首页 / 问题 / 正文

Python的Scrapy似乎无法从所有可用的URL获取数据

6 投票
2 回答
2874 浏览
提问于 2025-04-17 07:06

我正在尝试从 thesession.org 网站上抓取数据，想要制作一个表格，记录每首曲子被多少会员添加到他们的曲谱中，这样我就可以找到一些受欢迎的曲子来学习。我从 这个 scrapy 教程 开始，试着修改它以满足我的需求。问题是，虽然 thesession.org 网站上似乎有大约 10,390 首曲子，但我的抓取工具只返回了 10 首曲子的数据（就是在 http://www.thesession.org/tunes/index.php 上的那些）。我该如何获取所有曲子（或者前一百首曲子）的数据呢？任何建议都非常感谢。

这是我目前的进展：

items.py

from scrapy.item import Item, Field

class tuneItem(Item):
    url = Field()
    name1 = Field()
    name2 = Field()
    key = Field()
    count = Field() 
    pass

tune_spider.py

from scrapy.spider import BaseSpider
from scrapy.contrib.spiders import CrawlSpider, Rule
from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
from scrapy.selector import HtmlXPathSelector
from scrapy.item import Item
from tutorial.items import tuneItem
from scrapy.conf import settings

class tunesSpider(CrawlSpider):

    name = "irishtunes"
    allowed_domains = ["thesession.org"]
    start_urls = ["http://www.thesession.org/tunes"]
    rules = [Rule(SgmlLinkExtractor(allow=['/display/\d+'], deny=['/members/','/recordings/','/index/','/display/\d+/.']), 'parse_tune')]

    def parse_tune(self, response):
        x = HtmlXPathSelector(response)

        tune = tuneItem()
        tune['url'] = response.url
        tune['name1'] = x.select("//div[@id='details']//div[@class='box']/h1/text()").extract()
        tune['name2'] = x.select("//div[@id='details']//div[@class='box']/h2/text()").extract()
        tune['key']   = x.select("//div[@id='details']//div[@class='box']/p[1]/text()").extract()
        tune['count'] = x.select("//div[@id='details']//div[@class='box']/p[3]/text()").re('\d+')
        return tune

我通过打开控制台，进入包含教程配置文件的目录，然后运行 scrapy crawl irishtunes --set FEED_URI=scraped_data.csv --set FEED_FORMAT=csv 来启动抓取工具。

这是我得到的结果：

C:\Users\BM\Desktop\scrape\tutorial>scrapy crawl irishtunes --set FEED_URI=scrap
ed_data.csv --set FEED_FORMAT=csv
2011-11-25 22:45:47-0800 [scrapy] INFO: Scrapy 0.14.0.2841 started (bot: tutoria
l)
2011-11-25 22:45:47-0800 [scrapy] DEBUG: Enabled extensions: FeedExporter, LogSt
ats, TelnetConsole, CloseSpider, WebService, CoreStats, SpiderState
2011-11-25 22:45:48-0800 [scrapy] DEBUG: Enabled downloader middlewares: HttpAut
hMiddleware, DownloadTimeoutMiddleware, UserAgentMiddleware, RetryMiddleware, De
faultHeadersMiddleware, RedirectMiddleware, CookiesMiddleware, HttpCompressionMi
ddleware, ChunkedTransferMiddleware, DownloaderStats
2011-11-25 22:45:48-0800 [scrapy] DEBUG: Enabled spider middlewares: HttpErrorMi
ddleware, OffsiteMiddleware, RefererMiddleware, UrlLengthMiddleware, DepthMiddle
ware
2011-11-25 22:45:48-0800 [scrapy] DEBUG: Enabled item pipelines:
2011-11-25 22:45:48-0800 [irishtunes] INFO: Spider opened
2011-11-25 22:45:48-0800 [irishtunes] INFO: Crawled 0 pages (at 0 pages/min), sc
raped 0 items (at 0 items/min)
2011-11-25 22:45:48-0800 [scrapy] DEBUG: Telnet console listening on 0.0.0.0:602
3
2011-11-25 22:45:48-0800 [scrapy] DEBUG: Web service listening on 0.0.0.0:6080
2011-11-25 22:45:48-0800 [irishtunes] DEBUG: Redirecting (301) to <GET http://ww
w.thesession.org/tunes/> from <GET http://www.thesession.org/tunes>
2011-11-25 22:45:48-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/> (referer: None)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11602> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11602>
        {'count': [u'1'],
         'key': [u'Key signature: Dmajor'],
         'name1': [u"Brendan Begley's"],
         'name2': [u'polka'],
         'url': 'http://www.thesession.org/tunes/display/11602'}
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11593> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11593>
        {'count': [u'3'],
         'key': [u'Key signature: Amajor'],
         'name1': [u'Carleton County Breakdown'],
         'name2': [u'reel'],
         'url': 'http://www.thesession.org/tunes/display/11593'}
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11597> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11597>
        {'count': [u'3'],
         'key': [u'Key signature: Dmajor'],
         'name1': [u"Kasper's Rant"],
         'name2': [u'hornpipe'],
         'url': 'http://www.thesession.org/tunes/display/11597'}
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11594> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11594>
        {'count': [u'5'],
         'key': [u'Key signature: Gmajor'],
         'name1': [u'The Full Of The Bag'],
         'name2': [u'hornpipe'],
         'url': 'http://www.thesession.org/tunes/display/11594'}
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11599> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11599>
        {'count': [u'1'],
         'key': [u'Key signature: Adorian'],
         'name1': [u'The New Steamboat'],
         'name2': [u'reel'],
         'url': 'http://www.thesession.org/tunes/display/11599'}
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11598> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11598>
        {'count': [u'4'],
         'key': [u'Key signature: Gmajor'],
         'name1': [u"Galen's Arrival"],
         'name2': [u'reel'],
         'url': 'http://www.thesession.org/tunes/display/11598'}
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11596> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11596>
        {'count': [u'2'],
         'key': [u'Key signature: Amixolydian'],
         'name1': [u'Culloden Day'],
         'name2': [u'strathspey'],
         'url': 'http://www.thesession.org/tunes/display/11596'}
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11595> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11595>
        {'count': [u'2'],
         'key': [u'Key signature: Aminor'],
         'name1': [u'Miss Sine Flemington'],
         'name2': [u'barndance'],
         'url': 'http://www.thesession.org/tunes/display/11595'}
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11600> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11600>
        {'count': [u'2'],
         'key': [u'Key signature: Dmajor'],
         'name1': [u"Joan Martin's"],
         'name2': [u'polka'],
         'url': 'http://www.thesession.org/tunes/display/11600'}
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Crawled (200) <GET http://www.these
ssion.org/tunes/display/11601> (referer: http://www.thesession.org/tunes/)
2011-11-25 22:45:49-0800 [irishtunes] DEBUG: Scraped from <200 http://www.theses
sion.org/tunes/display/11601>
        {'count': [u'2'],
         'key': [u'Key signature: Gmajor'],
         'name1': [u'My Time Inside 2005'],
         'name2': [u'waltz'],
         'url': 'http://www.thesession.org/tunes/display/11601'}
2011-11-25 22:45:49-0800 [irishtunes] INFO: Closing spider (finished)
2011-11-25 22:45:49-0800 [irishtunes] INFO: Stored csv feed (10 items) in: scrap
ed_data.csv
2011-11-25 22:45:49-0800 [irishtunes] INFO: Dumping spider stats:
        {'downloader/request_bytes': 3655,
         'downloader/request_count': 12,
         'downloader/request_method_count/GET': 12,
         'downloader/response_bytes': 31620,
         'downloader/response_count': 12,
         'downloader/response_status_count/200': 11,
         'downloader/response_status_count/301': 1,
         'finish_reason': 'finished',
         'finish_time': datetime.datetime(2011, 11, 26, 6, 45, 49, 500000),
         'item_scraped_count': 10,
         'request_depth_max': 1,
         'scheduler/memory_enqueued': 12,
         'start_time': datetime.datetime(2011, 11, 26, 6, 45, 48, 10000)}
2011-11-25 22:45:49-0800 [irishtunes] INFO: Spider closed (finished)
2011-11-25 22:45:49-0800 [scrapy] INFO: Dumping global stats:
        {}

编辑： @reclosedev 的回答让我找到了方向。对于任何想知道结果的人，这里有个快照……

(1) 大多数曲子在 10 个会员的曲谱中都不到 10 个。

enter image description here

(2) 我从网站上抓取到的 10,379 首曲子的受欢迎程度（根据它们出现在多少个曲谱中来衡量）遵循幂律分布。

enter image description here

(3) 这里是那些在网站上超过 1000 个曲谱中的曲子，显示了排名前列的曲子名称以及它们出现在多少个曲谱中。

enter image description here

数据可视化 数据分析 网络爬虫 数据抓取 爬虫技术 网站结构 幂律分布 曲谱

2 个回答

0

有很多种方法可以做到这一点，我给你推荐一个最简单的方法：

你可以把你的代码运行十次，方法是替换掉start_urls，或者像这样循环：range(10, 100, 10)

http://www.thesession.org/tunes/index/new?new_start=10
http://www.thesession.org/tunes/index/new?new_start=20
http://www.thesession.org/tunes/index/new?new_start=30
http://www.thesession.org/tunes/index/new?new_start=40
http://www.thesession.org/tunes/index/new?new_start=50
http://www.thesession.org/tunes/index/new?new_start=60
http://www.thesession.org/tunes/index/new?new_start=70
http://www.thesession.org/tunes/index/new?new_start=80
http://www.thesession.org/tunes/index/new?new_start=90
回答于 2025-04-17 由 Python大师
分享 举报
6

你需要添加一个 Rule，这个规则会提取所有页面的链接，然后爬虫会 follow（跟随）这些链接：

rules = [
    ..., #your existing parse_tune rule
    Rule(
        SgmlLinkExtractor(
             allow=('/index/new\?new_start=\d+',)
        ),
        follow=True,
    ),
]

补充：

follow=True 其实不是必须的，因为 callback=None 的意思是默认情况下就是 follow=True。

回答于 2025-04-17 由 Python大师
分享 举报

撰写回答