如何使用 Scrapy 抓取网站所有页面上的链接

Question

我正在学习scrapy，我正在尝试提取所有包含：“http://lattes.cnpq.br/andasequenceofnumbers”的链接，例如：http://lattes.cnpq.br/0281123427918302 但我不知道网站上包含这些信息的页面是什么。 例如这个网站：

http://www.ppgcc.ufv.br/

我想要的链接在这个页面上：

http://www.ppgcc.ufv.br/?page_id=697

我能做什么？ 我正在尝试使用规则，但我不知道如何正确使用正则表达式。 谢谢

1 编辑----

我需要在主 (ppgcc.ufv.br) 站点的所有页面上搜索链接类型 (http://lattes.cnpq.br/asequenceofnumber)。我的目标是获取所有链接 lattes.cnpq.br/numbers 但我不知道它们在哪里。我使用的是一个简单的代码，实际上是这样的：

class ExampleSpider(scrapy.Spider):
    name = "example"
    allowed_domains = ["ppgcc.ufv.br"]
    start_urls = (
        'http://www.ppgcc.ufv.br/',
    )
    rules = [Rule(SgmlLinkExtractor(allow=[r'.*']), follow=True),
             Rule(SgmlLinkExtractor(allow=[r'@href']), callback='parse')]

    def parse(self, response):
        filename = str(random.randint(1, 9999))
        open(filename, 'wb').write(response.body)

#I'm trying to understand how to use rules correctly

2 编辑----

使用：

class ExampleSpider(CrawlSpider):
    name = "example"
    allowed_domains = [".ppgcc.ufv.br"]
    start_urls = (
        'http://www.ppgcc.ufv.br/',
    )
    rules = [Rule(SgmlLinkExtractor(allow=[r'.*']), follow=True),
            Rule(SgmlLinkExtractor(allow=[r'@href']), callback='parse_links')]
    def parse_links(self, response):
        filename = "Lattes.txt"
        arquivo = open(filename, 'wb')
        extractor = LinkExtractor(allow=r'lattes\.cnpq\.br/\d+')
        for link in extractor.extract_links(response):
            url = link.urlextractor = LinkExtractor(allow=r'lattes\.cnpq\.br/\d+')
            arquivo.writelines("%s\n" % url)                
            print url

它告诉我：

C:\Python27\Scripts\tutorial3>scrapy crawl example
2015-06-02 08:08:18-0300 [scrapy] INFO: Scrapy 0.24.6 started (bot: tutorial3)
2015-06-02 08:08:18-0300 [scrapy] INFO: Optional features available: ssl, http11
2015-06-02 08:08:18-0300 [scrapy] INFO: Overridden settings: {'NEWSPIDER_MODULE': 'tutorial3.spiders', 'SPIDER_MODULES': ['tutorial3
.spiders'], 'BOT_NAME': 'tutorial3'}
2015-06-02 08:08:19-0300 [scrapy] INFO: Enabled extensions: LogStats, TelnetConsole, CloseSpider, WebService, CoreStats, SpiderState

2015-06-02 08:08:19-0300 [scrapy] INFO: Enabled downloader middlewares: HttpAuthMiddleware, DownloadTimeoutMiddleware, UserAgentMidd
leware, RetryMiddleware, DefaultHeadersMiddleware, MetaRefreshMiddleware, HttpCompressionMiddleware, RedirectMiddleware, CookiesMidd
leware, ChunkedTransferMiddleware, DownloaderStats
2015-06-02 08:08:19-0300 [scrapy] INFO: Enabled spider middlewares: HttpErrorMiddleware, OffsiteMiddleware, RefererMiddleware, UrlLe
ngthMiddleware, DepthMiddleware
2015-06-02 08:08:19-0300 [scrapy] INFO: Enabled item pipelines:
2015-06-02 08:08:19-0300 [example] INFO: Spider opened
2015-06-02 08:08:19-0300 [example] INFO: Crawled 0 pages (at 0 pages/min), scraped 0 items (at 0 items/min)
2015-06-02 08:08:19-0300 [scrapy] DEBUG: Telnet console listening on 127.0.0.1:6023
2015-06-02 08:08:19-0300 [scrapy] DEBUG: Web service listening on 127.0.0.1:6080
2015-06-02 08:08:19-0300 [example] DEBUG: Crawled (200) <GET http://www.ppgcc.ufv.br/> (referer: None)
2015-06-02 08:08:19-0300 [example] DEBUG: Filtered offsite request to 'www.cgu.gov.br': <GET http://www.cgu.gov.br/acessoainformacao
gov/>
2015-06-02 08:08:19-0300 [example] DEBUG: Filtered offsite request to 'www.brasil.gov.br': <GET http://www.brasil.gov.br/>
2015-06-02 08:08:19-0300 [example] DEBUG: Filtered offsite request to 'www.ppgcc.ufv.br': <GET http://www.ppgcc.ufv.br/>
2015-06-02 08:08:19-0300 [example] DEBUG: Filtered offsite request to 'www.ufv.br': <GET http://www.ufv.br/>
2015-06-02 08:08:19-0300 [example] DEBUG: Filtered offsite request to 'www.dpi.ufv.br': <GET http://www.dpi.ufv.br/>
2015-06-02 08:08:19-0300 [example] DEBUG: Filtered offsite request to 'www.portal.ufv.br': <GET http://www.portal.ufv.br/?page_id=84
>
2015-06-02 08:08:19-0300 [example] DEBUG: Filtered offsite request to 'www.wordpress.org': <GET http://www.wordpress.org/>
2015-06-02 08:08:19-0300 [example] INFO: Closing spider (finished)
2015-06-02 08:08:19-0300 [example] INFO: Dumping Scrapy stats:
        {'downloader/request_bytes': 215,
         'downloader/request_count': 1,
         'downloader/request_method_count/GET': 1,
         'downloader/response_bytes': 18296,
         'downloader/response_count': 1,
         'downloader/response_status_count/200': 1,
         'finish_reason': 'finished',
         'finish_time': datetime.datetime(2015, 6, 2, 11, 8, 19, 912000),
         'log_count/DEBUG': 10,
         'log_count/INFO': 7,
         'offsite/domains': 7,
         'offsite/filtered': 42,
         'request_depth_max': 1,
         'response_received_count': 1,
         'scheduler/dequeued': 1,
         'scheduler/dequeued/memory': 1,
         'scheduler/enqueued': 1,
         'scheduler/enqueued/memory': 1,
         'start_time': datetime.datetime(2015, 6, 2, 11, 8, 19, 528000)}
2015-06-02 08:08:19-0300 [example] INFO: Spider closed (finished)

我在看网站的源代码，有更多的页面链接没有抓取到，可能是我的规则不正确

Answer 1

所以，首先有几件事：

1) rules 属性仅在您扩展 CrawlSpider 类时有效，如果您扩展更简单的 scrapy.Spider，它们将不起作用。

2) 如果你走rules 和CrawlSpider 路由，你不应该覆盖默认的parse 回调，因为默认实现是实际调用规则的东西——所以你想为你的回调。

3) 要实际提取您想要的链接，您可以在回调中使用LinkExtractor 从页面中抓取链接：

from scrapy.contrib.linkextractors import LinkExtractor

class MySpider(scrapy.Spider):
    ...

    def parse_links(self, response):
        extractor = LinkExtractor(allow=r'lattes\.cnpq\.br/\d+')
        for link in extractor.extract_links(response):
            item = LattesItem()
            item['url'] = link.url

希望对你有帮助。