On 01/12/2016 21:21, Nicolas George wrote: > It makes the API slave to changes in these standards and will cause no > end of problems if these standards add gaps (which is exactly what is > currently happening) or if the API wants to support values coming from > different standards.
Do you have a specific standard in mind? > I think a better design decision would be to use it as a real C enum, > with automatic consecutive values, and use tables whenever the values > for 23001-8_2013 are needed. Surely it gives more flexibility at the cost of 1 level of indirection and a function call. Do you have patch for it? lu _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
