patch feedback:depends_test - port test - port lint -append

[Kurt Hindenburg]

Hello,

> On Sep 4, 2015, at 3:02 PM, Joshua Root <jmr at macports.org> wrote:
> 
> On 2015-9-4 07:43 , Kurt Hindenburg wrote:
>> Hi,
>>  Can I get some feedback on these patches?
>> 
>> https://trac.macports.org/ticket/38208 - add depends_test
> 
> r+
> 
> LGTM. My one concern would be that end users may not want irrelevant (to
> them) test deps showing up in depends:, rdeps and so forth, but we
> should come up with a general solution to that, not just omit them.
> 

committed r140488

>> https://trac.macports.org/ticket/45010 - port test stops if test.run is
>> not yes
> 
> r-
> 
> This shouldn't be checked in portmain but rather somewhere higher up in
> the vicinity of mportexec, where the deps are being computed and the
> target is known.

A new patch attached to https://trac.macports.org/ticket/45010 <https://trac.macports.org/ticket/45010>

> 
>> https://trac.macports.org/ticket/28640 - port lint check when should use
>> -append
> 
> r-
> 
> Overriding the deps set by a portgroup is not always incorrect.

Do you mean to say that there are times that the Portfile should override the PortGroup?  Do you think changing this lint to a warning suggesting they should verify if they should use ‘-append’ ?

Thanks
  Kurt

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.macosforge.org/pipermail/macports-dev/attachments/20150920/b8271f28/attachment-0001.html>