XPath - Select All Text As Single Line

Question

I have the following HTML:

<p>
    Some cool text
    <a href="#">looks like</a>
    this.
</p>

I want to grab the text as a single line:

Some cool text looks like this.

I'm currently using the following XPath query:

//p//text()

And it returns all the text, but as separate lines:

Some cool text

looks like

this.

Any thoughts on how to modify my query so it can return on a single line?

The query needs to meet XPath1.0 requirements.

score 2 · Answer 1 · answered Feb 05 '19 at 21:27

2

text() selects individual text nodes, and there will always be at least one text node in between elements (where there is text). What you want is to convert the p to a string:

//p/string(.)

answered Feb 05 '19 at 21:27

wst

11,681
1
24
39

Good, +1, but note that it requires XPath 2.0. – kjhughes Feb 05 '19 at 21:34
@kjhughes The question is tagged `xquery`, and XPath 2 is part of the XQuery 1 spec. – wst Feb 05 '19 at 21:41
It's a bit surprising to see both [tag:xquery] and [tag:web-scraping] together, so a tagging mistake isn't inconceivable. Anyway, later readers might benefit from an explicit note. – kjhughes Feb 05 '19 at 21:46
@kjhughes Yeah, I am also curious :) – wst Feb 05 '19 at 21:49

score 2 · Answer 2 · answered Feb 05 '19 at 21:30

2

If you have XPath-2.0 or above available you can use string-join(...) to merge the text() values:

string-join(normalize-space(p))

answered Feb 05 '19 at 21:30

zx485

28,498
28
50
59

score 2 · Answer 3 · answered Feb 05 '19 at 21:33

2

This XPath (1.0 on up),

string(normalize-space())

will return

"Some cool text looks like this."

in a single line, as requested.

See also Testing text() nodes vs string values in XPath

answered Feb 05 '19 at 21:33

kjhughes

106,133
27
181
240

XPath - Select All Text As Single Line

3 Answers3