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

Make sequence files recognized as a taxonomy

Browse Source
This commit is contained in:
Eric Coissac
2025-03-14 14:22:22 +01:00
parent d1c31c54de
commit 8448783499
21 changed files with 657 additions and 467 deletions
Download Patch File Download Diff File Expand all files Collapse all files

68
pkg/obiformats/universal_read.go
View File

@ -3,11 +3,8 @@ package obiformats
import (
"bufio"
"bytes"
"encoding/csv"
"errors"
"io"
"path"
"regexp"
"github.com/gabriel-vasile/mimetype"
@ -41,70 +38,7 @@ type SequenceReader func(reader io.Reader, options ...WithOption) (obiiter.IBioS
// - io.Reader: A modified reader with the read data.
// - error: Any error encountered during the process.
func OBIMimeTypeGuesser(stream io.Reader) (*mimetype.MIME, io.Reader, error) {
csv := func(in []byte, limit uint32) bool {
in = dropLastLine(in, limit)
br := bytes.NewReader(in)
r := csv.NewReader(br)
r.Comma = ','
r.ReuseRecord = true
r.LazyQuotes = true
r.Comment = '#'
lines := 0
for {
_, err := r.Read()
if errors.Is(err, io.EOF) {
break
}
if err != nil {
return false
}
lines++
}
return r.FieldsPerRecord > 1 && lines > 1
}
fastaDetector := func(raw []byte, limit uint32) bool {
ok, err := regexp.Match("^>[^ ]", raw)
return ok && err == nil
}
fastqDetector := func(raw []byte, limit uint32) bool {
ok, err := regexp.Match("^@[^ ].*\n[A-Za-z.-]+", raw)
return ok && err == nil
}
ecoPCR2Detector := func(raw []byte, limit uint32) bool {
ok := bytes.HasPrefix(raw, []byte("#@ecopcr-v2"))
return ok
}
genbankDetector := func(raw []byte, limit uint32) bool {
ok2 := bytes.HasPrefix(raw, []byte("LOCUS "))
ok1, err := regexp.Match("^[^ ]* +Genetic Sequence Data Bank *\n", raw)
return ok2 || (ok1 && err == nil)
}
emblDetector := func(raw []byte, limit uint32) bool {
ok := bytes.HasPrefix(raw, []byte("ID "))
return ok
}
mimetype.Lookup("text/plain").Extend(fastaDetector, "text/fasta", ".fasta")
mimetype.Lookup("text/plain").Extend(fastqDetector, "text/fastq", ".fastq")
mimetype.Lookup("text/plain").Extend(ecoPCR2Detector, "text/ecopcr2", ".ecopcr")
mimetype.Lookup("text/plain").Extend(genbankDetector, "text/genbank", ".seq")
mimetype.Lookup("text/plain").Extend(emblDetector, "text/embl", ".dat")
mimetype.Lookup("text/plain").Extend(csv, "text/csv", ".csv")
mimetype.Lookup("application/octet-stream").Extend(fastaDetector, "text/fasta", ".fasta")
mimetype.Lookup("application/octet-stream").Extend(fastqDetector, "text/fastq", ".fastq")
mimetype.Lookup("application/octet-stream").Extend(ecoPCR2Detector, "text/ecopcr2", ".ecopcr")
mimetype.Lookup("application/octet-stream").Extend(genbankDetector, "text/genbank", ".seq")
mimetype.Lookup("application/octet-stream").Extend(emblDetector, "text/embl", ".dat")
mimetype.Lookup("application/octet-stream").Extend(csv, "text/csv", ".csv")
obiutils.RegisterOBIMimeType()
// Create a buffer to store the read data
mimetype.SetLimit(1024 * 1024)