Null pointer exception when calling webDriver.findElements

[ClintonMorrison]

Meta -

OS: OSX
Selenium Version: 3.3.1, IntelliJ
Browser: Chrome 60

Browser Version:
Version 60.0.3112.90 (Official Build) (64-bit)

Expected Behavior -

No null pointer exception when calling webDriver.findElements

Actual Behavior -

Occasionally, calling webDriver.findElements(By.cssSelector(query)) causes a null pointer exception. Here is the stack trace:

java.lang.NullPointerException
    at org.openqa.selenium.remote.RemoteWebDriver.findElements(RemoteWebDriver.java:450)
    at org.openqa.selenium.remote.RemoteWebDriver.findElementsByCssSelector(RemoteWebDriver.java:509)
    at org.openqa.selenium.By$ByCssSelector.findElements(By.java:441)
    at org.openqa.selenium.remote.RemoteWebDriver.findElements(RemoteWebDriver.java:402)
    at sun.reflect.GeneratedMethodAccessor7.invoke(Unknown Source)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at org.openqa.selenium.remote.Augmenter$CompoundHandler.intercept(Augmenter.java:189)
    at org.openqa.selenium.remote.RemoteWebDriver$$EnhancerByCGLIB$$88202d8c.findElements(<generated>)

Steps to reproduce -

It seems to happen only rarely happen so it's hard to reliably reproduce. I think it is a race condition. There was a recently a report of a similar issue in: #4375

[barancev]

Actually, it's a chromedriver issue. It must return a list of web elements (or an empty list) on this operation. But sometimes it breaks the contract and returns null. Sigh

So we have to be more defensive and convert null to an empty list on the client side. All language bindings can suffer from this issue.

[barancev]

Fixed in Java binding by 4a3ec2d

[ClintonMorrison]

Thanks @barancev!

[ClintonMorrison]

What release will this be in?

[p0deje]

Fixed in Ruby bindings in d4d8433.

[lmtierney]

Does this only affect findElements and not findChildElements?

[lmtierney]

Fixed in python 4fcce6e

[IvantsovAV]

A moment ago (but don't reproduce):

InternetExplorerDriver server (32-bit) 3.6.0.0
selenium-java 3.7.1

java.lang.NullPointerException at java.util.Objects.requireNonNull(Objects.java:203) at java.util.Optional.<init>(Optional.java:96) at java.util.Optional.of(Optional.java:108) at java.util.stream.FindOps$FindSink$OfRef.get(FindOps.java:193) at java.util.stream.FindOps$FindSink$OfRef.get(FindOps.java:190) at java.util.stream.FindOps$FindOp.evaluateSequential(FindOps.java:152) at java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:234) at java.util.stream.ReferencePipeline.findFirst(ReferencePipeline.java:464) at org.openqa.selenium.support.ui.ExpectedConditions.findElement(ExpectedConditions.java:881) at org.openqa.selenium.support.ui.ExpectedConditions.access$000(ExpectedConditions.java:43) at org.openqa.selenium.support.ui.ExpectedConditions$20.apply(ExpectedConditions.java:579) at org.openqa.selenium.support.ui.ExpectedConditions$20.apply(ExpectedConditions.java:575) at org.openqa.selenium.support.ui.FluentWait.until(FluentWait.java:208)

Maybe it returns list of null(s)?

[BrandonDudek]

In Selenium 3.14.0, I see cases where response is returning an empty string as the value.
Can this case be covered as well?

[diemol]

It seems it was implemented in all bindings except .NET, in case any one finds an issue there, please open a new issue.

[sushmanarla]

The problem for me was that the path to find chrome browser was missing ".exe" in it. I'm referring to the "VM options:" in Run>Edit Configurations>JDK Settings

-Dwebdriver.chrome.driver="D:\Work\Softwares\chromedriver_win32.\chromedriver.exe"