<div><br></div><div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">ons. 30. apr. 2025 kl. 14:13 skrev Jaikiran Pai <<a href="mailto:jpai@openjdk.org">jpai@openjdk.org</a>>:</div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;padding-left:1ex;border-left-color:rgb(204,204,204)">
<br>
Hello Eirik,<br>
<br>
> Before this change, two ZipFile instances opened using different (non-UTF-8) charsets would have equal keys, and thus be backed by the same Source and ZipCoder instance, whichever ZipFile constructed first would "win".<br>
> This seems like a separate bug, independent of the concurrency concerns described JDK-8347712.<br>
> For the benefit of future maintainers, I think this independent bug should be described in a separate JBS issue.<br>
<br>
That's right - when fixing 8347712 I saw this issue with `Key` class and addressed it in this PR. You have a good point that having a JBS issue specifically for this will be useful. I have now created <a href="https://bugs.openjdk.org/browse/JDK-8355975" rel="noreferrer" target="_blank">https://bugs.openjdk.org/browse/JDK-8355975</a> and linked that issue in this current PR. What helped was your test case example that uses `ISO-8859-15` and `ISO-8859-1` charsets to trigger this issue. I wasn't aware of that difference between `ISO-8859-15` and `ISO-8859-1` charsets. Thank you, I've now added that test in this PR.</blockquote><div dir="auto"><br></div><div dir="auto">Thanks Jaikiran for addressing this bad other comments. I’m limited to the mobile UI of GitHub right now, but everything looks good. Thanks for adding the test 👍 </div><div dir="auto"><br></div><div dir="auto">Eirik. </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;padding-left:1ex;border-left-color:rgb(204,204,204)" dir="auto"></blockquote></div></div>