On 3/4/19 10:34 PM, Eric Blake wrote:
> Upcoming patches will add new flags for increasing the amount of
> information present in <domain> dumpxml, but where the source
> of that information is tied to somewhere other than the active
> or offline domain sub-definition. Make those extensions easier
> by updating internal callers to pass in a struct, rather than
> adding new parameters for each extension, so that later patches
> only have to patch callers that care about a new member of the
> struct rather than all callers.
>
> I considered updating virDomainDefFormat(), but as there were
> so many existing callers, it was easier to just add a new
> wrapper function virDomainDefFormatFull() which takes the full
> struct, while making the existing interface forward on to the
> full one.
>
> Since all callers are being adjusted anyway, reorder the
> parameters of virDomainDefFormatInternal to put buf first, as
> that tends to be more typical.
>
> Signed-off-by: Eric Blake <ebl...@redhat.com>
> ---
> src/conf/domain_conf.h | 18 ++++++++++++++----
> src/conf/domain_conf.c | 33 ++++++++++++++++++++++++++-------
> src/conf/snapshot_conf.c | 5 ++++-
> src/network/bridge_driver.c | 2 +-
> src/qemu/qemu_domain.c | 12 ++++++------
> 5 files changed, 51 insertions(+), 19 deletions(-)
>
The structure formatting of _virDomainDefFormatData and friends is a bit
non-standard from what is typically done, but not incorrect.
Reviewed-by: John Ferlan <jfer...@redhat.com>
John
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
