Greetings! I attempted to see if there was any work or discussion on this, but brought up nothing, I hope I am not duplicating some effort/thought here. I recently developed a small script allowing me to assign ports to groups, using aliases, so that I may autoconnect these groups to ardour buses. However, I quickly discovered there is a limit of two aliases per port, which is admittedly workable, but not pleasantly so. perhaps there is a better method of achieving this functionality? but I wanted to continue using jack-matchmaker. So I've patched a local copy of jack2 to allocate an array for aliases per port, and allow its size to be specified via the compile time flag --aliases-per-port. However, the implementation is not the best, I needed to introduce a seperate version of jack_port_get_aliases, suffixed with _b, which can be used in conjunction a new function, jack_port_max_aliases, to safely retrieve aliases beyond the first two. applications using the old jack_port_get_aliases will still be shown only the first two aliases in the list. however the matching functions will match against all aliases. Additionally there is what looks like serialization&matching code in JackDriver, which i have not yet attempted to patch. I've also patched copys of jack-matchmaker and luajack to make use of this. I dont imagine such functionality would be accepted into jack in its current state? it seems there is sort of a bind here regarding external api design. I would love to contribute this to jack, so i'm emailing to see what the possibility of that is, and if anyone has any ideas about how to go about it. I am willing to patch jack1 as well for it, the relevant code looks near identical. I would also like to dynamically manage the array size or string memory allocation, perhaps within the limit, as statically allocating even port_num*(alias_limit=8)*name_size to store mostly unused aliases seems rather heavy, but it looked like everything regarding ports is statically allocated for a reason?