[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#455512: closed by sean finney <seanius@debian.org> (not a bug really)



reopen 455512
thanks

On Sat, Dec 22, 2007 at 10:06:04AM +0000, Debian Bug Tracking System wrote:
> 
> policy doesn't say anything about the bulid target needing to be idempotent by 
> itself.

I didn't say Policy requires it.  It's common to use "important" severity for
FTBFS bugs that are nevertheless not RC.  But feel free to adjust severity if
that bothers you.

> it does say that you can require debian/rules clean to be run 
> first-- which if you do should solve the problem with interrupted builds.  

Doing that can be really inconvenient when you're modifiing it.  Every little
change requires waiting a few minutes.

If there was a reason for not supporting it in this particular package, I
wouldn't bother, but running "mkdir -p" instead of "mkdir" is not a real
inconvenient.

> there are many easily constructable situations where running build twice in a 
> row might produce unwanted results (maybe if someone echo's or sed's onto 
> files in the build target and undoes them in the clean target).

Which is a *really* bad idea and also defeats most of the purpose of using make
in the first place.

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call, if you are unable to speak?
(as seen on /.)



Reply to: