xcut

#!/usr/bin/env python3
import click
from functools import reduce
import sys
import csv,json
import re


def rowsReader(file, delimiter):
    for line in file:
        row = line.strip().split(delimiter)
        yield row


def jsonReader(file):
    for line in file:
        row = line.strip()
        yield json.loads(row)



help_text = """
Example: 

$ cat test.csv
name,gender,job
Jack,male,coder
Lucy,female,artist

$ cat test.csv | xcut -f job,name 
job,name
coder,Jack
artist,Lucy

$ xcut -f job,name -od $'\\t' test.csv
job       name
--------------------
coder     Jack
artist    Lucy
    """


@click.command()
@click.option('-f', '--fields', default=None, help='Select fields')
@click.option('-d', '--delimiter', default=',', help='Set delimiter, `,` is default delimiter')
@click.option('--titles', default=None, help='Set all titles')
@click.option('--from-format', default='csv', help="Use csv parser")
@click.option('--to-csv', is_flag=True, help="Output with csv format")
@click.option('-t', '--tt', default='head', help='Set title type: head, index, kv')
@click.option('-h', '--help',  is_flag=True, help='help')
@click.option('--no-title', is_flag=True, help='Do not print title')
@click.option('-od', '--out-delimiter', default=',', help='Set output delimiter')
@click.option('--pretty', is_flag=True, help='Human pretty print')
@click.argument('file', default=sys.stdin, type=click.File('r'))
def xcut(file, tt, no_title, pretty, delimiter, out_delimiter, fields, titles, from_format, to_csv, help):
    f'''
    {help_text}
    '''
    if help:
        ctx = click.get_current_context()
        print(ctx.get_help())
        quit(help_text)
    if fields is None:
        quit(f'For help:\n xcut --help')

    tab = 8
    if from_format == 'csv': # index
        rows = csv.reader(file, delimiter=delimiter)
    elif from_format == 'json': # kvjson
        rows = jsonReader(file)
    else: # kvEqual: k=v
        rows = rowsReader(file, delimiter=delimiter)
    fields = fields.split(',')
    if from_format ==  'json':
        tt = 'json'
    if tt == 'index': # -f 1,2,3
        indexes = []
        for i in fields:
            if i.isdigit():
                indexes.append(int(i)-1)
            elif re.fullmatch(r'\d+-\d+', i):
                start, end = [int(i)-1 for i in i.split('-')]
                if end - start < 100:
                    indexes += list(range(start, end+1))
        fields = [str(i+1) for i in indexes]
    elif tt == 'kv': # -f:k1,k2 k1=v1,k2=v2
        pass
    elif tt == "head": # -f: name,age,gender
        if titles:
            titles = titles.split(',')
        else:
            titles = next(rows)
        indexes = [titles.index(f) if f in titles else None for f in fields]

    if to_csv:
        writer = csv.writer(sys.stdout, delimiter=out_delimiter[0],
                            quotechar='"', quoting=csv.QUOTE_MINIMAL, lineterminator='\n')

    # print titles and pretty '----'
    if not no_title:
        print(out_delimiter.join(fields))
        if pretty:
            print('-'*tab*len(fields))

    for row in rows:
        if tt == 'json': # kv
            # log format: key1=val1,key2=val2,key3=val3....
            data = [row.get(k, '') for k in fields]
        elif tt in ['index','head']: #head+indexes
            # log format: val1,val2,val3....
            #data = dict(zip(titles, row))
            data = ['' if i is None else (
                row[i:i+1] or [''])[0] for i in indexes]
        elif tt == 'kv': # kv
            # log format: key1=val1,key2=val2,key3=val3....
            data = {}
            for item in row:
                if not '=' in item:
                    continue
                k, v = item.split('=', 1)
                k = k.strip()
                if k in fields:
                    data[k] = v
            data = [data.get(k, '') for k in fields]
        else:
            quit(f'bad tt={tt}')
        if to_csv:
            writer.writerow(data)
        else:
            line = out_delimiter.join(data)  # .expandtabs(tab)
            print(line)


if __name__ == '__main__':
    xcut()