[
https://issues.apache.org/jira/browse/JCR-2327?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12759421#action_12759421
]
Thomas Mueller commented on JCR-2327:
-------------------------------------
+1
Probably you can even ignore some of the strict formatting rules, I propose:
long high = 0, low = 0;
int i = 0;
for (int j = 0; i < s.length() && j < 16; i++) {
char ch = s.charAt(i);
if (ch != '-') {
high = (high << 4) | Character.digit(ch, 16);
j++;
}
}
for (int j = 0; i < s.length() && j < 16; i++) {
char ch = s.charAt(i);
if (ch != '-') {
low = (low << 4) | Character.digit(ch, 16);
j++;
}
}
> java.util.UUID.fromString() too slow
> ------------------------------------
>
> Key: JCR-2327
> URL: https://issues.apache.org/jira/browse/JCR-2327
> Project: Jackrabbit Content Repository
> Issue Type: Improvement
> Components: jackrabbit-core, jackrabbit-jcr-commons
> Reporter: Marcel Reutegger
>
> Benchmarking shows that the java.util.UUID.fromString() method is 10 times
> slower than the previous version we used from jackrabbit-jcr-commons. This
> method is quite heavily used in the query section or more generally whenever
> a NodeId is created from a String.
> I'd like to introduce the custom String UUID parsing code again that we had
> in the jackrabbit-jcr-commons UUID class and use it in the NodeId(String)
> constructor.
> WDYT?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
