0. Reference
1. Initialization
In [325]: from scrapy import Selector In [326]: text=""" ...: <div> ...: <a>1a</a> ...: <p>2p</p> ...: <p>3p</p> ...: </div>""" In [327]: sel=Selector(text=text) In [328]: print(sel.extract()) <html><body><div> <a>1a</a> <p>2p</p> <p>3p</p> </div></body></html>
2. Parent node/previous next sibling node
In [329]: sel.xpath('//a/parent::*/p').extract() Out[329]: ['<p>2p</p>', '<p>3p</p>'] In [330]: sel.xpath('//p/preceding-sibling::a').extract() Out[330]: ['<a>1a</a>'] In [331]: sel.xpath('//a/following-sibling::p').extract() Out[331]: ['<p>2p</p>', '<p>3p</p>']
3. CSS first few child nodes
3.1 General
#Complete list of child nodes, counting from the first child node, and satisfying the child node tag limit In [332]: sel.css( ' a :nth-child(1) ' ).extract() Out[ 332]: [ ' <a>1a</a> ' ] #Complete list of child nodes, counting from the last child node, and satisfying the child node tag limit In [333]: sel.css( ' a :nth -last-child(1) ' ).extract() Out[333]: [] In [334]: sel.css('p:nth-child(1)').extract() Out[334]: [] In [335]: sel.css('p:nth-child(2)').extract() Out[335]: ['<p>2p</p>'] In [336]: sel.css('p:nth-child(3)').extract() Out[336]: ['<p>3p</p>'] In [337]: sel.css('p:nth-last-child(1)').extract() Out[337]: ['<p>3p</p>'] In [338]: sel.css('p:nth-last-child(2)').extract() Out[338]: ['<p>2p</p>'] In [339]: sel.css('p:nth-last-child(3)').extract() Out[339]: []
3.2 Special designation
In [340]: sel.css('a:first-child').extract() Out[340]: ['<a>1a</a>'] In [341]: sel.css('a:last-child').extract() Out[341]: [] In [342]: sel.css('p:first-child').extract() Out[342]: [] In [343]: sel.css('p:last-child').extract() Out[343]: ['<p>3p</p>']
3.3 The above -child is modified to -of-type, and only the filtered corresponding child node list is counted
4. The first few child nodes of Xpath
In [344]: sel.xpath('//div').extract() Out[344]: ['<div>\n <a>1a</a>\n <p>2p</p>\n <p>3p</p>\n</div>'] In [345]: sel.xpath('//div/*').extract() Out[345]: ['<a>1a</a>', '<p>2p</p>', '<p>3p</p>'] In [346]: sel.xpath('//div/node()').extract() Out[346]: ['\n ', '<a>1a</a>', '\n ', '<p>2p</p>', '\n ', '<p>3p</p>', '\n'] In [347]: sel.xpath('//div/a').extract() Out[347]: ['<a>1a</a>'] In [348]: sel.xpath('//div/p').extract() Out[348]: ['<p>2p</p>', '<p>3p</p>'] In [349]: In [349]: sel.xpath('//div/a[1]').extract() Out[349]: ['<a>1a</a>'] In [350]: sel.xpath('//div/a[last()]').extract() Out[350]: ['<a>1a</a>'] In [351]: In [ 351]: sel.xpath( ' //div/p [1] ' ).extract() #equivalent to the filtered list of child nodes Out[351]: [ ' <p>2p</p> ' ] In [352]: sel.xpath('//div/p[last()]').extract() Out[352]: ['<p>3p</p>'] In [353]: sel.xpath('//div/p[last()-1]').extract() Out[353]: ['<p>2p</p>'] In [354]: In [ 354]: sel.xpath( ' //div/*[1] ' ).extract() #Complete list of child nodes Out[354]: [ ' <a>1a</a> ' ] In [355]: sel.xpath('//div/*[last()]').extract() Out[355]: ['<p>3p</p>'] In [356]: In [ 356]: sel.xpath( ' //div/node()[1] ' ).extract() #Include plain text Out[356]: [ ' \n ' ] In [357]: sel.xpath('//div/node()[last()]').extract() Out[357]: ['\n']