The actual size of an image depends on the host filesystem.  Therefore,
it should be filtered from the test output if it's not part of the test
itself (and if it is, the test should take extra care, like 106 does).

So let's put the filtering sed line from 067 and 087 into a common
filter function and use it in 184 and 191.


Max Reitz (2):
  iotests: Pull _filter_actual_image_size from 67/87
  iotests: Filter actual image size in 184 and 191

 tests/qemu-iotests/067           |  2 +-
 tests/qemu-iotests/087           |  2 +-
 tests/qemu-iotests/184           |  3 ++-
 tests/qemu-iotests/184.out       |  6 +++---
 tests/qemu-iotests/191           |  4 ++--
 tests/qemu-iotests/191.out       | 46 ++++++++++++++++++++--------------------
 tests/qemu-iotests/common.filter |  6 ++++++
 7 files changed, 38 insertions(+), 31 deletions(-)

-- 
2.13.6

On 10/09/2017 11:34 AM, Max Reitz wrote:
> The actual size of an image depends on the host filesystem.  Therefore,
> it should be filtered from the test output if it's not part of the test
> itself (and if it is, the test should take extra care, like 106 does).
> 
> So let's put the filtering sed line from 067 and 087 into a common
> filter function and use it in 184 and 191.
> 
> 
> Max Reitz (2):
>   iotests: Pull _filter_actual_image_size from 67/87
>   iotests: Filter actual image size in 184 and 191

Reviewed-by: Eric Blake <eblake@redhat.com>

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org

On 2017-10-09 18:34, Max Reitz wrote:
> The actual size of an image depends on the host filesystem.  Therefore,
> it should be filtered from the test output if it's not part of the test
> itself (and if it is, the test should take extra care, like 106 does).
> 
> So let's put the filtering sed line from 067 and 087 into a common
> filter function and use it in 184 and 191.

Thanks for the reviews, applied to my block branch:

https://github.com/XanClic/qemu/commits/block

Max