7 years, 1 month ago.

Best practices for duplicate libraries?

So, is there some kind of consensus or best practice on the following...

Developer wants to create/publish new library 'A' 'A' depends on existing libraries 'B' & 'C' 'B' depends on 'D'. 'C' also depends on 'D'.

Is the only solution to fork a version of 'B' and/or 'C' that doesn't resolve it's dependencies automatically on inclusion into a program? In other words, 'B' and 'C' still depend on 'D', but 'D' has been removed from 'B' and 'C' and is now a dependency of 'A' (just as 'B' and 'C' are)? In this case, is there any automatic method that the forked version of 'B' and 'C' can track changes in the original?

Be the first to answer this question.