How to scrape contents from multiple tables in a w

Question

I want to scrape contents from multiple tables in a webpage and the HTML code goes like this :

<div class="fixtures-table full-table-medium" id="fixtures-data">             
    <h2 class="table-header"> Date 1    </h2>
    <table class="table-stats">
        <tbody>
            <tr class='preview' id='match-row-EFBO755307'>
                <td class='details'>
                    <p>
                        <span class='team-home teams'>
                            <a href='random_team'>team 1</a>                
                        </span>                 
                        <span class='team-away teams'>
                            <a href='random_team'>team 2</a>                
                        </span>
                    </p>
                </td>
            </tr>
            <tr class='preview' id='match-row-EFBO755307'>
                <td class='match-details'>
                    <p>
                        <span class='team-home teams'>
                            <a href='random_team'>team 3</a>                
                        </span>                 
                        <span class='team-away teams'>
                            <a href='random_team'>team 4</a>                
                        </span>
                    </p>
                </td>
            </tr>
        </tbody>
    </table>

    <h2 class="table-header"> Date 2    </h2>
    <table class="table-stats">
        <tbody>
            <tr class='preview' id='match-row-EFBO755307'>
                <td class='match-details'>
                    <p>
                        <span class='team-home teams'>
                            <a href='random_team'>team X</a>                
                        </span>                 
                        <span class='team-away teams'>
                            <a href='random_team'>team Y</a>                
                        </span>
                    </p>
                </td>
            </tr>
            <tr class='preview' id='match-row-EFBO755307'>
                <td class='match-details'>
                    <p>
                        <span class='team-home teams'>
                            <a href='random_team'>Team A</a>                
                        </span>                 
                        <span class='team-away teams'>
                            <a href='random_team'>Team B</a>                
                        </span>
                    </p>
                </td>
            </tr>
        </tbody>
    </table>
</div>

There are more matches under the dates (9 or 2 or 1 depending on the matches played on that date) and the no. of tables is 63 (which is equal to no. of days)

I want to extract, for each date, matches between teams and also which team is home and which team is away.

I was using the scrapy shell and tried following commands:

 title = sel.xpath("//td[@class = 'match-details']")[0] 
 l_home = title.xpath("//span[@class = 'team-home teams']/a/text()").extract()

This printed a list of the home teams and this printed a list of all the away teams,

 l_Away = title.xpath("//span[@class = 'team-away teams']/a/text()").extract()

This gave me a list for all the dates :

sel.xpath("/html/body/div[3]/div/div/div/div[4]/div[2]/div/h2/text()").extract()

What I want is for all dates get the matches that are played on a day (and also which team is home and away)

Should my items.py look like this:

date = Field()
home_team = Field()
away_team2 = Field()

Please help me to write the parse function and the Item class.

Thanks in advance.

Answer 1

Here's an example logic from scrapy shell:

>>> for table in response.xpath('//table[@class="table-stats"]'):
...     date = table.xpath('./preceding-sibling::h2[1]/text()').extract()[0]
...     print date
...     for match in table.xpath('.//tr[@class="preview" and @id]'):
...         home_team = match.xpath('.//span[@class="team-home teams"]/a/text()').extract()[0]
...         away_team = match.xpath('.//span[@class="team-away teams"]/a/text()').extract()[0]
...         print home_team, away_team
... 
 Date 1    
team 1 team 2
team 3 team 4
 Date 2    
team X team Y
Team A Team B

In the parse() method you would need to instantiate an Item instance in the inner loop and yield it:

def parse(self, response):
    for table in response.xpath('//table[@class="table-stats"]'):
        date = table.xpath('./preceding-sibling::h2[1]/text()').extract()[0]
        for match in table.xpath('.//tr[@class="preview" and @id]'):
            home_team = match.xpath('.//span[@class="team-home teams"]/a/text()').extract()[0]
            away_team = match.xpath('.//span[@class="team-away teams"]/a/text()').extract()[0]

            item = MyItem()
            item['date'] = date
            item['home_team'] = home_team
            item['away_team'] = away_team
            yield item

where Myitem would be:

class MyItem(Item):
    date = Field()
    home_team = Field()
    away_team = Field()