Re: pcl-cvs/xemacs bug

nix> OK, so code that calls PROCESS_LIVE_P is asking two different nix> questions in two different places. nix> In create_process(), read_process_output(), and nix> set_process_filter(), the question being asked is `can I get nix> data from this subprocess) (what the original definition nix> said). This may well be possible if the process has done some nix> output that isn't yet read, and then died. nix> In delete_process(), kill_buffer_processes(), and nix> process-live-p itself, the question is `can I send signals to nix> this process' or `is this process still running in the OS'. nix> I think we need two #defines; PROCESS_LIVE_P() as defined in nix> 21.4.7, and PROCESS_INPUTTABLE_P (crap name though). How about PROCESS_READABLE_P instead? nix> I'll roll a patch that does that. nix> Sorry for the inconvenience. nix> -- nix> `Blueshifters were moving away... Redshifters were moving in...' nix> --- Justina Robson, _Mappa Mundi_, with startling new nix> evidence for the gib gnab hypothesis -- Adrian Aichner mailto:adrian＠xemacs.org http://www.xemacs.org/