[RFC][PATCH] -d_lines

[Richard E Maine]

On Aug 17, 2005, at 7:09 PM, Brooks Moses wrote:

> In any case, I agree with you and Richard that this should only be on 
> by default in -std=legacy, if that.  (It may be worth having it on in 
> -std=legacy, given that that option really seems to mean "accept 
> anything that we can accept".)

The -std=legacy switch (if that's what it is called) is what I was 
referring to as an "accept a whole bunch of common nonstandard things" 
switch. As such, I'm neutral about whether that should change the 
default for d_lines (whatever it gets called) or not. It seems 
plausible, anyway. I will note that, since -d_lines has 3 options, it 
wouldn't be a matter of just having it "on", but rather of changing the 
default to a different one of the 3 values.

As an aside, is -std=legacy really very close to "accept anything that 
we can accept?" If so, I'd find it strangely named. I haven't checked 
in detail, but I just assumed from the name that it referred to 
accepting things that were... well... legacy extensions. For example, 
to the extent that new f2003 features are implemented, would one really 
expect acceptance of them to be turned on by -std=legacy? That 
terminology would confuse me.

I agree that this d_lines stuff counts as a legacy feature, so that 
aspect of my comment is more an abstract side note than directly to the 
point of this case.

-- 
Richard Maine                |  Good judgment comes from experience;
Richard.Maine@nasa.gov       |  experience comes from bad judgment.
                             |        -- Mark Twain