On Fri, 2019-05-03 at 11:57 +0200, Michal Privoznik wrote:
> On 5/3/19 11:18 AM, Andrea Bolognani wrote:
> > On Thu, 2019-04-18 at 14:11 +0200, Michal Privoznik wrote:
> > > +if (virBufferCurrentContent(&buf1) ||
> > > +!virBufferCurrentContent(&buf2)) {
> > > +VIR_TEST_DEBUG("U
On 5/3/19 11:18 AM, Andrea Bolognani wrote:
On Thu, 2019-04-18 at 14:11 +0200, Michal Privoznik wrote:
If an error occurs in a virBuffer* API the idea is to free the
content immediately and set @error member used in error reporting
later. Well, this is not what how virBufferAddBuffer works.
Sig
On Thu, 2019-04-18 at 14:11 +0200, Michal Privoznik wrote:
> If an error occurs in a virBuffer* API the idea is to free the
> content immediately and set @error member used in error reporting
> later. Well, this is not what how virBufferAddBuffer works.
>
> Signed-off-by: Michal Privoznik
> ---
>
On 5/3/19 10:40 AM, Erik Skultety wrote:
On Thu, Apr 18, 2019 at 02:11:23PM +0200, Michal Privoznik wrote:
If an error occurs in a virBuffer* API the idea is to free the
content immediately and set @error member used in error reporting
later. Well, this is not what how virBufferAddBuffer works.
On Thu, Apr 18, 2019 at 02:11:23PM +0200, Michal Privoznik wrote:
> If an error occurs in a virBuffer* API the idea is to free the
> content immediately and set @error member used in error reporting
> later. Well, this is not what how virBufferAddBuffer works.
>
> Signed-off-by: Michal Privoznik
>
If an error occurs in a virBuffer* API the idea is to free the
content immediately and set @error member used in error reporting
later. Well, this is not what how virBufferAddBuffer works.
Signed-off-by: Michal Privoznik
---
src/util/virbuffer.c | 2 +-
tests/virbuftest.c | 32 +++
