Re: [Qemu-devel] [PATCH v8 08/18] qapi: Lazy creation of array types

[Eric Blake <eblake@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 1421 bytes --]

On 10/14/2015 01:15 AM, Markus Armbruster wrote:
> Eric Blake <eblake@redhat.com> writes:
> 
>> Commit ac88219a had several TODO markers about whether we needed
>> to automatically create the corresponding array type alongside
>> any other type.  It turns out that most of the time, we don't!
>>
>> As part of lazy creation of array types, this patch now assigns
>> an 'info' to array types at their point of first instantiation,
>> rather than leaving it None.
> 
> I'm afraid this flips the value of .is_implicit() to False.  Currently
> harmless, but let's keep it correct anyway.
> 
> The obvious fix is to define the trivial override method:
> 
>     def is_implicit(self):
>         return True
> 
> But I'd rather do *all* the "give implicit types info" work in "qapi:
> Track location that created an implicit type", i.e. move the plumbing of
> info there, add the override method there, drop the "As part of"
> paragraph from the commit message here.  I append what's left of this
> patch then.  I like it, because the patch that actually changes
> generated code (this one) becomes really simple, and the lengthened
> patch remains mere info-plumbing that doesn't affect the generated code.

Makes sense, so I agree with how you've redone the current state of
qapi-next.

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 604 bytes --]