replacing internal spaces and tabs by _ in phylip conversion

align/align.go

@@ -7,6 +7,7 @@ import (
 	"log"
 	"math"
 	"math/rand"
+	"regexp"
 	"sort"
 	"strings"
 
@@ -69,7 +70,7 @@ type Alignment interface {
 	Rename(namemap map[string]string)
 	Pssm(log bool, pseudocount float64, normalization int) (pssm map[rune][]float64, err error) // Normalization: PSSM_NORM_NONE, PSSM_NORM_UNIF, PSSM_NORM_DATA
 	TrimNames(size int) (map[string]string, error)
-	CleanNames() // Removes spaces and tabs from sequence names
+	CleanNames() // Removes spaces and tabs at beginning and end of sequence names and replaces others by "_"
 	TrimSequences(trimsize int, fromStart bool) error
 	AppendSeqIdentifier(identifier string, right bool)
 	AvgAllelesPerSite() float64
@@ -678,9 +679,12 @@ func (a *align) AppendSeqIdentifier(identifier string, right bool) {
 
 // Removes spaces and tabs from sequence names
 func (a *align) CleanNames() {
+	firstlast := regexp.MustCompile("(^[\\s\\t]+|[\\s\\t]+$)")
+	inside := regexp.MustCompile("[\\s\\t]+")
+
 	for _, seq := range a.seqs {
-		seq.name = strings.Replace(seq.name, " ", "", -1)
-		seq.name = strings.Replace(seq.name, "\t", "", -1)
+		seq.name = firstlast.ReplaceAllString(seq.name, "")
+		seq.name = inside.ReplaceAllString(seq.name, "_")
 	}
 }
 

align/align_test.go

@@ -45,6 +45,30 @@ func TestAppendIdentifier(t *testing.T) {
 	})
 }
 
+func TestCleanNames(t *testing.T) {
+	a, err := RandomAlignment(AMINOACIDS, 300, 50)
+	if err != nil {
+		t.Error(err)
+
+	}
+	a2, _ := a.Clone()
+	a.AppendSeqIdentifier("\t \t", false)
+	a.AppendSeqIdentifier("\t \t", true)
+
+	a.CleanNames()
+	i := 0
+	a.IterateChar(func(name string, sequence []rune) {
+		expected, found := a2.GetSequenceNameById(i)
+		if !found {
+			t.Error("Unknown sequence name after clean")
+		}
+		if name != expected {
+			t.Error("Unexpected sequence name after clean")
+		}
+		i++
+	})
+}
+
 func TestRemoveOneGapSite(t *testing.T) {
 	a, err := RandomAlignment(AMINOACIDS, 300, 300)
 	if err != nil {

test.sh

@@ -166,20 +166,20 @@ rm -f expected result mapfile
 echo "->goalign reformat phylip spaces tabs"
 cat > expected <<EOF
    5   5
-Seq0000  GATTA
-Seq0001  ATTTG
-Seq0002  CCGTA
-Seq0003  GGCCA
-Seq0004  GAATC
+S_e_q_0_0_00  GATTA
+Se_q00_0_1  ATTTG
+Se_q00_0_2  CCGTA
+Seq00_03  GGCCA
+Seq_0004  GAATC
 EOF
 cat > input.fa <<EOF
 > S e q 0	0	00
 GATTA
->	Se    q00 0 1     
+> 	 Se    q00 0 1    	 
 ATTTG
->	Se    q00 0 2    
+>	 Se    q00 0 2    	 	
 CCGTA
->Seq00     03
+>		Seq00     03
 GGCCA
 >Seq		0004
 GAATC