Libguestfs - Mar 2022 - [libguestfs-common PATCH] mlcustomize/firstboot: take an optional priority in "add_firstboot_script"

This patch is fine, thanks ...
> +      it is taken as [default_prio], that is, 5000. If [prio] is smaller
than 0
> +      or greater than 9999, an Assert_failure is raised (the [prio]
parameter
> +      is not expected to depend on external data).
... but I'm confused by what you meant by "is not expected to depend
on external data"?

Anyway, ACK

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-df lists disk usage of guests without needing to install any
software inside the virtual machine.  Supports Linux and Windows.
http://people.redhat.com/~rjones/virt-df/

On Tue, Mar 08, 2022 at 04:00:15PM +0000, Richard W.M. Jones
wrote:
> This patch is fine, thanks ...
> 
> > +      it is taken as [default_prio], that is, 5000. If [prio] is
smaller than 0
> > +      or greater than 9999, an Assert_failure is raised (the [prio]
parameter
> > +      is not expected to depend on external data).
> 
> ... but I'm confused by what you meant by "is not expected to
depend
> on external data"?
> 
> Anyway, ACK
... but see my comment on patch 11:

https://listman.redhat.com/archives/libguestfs/2022-March/028400.html

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
libguestfs lets you edit virtual machines.  Supports shell scripting,
bindings from many languages.  http://libguestfs.org

On 03/08/22 17:00, Richard W.M. Jones wrote:
> This patch is fine, thanks ...
> 
>> +      it is taken as [default_prio], that is, 5000. If [prio] is
smaller than 0
>> +      or greater than 9999, an Assert_failure is raised (the [prio]
parameter
>> +      is not expected to depend on external data).
> 
> ... but I'm confused by what you meant by "is not expected to
depend
> on external data"?
In general, no assert() condition should depend on user-controlled data;
assert() is for catching programming bugs. If user input can lead to the
falsification of a condition (invariant etc), then that needs to be
caught with a different tool -- throw a different exception, report an
error, etc. So the idea here is that the priority should never depend on
user input.

(For example: in case the priority parameter were exposed by
virt-customize's --firstboot* options in the future, then the assert()
would no longer be OK. A different kind of exception would be necessary.)
> 
> Anyway, ACK
Thanks!
Laszlo