Re: [PATCH 05/10] mf: Get max depth of branches in input topology in _Load().

[Sergio Gómez Del Real]

On 22/06/20 6:44 a. m., Nikolay Sivov wrote:
> On 6/15/20 4:41 AM, Sergio Gómez Del Real wrote:
> > +static int topology_loader_get_branch_depth(IMFTopologyNode *node, int level)
> > +{
> > - IMFTopologyNode *current_node;
> > - MF_TOPOLOGY_TYPE node_type;
> > - static int max_level = 0;
> > -
> > - if (level > max_level)
> > -
> >    max_level = level;
> > -
> > - IMFTopologyNode_GetNodeType(node, &node_type);
> > - if (node_type == MF_TOPOLOGY_SOURCESTREAM_NODE || node_type == MF_TOPOLOGY_TRANSFORM_NODE)
> > - {
> > -
> >    int stream = 0;
> > -
> >    DWORD input_stream;
> > -
> > -
> >    while (SUCCEEDED(IMFTopologyNode_GetOutput(node, stream++, &current_node, &input_stream)))
> > -
> >        topology_loader_get_branch_depth(current_node, level + 1);
> > - }
> > -
> > - return max_level;
> >
> > +}
>
> I don't understand the purpose of this. Could you elaborate? I see it's
> used in 6/10, but why can't we
> iterate until output node is reached?

At any point in the branch, a node could have multiple output streams, 
issuing multiple sub-branches, leading to other nodes with potential for 
further sub-branching. This gives some generality without complicating 
matters too much. The idea is to get the deepest level present in the 
branch(es) starting at a source, hence the use of max_level.
> P.S. regardless of the above, static variable should not be used for this.

Why?