[vlmc-devel] [PATCH 5/7] GenericRenderer Implement setRenderTarget

[yikei lu]

well then the ownership of the pointer will change. And if
GenericRenderer should have the ownership, I think we don't have to
use unique_ptr.

Or we can use std::unique_ptr outside GenericRenderer, and pass it as
a raw pointer, which won't change the ownership


2016-04-17 17:29 GMT+09:00 Hugo Beauzée-Luyssen <hugo at beauzee.fr>:
> On 04/17/2016 10:27 AM, yikei lu wrote:
>>
>> I think it would be better to use shared_ptr rather than unique_ptr
>> because we we could call setRenderTarget outside the
>> GenericRenderer-based classes,
>>
>>
>>
>> 2016-04-17 17:00 GMT+09:00 Hugo Beauzée-Luyssen <hugo at beauzee.fr>:
>>>
>>> On 04/17/2016 09:45 AM, yikei lu wrote:
>>>>
>>>>
>>>> Yeah, I also once thought that it would be great to use unique_ptr,
>>>> but still I don't really know when we should use it since I don't use
>>>> it so much. :)
>>>>
>>>> I think this is simple enough to manage it, but it might be better to
>>>> use unique_pt because when we setRenderTarget, the pointer should be
>>>> used only in GenericRenderer. But the parent renderer has the
>>>> ownership of the pointer.
>>>>
>>>> Yes I think it's safe to use unique_pt :)
>>>> Regards
>>>>
>>>> 2016-04-17 16:34 GMT+09:00 Hugo Beauzée-Luyssen <hugo at beauzee.fr>:
>>>>>
>>>>>
>>>>> On 04/16/2016 08:22 AM, Yikai Lu wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>> ---
>>>>>>     src/Renderer/GenericRenderer.cpp | 8 ++++++++
>>>>>>     src/Renderer/GenericRenderer.h   | 5 +++++
>>>>>>     2 files changed, 13 insertions(+)
>>>>>>
>>>>>> diff --git a/src/Renderer/GenericRenderer.cpp
>>>>>> b/src/Renderer/GenericRenderer.cpp
>>>>>> index 598b59e..49ea01b 100644
>>>>>> --- a/src/Renderer/GenericRenderer.cpp
>>>>>> +++ b/src/Renderer/GenericRenderer.cpp
>>>>>> @@ -27,6 +27,7 @@
>>>>>>     GenericRenderer::GenericRenderer()
>>>>>>         : m_sourceRenderer( nullptr )
>>>>>>         , m_paused( false )
>>>>>> +    , m_renderTarget( nullptr )
>>>>>>     {
>>>>>>         m_eventWatcher = new RendererEventWatcher;
>>>>>>     }
>>>>>> @@ -35,6 +36,7 @@ GenericRenderer::~GenericRenderer()
>>>>>>     {
>>>>>>         delete m_sourceRenderer;
>>>>>>         delete m_eventWatcher;
>>>>>> +    delete m_renderTarget;
>>>>>>     }
>>>>>>
>>>>>>     bool
>>>>>> @@ -49,6 +51,12 @@ GenericRenderer::isRendering() const
>>>>>>         return m_isRendering;
>>>>>>     }
>>>>>>
>>>>>> +void
>>>>>> +GenericRenderer::setRenderTarget( Backend::IRenderTarget *target )
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> I'm enclined to think this should take a std::unique_ptr instead of a
>>>>> raw
>>>>> one, opinion?
>>>>>
>>>>>
>>>>>> +{
>>>>>> +    m_renderTarget = target;
>>>>>> +}
>>>>>> +
>>>>>>     #ifdef WITH_GUI
>>>>>>     RendererEventWatcher*
>>>>>>     GenericRenderer::eventWatcher()
>>>>>> diff --git a/src/Renderer/GenericRenderer.h
>>>>>> b/src/Renderer/GenericRenderer.h
>>>>>> index 6fb326a..fbc9a5e 100644
>>>>>> --- a/src/Renderer/GenericRenderer.h
>>>>>> +++ b/src/Renderer/GenericRenderer.h
>>>>>> @@ -29,6 +29,7 @@
>>>>>>
>>>>>>     #include "Workflow/Types.h"
>>>>>>     #include "Tools/RendererEventWatcher.h"
>>>>>> +#include "Backend/IRenderTarget.h"
>>>>>>
>>>>>>     class   Clip;
>>>>>>     class   Media;
>>>>>> @@ -135,6 +136,8 @@ public:
>>>>>>          */
>>>>>>         bool                            isRendering() const;
>>>>>>
>>>>>> +    void                            setRenderTarget(
>>>>>> Backend::IRenderTarget* target );
>>>>>> +
>>>>>>         RendererEventWatcher*           eventWatcher();
>>>>>>     protected:
>>>>>>         Backend::ISourceRenderer*       m_sourceRenderer;
>>>>>> @@ -150,6 +153,8 @@ protected:
>>>>>>         bool                            m_paused;
>>>>>>         bool                            m_isRendering;
>>>>>>
>>>>>> +    Backend::IRenderTarget*         m_renderTarget;
>>>>>> +
>>>>>>     public slots:
>>>>>>         /**
>>>>>>          *  \brief      This SLOT will be called when the time cursor
>>>>>> has
>>>>>> changed.
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Vlmc-devel mailing list
>>>>> Vlmc-devel at videolan.org
>>>>> https://mailman.videolan.org/listinfo/vlmc-devel
>>>>
>>>>
>>>> _______________________________________________
>>>> Vlmc-devel mailing list
>>>> Vlmc-devel at videolan.org
>>>> https://mailman.videolan.org/listinfo/vlmc-devel
>>>>
>>> Ok, I'll wait for your updated patch then :)
>>> Thanks!
>>>
>>> _______________________________________________
>>> Vlmc-devel mailing list
>>> Vlmc-devel at videolan.org
>>> https://mailman.videolan.org/listinfo/vlmc-devel
>>
>> _______________________________________________
>> Vlmc-devel mailing list
>> Vlmc-devel at videolan.org
>> https://mailman.videolan.org/listinfo/vlmc-devel
>>
> I don't see why that would change with unique_ptr ?
>
> _______________________________________________
> Vlmc-devel mailing list
> Vlmc-devel at videolan.org
> https://mailman.videolan.org/listinfo/vlmc-devel