The CodecPipeline class uses the character 'underscore' to split the different encoding schemes in the format parameter used to initialize the class. However, in some situations one may wonder to select a different separator character to split the different schemes pipelined e.g. as in ticket [#275], the VIDEO_IMAGE keyword comes from an external source (Lima by ESRF). In this situation, the encoding format could cause problems if the separator 'underscore' is being used for pipelining, VIDEO and IMAGE will be splitted in two and interpreted as two different encoding schemes.
The proposal is to define a global configuration for this separator, in the way that could be changed on the future if any codec scheme names collides with the current separator.
OLD Tickets: #275
OLD Tickets: #306
In an internal ALBA discusion we agred to modify the API for manage several encoding schemes (use a list instead of a string separated by any character). So I will reject this feature and open a bug to adapt CodecPipeline class.