重构积分统计程序

原程序:

#!/usr/bin/python
# -*- coding: utf-8 -*-

import json


def read_config(path):
    with open(path, 'r') as load_f:
        load_dict = json.load(load_f)
    # print path
    # print load_dict
    return load_dict


def printData(aggr):
    for data in sorted(aggr.items(), key=lambda item:item[1]):
        print data[0], ":", data[1]


def calculateAge(hukou):
    aggr = {}

    for person in hukou["rows"]:
        year_str = person["idCard"][6:10]
        if aggr.has_key(year_str):
            aggr[year_str] += 1
        else:
            aggr[year_str] = 1

    print printData(aggr)


def calculateCompany(hukou):
    aggr = {}

    for person in hukou["rows"]:
        company = person["unit"]
        if aggr.has_key(company):
            aggr[company] += 1
        else:
            aggr[company] = 1

    printData(aggr)


def calculateLevel(hukou):
    aggr = {}
    aggr["博士"] = 0
    aggr["硕士"] = 0
    aggr["本科"] = 0
    aggr["其他"] = 0

    for person in hukou["rows"]:
        degree = person["s3"]
        if degree == 15:
            aggr["本科"] += 1
        elif degree == 26:
            aggr["硕士"] += 1
        elif degree == 37:
            aggr["博士"] += 1
        else:
            aggr["其他"] += 1

    printData(aggr)


def calculateScore(hukou):
    aggr = {}

    for person in hukou["rows"]:
        score = int(person["score"])
        if aggr.has_key(score):
            aggr[score] += 1
        else:
            aggr[score] = 1

    printData(aggr)


if __name__ == '__main__':
    hukou = read_config('./beijing.json')
    print u"总人数: ", hukou["currentPageSize"]

    # calculateAge(hukou)
    calculateCompany(hukou)
    # calculateLevel(hukou)
    # calculateScore(hukou)

里面重复的分组统计代码看起来实在不舒服, 忍不住要改一下:

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

from collections import Counter
import json


def read_config(path):
    with open(path, 'r') as load_f:
        return json.load(load_f)


def print_data(aggr):
    for data in sorted(aggr.items(), key=lambda item: item[1]):
        print(data[0], ':', data[1])


def iter_age(hukou):
    return (person['idCard'][6:10] for person in hukou['rows'])


def iter_company(hukou):
    return (person['unit'] for person in hukou['rows'])


def iter_level(hukou):
    for person in hukou['rows']:
        degree = person['s3']
        if degree == 15:
            yield '本科'
        elif degree == 26:
            yield '硕士'
        elif degree == 37:
            yield '博士'
        else:
            yield '其他'


def iter_score(hukou):
    return (int(person['score']) for person in hukou['rows'])


def stat(hukou, iter_func):
    return Counter(iter_func(hukou))


if __name__ == '__main__':
    hukou = read_config('./beijing.json')
    print('总人数: ', hukou['currentPageSize'])

    print_data(stat(hukou, iter_age))
    #print_data(stat(hukou, iter_company))
    #print_data(stat(hukou, iter_level))
    #print_data(stat(hukou, iter_score))

改动有:

1. 修改Shebang, 使其适应更多的Posix系统. 不同的环境, python解释器的路径是不一样的.
   默认是 python2 还是 python3 也不一定, 新的 Shebang 指定了 python 的版本.
2. 从 python2 修改为 python3, python3 默认使用 unicode 内码, 中文前面不需要使用 u 前缀了
3. 抽象出提取统计信息的 iter_xxx 函数, 复用统计程序 stat.
4. 使用python3内置的 Counter 类完成计数, 避免手工的循环
5. 使用 snake_case 风格的命名规范代替 camelCase.