Re: [PATCH for-4.5 RFC] pygrub: Fix regression from c/s d1b93ea, attempt 2

[Boris Ostrovsky <boris.ostrovsky@oracle.com>]

On 11/17/2014 11:59 AM, Andrew Cooper wrote:
> On 17/11/14 16:58, Ian Campbell wrote:
>> On Mon, 2014-11-17 at 15:19 +0000, Andrew Cooper wrote:
>>> c/s d1b93ea causes substantial functional regressions in pygrub's ability to
>>> parse bootloader configuration files.
>> Please can you and Boris both provide examples of (ideally real-world)
>> configuration files which exhibit these failures as patches against
>> tools/pygrub/examples/.
>>
>> Boris, in your case I mean the one which caused you to write the
>> original patch, which I should have remembered to ask for at the time.


I wanted to be able to parse grub2's default values set as strings, such as

    set default="Fedora, with Xen xen and Linux 3.17.0-rc3"

or, for submenus

    set default="Advanced options for Fedora (with Xen 
hypervisor)>Fedora, with Xen xen and Linux 3.17.0-rc3"

Original pygrub would not be able to understand the string and default 
to zero.

-boris


>>
>> Andy, in your case it would make sense to include at least one in this
>> patch I think.
>>
>> Ian.
>>
> examples/ doesn't help.  The parsers themselves don't exhibit the bug.
> It is only when pygrub itself attempts to interact with the parsed
> config does the issue exhibits itself.
>
> I would have provided an example if it would have helped.
>
> ~Andrew