2022-05-28 03:46:06 maw 2022-05-28 07:04:09 whats best material to learn forth? ive already read Start Forth tutorial 2022-05-28 07:15:06 there is starting forth book: https://www.forth.com/wp-content/uploads/2018/01/Starting-FORTH.pdf 2022-05-28 07:15:18 than thinking forth book: https://downloads.sourceforge.net/thinking-forth/thinking-forth-color.pdf 2022-05-28 07:15:21 and* 2022-05-28 07:16:09 i dived in to jonesforth, and read through brad rodriguez site 2022-05-28 07:18:02 a mirror of jonesforth is on github: https://github.com/nornagon/jonesforth 2022-05-28 07:18:55 this is brad rodriguez moving forth: https://www.bradrodriguez.com/papers/moving1.htm 2022-05-28 07:56:53 ty dave0 2022-05-28 10:57:29 Good morning gents - happy Saturday! 2022-05-28 10:57:45 why is saturday happy 2022-05-28 10:59:03 Well, Saturdays are always nice in my book, but this one is particularly nice because it's a Saturday kicking off a three-day weekend. Monday is a holiday in the States - "Memorial Day." 2022-05-28 10:59:10 So, I'm in an extra chipper mood. 2022-05-28 11:00:26 I've been exercising well lately and I'm in that "feel good" phase that comes in the early parts of such work. 2022-05-28 11:01:02 You get used to it and stop noticing later on, but in that early time it's actually evident that you feel better than you did a month prior. 2022-05-28 11:26:54 ooooh pforth DUMP is in units of 16 not the cell size so can easily try to dump more memory and thus segfaults when it wanders off into ... 2022-05-28 11:32:20 ? But I thought RAM protection was on 4k boundaries. You mean it does 16 byte chunks regardless of whether you start on a 16-byte boundary or not? 2022-05-28 11:33:28 If you start on a 16-byte boundary then I'd think all of each 16-byte chunk would have the same protection settings. 2022-05-28 11:34:46 but you can 1+ the start address and then request 256 from there 2022-05-28 11:35:22 I see. So if you start in the middle then you can't dump all the way to your last readable byte - it'll overrun. 2022-05-28 11:35:43 or actually ' + 1 dump blow it up 2022-05-28 11:36:33 : my_dump swap 0xF not and swap dump ; 2022-05-28 11:38:06 I'm guessing a little at the parameters for dump; that assumes it's (addr count --) 2022-05-28 11:38:30 it is 2022-05-28 11:38:39 I need to write a dump for mine. 2022-05-28 11:39:06 One of those things that's "easy," but slightly tedious if you want the standard display with both the hex and ascii. 2022-05-28 11:39:47 I haven't done my formatted output stuff yet, so I'll likely wait until I have that to draw on. 2022-05-28 11:40:06 pforth does it in C 2022-05-28 11:40:20 Yes - I used pforth for a consulting project back in the early 2000's. 2022-05-28 11:40:35 My memory of it was that it was fairly easy to build it. 2022-05-28 11:41:04 But at that time at least its default prompt and input line handling wasn't Forth "standard." 2022-05-28 11:41:19 I tried buildin gforth and then there was a clang process taking a gig of memory and the one cpu fan that still works spun up 2022-05-28 11:41:24 Looked like they didn't bother to do the termios dance that's required to get it all to work in the standard way. 2022-05-28 11:41:35 :-) 2022-05-28 11:41:41 I've never attempted to build gforth. 2022-05-28 11:42:06 I also want to remember that the pforth source code was fairly well organized - it was easy to find places I needed to change things. 2022-05-28 11:42:18 ... using gcc results in a "wtf is that flag?? compiler error" and if you remove the flag it blows up somewhere else 2022-05-28 11:42:59 That kind of thing plagues a lot of things I try to build. I've never understood why they've just kept on making build tools more and more involved. 2022-05-28 11:43:17 I was perfectly happy with old school make, but they just kept piling new tools on. 2022-05-28 11:43:32 I'm talking about the old, very straightforward and simple make. 2022-05-28 11:44:01 Back then I could open a makefile and see exactly what it was doing, but these days I just don't follow it all. 2022-05-28 11:44:09 well if you want portability to HP/UX, or ... 2022-05-28 11:45:24 I know. I know they've had "reasons." 2022-05-28 11:46:00 so you end up with a 25,000 line Configure shell script, or autotools, or cmake 2022-05-28 11:46:06 Exactly. 2022-05-28 11:46:26 or a very short shell script, say for rustup, that is hilariously unportable 2022-05-28 11:46:58 Well, I'm not sure I see the need for a single thing to run on 500 different operating systems. 2022-05-28 11:47:12 But I see why someone might want it. 2022-05-28 11:47:22 Makes it easier for the developers - harder for the users. 2022-05-28 11:47:29 If something doesn't work, that is. 2022-05-28 11:47:37 As long as it "just works" without any problems, it's great. 2022-05-28 11:47:48 But if you need to tinker with something, you can't figure out what the hell is going on. 2022-05-28 11:47:58 A whole "priesthood" has grown up around those tools. 2022-05-28 19:58:02 thrig: did you manage to build gforth? 2022-05-28 19:58:53 The build process is fairly standard autoconf/autotools/automake and the only extra dependency is m4 https://github.com/NixOS/nixpkgs/blob/nixos-21.11/pkgs/development/compilers/gforth/default.nix 2022-05-28 19:59:37 if you want to enable the swig bindings it's a bit more involved but not bad