爬取知名社区技术文章_items_2

item中定义获取的字段和原始数据进行处理并合法化数据

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

import scrapy
import hashlib
import re
from scrapy.loader.processors import (MapCompose, TakeFirst, Join)
from scrapy.loader import ItemLoader

def go_md5(value):
    # 对cont_url进行md5，作为该表的主键
    m = hashlib.md5()
    if isinstance(value, str):
        m.update(bytes(value, encoding='utf-8'))
        # print(type(m.hexdigest()))
        return m.hexdigest()

def go_time(value):
    # 获取时间，并且格式化时间，raw_t为原始数据，new_t为符合mysql中data类型数据
    raw_t = value.strip()
    if raw_t:
        median_t = raw_t.replace('·', '')
        if median_t:
            time_l = median_t.split('/')
            new_t = '-'.join(time_l)
            return new_t.strip()
        return median_t
    else:
        return raw_t

def go_cont(value):
    # 把文章内容中换行和空格去掉
    return value.strip()

def go_img(value):
    # 确定图片下载器获取的是列表，下载器获取的图片url对象为列表形式
    return value

def get_num(value):
    # 获取评论、点赞、收藏数
    num = re.match(r'.*?(\d+).*?', value)
    if num:
        return int(num.group(1))
    else:
        return 0

class ArticleItemLoader(ItemLoader):
    """
    自定义ItemLoader，要求取每个字段列表中第一个值
    """
    default_output_processor = TakeFirst()

class JobboleItem(scrapy.Item):
    """
    input_processor 数据预处理
    output_processor 数据返回item数据处理
    """

    cont_id = scrapy.Field(
        input_processor=MapCompose(go_md5)

    )
    cont_url = scrapy.Field() #
    title = scrapy.Field()
    publish_time = scrapy.Field(
        input_processor=MapCompose(go_time)
    )
    cont = scrapy.Field(
        input_processor=MapCompose(go_cont),
        output_processor=Join('')
    )
    img_url = scrapy.Field(
        output_processor=MapCompose(go_img)
    )
    link_num = scrapy.Field(
        input_processor=MapCompose(get_num)
    )
    collection_num = scrapy.Field(
        input_processor=MapCompose(get_num)
    )
    comment_num = scrapy.Field(
        input_processor=MapCompose(get_num)
    )
    img_path = scrapy.Field()

# 测试
if __name__ == '__main__':
    result = get_num(' s ss 14 ssss')
    print(result)