On 06/18/2012 08:48 AM, Peter Krempa wrote:
On 06/15/12 06:18, Eric Blake wrote:
> The generator doesn't handle lists of virDomainSnapshotPtr, so
> this commit requires a bit more work than some RPC additions.
>
> +struct remote_domain_snapshot_list_all_children_ret {
> + remote_nonnull_domain_snapshot snapshots<>;
Hopefully we'll go with unbounded arrays in the future, but as with my
listing function: Shouldn't we use the limit that was imposed on name
list length also on this list? (REMOTE_DOMAIN_SNAPSHOT_LIST_NAMES_MAX).
Here's a case where I was copying from your series :)
Whichever way you go in your commit, I will copy.
> + int ret;
> +};
> +
> struct remote_domain_snapshot_lookup_by_name_args {
> remote_nonnull_domain dom;
> remote_nonnull_string name;
ACK with the limit added. I'm a fan of unbounded arrays, but I think we
should be consistent.
I'm starting to agree that unbounded intermediates are okay because we
still have the overall bound on the max RPC; but I'm also thinking that
a global swap to unbounded would be better than introducing one instance
of an unbounded use. So being consistent by having this commit preserve
a bound is fine by me.
--
Eric Blake eblake(a)redhat.com +1-919-301-3266
Libvirt virtualization library
http://libvirt.org