I appreciate the commitment to consistently but agree where it isn't needed, it's reasonable for us as a group to decline to use final where not strictly required.

On Jul 9, 2013 5:53 PM, "Ralph Goers" <ralph.goers@dslextreme.com> wrote:
This topic was covered in an email thread that started with this email - http://mail-archives.apache.org/mod_mbox/logging-log4j-dev/201301.mbox/%3CCACZkXPwsDyD-X%2B5wHuq4X26wzCZnsHJ%3Ds5rLJNKe71XYyg6LfA%40mail.gmail.com%3E.  Unfortunately I don't know how to link the whole thread from the apache archives but you can find it at http://marc.info/?t=134978818500007&r=1&w=2.

I happen to agree with you, as you will see from the thread.  I find it more annoying that many, many files were changed at once with no warning.  I also suspect Gary does this through something automated in Eclipse, which means he doesn't see the comments that say "don't do that".

Ralph

On Jul 9, 2013, at 4:41 PM, Nick Williams wrote:

Gary,

Earlier today you committed r1501104 with the following comment:

Add final modifier to private fields.
Add final modifier to method parameters.
Add final modifier to local variables.

I've seen you do similar things before, too. Can you explain a little bit why you do this? There's certainly no compiler value--the compiler for a long time has been capable of determining whether a variable is ever changed and optimizing appropriately. All it seems to do to me is clutter the code...significantly. It makes the code harder to read, IMO.

Can you justify the intentional mass change of this nature?

Thanks,

Nick
---------------------------------------------------------------------
To unsubscribe, e-mail: log4j-dev-unsubscribe@logging.apache.org
For additional commands, e-mail: log4j-dev-help@logging.apache.org