func initBufferedReader(file string, bufferSize int, chunkSize int, fn func(line string) (interface{}, bool, error)) (*BufferedReader, error) {
if bufferSize < 1 {
bufferSize = 1
}
if chunkSize < 1 {
chunkSize = 1
}
reader := new(BufferedReader)
fh, err := xopen.Ropen(file)
if err != nil {
return nil, err
}
reader.reader = fh
reader.BufferSize = bufferSize
reader.ChunkSize = chunkSize
reader.ProcessFunc = fn
reader.Ch = make(chan Chunk, bufferSize)
reader.done = make(chan struct{})
reader.finished = false
reader.cancelled = false
return reader, nil
}
// NewReader is constructor of FASTX Reader.
//
// Parameters:
//
// t sequence alphabet
// if nil is given, it will guess alphabet by the first record
// file file name, "-" for stdin
// idRegexp id parsing regular expression string, must contains "(" and ")" to capture matched ID
// "" for default value: `^([^\s]+)\s?`
// if record head does not match the idRegxp, whole name will be the id
//
func NewReader(t *seq.Alphabet, file string, idRegexp string) (*Reader, error) {
var r *regexp.Regexp
if idRegexp == "" {
r = regexp.MustCompile(DefaultIDRegexp)
isUsingDefaultIDRegexp = true
} else {
if !reCheckIDregexpStr.MatchString(idRegexp) {
return nil, fmt.Errorf(`fastx: regular expression must contain "(" and ")" to capture matched ID. default: %s`, DefaultIDRegexp)
}
var err error
r, err = regexp.Compile(idRegexp)
if err != nil {
return nil, fmt.Errorf("fastx: fail to compile regexp: %s", idRegexp)
}
if idRegexp == DefaultIDRegexp {
isUsingDefaultIDRegexp = true
}
}
fh, err := xopen.Ropen(file)
if err != nil {
return nil, fmt.Errorf("fastx: %s", err)
}
fastxReader := &Reader{
fh: fh,
buf: make([]byte, pageSize),
t: t,
IDRegexp: r,
firstseq: true,
checkSeqType: true,
}
fastxReader.buffer = bytes.NewBuffer(make([]byte, 0, defaultBytesBufferSize))
fastxReader.seqBuffer = bytes.NewBuffer(make([]byte, 0, defaultBytesBufferSize))
fastxReader.qualBuffer = bytes.NewBuffer(make([]byte, 0, defaultBytesBufferSize))
return fastxReader, nil
}
