Why we use different logic with openjdk java Proxy when it comes to http.nonProxyHosts system property

[cccccccmcho]

Hi guys.
Thank you for reaching out first.

I've been working on applying a proxy to my server recently.
However, I've noticed that the way non-proxy hosts are handled in the AWS SDK is different from the way they are handled in the actual http client, and I'd like to know why.

Does anyone know?

reference

openjdk DefaultProxySeleector.java

• https://github.com/openjdk/jdk/blob/master/src/java.base/share/classes/sun/net/spi/DefaultProxySelector.java#L345

aws sdk

• aws-sdk-java-v2/utils/src/main/java/software/amazon/awssdk/utils/http/SdkHttpUtils.java

  Line 437 in 9530396

  private static Set<String> extractNonProxyHosts(String nonProxyHosts) {

The logic for evaluating the regex is different, so it actually gives different results in the two cases.
Also, aws allows * in the middle and openjdk does not, why?