On Fri, 19 Dec 2025 08:54:35 GMT, John Hendrikx <[email protected]> wrote:
>> I think a second reviewer couldn't hurt. I only have two questions: >> >> 1. I presume that moving the local listeners to be instance variables is >> needed? At first glance, I couldn't figure out why. >> 2. Is a test feasible? If not, we should consider filing a follow-up. > >> I think a second reviewer couldn't hurt. I only have two questions: >> >> 1. I presume that moving the local listeners to be instance variables is >> needed? At first glance, I couldn't figure out why. >> 2. Is a test feasible? If not, we should consider filing a follow-up. > > 1. It is not needed, but they also don't need to be recreated each time as > they're constant > 2. I think a test may be possible, we can check if the data objects get GC'd > > I'm neutral on the get property thing > BTW, I want to start an informal knowledge base - your comments @hjohn are > too good to just leave in the PRs :-) or JBS in tickets > https://bugs.openjdk.org/browse/JDK-8373908?focusedId=14842224&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14842224 Yeah, I don't know where it could be stored, there are many insightful comments in the JBS by developers that have worked on FX over the years. Perhaps we could look if such things can just be added to the documentation in a logical spot, on a class or package. ------------- PR Comment: https://git.openjdk.org/jfx/pull/2013#issuecomment-3674709646
