On Dec 7, 2007, at 9:56 AM, Anders F Björklund wrote:

> Jeff Johnson wrote:
>
>> Which is not true with the (arguably "standard" rpmbuild  
>> invocation here,
>> at least as consistent as possible with what is documented in  
>> "Maximum RPM" and
>> "The RedHat RPM Guide" and with the "default" distributed rpm  
>> macro configuration):
>
>> /var/tmp/devtool-test-root: No such file or directory
>
> Maybe it needs a "mkdir -p" or "install -d" on the $RPM_BUILD_ROOT,
> or perhaps even some files in order to test-drive payload as well ?
>

Either that or changing the brp-foo scripts to be more careful about
non-existent %{buildroot}. Easier to add the lazy mkdir than it
is to change everyone's pwecious waluable build scwipts, sigh.

The typical packaging paradigm (on linux anyways) is

...
%install
rm -rf %{buildroot}
mkdir -p %{buildroot}

(aside) I've already made the explicit "rm -rf %{buildroot}"
superfluous. I'm already trying to figger how disruptive
a lazy "mkdir -p %{buildroot}" would be ...

There's always some packaging pathology that any *ahem*
rpmbuild helpfulness like "mkdir -p %{buildroot}" runs afoul of ...

Truly, trying to understand whay you don't see is very mysterious.
Consider the current segfault backtrace ... ;-)

73 de Jeff