On Thu, 9 Apr 2026 14:40:26 GMT, Daniel Fuchs <[email protected]> wrote:

>> Following tests are refactored to use JUnit Jupiter API :
>> sun/net/ext/ExtendedSocketOptionsTest.java
>> sun/net/www/protocol/file/DirPermissionDenied.java
>> sun/net/www/protocol/jar/MultiReleaseJarURLConnection.java
>> sun/net/www/protocol/jrt/Basic.java
>> sun/net/www/protocol/http/TestTransparentNTLM.java
>> sun/net/www/protocol/http/HttpHeaderParserTest.java
>> sun/net/www/http/RequestMethodCheck/RequestMethodEquality.java
>> sun/net/www/http/KeepAliveStreamCleaner/java.base/sun/net/www/http/KeepAliveStreamCleanerTest.java
>> sun/net/www/http/KeepAliveStreamCleaner/KeepAliveStreamCleanerTestDriver.java
>> sun/net/www/MessageHeaderTest.java
>> sun/net/spi/DefaultProxySelectorTest.java
>
> test/jdk/sun/net/www/http/RequestMethodCheck/RequestMethodEquality.java line 
> 115:
> 
>> 113:             // If both connectTimeout values are equal, it means the 
>> test retrieved the same broken
>> 114:             // HttpClient from the cache and is trying to re-use it.
>> 115:             assertNotEquals(originalConnectTimeout, 
>> cachedConnectTimeout, "Both connectTimeout values are equal.\nThis means the 
>> test is reusing a broken HttpClient rather than creating a new one.");
> 
> arguments should be swapped. Could you make a pass over all the files and 
> double check this please?
> The argument order for JUnit is `(expected, actual [, msg])`

Here second httpclient (cachedClient) should not be retrieved from cache, 
arguments order is correct here. Not expected is originalConnectTimeout and 
actual is from new httpclient (cachedclient)

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/30645#discussion_r3065755916

Reply via email to