Re: [Bug-zile] astr_cstr

[Gary V. Vaughan]

Hi Reuben,

On 8 Sep 2011, at 04:27, Reuben Thomas wrote:
> On 7 September 2011 19:50, Gary V. Vaughan <address@hidden> wrote:
>> 
>>> commit 5171aa758f7354e8db522f341a2dc55563e82a62
>>> Author: Gary V. Vaughan <address@hidden>
>>> Date:   Thu Sep 1 21:57:27 2011 +0700
>> 
>> Feel free to revert.
> 
> Could you possibly do that?

Done.

>> Maybe this manywarnings macro is really far more trouble than it's worth?
> 
> I think it's a great pity that there's no way to annotate specific bits
> of code as not passing a particular warning; at present, the best one
> can do is to turn off a particular warning for a particular file. I
> think we should do that for now.

Agreed.

> The only other thing I can think of
> doing is turning off -Werror and putting up with a few warnings.

Agreed. Reluctantly. To me, manywarnings is more of a nuisance than a tool,
but it's easy enough to not use --enable-gcc-warnings if I don't want to
be bothered by a list of warnings -- some of which I have to remember are
not important and some of which I have to find out whether they are new
and whether they need investigation or corrective action.

Worse, what are we to do with warnings like this though, which are spewed
out by the compiler without additional warning flags enabled?

$ git clone --reference ../Devo/zile--savannah--0 git://git.sv.gnu.org/zile zile
...
$ cd zile
$ ./bootstrap
...
$ ./configure
...
$ make
...
src/astr.c: In function 'castr_new_nstr':
src/astr.c:67: warning: assignment discards qualifiers from pointer target type
...

At the moment, there is only one (which makes it all the more annoying!)
but it's easy to start ignoring the real warnings when we end up with a few
noisy warnings that we have to learned to ignore.

Cheers,
-- 
Gary V. Vaughan (gary AT gnu DOT org)