Re: [Qemu-devel] [PATCH 1/2] qemu-iotests: add _filter_qmp_events() for filtering out QMP events

[Max Reitz <mreitz@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 1786 bytes --]

On 09.02.2016 14:23, Sascha Silbe wrote:
> The order of some QMP events may depend on the architecture being
> tested. Add support for filtering out QMP events so we can use a
> single reference output for all architecture when the test doesn't
> care about the events.
> 
> Signed-off-by: Sascha Silbe <silbe@linux.vnet.ibm.com>
> ---
>  tests/qemu-iotests/common.filter | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
> index 84b7434..b908aa2 100644
> --- a/tests/qemu-iotests/common.filter
> +++ b/tests/qemu-iotests/common.filter
> @@ -178,6 +178,12 @@ _filter_qmp()
>          -e '    QMP_VERSION'
>  }
>  
> +# remove QMP events from output
> +_filter_qmp_events()
> +{
> +    sed -e '/^{\(.*, \)"event": ".*}$/ d'
> +}

There is a pretty good reason test 067 uses -qmp-pretty (as you yourself
say, the lines get pretty long otherwise, and if we have any change
within, the whole line needs to be changed). Using the following ugly
piece of code here instead, we would still be able to use it:

tr '\n' '\t' \
  | sed -e
's/{\s*"timestamp":\s*{[^}]*},\s*"event":[^,}]*\(,\s*"data":\s*{[^}]*}\)\?\s*}\s*//g'
\
  | tr '\t' '\n'

(I'm too lazy for multi-line sed, obviously; and this will break if the
data contains any JSON objects in turn.)

The correct way would be to actually parse the JSON (using perl, python
or whatever) and remove all the top-level objects containing an "event"
key, obviously... But that's probably too much.

I'm not strictly against just dropping -qmp-pretty in 067, but there is
a good reason it's there.

Max

> +
>  # replace driver-specific options in the "Formatting..." line
>  _filter_img_create()
>  {
> 



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]