Currently we are "forcing" extractors to have a common interface to each of their own support file types, e.g., yadg supports a parameter defining the file type. Other libraries (particularly those developed outside of this WG) probably will not want to put in the work to do this, so we may consider allowing usage instructions to also be broken down by supported file type. In our current scheme this leads to a combinatorial explosion of usage instructions, so we may need to give it some thought...
Currently we are "forcing" extractors to have a common interface to each of their own support file types, e.g., yadg supports a parameter defining the file type. Other libraries (particularly those developed outside of this WG) probably will not want to put in the work to do this, so we may consider allowing usage instructions to also be broken down by supported file type. In our current scheme this leads to a combinatorial explosion of usage instructions, so we may need to give it some thought...