Why is XPath counting the wrong number of substrings?

Question

I am trying to count the number of times a element with a partial text of "25516.B1-" appears on the page.

Right now I am trying this. However it returns 17 even though there are only 3 instances of the element on the page.

Count 3 basiskerntaken
${count}=  Get Element Count   //*[contains(.,"25516.B1-")]
Log To Console  ${count}

`//*[contains(.,"25516.B1-")]` includes child nodes. Try `//*[contains(text(),"25516.B1-")]` — Trapli, Nov 12 '21 at 15:34
@Trapli: Closer, but see [Why is XPath contains(text(),'substring') not working as expected?](https://stackoverflow.com/q/69909751/290085) for a problem with that approach and [my answer below](https://stackoverflow.com/a/69945660/290085) for a better way to count substrings within text nodes. — kjhughes, Nov 12 '21 at 16:28
@kjhughes your answer will not work with python + selenium, since the selenium insists to return webelements only, it won't return text nodes (https://stackoverflow.com/questions/20375096). It might work if the xml/html source is an lxml tree or something, which is not _that picky_ — Trapli, Nov 12 '21 at 19:57
@Trapli: Selenium is a handicapped host of XPath if it has such a limitation. Alright, well, I've updated my answer for a workaround for XPath hosts that can only evaluate XPaths that return elements. Thanks for the information. — kjhughes, Nov 12 '21 at 23:05

kjhughes · Answer 1 · 2021-11-12T23:02:40.860

_{^{I will use generic XPath and XML for the benefit of future readers, but the concepts are the same and can be applied to any framework that uses XPath, including robotframework. See addendum regarding selenium's limitations, however.}}

Counting the number of elements whose string value contains a substring,

count(//*[contains(.,"substring")])

will overcount the number of occurrences of the "substring" due to the nesting of elements. For example, both e and r have string values that contain "substring":

<r>
  <e>substring</e>
</r>

Instead, count the number of text nodes that contain the substring:

count(//text()[contains(.,"substring")])

There is only one such text node in the above example.

Note that this assumes that text is not partitioned across child elements:

<r>
  <e><b>subs</b>tring</e>
</r>

Selenium-specific Addendum

@Trapli reports that "selenium insists to return webelements only, it won't return text nodes", here's a work-around that will return elements that selenium can then count:

//*[text()[contains(.,"substring")]]

This returns all elements that have a text node child that contains the provided "substring". It is unfortunate that selenium is limited in how it hosts XPath — most hosts are not so restrictive.

score 0 · Answer 2 · edited Nov 12 '21 at 19:35

0

If you know you have elements with that partial text, I would say using FindElements method to find all occurences of that method, it returns a list of elements , you can simply call List.count ();

List<WebElement> elementName = driver.findElements(By.LocatorStrategy("LocatorValue"));

elementname.count is your answer.

edited Nov 12 '21 at 19:35

undetected Selenium

183,867
41
278
352

answered Nov 12 '21 at 15:55

praveen kumar

11

Your answer could be improved with additional supporting information. Please [edit] to add further details, such as citations or documentation, so that others can confirm that your answer is correct. You can find more information on how to write good answers [in the help center](/help/how-to-answer). – jahantaila Nov 12 '21 at 16:05

score 0 · Accepted Answer · answered Nov 15 '21 at 07:52

0

I fixed it by doing the following

${count}=  Get Element Count     //td[contains(text(),'25516.B1-')]
Should Be Equal As Integers     ${count}    3

This returns 3 elements and passed the test.

Thanks for all the answers!

answered Nov 15 '21 at 07:52

Mwiendels

47
5

score -1 · Answer 4 · answered Nov 12 '21 at 15:45

-1

Update the generic xpath to a bit more specific xpath as follows:

${count}=  Get Element Count   //tag_name[contains(.,"25516.B1-")]

More canonically:

${count}=  Get Element Count   //tag_name[contains(.,"25516") and contains(.,"B1-")]

answered Nov 12 '21 at 15:45

undetected Selenium

183,867
41
278
352

This is good if `tag_name` is known (+1). When unknown, see [my answer](https://stackoverflow.com/a/69945660/290085) for an explanation of why OP's XPath was overcounting and how to correct the problem. – kjhughes Nov 12 '21 at 16:04

Why is XPath counting the wrong number of substrings?

4 Answers4

See also

Selenium-specific Addendum