[ https://issues.jboss.org/browse/GTNPORTAL-3537?page=com.atlassian.jira.pl... ] Juraci Paixão Kröhling commented on GTNPORTAL-3537: --------------------------------------------------- Ok, we have actually thought about clashes on the hashCode when doing this, but the chances of this occurring were thought to be really small. It seems it's not the case :-) The solution you mentioned would work for your case, but wouldn't work for a hashcode which is one digit longer, with all but one digit being the same. Example: [code] local._juzu-tutorial163057421112 local._juzu-tutorial163057421110 [code] Those would be truncated to: [code] local._juzu-tutorial1630574211 local._juzu-tutorial1630574211 [code] I remember that [~ppalaga] provided some options during the review. He would perhaps like to chime in as well. All in all, I agree with your idea of refactoring this part, as there's no reasonable way to a really unique ID with only a few random chars. -- This message was sent by Atlassian JIRA (v6.3.1#6329)