0

I did the following search

parts.get(i).findElements(By.xpath("//li[starts-with(@class, '_lessons--row-')]"))

and it returned dozens of results, while I see in Developer Tools, that there are no more than 3 of them.

parts.get(i) returns single WebElement.

Looks like it searches not children of a given element, but over entire page. Can double slash cause this? What double slash means in XPath?

Dims
  • 42,427
  • 94
  • 291
  • 543
  • `//` is short for `/descendant-or-self::node()/`. See https://www.w3.org/TR/xpath/#path-abbrev for more info. – Daniel Haley Jan 12 '17 at 23:22
  • "descendant or self" sounds contradicting with what other people saying (from root)? – Dims Jan 13 '17 at 07:38
  • Take a look at the link to the spec above. Scroll past the examples in the beginning of that section and read the last few paragraphs. It explains both `//` and `.` and how they work together. – Daniel Haley Jan 13 '17 at 09:53
  • This might also help: http://stackoverflow.com/q/33918010/317052 – Daniel Haley Jan 13 '17 at 15:50

3 Answers3

2

Your xpath is faulty here.

"//li[starts-with(@class, '_lessons--row-')]"

// searches from root level, to search from node preappend .:

".//li[starts-with(@class, '_lessons--row-')]"
Granitosaurus
  • 19,024
  • 4
  • 51
  • 73
0

Try your xpath with .// , normally you should start xpath with "." to stop finding elements from root. 

.//li[starts-with(@class, '_lessons--row-')]
promitheus
  • 47
  • 1
  • 13
0

// match relative data. which starts at the document root. In your case you are trying to locate using 

//li[starts-with(@class, '_lessons--row-')]

So it will return all the match in your html. If you want to locate some specific portion of element with class have start text_lessons--row- . You have to make your xpath more specific.

e.g 

 //div[@id='someid']//li[starts-with(@class, '_lessons--row-')]
NarendraR
  • 7,242
  • 9
  • 41
  • 75