PuppeteerJS - how can I scrape text content from a td element based on the text of the adjacent td? - element

Question

I am attempting to scrape a link from a td cell adjacent to another td labeling the type or description of the link using puppeteer. There are no classes or id distinguishing these td cells other than the text content
<tr>
<td scope="row">1</td>
<td scope="row">10-Q</td>
<td scope="row">nflx-093018x10qxdoc.htm</td>
<td scope="row">10-Q</td>
<td scope="row">1339833</td>
</tr>
<tr class="blueRow">
<td scope="row">2</td>
<td scope="row">EXHIBIT 31.1</td>
<td scope="row">nflx311_q32018.htm</td>
<td scope="row">EX-31.1</td>
<td scope="row">14914</td>
</tr>
<tr>
<td scope="row">3</td>
<td scope="row">EXHIBIT 31.2</td>
<td scope="row">nflx312_q32018.htm</td>
<td scope="row">EX-31.2</td>
<td scope="row">14553</td>
</tr>
<tr class="blueRow">
<td scope="row">4</td>
<td scope="row">EXHIBIT 32.1</td>
<td scope="row">nflx321_q32018.htm</td>
<td scope="row">EX-32.1</td>
<td scope="row">12406</td>
</tr>
the link after td containing '10Q'

Answer 1

XPath expressions
This is where XPath expression are great:
//td[contains(., '10-Q')]/following-sibling::td[1]/a[1]
This XPath expression queries for a td element containing the text 10-Q. Then it will take the following td element and return the first link (a) inside. Alternatively, you could use //td[text()='10-Q']/ in the beginning, if you don't just want the element to contain the text, but to exactly match it.
Usage within puppeteer
To get the element with puppeteer, use the page.$x function. To extract information (like href) from the queried node, use page.evaluate.
Putting all together, the code looks like this:
const [linkHandle] = await page.$x("//td[contains(., '10-Q')]/following-sibling::td[1]/a[1]");
const address = await page.evaluate(link => link.href, linkHandle);

Answer 2

You can do this with vanila javascript,
// find all tr elements
[...document.querySelectorAll('tr')]
// check which one of them includes the word
.find(e=>e.innerText.includes('10-Q'))
// get the link inside
.querySelector('a')
With puppeteer $eval, this can be simplified,
page.$$eval('tr', eachTr=> eachTr.find(e=>e.innerText.includes('10-Q')).querySelector('a'))
Or page.evaluate,
page.evaluate(()=> {
// find all tr elements
return [...document.querySelectorAll('tr')]
// check which one of them includes the word
.find(e=>e.innerText.includes('10-Q'))
// get the link inside
.querySelector('a')
// do whatever you want to do with this
.href
})
Readable solution.