[30677] trunk/dports/security/authforce

[Ryan Schmidt]

On Nov 4, 2007, at 16:54, Mark Grimes wrote:

> On Nov 4, 2007, at 13:42, Ryan Schmidt wrote:
>
>> On Nov 4, 2007, at 03:31, Mark Grimes wrote:
>>
>>> On Nov 4, 2007, at 1:04 AM, Ryan Schmidt wrote:
>>>
>>>> On Nov 3, 2007, at 15:38, source_changes at macosforge.org wrote:
>>>>
>>>>> --- trunk/dports/security/authforce/Portfile    2007-11-03  
>>>>> 20:15:07 UTC (rev 30676)
>>>>> +++ trunk/dports/security/authforce/Portfile    2007-11-03  
>>>>> 20:38:14 UTC (rev 30677)
>>>>> @@ -22,6 +22,9 @@
>>>>> configure.args    --mandir=${prefix}/share/man --infodir=$ 
>>>>> {prefix}/share/info
>>>>> configure.cppflags-append "-L${prefix}/lib"
>>>>> configure.cflags-append "-no-cpp-precomp -flat_namespace - 
>>>>> undefined suppress -lintl -L${prefix}/lib"
>>>>> +
>>>>> +patchfiles  patch-http.c
>>>>> +
>>>>
>>>>
>>>> FYI: The patchfile should be named "patch-http.c.diff". "port  
>>>> lint" has some other recommendations for this portfile as well:
>>>>
>>>>
>>>> $ sudo port lint
>>>> Password:
>>>> --->  Verifying Portfile for authforce
>>>> Warning: Line 2 should be a newline (after RCS tag)
>>>> Warning: Line 3 should be a newline (after PortSystem)
>>>> Warning: Line 16 has trailing whitespace before newline
>>>> Warning: Line 24 has trailing whitespace before newline
>>>> Warning: Line 32 has trailing whitespace before newline
>>>> Warning: Patchfile patch-http.c does not follow the source patch  
>>>> naming policy "patch-*.diff"
>>>> --->  0 errors and 6 warnings found.
>>>> $
>>>
>>> Aside from the strange "mandated" naming conventions, port lint  
>>> is suggesting (as warnings) that diff -p -u is insufficient,  
>>> which I find a little disturbing. Is there an alternative set of  
>>> diff flags that will make lint happy?
>>
>> I don't really understand. What is strange about the naming  
>> conventions? And what do you mean "diff -p -u is insufficient"?  
>> What does diff have to do with lint?
>
> You said lint had other suggestions aside from the .diff and I'm  
> saying this patch is made with diff -p -u so I am assuming all  
> these suggestions (lint warnings) can be dismissed.

Ah, I see. The port has no maintainer. I thought maybe you were its  
maintainer; I didn't look closely enough. In any case, I wanted to  
point out to you the patchfile naming situation, and rather than just  
state it, I wanted to show you (and others) how you can learn of this  
yourself (by using "port lint" before committing). In this case "port  
lint" also showed other errors unrelated to the changes you made, and  
if you were the maintainer, you might want to take care of those too.  
But since the port has no maintainer I went ahead and made the  
whitespace changes in r30700.