Issue 90 Further Clarifications

Dustan B Helm dustan.helm at utexas.edu
Thu Dec 3 22:03:55 UTC 2020


On Wed, Nov 25, 2020 at 12:44 AM Peter Krempa <pkrempa at redhat.com> wrote:

> On Tue, Nov 24, 2020 at 15:25:01 -0600, Dustan B Helm wrote:
> > On Tue, Nov 24, 2020 at 12:59 AM Peter Krempa <pkrempa at redhat.com>
> wrote:
> > > On Mon, Nov 23, 2020 at 17:17:15 -0600, Dustan B Helm wrote:
>
> [...]
>
> - any XML addition requires tests to be added
>
>     In this case you'll be also needing to add a qemu commandline test
>     later so I suggest you add your XML test as a new file to:
>     tests/qemuxml2argvdata, and enable the test in
>     tests/qemuxml2xmltest.c (output file is stored in
>     tests/qemuxml2xmloutdata, you can also run the test with
>     VIR_TEST_REGENREATE_OUTPUT=1 env variable set which generates the
>     proper output file)
>
>     Always use one of the _CAPS_LATEST macros for invoking the test
>     regardless of prior art.
>
>     You can add multiple disks to the tested XML to test any combination
>     of the <nfs> parameters.
>
>     Please make sure to list at least one example of the NFS disk being
>     a <backingStore> of a <source>
>     (see tests/qemuxml2argvdata/disk-backing-chains.xml for inspiration)
>
>     Don't forget to also document the new elements in
>     docs/formatdomain.rst
>
We don’t understand exactly how the testing frameworks are functioning.
With regards to our XML testing, the tests/qemuxml2argvdata/ directory
contains for each test both an xml file and an args file- is this .args
file the expected output qemu command we want to generate from the input
.xml file?

If so, how do we invoke the VIR_TEST_REGENERATE_OUTPUT flag to create this
.args file, and how concerned should we be about non-disk information
provided in the .xml file as input? Additionally, since our source name (a
file path) needs to include the NFS export at the beginning, how would we
set up the XML input? Can the file path and NFS export be chosen
arbitrarily?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/libvir-list/attachments/20201203/ad510a8a/attachment-0001.htm>


More information about the libvir-list mailing list