scrapy修改源码生成Pipelines、spider文件添加作者时间注释

修改pipeline、item

1、当我们编写的爬虫量非常大时，而且他们都具有特定的规律，每次我们都需要复制相同的pipelines元素，或者其他重复的工作，那么我们就可以通过修改scrapy源码中的template模板，让他创建时自动生成你需要的Pipelines文件格式，简化工作

class ExcelPipeline(object):
    def __init__(self):
        self.wb = Workbook()
        self.ws = self.wb.active
        self.ws.append(['标题', '来源', '发布时间', '内容链接', '内容'])

    def process_item(self, item, spider):
        line = [
            item['title'], item['source'], item['publicTime'],
            item['url'], item['content']
        ]
        self.ws.append(line)
        return item

    def close_spider(self, spider):
        self.wb.save(spider.zh_name + '(' + str(datetime.date.today()) + ').xlsx')
        self.wb.close()


class EmailPipeline(object):
    def process_item(self, item, spider):
        return item

    def close_spider(self, spider):
        mailer = MailSender(smtphost="smtp.163.com",
                            mailfrom="[email protected]",
                            smtppass="FVDYFCDKVXGTVKNA",
                            smtpuser="[email protected]",
                            smtpport=25,
                            smtptls=True
                            )
        body = '招标邮件，及时查收'
        subject = spider.output_excel_filename
        attach_mime = "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
        f = open(spider.output_excel_filename, "rb")
        return mailer.send(to=["[email protected]"], subject=subject, body=body, cc=["[email protected]"],
                           attachs=[(spider.output_excel_filename, attach_mime, f)], mimetype="text/plain", charset='utf-8')


class KafkaPipeline(object):
    def open_spider(self, spider):
        self.producer = KafkaProducer(bootstrap_servers=['sentiment01:9092', 'sentiment03:9092'], value_serializer=lambda m: json.dumps(m).encode('ascii'))

    def process_item(self, item, spider):
        item['index'] = ELASTICSEARCH_INDEX
        self.producer.send('sentiment', dict(item))
        return item

    def close_spider(self, spider):
        self.producer.flush()
        self.producer.close()


class MongoPipeline(object):
    def open_spider(self, spider):
        pass

    def process_item(self, item, spider):
        return item

    def close_spider(self, spider):
        pass


class ElasticsearchPipeline(object):
    def open_spider(self, spider):
        self.es = Elasticsearch(([{"host": ELASTICSEARCH_HOST, "port": str(ELASTICSEARCH_PORT)}]))

    def process_item(self, item, spider):
        actions = [
            {
                '_op_type': 'index',
                '_index': ELASTICSEARCH_INDEX,
                '_type': ELASTICSEARCH_TYPE,
                '_source': dict(item)
            }
        ]
        elasticsearch.helpers.bulk(self.es, actions)  
        return item

    def close_spider(self, spider):
        pass


class RedisPipeline(object):
    def open_spider(self, spider):
        spider.duplicate = Duplicate(spider.name)
        spider.duplicate.find_all_url(index=ELASTICSEARCH_INDEX, doc_type=ELASTICSEARCH_TYPE, source='url')

    def process_item(self, item, spider):
        return item

    def close_spider(self, spider):
        print('爬虫关闭')
        r = redis.Redis(host=REDIS_HOST, port=str(REDIS_PORT), db=0)
        r.delete(spider.name)
分析：如果每个爬虫都需要这些pipelines，那么我们可以在源码中提前加上

我们可以直接在这里，提前修改好你需要的文件格式。

方式二：

你可以把公共的代码抽离出来，放在一个包下，然后在pipelines文件中直接饮用即可

添加作者注释

scrapy-commands-genspiders.py文件下修改如下：

添加时间注释：

三、生成spider

背景：scrapy startproject xxx时创建自己的模板文件或修改原有模板
1.在python37\Lib\site-packages\scrapy\templates\project\module添加或修改模板文件


2.在startproject时添加须在D:\tools\python\python37\Lib\site-packages\scrapy\commands\startproject.py

1. TEMPLATES_TO_RENDER = (
2. ('scrapy.cfg',),
3. ('${project_name}', 'settings.py.tmpl'),
4. ('${project_name}', 'items.py.tmpl'),
5. ('${project_name}', 'pipelines.py.tmpl'),
6. ('${project_name}', 'middlewares.py.tmpl'),
7. //新添加（spider文件）
8. ('${project_name}', 'spiders\\spider_news.py.tmpl'),
9. ('${project_name}', 'spiders\\spider_paper.py.tmpl'),
10. )