获取一系列行的正确 nokogiri xpath 是什么？

Question

我有一个表格，格式如下:

<tr class="style6"><td>SomeStuff</td></tr>
<tr><td>Some other stuff</td></tr>
<tr><td>Some other stuff</td></tr>
<tr><td>Some other stuff</td></tr>
<tr><td>Some other stuff</td></tr>
<tr><td>Some other stuff</td></tr>
<tr class="style6"><td>SomeStuff</td></tr>
<tr><td>Some other stuff</td></tr>
<tr><td>Some other stuff</td></tr>
<tr><td>Some other stuff</td></tr>
<tr><td>Some other stuff</td></tr>
<tr><td>Some other stuff</td></tr>

我想要一个行块(从style6类开始到下一个style6发生之前的最后一行)分成我可以迭代的组。有没有办法把它分割成块呢?我知道Xpath position函数，但不确定它在这种情况下是否有意义。

任何想法?

Answer 1

一个有用的模式是计算前面的<tr class="style6"><td>SomeStuff</td></tr>。

对于示例中的第一个组，它将是:

//tr[not(@class="style6")][count(preceding-sibling::tr[@class="style6"])=1]

第二组:

//tr[not(@class="style6")][count(preceding-sibling::tr[@class="style6"])=2]

等。

我不使用nokogiri，所以这里有一个使用Python和lxml的例子:

>>> import lxml.html
>>> from pprint import pprint
>>> doc = lxml.html.fromstring('''<tr class="style6"><td>SomeStuff</td></tr>
... <tr><td>Some other stuff group 1</td></tr>
... <tr><td>Some other stuff group 1</td></tr>
... <tr><td>Some other stuff group 1</td></tr>
... <tr><td>Some other stuff group 1</td></tr>
... <tr><td>Some other stuff group 1</td></tr>
... <tr class="style6"><td>SomeStuff</td></tr>
... <tr><td>Some other stuff group 2</td></tr>
... <tr><td>Some other stuff group 2</td></tr>
... <tr><td>Some other stuff group 2</td></tr>
... <tr><td>Some other stuff group 2</td></tr>
... <tr><td>Some other stuff group 2</td></tr>
... <tr class="style6"><td>SomeStuff</td></tr>
... <tr><td>Some other stuff group 3</td></tr>
... <tr><td>Some other stuff group 3</td></tr>
... <tr><td>Some other stuff group 3</td></tr>
... <tr><td>Some other stuff group 3</td></tr>
... <tr><td>Some other stuff group 3</td></tr>''')
>>> pprint(list(lxml.html.tostring(row)
...            for row in doc.xpath('''
...                 //tr[not(@class="style6")]
...                     [count(preceding-sibling::tr[@class="style6"])=1]''')))
[b'<tr><td>Some other stuff group 1</td></tr>n',
 b'<tr><td>Some other stuff group 1</td></tr>n',
 b'<tr><td>Some other stuff group 1</td></tr>n',
 b'<tr><td>Some other stuff group 1</td></tr>n',
 b'<tr><td>Some other stuff group 1</td></tr>n']
>>> pprint(list(lxml.html.tostring(row)
...            for row in doc.xpath('''
...                 //tr[not(@class="style6")]
...                     [count(preceding-sibling::tr[@class="style6"])=2]''')))
[b'<tr><td>Some other stuff group 2</td></tr>n',
 b'<tr><td>Some other stuff group 2</td></tr>n',
 b'<tr><td>Some other stuff group 2</td></tr>n',
 b'<tr><td>Some other stuff group 2</td></tr>n',
 b'<tr><td>Some other stuff group 2</td></tr>n']
>>> pprint(list(lxml.html.tostring(row)
...            for row in doc.xpath('''
...                 //tr[not(@class="style6")]
...                     [count(preceding-sibling::tr[@class="style6"])=3]''')))
[b'<tr><td>Some other stuff group 3</td></tr>n',
 b'<tr><td>Some other stuff group 3</td></tr>n',
 b'<tr><td>Some other stuff group 3</td></tr>n',
 b'<tr><td>Some other stuff group 3</td></tr>n',
 b'<tr><td>Some other stuff group 3</td></tr>']
>>>