expect-puppeteer: trim out redundant whitespace in textContent when doing text matching

[rkoval]

When matching on textContent, a lot of the time, there will be extra whitespace/newlines surrounding the text you're matching based on how the HTML looks within the element you're matching. It could be nice to strip all of the redundant whitespace out when doing the comparison within this library as well such that consumers of it don't need to account for that whitespace in their comparison strings.

For example, a textContent of

'    white  

  space   '

could be stripped to just

'white space'

to allow matching by passing just 'white space' without needing wildcards in the comparison string. This could vastly improve test cases asserting that text exists. This could also perhaps be configurable too in case someone actually needed the textContent as-is in the HTML.

[gregberge]

I prefer to not change the textContent, you can use a RegExp like white.*space and it should work.

[rkoval]

Unfortunately, not when targeting any nested trees that result in newlines between strings to match (which, in a lot of projects i imagine, is most of the meaningful text). You have to expand the regex to appease every newline and surrounding whitespace on every line until you get to the strings you want to match.

Would you at least be open to adding a secondary API for this sort of behavior that will only attempt to strip when you pass the right argument? I would be happy to implement, as it's very easy to do. Example:

const textContent = `    white

  space   `
const strippedTextContent = textContent.replace(/\s+/g,' ').trim() // results in 'white space'

[gregberge]

@rkoval OK can you implement it?