomically.
+func (n *Number) AsNumberAtomic() Number {
+	return NewNumberFromRaw(n.AsRawAtomic())
+}
+
+// AsRawAtomic gets the uninterpreted raw value atomically. Might be
+// useful for some atomic operations.
+func (n *Number) AsRawAtomic() uint64 {
+	return atomic.LoadUint64(n.AsRawPtr())
+}
+
+// AsInt64Atomic assumes that the number contains an int64 and returns
+// it as such atomically.
+func (n *Number) AsInt64Atomic() int64 {
+	return atomic.LoadInt64(n.AsInt64Ptr())
+}
+
+// AsFloat64Atomic assumes that the measurement value contains a
+// float64 and returns it as such atomically.
+func (n *Number) AsFloat64Atomic() float64 {
+	return internal.RawToFloat64(n.AsRawAtomic())
+}
+
+// - as x ptr
+
+// AsRawPtr gets the pointer to the raw, uninterpreted raw
+// value. Might be useful for some atomic operations.
+func (n *Number) AsRawPtr() *uint64 {
+	return (*uint64)(n)
+}
+
+// AsInt64Ptr assumes that the number contains an int64 and returns a
+// pointer to it.
+func (n *Number) AsInt64Ptr() *int64 {
+	return internal.RawPtrToInt64Ptr(n.AsRawPtr())
+}
+
+// AsFloat64Ptr assumes that the number contains a float64 and returns a
+// pointer to it.
+func (n *Number) AsFloat64Ptr() *float64 {
+	return internal.RawPtrToFloat64Ptr(n.AsRawPtr())
+}
+
+// - coerce
+
+// CoerceToInt64 casts the number to int64. May result in
+// data/precision loss.
+func (n *Number) CoerceToInt64(kind NumberKind) int64 {
+	switch kind {
+	case Int64NumberKind:
+		return n.AsInt64()
+	case Float64NumberKind:
+		return int64(n.AsFloat64())
+	default:
+		// you get what you deserve
+		return 0
+	}
+}
+
+// CoerceToFloat64 casts the number to float64. May result in
+// data/precision loss.
+func (n *Number) CoerceToFloat64(kind NumberKind) float64 {
+	switch kind {
+	case Int64NumberKind:
+		return float64(n.AsInt64())
+	case Float64NumberKind:
+		return n.AsFloat64()
+	default:
+		// you get what you deserve
+		return 0
+	}
+}
+
+// - set
+
+// SetNumber sets the number to the passed number. Both should be of
+// the same kind.
+func (n *Number) SetNumber(nn Number) {
+	*n.AsRawPtr() = nn.AsRaw()
+}
+
+// SetRaw sets the number to the passed raw value. Both number and the
+// raw number should represent the same kind.
+func (n *Number) SetRaw(r uint64) {
+	*n.AsRawPtr() = r
+}
+
+// SetInt64 assumes that the number contains an int64 and sets it to
+// the passed value.
+func (n *Number) SetInt64(i int64) {
+	*n.AsInt64Ptr() = i
+}
+
+// SetFloat64 assumes that the number contains a float64 and sets it
+// to the passed value.
+func (n *Number) SetFloat64(f float64) {
+	*n.AsFloat64Ptr() = f
+}
+
+// - set atomic
+
+// SetNumberAtomic sets the number to the passed number
+// atomically. Both should be of the same kind.
+func (n *Number) SetNumberAtomic(nn Number) {
+	atomic.StoreUint64(n.AsRawPtr(), nn.AsRaw())
+}
+
+// SetRawAtomic sets the number to the passed raw value
+// atomically. Both number and the raw number should represent the
+// same kind.
+func (n *Number) SetRawAtomic(r uint64) {
+	atomic.StoreUint64(n.AsRawPtr(), r)
+}
+
+// SetInt64Atomic assumes that the number contains an int64 and sets
+// it to the passed value atomically.
+func (n *Number) SetInt64Atomic(i int64) {
+	atomic.StoreInt64(n.AsInt64Ptr(), i)
+}
+
+// SetFloat64Atomic assumes that the number contains a float64 and
+// sets it to the passed value atomically.
+func (n *Number) SetFloat64Atomic(f float64) {
+	atomic.StoreUint64(n.AsRawPtr(), internal.Float64ToRaw(f))
+}
+
+// - swap
+
+// SwapNumber sets the number to the passed number and returns the old
+// number. Both this number and the passed number should be of the
+// same kind.
+func (n *Number) SwapNumber(nn Number) Number {
+	old := *n
+	n.SetNumber(nn)
+	return old
+}
+
+// SwapRaw sets the number to the passed raw value and returns the old
+// raw value. Both number and the raw number should represent the same
+// kind.
+func (n *Number) SwapRaw(r uint64) uint64 {
+	old := n.AsRaw()
+	n.SetRaw(r)
+	return old
+}
+
+// SwapInt64 assumes that the number contains an int64, sets it to the
+// passed value and returns the old int64 value.
+func (n *Number) SwapInt64(i int64) int64 {
+	old := n.AsInt64()
+	n.SetInt64(i)
+	return old
+}
+
+// SwapFloat64 assumes that the number contains an float64, sets it to
+// the passed value and returns the old float64 value.
+func (n *Number) SwapFloat64(f float64) float64 {
+	old := n.AsFloat64()
+	n.SetFloat64(f)
+	return old
+}
+
+// - swap atomic
+
+// SwapNumberAtomic sets the number to the passed number and returns
+// the old number atomically. Both this number and the passed number
+// should be of the same kind.
+func (n *Number) SwapNumberAtomic(nn Number) Number {
+	return NewNumberFromRaw(atomic.SwapUint64(n.AsRawPtr(), nn.AsRaw()))
+}
+
+// SwapRawAtomic sets the number to the passed raw value and returns
+// the old raw value atomically. Both number and the raw number should
+// represent the same kind.
+func (n *Number) SwapRawAtomic(r uint64) uint64 {
+	return atomic.SwapUint64(n.AsRawPtr(), r)
+}
+
+// SwapInt64Atomic assumes that the number contains an int64, sets it
+// to the passed value and returns the old int64 value atomically.
+func (n *Number) SwapInt64Atomic(i int64) int64 {
+	return atomic.SwapInt64(n.AsInt64Ptr(), i)
+}
+
+// SwapFloat64Atomic assumes that the number contains an float64, sets
+// it to the passed value and returns the old float64 value
+// atomically.
+func (n *Number) SwapFloat64Atomic(f float64) float64 {
+	return internal.RawToFloat64(atomic.SwapUint64(n.AsRawPtr(), internal.Float64ToRaw(f)))
+}
+
+// - add
+
+// AddNumber assumes that this and the passed number are of the passed
+// kind and adds the passed number to this number.
+func (n *Number) AddNumber(kind NumberKind, nn Number) {
+	switch kind {
+	case Int64NumberKind:
+		n.AddInt64(nn.AsInt64())
+	case Float64NumberKind:
+		n.AddFloat64(nn.AsFloat64())
+	}
+}
+
+// AddRaw assumes that this number and the passed raw value are of the
+// passed kind and adds the passed raw value to this number.
+func (n *Number) AddRaw(kind NumberKind, r uint64) {
+	n.AddNumber(kind, NewNumberFromRaw(r))
+}
+
+// AddInt64 assumes that the number contains an int64 and adds the
+// passed int64 to it.
+func (n *Number) AddInt64(i int64) {
+	*n.AsInt64Ptr() += i
+}
+
+// AddFloat64 assumes that the number contains a float64 and adds the
+// passed float64 to it.
+func (n *Number) AddFloat64(f float64) {
+	*n.AsFloat64Ptr() += f
+}
+
+// - add atomic
+
+// AddNumberAtomic assumes that this and the passed number are of the
+// passed kind and adds the passed number to this number atomically.
+func (n *Number) AddNumberAtomic(kind NumberKind, nn Number) {
+	switch kind {
+	case Int64NumberKind:
+		n.AddInt64Atomic(nn.AsInt64())
+	case Float64NumberKind:
+		n.AddFloat64Atomic(nn.AsFloat64())
+	}
+}
+
+// AddRawAtomic assumes that this number and the passed raw value are
+// of the passed kind and adds the passed raw value to this number
+// atomically.
+func (n *Number) AddRawAtomic(kind NumberKind, r uint64) {
+	n.AddNumberAtomic(kind, NewNumberFromRaw(r))
+}
+
+// AddInt64Atomic assumes that the number contains an int64 and adds
+// the passed int64 to it atomically.
+func (n *Number) AddInt64Atomic(i int64) {
+	atomic.AddInt64(n.AsInt64Ptr(), i)
+}
+
+// AddFloat64Atomic assumes that the number contains a float64 and
+// adds the passed float64 to it atomically.
+func (n *Number) AddFloat64Atomic(f float64) {
+	for {
+		o := n.AsFloat64Atomic()
+		if n.CompareAndSwapFloat64(o, o+f) {
+			break
+		}
+	}
+}
+
+// - compare and swap (atomic only)
+
+// CompareAndSwapNumber does the atomic CAS operation on this
+// number. This number and passed old and new numbers should be of the
+// same kind.
+func (n *Number) CompareAndSwapNumber(on, nn Number) bool {
+	return atomic.CompareAndSwapUint64(n.AsRawPtr(), on.AsRaw(), nn.AsRaw())
+}
+
+// CompareAndSwapRaw does the atomic CAS operation on this
+// number. This number and passed old and new raw values should be of
+// the same kind.
+func (n *Number) CompareAndSwapRaw(or, nr uint64) bool {
+	return atomic.CompareAndSwapUint64(n.AsRawPtr(), or, nr)
+}
+
+// CompareAndSwapInt64 assumes that this number contains an int64 and
+// does the atomic CAS operation on it.
+func (n *Number) CompareAndSwapInt64(oi, ni int64) bool {
+	return atomic.CompareAndSwapInt64(n.AsInt64Ptr(), oi, ni)
+}
+
+// CompareAndSwapFloat64 assumes that this number contains a float64 and
+// does the atomic CAS operation on it.
+func (n *Number) CompareAndSwapFloat64(of, nf float64) bool {
+	return atomic.CompareAndSwapUint64(n.AsRawPtr(), internal.Float64ToRaw(of), internal.Float64ToRaw(nf))
+}
+
+// - compare
+
+// CompareNumber compares two Numbers given their kind.  Both numbers
+// should have the same kind.  This returns:
+//    0 if the numbers are equal
+//    -1 if the subject `n` is less than the argument `nn`
+//    +1 if the subject `n` is greater than the argument `nn`
+func (n *Number) CompareNumber(kind NumberKind, nn Number) int {
+	switch kind {
+	case Int64NumberKind:
+		return n.CompareInt64(nn.AsInt64())
+	case Float64NumberKind:
+		return n.CompareFloat64(nn.AsFloat64())
+	default:
+		// you get what you deserve
+		return 0
+	}
+}
+
+// CompareRaw compares two numbers, where one is input as a raw
+// uint64, interpreting both values as a `kind` of number.
+func (n *Number) CompareRaw(kind NumberKind, r uint64) int {
+	return n.CompareNumber(kind, NewNumberFromRaw(r))
+}
+
+// CompareInt64 assumes that the Number contains an int64 and performs
+// a comparison between the value and the other value. It returns the
+// typical result of the compare function: -1 if the value is less
+// than the other, 0 if both are equal, 1 if the value is greater than
+// the other.
+func (n *Number) CompareInt64(i int64) int {
+	this := n.AsInt64()
+	if this < i {
+		return -1
+	} else if this > i {
+		return 1
+	}
+	return 0
+}
+
+// CompareFloat64 assumes that the Number contains a float64 and
+// performs a comparison between the value and the other value. It
+// returns the typical result of the compare function: -1 if the value
+// is less than the other, 0 if both are equal, 1 if the value is
+// greater than the other.
+//
+// Do not compare NaN values.
+func (n *Number) CompareFloat64(f float64) int {
+	this := n.AsFloat64()
+	if this < f {
+		return -1
+	} else if this > f {
+		return 1
+	}
+	return 0
+}
+
+// - relations to zero
+
+// IsPositive returns true if the actual value is greater than zero.
+func (n *Number) IsPositive(kind NumberKind) bool {
+	return n.compareWithZero(kind) > 0
+}
+
+// IsNegative returns true if the actual value is less than zero.
+func (n *Number) IsNegative(kind NumberKind) bool {
+	return n.compareWithZero(kind) < 0
+}
+
+// IsZero returns true if the actual value is equal to zero.
+func (n *Number) IsZero(kind NumberKind) bool {
+	return n.compareWithZero(kind) == 0
+}
+
+// - misc
+
+// Emit returns a string representation of the raw value of the
+// Number. A %d is used for integral values, %f for floating point
+// values.
+func (n *Number) Emit(kind NumberKind) string {
+	switch kind {
+	case Int64NumberKind:
+		return fmt.Sprintf("%d", n.AsInt64())
+	case Float64NumberKind:
+		return fmt.Sprintf("%f", n.AsFloat64())
+	default:
+		return ""
+	}
+}
+
+// AsInterface returns the number as an interface{}, typically used
+// for NumberKind-correct JSON conversion.
+func (n *Number) AsInterface(kind NumberKind) interface{} {
+	switch kind {
+	case Int64NumberKind:
+		return n.AsInt64()
+	case Float64NumberKind:
+		return n.AsFloat64()
+	default:
+		return math.NaN()
+	}
+}
+
+// - private stuff
+
+func (n *Number) compareWithZero(kind NumberKind) int {
+	switch kind {
+	case Int64NumberKind:
+		return n.CompareInt64(0)
+	case Float64NumberKind:
+		return n.CompareFloat64(0.)
+	default:
+		// you get what you deserve
+		return 0
+	}
+}