httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Guenter Knauf <fua...@apache.org>
Subject Re: svn commit: r1040177 - /httpd/httpd/trunk/modules/http/http_protocol.c
Date Fri, 03 Dec 2010 08:52:06 GMT
Am 02.12.2010 10:39, schrieb Joe Orton:
> On Mon, Nov 29, 2010 at 04:37:49PM -0000, fuankg@apache.org wrote:
>> URL: http://svn.apache.org/viewvc?rev=1040177&view=rev
>> Log:
>> Supress compiler warning.
> ...
>> ==============================================================================
>> --- httpd/httpd/trunk/modules/http/http_protocol.c (original)
>> +++ httpd/httpd/trunk/modules/http/http_protocol.c Mon Nov 29 16:37:49 2010
>> @@ -204,7 +204,12 @@ AP_DECLARE(int) ap_set_keepalive(request
>>        *   THEN we can be persistent, which requires more headers be output.
>>        *
>>        * Note that the condition evaluation order is extremely important.
>> +     *
>> +     * Silent compiler warnings for (r->chunked = 1) with #pragma
>>        */
>> +#ifdef __WATCOMC__
>> +#pragma disable_message(105)
>> +#endif
>
> Eww.  Do you really need to litter the source code with this stuff?  Can
> you not flip compiler switches in the Makefiles?  Regards, Joe
no. If I use a compiler switch it aplies for all source files which is 
certainly not what I want - instead I only want to suppress this one 
warning about assignment inside condition where the comment states that 
its ugly but wanted, while at all other places the very same warning is 
important and useful since you can just acciedently miss a 2nd '=' where 
you dont want to assign but only compare.
I even expect to add more pragmas for other compilers, f.e. gcc, at this 
place if we would compile with all possible warnings (maybe -Wall is 
already enough).

Gün.



Mime
View raw message