On 16 April 2015 at 15:50, Mike Holmes <[email protected]> wrote:

>
>
> On 16 April 2015 at 09:41, Taras Kondratiuk <[email protected]>
> wrote:
>
>> On 04/16/2015 04:29 PM, Mike Holmes wrote:
>>
>>> We have an 80 char limit that is frequently an issue for strings and we
>>> just accept that we ignore the warning.
>>>
>>> Allow split strings so that there is a valid alternative
>>> Thus the following example becomes a legal alternative to the > 80 chars
>>> warning.
>>>
>>> printf("\nThread %u (id=%d core=%d) had %u sync_failures"
>>>         " in %u iterations\n", thread_num,
>>>         .....
>>>
>>>
>> The reason to have an exception for printed strings length is to have
>> them in one line to be searchable in a codebase.
>>
>> In this example grep'ing for 'sync_failures in' won't find this string.
>>
> The above format string would better be expressed as
 printf("\nThread %u (id=%d core=%d) had %u sync_failures in "
        "%u iterations\n", thread_num,
E.g. break at a formatting directive.
But grepping using regular expressions ("had .* sync_failures in .*
iterations") would still fail.
So allowing for lines (with strings) longer than 80 chars is preferable.



> As with Google you just take a portion of the string if you really have an
> issue.
>
> I rarely include syntactic elements in a search, in the same way that if
> you search for this string verbatim with he specific thread id, core and
> number of failures set in the actual string seen on stdout you would not
> find the code either.
>
> --
> Mike Holmes
> Technical Manager - Linaro Networking Group
> Linaro.org <http://www.linaro.org/> *│ *Open source software for ARM SoCs
>
>
>
> _______________________________________________
> lng-odp mailing list
> [email protected]
> https://lists.linaro.org/mailman/listinfo/lng-odp
>
>
_______________________________________________
lng-odp mailing list
[email protected]
https://lists.linaro.org/mailman/listinfo/lng-odp

Reply via email to