Re: [PATCH] zsh/random module [UPDATED]

[Clinton Bunch <cdb_zsh@xxxxxxxxxxx>]

On 11/23/2022 3:42 PM, dana wrote:
[snip]
> On Wed 23 Nov 2022, at 13:46, Daniel Shahaf wrote:
> > > +%test
> > > +  getrandom -U 56
> > > +1f:Checking if system has a kernel random source
> > > +?(eval):1: No kernel random pool found.
> > > +
> > The test point's code has nothing to do with the test point's
> > description and expected errput.
> The assertion is backwards but i guess this is just testing to see if the
> module works on the system. But should that even be a test? Shouldn't there
> instead be a %prep that just leaves the whole file unimplemented in this case?
> Unless we're expecting the module to be both available and usable on every zsh
> build
The test is because the module is designed to be run on operating 
systems where the kernel module for urandom is an add-on package. 
(Specifically, my problem child for the last decade, HP-UX), so may be 
available on the build system, but the zsh package installed on one 
where it isn't available.  On further thought, since the reverse is also 
true, though, the test may be misguided.


[snip]

> dana