Actually... for at least builds on 14.04 the majority of warnings had
been cleared up.

The majority of what had remained were in external libs like libcoap.


I think the most common way for people to have Jenkins flag warnings is
to build with the warnings-as-errors flag. We're probably not to that
point yet.


On 08/04/2015 08:07 AM, Light, John J wrote:
> I wasn't complaining about the new warning level.  I was pointing out the 
> irony that we hadn't eliminated the warnings that appeared BEFORE we changed 
> the warning level.
> 
> I'm surprised Jenkins doesn't complain about warnings.
> 
> John
> 
> -----Original Message-----
> From: Jon A. Cruz [mailto:jonc at osg.samsung.com] 
> Sent: Monday, August 03, 2015 3:35 PM
> To: Keane, Erich; Light, John J
> Cc: iotivity-dev at lists.iotivity.org
> Subject: Re: [dev] warnings!
> 
> FYI, the increased warnings had already flagged at least one bug where 
> network send errors were being missed due to a check that looked at a signed 
> value being less than zero. Catching such a bug in code review that should 
> have been flagged by the compiler was the reason the lack of -Wextra was 
> noticed in the first place.
> 
> First cleanups fixed Ubuntu 14.04 from 9.0k warnings down to 0.6k warnings. 
> Similar passes for 12.04 will follow, but as it stands builds were only 
> seeing about 2k warnings.
> 
> The velocity on warning cleanup should fairly quickly get to the point again 
> where it is easy to spot significant issues as (or before) they are 
> introduced.
> 
> 
> On 08/03/2015 03:00 PM, Keane, Erich wrote:
>> As far as the 'sea of non-critical warnings', you aren't wrong.
>> However, NOW is sorta the best time to do this, since we are between 
>> releases, and it gives us as much time as possible to fix them.  All 
>> patches to fix warnings are looked on quite favorably :)
>>
>> -Erich
>>
>>
>> On Mon, 2015-08-03 at 21:55 +0000, Light, John J wrote:
>>> I?ve noticed a great increase in the number of warnings during build.
>>> There have been more warnings in recently merged code, but this 
>>> lastest increase seems to be the result of ratcheting up the warning 
>>> threshold.
>>>
>>>  
>>>
>>> I suspect we are well into the territory where critical warnings 
>>> won?t be seen because they will be lost in a sea of non-critical warnings.
>>>
>>>  
>>>
>>> Leaving that aside, I have a coding question in this new regime.
>>>
>>>  
>>>
>>> A C file I am modifying but didn?t write has the following code:
>>>
>>>  
>>>
>>>     OCPersistentStorage ps = {};
>>>
>>>     ps.open = client_fopen;
>>>
>>>     ps.read = fread;
>>>
>>>     ps.write = fwrite;
>>>
>>>     ps.close = fclose;
>>>
>>>     ps.unlink = unlink;
>>>
>>>  
>>>
>>> This gets a warning about each line, like:
>>>
>>>  
>>>
>>>     warning: missing initializer for member 
>>> 'OCPersistentStorage::open'
>>>
>>>  
>>>
>>> I can eliminate the warnings by coding it thus:
>>>
>>>  
>>>
>>>     OCPersistentStorage ps = { client_fopen, fread, fwrite, fclose, 
>>> unlink };
>>>
>>>     OCRegisterPersistentStorageHandler(&ps);
>>>
>>>  
>>>
>>> But this seems more fragile since the ordering matters.
>>>
>>>  
>>>
>>> Is there a C initialization method I?m missing?
>>>
>>>  
>>>
>>> John
>>>
>>>
>>> _______________________________________________
>>> iotivity-dev mailing list
>>> iotivity-dev at lists.iotivity.org
>>> https://lists.iotivity.org/mailman/listinfo/iotivity-dev
>>
>> _______________________________________________
>> iotivity-dev mailing list
>> iotivity-dev at lists.iotivity.org
>> https://lists.iotivity.org/mailman/listinfo/iotivity-dev
>>
> 
> --
> Jon A. Cruz - Senior Open Source Developer Samsung Open Source Group jonc at 
> osg.samsung.com
> _______________________________________________
> iotivity-dev mailing list
> iotivity-dev at lists.iotivity.org
> https://lists.iotivity.org/mailman/listinfo/iotivity-dev
> 

-- 
Jon A. Cruz - Senior Open Source Developer
Samsung Open Source Group
jonc at osg.samsung.com

Reply via email to