obitools/obitools4
2
0
Fork 0
mirror of https://github.com/metabarcoding/obitools4.git synced 2025-06-29 16:20:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

Adds the hash option to obidistribute

Browse Source
This commit is contained in:
Eric Coissac
2022-02-14 09:12:57 +01:00
parent eb32620bb3
commit b931321ba1
3 changed files with 32 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored
View File

@ -12,3 +12,4 @@ vendor
/obipairing
/obipcr
/obifind
/obidistribute

30
pkg/obiformats/dispatcher.go
View File

@ -4,6 +4,7 @@ import (
"fmt"
"log"
"sync"
"sync/atomic"
"git.metabarcoding.org/lecasofts/go/obitools/pkg/obiseq"
)
@ -21,26 +22,25 @@ func WriterDispatcher(prototypename string,
jobDone.Add(1)
go func() {
n := 0
n := int32(0)
for newflux := range dispatcher.News() {
data, _ := dispatcher.Outputs(newflux)
out, err := formater(data,
fmt.Sprintf(prototypename, newflux),
options...)
if err != nil {
log.Fatalf("cannot open the output file for key %s", newflux)
}
go func(newflux string) {
data, _ := dispatcher.Outputs(newflux)
out, err := formater(data,
fmt.Sprintf(prototypename, newflux),
options...)
if err != nil {
log.Fatalf("cannot open the output file for key %s", newflux)
}
n++
atomic.AddInt32(&n, 1)
if n > 1 {
jobDone.Add(1)
}
go func() {
if atomic.LoadInt32(&n) > 1 {
jobDone.Add(1)
}
out.Recycle()
jobDone.Done()
}()
}(newflux)
}
}()

22
pkg/obitools/obidistribute/options.go
View File

@ -13,20 +13,29 @@ import (
var _FilenamePattern = ""
var _SequenceClassifierTag = ""
var _BatchCount = 0
var _HashSize = 0
func DistributeOptionSet(options *getoptions.GetOpt) {
options.StringVar(&_FilenamePattern, "pattern", _FilenamePattern,
options.Alias("p"),
options.Required("You must provide at pattern for the file names "),
options.Description("The N first sequence records of the file are discarded from the analysis and not reported to the output file."))
options.Description("The template used to build the names of the output files. "+
"The variable part is represented by '%s'. "+
"Example : toto_%s.fastq."))
options.StringVar(&_SequenceClassifierTag, "classifier", _SequenceClassifierTag,
options.Alias("c"),
options.Description("The N first sequence records of the file are discarded from the analysis and not reported to the output file."))
options.Description("The name of a tag annotating thes sequences. "+
"The name must corresponds to a string, a integer or a boolean value. "+
"That value will be used to dispatch sequences amoong the different files"))
options.IntVar(&_BatchCount, "batch", 0,
options.IntVar(&_BatchCount, "batches", 0,
options.Alias("n"),
options.Description("The N first sequence records of the file are discarded from the analysis and not reported to the output file."))
options.Description("Indicates in how many batches the input file must bee splitted."))
options.IntVar(&_HashSize, "hash", 0,
options.Alias("H"),
options.Description("Indicates to split the input into at most <n> batch based on a hash code of the seequence."))
}
func OptionSet(options *getoptions.GetOpt) {
@ -41,10 +50,11 @@ func CLISequenceClassifier() obiseq.SequenceClassifier {
return obiseq.AnnotationClassifier(_SequenceClassifierTag)
case _BatchCount > 0:
return obiseq.RotateClassifier(_BatchCount)
case _HashSize > 0:
return obiseq.HashClassifier(_HashSize)
}
log.Fatal("one of the options --classifier or --batch must be specified")
log.Fatal("one of the options --classifier, -- hash or --batch must be specified")
return nil
}