Patch Set 1:

> Regarding deprecated, I think it's good that it fails when using a
 > deprecated symbol, this way it forces us to fix it.

That part is good, yes, but the pattern that has emerged recently is: our 
jenkins builds use -Werror. That means as soon as we create some foo2() and 
deprecate foo(), all of the builds that still use foo() flare up red. It's 
still ok for that other project to use foo(), we haven't yet seen a need for it 
to move the newer API and use the new, safer features / the clarified signature 
/ the more universal application. So what do I do? I *don't* mark foo() as 
deprecated. I make a note in my mind to first look at all the dependant 
projects and get rid of foo(), and *then* mark it deprecated. And then that 
never happens. The conclusion is we will refrain from marking anything 
deprecated if our builds fail because of deprecation warnings.

So my opinion is quite strongly against failing for deprecation. We really need 
all those other warnings as errors, yes, deprecation as error breaks the 
workflow.

 > Regarding the warning pragmas, I'd bet -Wno-error=cpp also disabled
 > other C preprocessor warnings which we may not want to disable. If
 > you want to add explicit warning messages to show during the build,
 > use #pragma message instead.

man gcc:

        -Wno-cpp
           (C, Objective-C, C++, Objective-C++ and Fortran only)

           Suppress warning messages emitted by "#warning" directives.

and

       -Wno-pragmas
           Do not warn about misuses of pragmas, such as incorrect parameters, 
invalid syntax, or conflicts between pragmas.  See also -Wunknown-pragmas.

So clearly =pragma is wrong and =cpp has exactly the desired effect. (Unless 
the manpage is wrong.)

-- 
To view, visit https://gerrit.osmocom.org/7096
To unsubscribe, visit https://gerrit.osmocom.org/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Ibcc8238584a585434b39a046cd2d7e18ddaf7f8c
Gerrit-PatchSet: 1
Gerrit-Project: libasn1c
Gerrit-Branch: master
Gerrit-Owner: Neels Hofmeyr <nhofm...@sysmocom.de>
Gerrit-Reviewer: Harald Welte <lafo...@gnumonks.org>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: Neels Hofmeyr <nhofm...@sysmocom.de>
Gerrit-Reviewer: Pau Espin Pedrol <pes...@sysmocom.de>
Gerrit-HasComments: No

Reply via email to