Penser à gérer le case de délimiteurs multiples qui s'enchaînent et qui ne doivent être pris en compte que pour 1 seul

csvparser.go

@@ -7,28 +7,22 @@ import (
 )
 
 type CsvParser struct {
-	enclosers []string
-	delimiter rune
-	fields    []string
+	enclosers  []string
+	delimiters string
+	fields     []string
 }
 
-func (p *CsvParser) Initialize(delimiter string, enclosers []string, lineFormat string) error {
-	if utf8.RuneCountInString(delimiter) != 1 {
-		return fmt.Errorf("delimiter shoud be one character")
+func (p *CsvParser) Initialize(delimiters string, enclosers []string, lineFormat string) error {
+	if utf8.RuneCountInString(delimiter) == 0 {
+		return fmt.Errorf("delimiter shoud be at least one character")
 	}
-	p.enclosers = make([]string, 0)
+	p.delimiters = delimiters
 	for _, encloser := range enclosers {
 		if utf8.RuneCountInString(encloser) != 2 {
 			return fmt.Errorf("encolser should have to characters")
 		}
 	}
 	p.enclosers = enclosers
-	p.delimiter = []rune(delimiter)[0]
-	for _, pair := range enclosers {
-		if utf8.RuneCountInString(pair) != 2 {
-			return fmt.Errorf("encoloser should contain two characters: %s", pair)
-		}
-	}
 	// line format is in the form of: field1 field2 ignore ...
 	// if field name is ignore, it is parsed but not retained
 	p.fields = strings.Split(lineFormat, " ")
@@ -40,56 +34,60 @@ func (p *CsvParser) Parse(line string) (map[string]string, error) {
 	currentFieldIndex := 0
 	valueStart := 0
 	escape := false
-  enclosed := false
+	enclosed := false
 
 	ret := make(map[string]string)
 	indexMax := len(line) - 1
-  maxFieldIndex := len(p.fields) - 1
-  delimiter := p.delimiter
+	maxFieldIndex := len(p.fields) - 1
+	delimiters := p.delimiters
 	for index, r := range line {
-    if r == '\\' {
-      // Check if EOL before continue
-      escape=true
-      continue
-    }
-    if escape {
-      escape=false
-      continue
-    }
-    if r == delimiter {
-      if p.fields[currentFieldIndex] != "ignore" {
-        ret[p.fields[currentFieldIndex]] = line[valueStart : index]
-      }
-      currentFieldIndex++
-      if currentFieldIndex > maxFieldIndex {
-        break
-      }
-      valueStart = index + 1
-      if enclosed {
-        enclosed=false
-        delimiter = p.delimiter
-        // Omit next delimiter
-        escape=true
-      }
-      continue
-    }
-    if index >= indexMax {
-      if p.fields[currentFieldIndex] != "ignore" {
-			  ret[p.fields[currentFieldIndex]] = line[valueStart:]
-      }
-      continue
-    }
+		if r == '\\' {
+			// Check if EOL before continue
+			escape = true
+			continue
+		}
+		if escape {
+			escape = false
+			continue
+		}
+		for _, d := range delimiters {
+      delimiter = true
+		}
+		if delimiter {
+			if p.fields[currentFieldIndex] != "ignore" {
+				ret[p.fields[currentFieldIndex]] = line[valueStart:index]
+			}
+			currentFieldIndex++
+			if currentFieldIndex > maxFieldIndex {
+				break
+			}
+			valueStart = index + 1
+			if enclosed {
+				enclosed = false
+				delimiter = p.delimiter
+				// Omit next delimiter
+				escape = true
+			}
+			continue
+		}
 
-    for _, encloser := range p.enclosers {
-      runes := []rune(encloser)
-      if r == runes[0] {
-        // opening encloser
-        enclosed = true
-        delimiter = runes[1]
-        valueStart++
-        break
-      }
-    }
-  }
+		if index >= indexMax {
+			if p.fields[currentFieldIndex] != "ignore" {
+				ret[p.fields[currentFieldIndex]] = line[valueStart:]
+			}
+			continue
+		}
+
+		for _, encloser := range p.enclosers {
+			runes := []rune(encloser)
+			if r == runes[0] {
+				// opening encloser
+				enclosed = true
+				delimiters = runes[1]
+				valueStart++
+				break
+			}
+		}
+	}
 	return ret, nil
 }