https://bz.apache.org/bugzilla/show_bug.cgi?id=61253
Coty Sutherland changed:
What|Removed |Added
Status|NEW |RESOLVED
https://bz.apache.org/bugzilla/show_bug.cgi?id=61253
--- Comment #5 from Coty Sutherland ---
Created attachment 35123
--> https://bz.apache.org/bugzilla/attachment.cgi?id=35123=edit
Patch proposal that logs the attribute name and value that failed to update
--
You are
https://bz.apache.org/bugzilla/show_bug.cgi?id=61253
--- Comment #4 from Coty Sutherland ---
Created attachment 35109
--> https://bz.apache.org/bugzilla/attachment.cgi?id=35109=edit
PropertySourceImpl to reproduce with
I didn't add a reproducer earlier, but here's one in
https://bz.apache.org/bugzilla/show_bug.cgi?id=61253
--- Comment #3 from Remy Maucherat ---
As I see it, when a property doesn't exist (without errors) in the property
source (SystemPropertySource), it should return null and the substitution isn't
made. The source throwing an
https://bz.apache.org/bugzilla/show_bug.cgi?id=61253
--- Comment #2 from Konstantin Kolinko ---
Bad idea.
The end user does not need to know that somebody tried a property substitution
here. It is not an error to write literal ${foo}. The correct substitution for
https://bz.apache.org/bugzilla/show_bug.cgi?id=61253
--- Comment #1 from Christopher Schultz ---
+1000
I believe that any empty catch block is a bug, unless it's one of those "never
happen" exceptions, and even those should be logged at an ERROR/FATAL level.
--
https://bz.apache.org/bugzilla/show_bug.cgi?id=61253
Coty Sutherland changed:
What|Removed |Added
Keywords||Beginner
--