Adds a first version of a new obidistribute command

pkg/obiseq/batchiterator.go

@@ -1,8 +1,12 @@
 package obiseq
 
 import (
+	"fmt"
 	"log"
 	"sync"
+	"sync/atomic"
+
+	"github.com/tevino/abool/v2"
 )
 
 type BioSequenceBatch struct {
@@ -36,40 +40,45 @@ func (batch BioSequenceBatch) IsNil() bool {
 
 // Structure implementing an iterator over bioseq.BioSequenceBatch
 // based on a channel.
-type __ibiosequencebatch__ struct {
-	channel     chan BioSequenceBatch
-	current     BioSequenceBatch
-	pushBack    bool
-	all_done    *sync.WaitGroup
-	buffer_size int
-	finished    bool
-	p_finished  *bool
+type _IBioSequenceBatch struct {
+	channel         chan BioSequenceBatch
+	current         BioSequenceBatch
+	pushBack        *abool.AtomicBool
+	all_done        *sync.WaitGroup
+	lock            *sync.RWMutex
+	buffer_size     int32
+	batch_size      int32
+	sequence_format string
+	finished        *abool.AtomicBool
 }
 
 type IBioSequenceBatch struct {
-	pointer *__ibiosequencebatch__
+	pointer *_IBioSequenceBatch
 }
 
 var NilIBioSequenceBatch = IBioSequenceBatch{pointer: nil}
 
 func MakeIBioSequenceBatch(sizes ...int) IBioSequenceBatch {
-	buffsize := 1
+	buffsize := int32(1)
 
 	if len(sizes) > 0 {
-		buffsize = sizes[0]
+		buffsize = int32(sizes[0])
 	}
 
-	i := __ibiosequencebatch__{
-		channel:     make(chan BioSequenceBatch, buffsize),
-		current:     NilBioSequenceBatch,
-		pushBack:    false,
-		buffer_size: buffsize,
-		finished:    false,
-		p_finished:  nil,
+	i := _IBioSequenceBatch{
+		channel:         make(chan BioSequenceBatch, buffsize),
+		current:         NilBioSequenceBatch,
+		pushBack:        abool.New(),
+		buffer_size:     buffsize,
+		batch_size:      -1,
+		sequence_format: "",
+		finished:        abool.New(),
 	}
-	i.p_finished = &i.finished
+
 	waiting := sync.WaitGroup{}
 	i.all_done = &waiting
+	lock := sync.RWMutex{}
+	i.lock = &lock
 	ii := IBioSequenceBatch{&i}
 	return ii
 }
@@ -82,6 +91,22 @@ func (iterator IBioSequenceBatch) Done() {
 	iterator.pointer.all_done.Done()
 }
 
+func (iterator IBioSequenceBatch) Unlock() {
+	iterator.pointer.lock.Unlock()
+}
+
+func (iterator IBioSequenceBatch) Lock() {
+	iterator.pointer.lock.Lock()
+}
+
+func (iterator IBioSequenceBatch) RLock() {
+	iterator.pointer.lock.RLock()
+}
+
+func (iterator IBioSequenceBatch) RUnlock() {
+	iterator.pointer.lock.RUnlock()
+}
+
 func (iterator IBioSequenceBatch) Wait() {
 	iterator.pointer.all_done.Wait()
 }
@@ -95,29 +120,48 @@ func (iterator IBioSequenceBatch) IsNil() bool {
 }
 
 func (iterator IBioSequenceBatch) BufferSize() int {
-	return iterator.pointer.buffer_size
+	return int(atomic.LoadInt32(&iterator.pointer.buffer_size))
+}
+
+func (iterator IBioSequenceBatch) BatchSize() int {
+	return int(atomic.LoadInt32(&iterator.pointer.batch_size))
+}
+
+func (iterator IBioSequenceBatch) SetBatchSize(size int) error {
+	if size >= 0 {
+		atomic.StoreInt32(&iterator.pointer.batch_size, int32(size))
+		return nil
+	}
+
+	return fmt.Errorf("size (%d) cannot be negative", size)
 }
 
 func (iterator IBioSequenceBatch) Split() IBioSequenceBatch {
-	i := __ibiosequencebatch__{
-		channel:     iterator.pointer.channel,
-		current:     NilBioSequenceBatch,
-		pushBack:    false,
-		all_done:    iterator.pointer.all_done,
-		buffer_size: iterator.pointer.buffer_size,
-		finished:    false,
-		p_finished:  iterator.pointer.p_finished}
+	iterator.pointer.lock.RLock()
+	defer iterator.pointer.lock.RUnlock()
+	i := _IBioSequenceBatch{
+		channel:         iterator.pointer.channel,
+		current:         NilBioSequenceBatch,
+		pushBack:        abool.New(),
+		all_done:        iterator.pointer.all_done,
+		buffer_size:     iterator.pointer.buffer_size,
+		batch_size:      iterator.pointer.batch_size,
+		sequence_format: iterator.pointer.sequence_format,
+		finished:        iterator.pointer.finished}
+	lock := sync.RWMutex{}
+	i.lock = &lock
+
 	newIter := IBioSequenceBatch{&i}
 	return newIter
 }
 
 func (iterator IBioSequenceBatch) Next() bool {
-	if *(iterator.pointer.p_finished) {
+	if iterator.pointer.finished.IsSet() {
 		return false
 	}
 
-	if iterator.pointer.pushBack {
-		iterator.pointer.pushBack = false
+	if iterator.pointer.pushBack.IsSet() {
+		iterator.pointer.pushBack.UnSet()
 		return true
 	}
 
@@ -129,13 +173,13 @@ func (iterator IBioSequenceBatch) Next() bool {
 	}
 
 	iterator.pointer.current = NilBioSequenceBatch
-	*iterator.pointer.p_finished = true
+	iterator.pointer.finished.Set()
 	return false
 }
 
 func (iterator IBioSequenceBatch) PushBack() {
 	if !iterator.pointer.current.IsNil() {
-		iterator.pointer.pushBack = true
+		iterator.pointer.pushBack.Set()
 	}
 }
 
@@ -150,7 +194,7 @@ func (iterator IBioSequenceBatch) Get() BioSequenceBatch {
 // Finished returns 'true' value if no more data is available
 // from the iterator.
 func (iterator IBioSequenceBatch) Finished() bool {
-	return *iterator.pointer.p_finished
+	return iterator.pointer.finished.IsSet()
 }
 
 func (iterator IBioSequenceBatch) IBioSequence(sizes ...int) IBioSequence {
@@ -378,7 +422,7 @@ func (iterator IBioSequenceBatch) DivideOn(predicate SequencePredicate,
 	buffsize := iterator.BufferSize()
 
 	if len(sizes) > 0 {
-		buffsize = sizes[1]
+		buffsize = sizes[0]
 	}
 
 	trueIter := MakeIBioSequenceBatch(buffsize)

pkg/obiseq/class.go

@@ -0,0 +1,65 @@
+package obiseq
+
+import (
+	"fmt"
+	"hash/crc32"
+	"strconv"
+)
+
+type SequenceClassifier func(sequence BioSequence) string
+
+func AnnotationClassifier(key string) SequenceClassifier {
+	f := func(sequence BioSequence) string {
+		if sequence.HasAnnotation() {
+			value, ok := sequence.Annotations()[key]
+
+			if ok {
+				switch value := value.(type) {
+				case string:
+					return value
+				default:
+					return fmt.Sprint(value)
+				}
+			}
+		}
+		return ""
+	}
+
+	return SequenceClassifier(f)
+}
+
+var SampleClassifier = AnnotationClassifier("sample")
+
+func PredicateClassifier(predicate SequencePredicate) SequenceClassifier {
+	f := func(sequence BioSequence) string {
+		if predicate(sequence) {
+			return "true"
+		} else {
+			return "false"
+		}
+	}
+
+	return SequenceClassifier(f)
+}
+
+// Builds a classifier function based on CRC32 of the sequence
+//
+func HashClassifier(size int) SequenceClassifier {
+	f := func(sequence BioSequence) string {
+		h := crc32.ChecksumIEEE(sequence.Sequence()) % uint32(size)
+		return strconv.Itoa(int(h))
+	}
+
+	return SequenceClassifier(f)
+}
+
+func RotateClassifier(size int) SequenceClassifier {
+	n := 0
+	f := func(sequence BioSequence) string {
+		h := n % size
+		n++
+		return strconv.Itoa(int(h))
+	}
+
+	return SequenceClassifier(f)
+}

pkg/obiseq/distribute.go

@@ -0,0 +1,107 @@
+package obiseq
+
+import (
+	"fmt"
+	"sync"
+)
+
+type IDistribute struct {
+	outputs map[string]IBioSequenceBatch
+	news    chan string
+	lock    *sync.Mutex
+}
+
+func (dist *IDistribute) Outputs(key string) (IBioSequenceBatch, error) {
+	dist.lock.Lock()
+	iter, ok := dist.outputs[key]
+	dist.lock.Unlock()
+
+	if !ok {
+		return NilIBioSequenceBatch, fmt.Errorf("key %s unknown", key)
+	}
+
+	return iter, nil
+}
+
+func (dist *IDistribute) News() chan string {
+	return dist.news
+}
+
+func (iterator IBioSequenceBatch) Distribute(class SequenceClassifier, sizes ...int) IDistribute {
+	batchsize := 5000
+	buffsize := 2
+
+	outputs := make(map[string]IBioSequenceBatch, 100)
+	slices := make(map[string]*BioSequenceSlice, 100)
+	orders := make(map[string]int, 100)
+	news := make(chan string)
+
+	if len(sizes) > 0 {
+		batchsize = sizes[0]
+	}
+
+	if len(sizes) > 1 {
+		buffsize = sizes[1]
+	}
+
+	jobDone := sync.WaitGroup{}
+	lock := sync.Mutex{}
+
+	jobDone.Add(1)
+
+	go func() {
+		jobDone.Wait()
+		close(news)
+		for _, i := range outputs {
+			close(i.Channel())
+		}
+	}()
+
+	go func() {
+		iterator = iterator.SortBatches()
+
+		for iterator.Next() {
+			seqs := iterator.Get()
+			for _, s := range seqs.slice {
+				key := class(s)
+				slice, ok := slices[key]
+
+				if !ok {
+					s := make(BioSequenceSlice, 0, batchsize)
+					slice = &s
+					slices[key] = slice
+					orders[key] = 0
+
+					lock.Lock()
+					outputs[key] = MakeIBioSequenceBatch(batchsize, buffsize)
+					lock.Unlock()
+
+					news <- key
+				}
+
+				*slice = append(*slice, s)
+				if len(*slice) == batchsize {
+					outputs[key].Channel() <- MakeBioSequenceBatch(orders[key], *slice...)
+					orders[key]++
+					s := make(BioSequenceSlice, 0, batchsize)
+					slices[key] = &s
+				}
+			}
+		}
+
+		for key, slice := range slices {
+			if len(*slice) > 0 {
+				outputs[key].Channel() <- MakeBioSequenceBatch(orders[key], *slice...)
+			}
+		}
+
+		jobDone.Done()
+
+	}()
+
+	return IDistribute{
+		outputs,
+		news,
+		&lock}
+
+}

pkg/obiseq/merge.go

@@ -0,0 +1,116 @@
+package obiseq
+
+import (
+	"fmt"
+	"log"
+)
+
+type StatsOnValues map[string]int
+
+func (sequence BioSequence) HasStatsOn(key string) bool {
+	if !sequence.HasAnnotation() {
+		return false
+	}
+
+	mkey := "merged_" + key
+	annotations := sequence.Annotations()
+	_, ok := annotations[mkey]
+
+	return ok
+}
+
+func (sequence BioSequence) StatsOn(key string) StatsOnValues {
+	mkey := "merged_" + key
+	annotations := sequence.Annotations()
+	istat, ok := annotations[mkey]
+
+	var stats StatsOnValues
+	var newstat bool
+
+	if ok {
+		switch istat := istat.(type) {
+		case StatsOnValues:
+			stats = istat
+			newstat = false
+		default:
+			stats = make(StatsOnValues, 100)
+			annotations[mkey] = stats
+			newstat = true
+		}
+	} else {
+		stats = make(StatsOnValues, 100)
+		annotations[mkey] = stats
+		newstat = true
+	}
+
+	if newstat && sequence.StatsPlusOne(key, sequence) {
+		delete(sequence.Annotations(), key)
+	}
+
+	return stats
+}
+
+func (sequence BioSequence) StatsPlusOne(key string, toAdd BioSequence) bool {
+	if toAdd.HasAnnotation() {
+		stats := sequence.StatsOn(key)
+		value, ok := toAdd.Annotations()[key]
+
+		if ok {
+			var sval string
+
+			switch value := value.(type) {
+			case string:
+				sval = value
+			case int,
+				uint8, uint16, uint32, uint64,
+				int8, int16, int32, int64, bool:
+				sval = fmt.Sprint(value)
+			default:
+				log.Fatalf("Trying to make stats on a none string, integer or boolean value (%v)", value)
+			}
+			old, ok := stats[sval]
+			if !ok {
+				old = 0
+			}
+			stats[sval] = old + 1
+
+			return true
+		}
+	}
+
+	return false
+}
+
+func (stats StatsOnValues) Merge(toMerged StatsOnValues) StatsOnValues {
+	for k, val := range toMerged {
+		old, ok := stats[k]
+		if !ok {
+			old = 0
+		}
+		stats[k] = old + val
+	}
+
+	return stats
+}
+
+func (sequence BioSequence) Merge(tomerge BioSequence, inplace bool, keys ...string) BioSequence {
+	if !inplace {
+		sequence = sequence.Copy()
+	}
+
+	annotation := sequence.Annotations()
+
+	annotation["count"] = tomerge.Count() + sequence.Count()
+
+	for _, key := range keys {
+		if tomerge.HasStatsOn(key) {
+			smk := sequence.StatsOn(key)
+			mmk := tomerge.StatsOn(key)
+			smk.Merge(mmk)
+		} else {
+			sequence.StatsPlusOne(key, tomerge)
+		}
+	}
+
+	return sequence
+}