readme.md

動きを試してみるとき

$ make one
python fetch.py -c config.json  -x 1 -y 2
id,x,y,x_y_rate
1,1,2,0.5

$ make many
python fetch.py -c config.json -x 1 -x 2 -x 3 -x 4 -x 5 -y 2 -y 1 -y 0 -y 1 -y 2
id,x,y,x_y_rate
1,1,2,0.5
2,4,1,4.0
3,9,0,-
4,16,1,16.0
5,25,2,12.5

$ SIZE=3 make
python fetch.py -c config.json --size=3 --input ./test-input.csv | tee dist/test-output.csv
id,x,y,x_y_rate
1,1,-50,-0.02
2,4,-49,-0.08163265306122448
3,9,-48,-0.1875

全部実行する時

# どういうコマンドが実行されるかは -n 付きで実行
$ make -n
mkdir -p dist
python fetch.py -c config.json  --input ./test-input.csv | tee dist/test-output.csv

# 全部実行
$ PREFIX=test make
id,x,y,x_y_rate
1,1,-50,-0.02
2,4,-49,-0.08163265306122448
3,9,-48,-0.1875
4,16,-47,-0.3404255319148936
5,25,-46,-0.5434782608695652
...
40,1600,-11,-145.45454545454547
41,1681,-10,-168.1
42,1764,-9,-196.0
43,1849,-8,-231.125
ERROR

# resume
$ RESUME=1 PREFIX=test make
44,1936,-7,-276.57142857142856
45,2025,-6,-337.5
46,2116,-5,-423.2
47,2209,-4,-552.25
48,2304,-3,-768.0
49,2401,-2,-1200.5
50,2500,-1,-2500.0
...
96,9216,45,204.8
97,9409,46,204.54347826086956
98,9604,47,204.3404255319149
99,9801,48,204.1875
100,10000,49,204.08163265306123
101,10201,50,204.02

fetch.py

import sys
import itertools
try:
    import csvresumable as csv
except ImportError:
    import csv


def collect(itr):
    for x, y in itr:
        x = int(x)
        y = int(y)
        yield {"id": x, "x": x * x, "y": y}


def run(*, config, size=None, x_list=None, y_list=None, input=None, decoration=None):
    if input is not None:
        r = ((row["x"], row["y"]) for row in csv.DictReader(input))
    else:
        r = zip((x_list or []), (y_list or []))

    if size is not None:
        r = itertools.islice(r, size)

    itr = collect(r)
    if decoration is not None:
        itr = decoration(itr)
    itr = decoration_for_calculation(itr)
    yield from itr


def decoration_for_calculation(itr):
    for row in itr:
        if row["y"] == 0:
            yield row
            continue
        row["x_y_rate"] = row["x"] / row["y"]
        yield row


def main():
    import argparse
    parser = argparse.ArgumentParser()
    parser.add_argument("-c", "--config", required=True)
    parser.add_argument("-x", action="append", dest="x_list")
    parser.add_argument("-y", action="append", dest="y_list")
    parser.add_argument("--size", type=int)
    parser.add_argument("--input", type=argparse.FileType("r"))

    args = parser.parse_args()
    itr = run(**vars(args))

    w = csv.DictWriter(sys.stdout, fieldnames=["id", "x", "y", "x_y_rate"], restval="-")
    w.writeheader()
    w.writerows(itr)


if __name__ == "__main__":
    main()

Makefile

PREFIX ?= test
SIZE ?=
CMD := python fetch.py -c config.json $(addprefix --size=,${SIZE})

default:
	mkdir -p dist
	${CMD} --input ./${PREFIX}-input.csv | tee dist/${PREFIX}-output.csv

one:
	${CMD} -x 1 -y 2

many:
	${CMD} -x 1 -x 2 -x 3 -x 4 -x 5 -y 2 -y 1 -y 0 -y 1 -y 2

test-input.csv