XPath: select element based on inner text ignore child

Question

<a href="editauthority.jsp" title="Add a new connection" class="link btn btn-primary" role="button">
    <i class="fa fa-plus-circle fa-fw" aria-hidden="true"></i>
Add a new connection</a>

I have an HTML element as above with an font-awesome icon. I want to find an element from html using its text.

This is what I have tried so far

//a[contains(@class,'btn') and contains(normalize-space(text()),'Add a new connection')]

But the problem is //a[contains(@class,'btn') and contains(normalize-space(text()),'')] returns the expected value with a new line before and few space at the start, which doesn't match with the expected value of Add a new connection

How can I match the element text ignore all new line and extra spaces.

I am using the above XPATH to find elements in selenium testing.

Update

Adding Selenium code

/**
* Clicks a button based on visible text, this type of button is created using anchor tag with .btn class
* @param text
*/
public void clickButton(String text)
{
    WebElement element =
        waitElementClickable(
            By.xpath("//a[contains(@class,'btn') and contains(normalize-space(text()),'" + text + "')]"));
    element.click();

    if (!isAlertPresent())
    {
      wait.until(ExpectedConditions.invisibilityOfElementLocated(By.id("loader")));
    }
}

Oh, so you want just to select element? I thought you need to extract text... this part about `XPath` that `"returns the expected value"` is confusing :) — Andersson, Mar 21 '17 at 21:21
Have you tried something simple like `By.linkText("Add a new connection")` or `By.partialLinkText("Add a new connection")`? One or both of them should work without all the XPath. — JeffC, Mar 22 '17 at 01:33

score 3 · Answer 1 · edited May 23 '17 at 12:09

3

It is important to realize the difference between Testing text() nodes vs string values in XPath.

Change this clause,

contains(normalize-space(text()),'Add a new connection')

to this

normalize-space()='Add a new connection'

So your complete XPath would read,

//a[contains(@class,'btn') and normalize-space()='Add a new connection']

You might also want to avoid inadvertent substring matching on @class by using this technique.

edited May 23 '17 at 12:09

Community

1
1

answered Mar 21 '17 at 21:14

kjhughes

89,675
16
141
199

The above code doesn't code when there a ` ` along with text in the prefix – Kishore Kumar Mar 21 '17 at 21:22
Your question said *nothing* about ` `, which presents its own set of issues and merits its own question. If this answer helps you with the question you asked, then please [**accept**](http://meta.stackoverflow.com/q/5234/234215) this answer and ask a new question about normalizing away ` `. Thanks. – kjhughes Mar 21 '17 at 22:57

score 0 · Answer 2 · answered Mar 21 '17 at 22:06

0

You can try this expression:

//a[contains(@class, "btn") and contains(.,"Add a new connection")]

Example

answered Mar 21 '17 at 22:06

Vlad

308
3
10

The problem with this is, it will also find match any button that starts with Add – Kishore Kumar Mar 21 '17 at 22:16
You can more specific point to a needed button e.g. change `a[contains(@class, "btn")` to `a[@class="link btn btn-primary"]` – Vlad Mar 21 '17 at 22:32

XPath: select element based on inner text ignore child

2 Answers2