tsvのファイル名とカラム名の一覧をtsvで出力するコマンド

tsv_column_print.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
import os
import re
import sys


def read_first_line(filename, filepath):
    filepath = os.path.join(filepath, filename)
    try:
        with open(filepath, mode='r') as f:
            line = f.readline()
            while line and line[0] == '#':
                line = f.readline()
            return line
    except IOError:
        print(u'Not allowed to read file to this path "{}".'
                .format(filepath))


if __name__ == '__main__':
    scriptdir = os.path.abspath(os.path.dirname(__file__))
    masterdir = os.path.join(scriptdir, u'your/tsv/dir')
    for masterfile in [f for f in os.listdir(masterdir) if re.match('^[a-z_]+\.tsv$', f)]:
        firstline = read_first_line(masterfile, masterdir)
        if not firstline:
            continue
        columns = firstline.rstrip('\n').split('\t')
        res = ['{}\t{}'.format(masterfile, columns[0])]
        for column in columns[1:]:
            res.append('\t{}'.format(column))
        print '\n'.join(res)