RE: PATCH for java.nio FileChannel and MappedByteBuffer

[Andrew Haley <aph at redhat dot com>]

Boehm, Hans writes:
 > Isn't this argument backwards?
 > 
 > As far as I'm concerned, unindented "ifdef"s are just as bad as
 > unindented "if"s.  I can't think of any reason to indent one and
 > not the other.

 > (There is an issue with #ifdefs that don't nest correctly inside
 > other program structure, and are thus unavoidably hard to read.
 > But I don't think that was the topic here.)

 > If emacs doesn't indent "ifdef"s, it needs to be fixed.

Okay.  Really, I don't care/mind.  But before any changes are made to
coding standards, it's imperative to ensure that tools are in place.
We want to get the utmost out of our volunteers, both in quality and
quantity.  We aren't going to do that if we make life hard for them.

To summarize: I'm happy to help with the building work.  But I'm not
prepared to mix concrete with my bare hands.

 > I don't use emacs, so I don't know.  But I'm also not a great
 > believer in fully automatic indentation:
 > 
 > 		do_something;
 > 		/* informal assertion */
 > 		do_something_else;
 > 		and_something_else;
 > 
 > looks to a formatter just like
 > 
 > 		do_something;
 > 		/* Informal description of big task	*/
 > 		  step1_of_big_task;
 > 		  step2_of_big_task;

I don't see your point here: the former is right, the latter is wrong.
If you really want the block to be indented, put in some braces.

Andrew.