xpath selecting text from link in <td> & text from <td> - ruby

Question

I have the following code which works very well:
rows = diary_HTML.xpath('//*[#id="main"]/div[2]/table/tbody/tr')
food_diary = rows.collect do |row|
detail = {}
[
["Food", 'td[1]/text()'],
["Calories", 'td[2]/text()'],
["Carbs", 'td[3]/text()'],
["Fat", 'td[4]/text()'],
["Protein", 'td[5]/text()'],
["Cholest", 'td[6]/text()'],
].each do |name, xpath|
detail[name] = row.at_xpath(xpath).to_s.strip
end
detail
end
However the "Food" td does not only include text, but also a link from which I want to get the text.
I know I can use 'td[1]/a/text()'to get the link text, but how do I do both?
'td[1]/a/text()' or 'td[1]/text()'
EDITED - Added Snippet.
I am trying to include the <tr class="meal_header">
<td class="first alt">Breakfast</td> on the first row, all lines with other regular tds on other rows whilst excluding td1 on the bottom row.
<tr class="meal_header">
<td class="first alt">Breakfast</td>
<td class="alt">Calories</td>
<td class="alt">Carbs</td>
<td class="alt">Fat</td>
<td class="alt">Protein</td>
<td class="alt">Sodium</td>
<td class="alt">Sugar</td>
</tr>
<tr>
<td class="first alt">
<a onclick="showEditFood(3992385560);" href="#">Hovis (Uk - White Bread (40g) Toasted With Flora Light Marg, 2 slice</a> </td>
<td>262</td>
<td>36</td>
<td>9</td>
<td>7</td>
<td>0</td>
<td>3</td>
</tr>
<tr class="bottom">
<td class="first alt" style="z-index: 10">
Add Food
<div class="quick_tools">
Quick Tools
<div id="quick_tools_0" class="quick_tools_options hidden">
<ul>
<li><a onclick="showLightbox(200, 250, '/food/quick_add?meal=0&date=2013-04-15'); return false;">Quick add calories</a></li>
<li>Remember meal</li>
<li>Copy yesterday</li>
<li>Copy from date</li>
<li>Copy to date</li>
</ul>
</div>
<div id="recent_meals_0" class="recent_meal_options hidden">
<ul id="recent_meal_options_0">
<li class="header">Copy from which date?</li>
<li>Sunday, April 14</li>
<li>Saturday, April 13</li>
</ul>
</div>
</div>
</td>
<td>285</td>
<td>39</td>
<td>9</td>
<td>10</td>
<td>0</td>
<td>3</td>
<td></td>

Answer 1

The short answer is: use Nokogiri::XML::Element#text, it will give the text of the element plus subelements (your a for example).
You can also clean that code up quite a bit:
keys = ["Food", "Calories", "Carbs", "Fat", "Protein", "Cholest"]
food_diary = rows.collect do |row|
Hash[keys.zip row.search('td').map(&:text)]
end
And as a final tip, avoid using xpath with html, css is so much nicer.

Answer 2

I think you can achieve this by altering the logic to look at element content when you don't have an explicit text() extraction in the xpath
rows = diary_HTML.xpath('//*[#id="main"]/div[2]/table/tbody/tr')
food_diary = rows.collect do |row|
detail = {}
[
["Food", 'td[1]'],
["Calories", 'td[2]/text()'],
["Carbs", 'td[3]/text()'],
["Fat", 'td[4]/text()'],
["Protein", 'td[5]/text()'],
["Cholest", 'td[6]/text()'],
].each do |name, xpath|
if xpath.include?('/text()')
detail[name] = row.at_xpath(xpath).to_s.strip
else
detail[name] = row.at_xpath(xpath).content.strip
end
end
detail
end
You could also add e.g. a symbol to the array, to describe how you were extracting the data, and have a case block which handled items depending on what the last stage was to do following the xpath
Note you could also do what you want by walking the node structure returned by xpath recursively, but that seems like overkill if you just want to ignore markup, links etc.