[Bug 66627] Regression due to MessageBytes refactoring

2023-06-08 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66627 --- Comment #5 from WJCarpenter --- Thanks for the quick turnaround, Mark! -- You are receiving this mail because: You are the assignee for the bug. - To unsubscribe, e-mail:

[Bug 66627] Regression due to MessageBytes refactoring

2023-06-08 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66627 Mark Thomas changed: What|Removed |Added Resolution|--- |FIXED Status|NEW

[Bug 66627] Regression due to MessageBytes refactoring

2023-06-07 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66627 --- Comment #3 from Mark Thomas --- The API is also documented not to be stable. See the section on API stability in the RELEASE-NOTES.txt file that should be at the root of every Tomcat 9 distribution. That said, I do plan to look at the feas

[Bug 66627] Regression due to MessageBytes refactoring

2023-06-07 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66627 --- Comment #2 from WJCarpenter --- I agree in general with the point about calling internals, but it is, after all, a documented API. Thanks for the pointer about OpenTelemetry's fix. We'll pursue that. -- You are receiving this mail becaus

[Bug 66627] Regression due to MessageBytes refactoring

2023-06-07 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66627 --- Comment #1 from Mark Thomas --- The assumption in CoyoteAdater is still valid - assuming no-one is calling Tomcat internals. As soon as components start accessing Tomcat internals then all sorts of things can go wrong unless those component