采集人物经历来佐证子平术

最新推荐文章于 2024-09-30 02:55:07 发布

warrah

最新推荐文章于 2024-09-30 02:55:07 发布

阅读量563

点赞数

分类专栏：岁月云——python 文章标签： python

本文链接：https://blog.csdn.net/warrah/article/details/122736653

版权

岁月云——python 专栏收录该内容

66 篇文章 0 订阅

订阅专栏

见《宋书·范晔传》：“言之皆有实证，非为空谈。”子平有较高的或然率，但如果没有人物经历来佐证，就变成三教九流，成为“玄学”实在可惜。老外搞个mbti性格测试就巴巴说是科学，有智慧的老前辈总结的经验，因为不懂而无法传承，散落在明间成为偷偷么么被人看不起，实在是看不过去。
有时候感觉西方人很笨，调研70多个家庭跟踪他们的一生，然后给出结论发表论文。然后中国的学生就认为人家严谨有科学研究精神，何曾想过这70个样本想对人类这么大基数根本不值得一提。另外一个视角，研究问题真的需要采用这么笨的方法吗？梁湘润大师等都说看一个人生的70%就不错了，是的人的极限估计也差不多是这些，另外20~30%还需要靠国运、环境和个人修养，人之一生岂是几百页的子平能讲完的呢，那岂不是白活。
我是会一些计算机，另外对梁老的一些观点特别认同，大部分人都是普通人，每天为了家计小常奔波，为了妻财子禄寿而焦虑，要想做人上人，你真的能承受他们所经历的吗？
这篇博客主要是从百度百科上采集人物经历，后面再结合子平理论进行实证。

1 采集人物信息

# -*- coding: utf-8 -*-
# @time    : 2022/1/22 11:03
# @author  : dzm
# @dsec    : 百度娱乐人物
from sqlalchemy.engine import create_engine
from personspider.settings import MYSQL_CONFIG
import scrapy
from pyquery import PyQuery as pq
from personspider.utils import str_util,person_util
from personspider.items.baidu_person import BaiduPersonItem,BaiduPersonExperienceItem\
    ,BaiduUrlItem,BaiduPersonRelationItem
from personspider.service.baidu_service import BaiduUrlService
import re

class yulespider(scrapy.Spider):
    name = 'baidu_yule'

    def start_requests(self):
        # 从数据库中读取链接
        engine = create_engine('mysql+pymysql://{}:{}@{}:3306/{}'.format(MYSQL_CONFIG['user'], MYSQL_CONFIG['password'],
                                                                         MYSQL_CONFIG['host'], MYSQL_CONFIG['db']),
                               connect_args={'charset': 'utf8'}, pool_size=MYSQL_CONFIG['pool_size'])
        baiduUrlService = BaiduUrlService(engine)
        urls = baiduUrlService.get_urls()
        if urls:
            for url in urls:
                yield scrapy.Request(url=url.url, callback=self.parse,dont_filter=True)

    def parse(self, response):
        cur_url = response.request.url
        cur_url = person_util.get_url(cur_url)
        soup = pq(response.text)
        # 人物基本信息
        basicInfo_blocks = soup('.basic-info .basicInfo-block')
        item = {}
        for basicInfo_block in basicInfo_blocks:
            size = pq(basicInfo_block)('dt').size()
            for i in range(size):
                name = pq(basicInfo_block)('dt:eq({})'.format(i)).text()
                name = str_util.clear(name)
                value = pq(basicInfo_block)('dd:eq({})'.format(i)).text()
                value = str_util.clear(value)
                item[name] = value
        person_item = self.pack_person(item, r'百度百科', cur_url)
        cur_name = person_item['cn_name']
        if 'birthday' in person_item.keys():
            print('中文名：{}, 出生日期:{},链接:{}'.format(person_item['cn_name'],person_item['birthday'],person_item['url']))
        else:
            print('中文名：{}, 链接:{}'.format(person_item['cn_name'],person_item['url']))
        # 出生日期不能为空，且需要有完整的年月日从中获取人物经历
        valid_person = person_item and 'birthday' in person_item.keys() \
                       and person_item['birthday'] \
                       and re.match('\d{4}[年\-]\d{1,2}[月\-]\d{1,2}日?',person_item['birthday'])
        if valid_person:
            person_id = person_item['id']
            yield person_item
            # 人物经历
            paras = soup('.para')
            for para in paras:
                content = pq(para).text()
                if re.match('^\d{4}年',content) and not re.match('^\d{4}年\d{1,2}月\d{1,2}日',content):
                    experiences = person_util.get_experience(content)
                    if experiences:
                        for experience in experiences:
                            if experience and experience['experience']:
                                exp_item = BaiduPersonExperienceItem()
                                exp_item['id'] = str_util.gen_md5(experience)
                                exp_item['person_id'] = person_id
                                exp_item['year'] = experience['year']
                                if 'month' in experience.keys():
                                    exp_item['month'] = experience['month']
                                exp_item['experience'] = experience['experience']
                                yield exp_item
        # 正在采集的链接
        curl_url_item = BaiduUrlItem()
        curl_url_item['id'] = str_util.gen_md5(cur_url)
        curl_url_item['url'] = cur_url
        curl_url_item['status'] = '1'
        curl_url_item['name'] = name
        yield curl_url_item
        # 人物关系
        relations = soup('.relations li')
        if relations:
            for i in range(len(relations)):
                relation = relations[i]
                url = 'https://baike.baidu.com' + pq(relation)('a').attr('href')
                url = person_util.get_url(url)
                # 人物关系取值各有不同
                name = pq(relation)('.title').text()
                if name:
                    tag = pq(relation)('.name').text()
                else:
                    name = pq(relation)('.name').attr('title')
                    tag = pq(relation)('.name').text()
                    tag = tag[:len(tag)-len(name)]
                if valid_person:
                    # 关系
                    relation_item = BaiduPersonRelationItem()
                    relation_item['one'] = person_id
                    relation_item['one_name'] = cur_name
                    relation_item['one_url'] = cur_url
                    relation_item['two'] = str_util.gen_md5(url)
                    relation_item['two_name'] = name
                    relation_item['two_url'] = url
                    relation_item['relation'] = tag
                    yield relation_item
                # 链接
                url_item = BaiduUrlItem()
                url_item['id'] = str_util.gen_md5(url)
                url_item['url'] = url
                url_item['status'] = '0'
                url_item['name'] = name
                yield url_item



    def pack_person(self,content,source, url):
        if content:
            item = BaiduPersonItem()
            item['source'] = source
            item['url'] = url
            item['id'] = str_util.gen_md5(url)
            for key in content.keys():
                if key == '中文名':
                    item['cn_name'] = content[key]
                elif key == '外文名':
                    item['en_name'] = content[key]
                elif key == '性别':
                    item['sex'] = content[key]
                elif key == '国籍':
                    item['nation'] = content[key]
                elif key == '出生日期':
                    birthday = re.search(r'\d{4}[年\-]\d{1,2}[月\-]\d{1,2}日?',content[key], re.S)
                    if birthday:
                        item['birthday'] = birthday.group(0)
                elif key == '出生地':
                    item['birthplace'] = content[key]
                elif key == '外文名':
                    item['deathday'] = content[key]
                elif key == '身高':
                    item['height'] = person_util.get_height(content[key])
                elif key == '毕业院校':
                    item['school'] = content[key]
                elif key == '职业':
                    item['occupation'] = content[key]
                elif key == '主要成就':
                    item['achievements'] = content[key]
                elif key == '代表作品':
                    item['representation'] = content[key]
            return item
        return None



if __name__ == '__main__':
    pass

2 提取内容

import re
from personspider.utils import str_util


def get_url(url):
    index = url.index('?') if '?' in url else None
    if index:
        return url[:index]
    else:
        return url

def get_experience(text):
    '''
    获取经历
    '''
    pattern = r'(\d{4}年)'
    results = re.split(pattern,text,re.S)
    size = len(results)
    contents = []
    i = 1
    while i < size-1:
        # 年份
        year = results[i]
        # 经历
        experience = results[i+1].strip('，') #去掉首字母，
        experience = str_util.clear(experience)
        # 月
        result = re.search(r'\d{1,2}月', experience)
        if result:
            months = re.split(r'(\d{1,2}月)',experience,re.S)
            j = 1
            while j<len(months)-1:
                month = months[j]
                experience = months[j+1].strip('，')
                experience = str_util.clear(experience)
                contents.append({'year':year,'month':month,'experience':experience})
                j = j+2
        else:
            contents.append({'year':year,'experience':experience})
        i = i+2
    return contents

def get_height(height):
    height = re.search(r'\d{1,4}(cm)?',height, re.S)
    if height:
        return height.group(0)
    else:
        return height

import hashlib
import re

def gen_md5(item):
    '''
    将字符串转md5
    '''
    m = hashlib.md5()
    md5 = str(item).encode('utf-8')
    m.update(md5)
    md5 = m.hexdigest()
    return md5

def remove_xa0(value):
    '''
    \xa0 是不间断空白符 &nbsp;
    '''
    return value.replace(u'\xa0',u'')

def remove_quote(value):
    p = re.compile('\[[\d\-\]]+')
    return p.sub("",value)

def remove_blank(value):
    return value.replace(' ','')

def clear(value):
    value = remove_xa0(value)
    value = remove_quote(value)
    value = remove_blank(value)
    return value

3 数据管道

from sqlalchemy.engine import create_engine
from personspider.items.baidu_person import BaiduPersonItem,BaiduPersonExperienceItem\
    ,BaiduUrlItem,BaiduPersonRelationItem
from personspider.service.baidu_service import BaiduPersonService,BaiduPersonExperienceService\
    ,BaiduPerson,BaiduPersonExperience,BaiduPersonRelation,BaiduUrl\
    ,BaiduPersonRelationService,BaiduUrlService


class MysqlPipeline(object):

    def __init__(self, engine):
        self.baiduPersonService = BaiduPersonService(engine)
        self.baiduPersonExperienceService = BaiduPersonExperienceService(engine)
        self.baiduUrlService = BaiduUrlService(engine)
        self.baiduPersonRelationService = BaiduPersonRelationService(engine)

    def process_item(self, item, spider):
        if type(item) == BaiduPersonItem:
            record = BaiduPerson(**item)
            self.baiduPersonService.insert(record)
        elif type(item) == BaiduPersonExperienceItem:
            record = BaiduPersonExperience(**item)
            self.baiduPersonExperienceService.insert(record)
        elif type(item) == BaiduUrlItem:
            record = BaiduUrl(**item)
            self.baiduUrlService.insert(record)
        elif type(item) == BaiduPersonRelationItem:
            record = BaiduPersonRelation(**item)
            self.baiduPersonRelationService.insert(record)




    @classmethod
    def from_settings(cls,settings):
        mysql_config = settings.get('MYSQL_CONFIG')
        engine = create_engine('mysql+pymysql://{}:{}@{}:3306/{}'.format(mysql_config['user'], mysql_config['password'],
                                                                         mysql_config['host'], mysql_config['db']),
                               connect_args={'charset': 'utf8'}, pool_size=mysql_config['pool_size'])
        return cls(engine)

4 写数据库

class BaiduPersonService(object):

    def __init__(self, engine):
        self.engine = engine
        Session = sessionmaker(engine)
        self.session = Session()
        self.emailService = EmailService()

    def exist(self, id):
        query = self.session.query(BaiduPerson).filter(BaiduPerson.id==id)
        return query.count()>0

    def insert(self, record):
        if not self.exist(record.id):
            try:
                record.create_time = datetime.datetime.now()
                self.session.add(record)
                self.session.commit()
            except Exception as e:
                title = r'{}写入数据库失败'.format(record.cn_name)
                content = r'ERROR {}'.format(str(e))
                self.emailService.sendEmail(title,content)

5 异常邮件发送
如果在解析过程中出现异常，总不需要时刻盯着吧，写个发邮件告知我，岂不是很安逸

import smtplib
from email.header import Header
from email.mime.text import MIMEText
from personspider.settings import EMAIL_CONFIG

class EmailService(object):

    def sendEmail(self,title, content):

        message = MIMEText(content, 'plain', 'utf-8')  # 内容, 格式, 编码
        message['From'] = "{}".format(EMAIL_CONFIG['sender'])
        message['To'] = ",".join(EMAIL_CONFIG['receivers'])
        message['Subject'] = title

        try:
            smtpObj = smtplib.SMTP_SSL(EMAIL_CONFIG['smtp']['host'], 465)  # 启用SSL发信, 端口一般是465
            smtpObj.login(EMAIL_CONFIG['smtp']['user'], EMAIL_CONFIG['smtp']['password'])  # 登录验证
            smtpObj.sendmail(EMAIL_CONFIG['sender'], EMAIL_CONFIG['receivers'], message.as_string())  # 发送
            print("mail has been send successfully.")
        except smtplib.SMTPException as e:
            print(e)

    def send_email2(SMTP_host, from_account, from_passwd, to_account, subject, content):
        email_client = smtplib.SMTP(SMTP_host)
        email_client.login(from_account, from_passwd)
        # create msg
        msg = MIMEText(content, 'plain', 'utf-8')
        msg['Subject'] = Header(subject, 'utf-8')  # subject
        msg['From'] = from_account
        msg['To'] = to_account
        email_client.sendmail(from_account, to_account, msg.as_string())

        email_client.quit()

if __name__ == '__main__':
    emailService = EmailService()
    title = r'数据库异常'
    content = r'很多多多问题'
    emailService.sendEmail(title, content)