That is correct. Lot of other stuff going on so I haven't made progress in a couple of weeks. (More medical problems to deal with for my partner, and that's not easy with the pandemic still raging, since she *got infected at a medical office* in December, so we're being ultra-paranoid -- getting the fourth dose, going off the immune suppressants, scheduling the appointments about 2 weeks after dose #4, wearing even better-fitting N95s than she wore last time...)
I'd say I'm three to five full days' work away from having a program which produces a pretty nice HTML timetable from GTFS data (looking similar to the ones from the classic timetable); it uses a little manual intervention in the form of templates (which can be prepped in a spreadsheet program as CSV) for the more complicated timetables ( while the templates can be autogenerated for the simple cases). I am quite sure I've got the architecture right so that I can generate nearly all the different styles of timetables which were used in prior Amtrak timetables. A few persnickety items will take longer.
The same program should be able to put it out in a raw CSV format if people want to present it differently, and the HTML should be suitable to feed through an HTML-to-PDF translater. If I get it finished I'll put it at the service of Rail Passengers Assocation and we should be able to pump out a full set of timetables with a script.
Manual work updating the applicable template would be required when train service actually changes significantly (new train numbers, cancelled train numbers, stations added or removed), but should be on the order of one minute's work -- and we all know that sort of change doesn't happen that often.
Changes in exactly when trains arrive should just be a matter of running "make new timetable", push a button, get a timetable. The heavy manual effort which was error-prone was copying the times into the spreadsheet, and I have it from inside Amtrak that they did not like the fact that occasionally copying errors crept in. So I figured I'd solve that problem.
I still need Amtrak to release up-to-date GTFS data to one of the standard sites like openmobilitydata, but I'm hearing that that is likely soon.
The program's in a "pieces lying on the floor" state at the moment, because I've gone through three rounds of prototypes in order to get the architecture right, and now I have to port the detail code from the first two prototypes to the third architecture, but that's pretty cut-and-dried work. The first prototype was producing a pretty nice timetable for the Cardinal but I needed to re-architect to do some of the other timetables.