featurebase/tracker.go
Seebs bb1d52a385 ingest and ingest/codec testing work
This is a design to let us write test cases for ingest with schema setup
and data in the json formats we want to use, and results as alternating
queries and expected results, so we can just create new test files and
run the tests against them. We also have to report back what we created
when creating things.

In the process of developing this, I noticed that the documentation describes
ingest schema as allowing more than one schema operation, but we didn't support
this, and also it wouldn't do much good because there was no way to do partial
things like "just add a field". Fixed.

Also we implement comparison for ops, so the test output is actually
a test rather than just some data to visually eyeball.

In the process, realize that the handling of timestamps was wrong; we said that we
take them as raw numbers relative to the epoch, not as raw Unix timestamps.

Also a couple of related cleanups caught by doing the testing.
2021-08-19 09:50:59 -05:00

203 lines
4.8 KiB
Go

// Copyright 2020 Pilosa Corp.
//
// 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 pilosa
import (
"sort"
"sync"
"time"
)
type ActiveQueryStatus struct {
PQL string `json:"PQL"`
SQL string `json:"SQL,omitempty"`
Node string `json:"node"`
Index string `json:"index"`
Age time.Duration `json:"age"`
}
type PastQueryStatus struct {
PQL string `json:"PQL"`
SQL string `json:"SQL,omitempty"`
Node string `json:"nodeID"`
Index string `json:"index"`
Start time.Time `json:"start"`
Runtime time.Duration `json:"runtime"` // deprecated
RuntimeNs time.Duration `json:"runtimeNanoseconds"`
}
type activeQuery struct {
PQL string
SQL string
node string
index string
started time.Time
}
type pastQuery struct {
PQL string
SQL string
node string
index string
started time.Time
runtime time.Duration
}
type queryStatusUpdate struct {
q *activeQuery
end bool
endTime time.Time
}
type queryTracker struct {
updates chan<- queryStatusUpdate
checks chan<- chan<- []*activeQuery
history *ringBuffer
wg sync.WaitGroup
stop chan struct{}
}
type ringBuffer struct {
queries []pastQuery
start int
count int
mu sync.Mutex
}
// newRingBuffer initializes an empty RingBuffer of specified capacity.
func newRingBuffer(n int) *ringBuffer {
return &ringBuffer{
queries: make([]pastQuery, n),
start: 0,
count: 0,
}
}
// add adds a new element to the queue, overwriting the oldest if it is already full.
func (b *ringBuffer) add(q pastQuery) {
b.mu.Lock()
defer b.mu.Unlock()
// len(b.queries) is used here as the *capacity* of the ringBuffer
b.queries[(b.start+b.count)%len(b.queries)] = q
if b.count == len(b.queries) {
b.start = (b.start + 1) % len(b.queries)
} else {
b.count++
}
}
// slice returns the contents of the RingBuffer, in insertion order.
func (b *ringBuffer) slice() []pastQuery {
b.mu.Lock()
defer b.mu.Unlock()
return append(b.queries[b.start:b.count], b.queries[0:b.start]...)
}
func newQueryTracker(historyLength int) *queryTracker {
done := make(chan struct{})
updates := make(chan queryStatusUpdate, 128)
checks := make(chan chan<- []*activeQuery)
history := newRingBuffer(historyLength)
tracker := &queryTracker{
updates: updates,
checks: checks,
history: history,
stop: done,
}
tracker.wg.Add(1)
go func() {
defer tracker.wg.Done()
activeQueries := make(map[*activeQuery]struct{})
for {
select {
case update := <-updates:
if update.end {
pq := pastQuery{update.q.PQL, update.q.SQL, update.q.node, update.q.index, update.q.started, update.endTime.Sub(update.q.started)}
tracker.history.add(pq)
delete(activeQueries, update.q)
} else {
activeQueries[update.q] = struct{}{}
}
case check := <-checks:
out := make([]*activeQuery, len(activeQueries))
i := 0
for q := range activeQueries {
out[i] = q
i++
}
check <- out
close(check)
case <-done:
return
}
}
}()
return tracker
}
func (t *queryTracker) Start(pql, sql, nodeID, index string, start time.Time) *activeQuery {
q := &activeQuery{pql, sql, nodeID, index, start}
t.updates <- queryStatusUpdate{q, false, time.Time{}}
return q
}
func (t *queryTracker) Finish(q *activeQuery) {
t.updates <- queryStatusUpdate{q, true, time.Now()}
}
func (t *queryTracker) ActiveQueries() []ActiveQueryStatus {
ch := make(chan []*activeQuery, 1)
t.checks <- ch
queries := <-ch
sort.Slice(queries, func(i, j int) bool {
switch {
case queries[i].started.Before(queries[j].started):
return true
case queries[i].started.After(queries[j].started):
return false
case queries[i].PQL < queries[j].PQL:
return true
case queries[i].PQL > queries[j].PQL:
return false
default:
return false
}
})
now := time.Now()
out := make([]ActiveQueryStatus, len(queries))
for i, v := range queries {
out[i] = ActiveQueryStatus{v.PQL, v.SQL, v.node, v.index, now.Sub(v.started)}
}
return out
}
func (t *queryTracker) PastQueries() []PastQueryStatus {
queries := t.history.slice()
out := make([]PastQueryStatus, len(queries))
for i, v := range queries {
out[i] = PastQueryStatus{v.PQL, v.SQL, v.node, v.index, v.started, v.runtime, v.runtime}
}
return out
}
func (t *queryTracker) Stop() {
close(t.stop)
t.wg.Wait()
}