Re: [Qemu-devel] [PATCH 07/17] json: learn to parse uint64 numbers

[Markus Armbruster <armbru@redhat.com>]

Marc-André Lureau <marcandre.lureau@gmail.com> writes:

> Hi
>
> On Mon, May 15, 2017 at 6:00 PM Markus Armbruster <armbru@redhat.com> wrote:
>
>> Marc-André Lureau <marcandre.lureau@redhat.com> writes:
>>
>> > Switch strtoll() usage to qemu_strtoi64() helper while at it.
>> >
>> > Replace temporarily the error in qnum_get_int() with values >INT64_MAX
>> > until the visitor is updated.
>> >
>> > Add a few tests for large numbers.
>> >
>> > Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
[...]
>> > diff --git a/qobject/qnum.c b/qobject/qnum.c
>> > index be6307accf..2f87952db8 100644
>> > --- a/qobject/qnum.c
>> > +++ b/qobject/qnum.c
>> > @@ -76,8 +76,8 @@ int64_t qnum_get_int(const QNum *qn, Error **errp)
>> >          return qn->u.i64;
>> >      case QNUM_U64:
>> >          if (qn->u.u64 > INT64_MAX) {
>> > -            error_setg(errp, "The number is too large, use qnum_get_uint()");
>> > -            return 0;
>> > +            /* temporarily accepts to cast to i64 until visitor is switched */
>> > +            error_report("The number is too large, use qnum_get_uint()");
>>
>> Awkward.  Can we avoid this somehow?
>>
>>
> by merging the next two patches? I think it's easier to understand why and
> how I came up with the following fix.

I played with it a bit.  Here's something that appears to do the trick:

PATCH 07: Drop changes to qobject/qnum.c tests/check-qnum.c
          Move change to tests/test-qobject-input-visitor.c to PATCH 08
PATCH 08: Move PATCH 09's change to hw/i386/acpi-build.c here
PATCH 09: No need to revert the changes dropped in PATCH 07; patch is
          now empty

Less churn, no temporary breakage, same end result.

Pushed as branch hack-qapi-qnum for your convenience:
https://repo.or.cz/qemu/armbru.git/shortlog/refs/heads/hack-qapi-qnum

Original is on branch review-qapi-qnum.

>> >          }
>> >          return qn->u.u64;
>> >      case QNUM_DOUBLE:
[...]