00:00:00 --- log: started forth/10.12.22 00:14:13 --- quit: tgunr (Ping timeout: 250 seconds) 00:19:11 --- quit: ygrek (Remote host closed the connection) 00:24:38 --- join: ygrek (debian-tor@gateway/tor-sasl/ygrek) joined #forth 00:35:42 --- quit: ygrek (Ping timeout: 245 seconds) 00:42:31 --- quit: TreyB (Ping timeout: 265 seconds) 00:48:08 --- join: TreyB (~Adium@adsl-75-53-203-31.dsl.hstntx.sbcglobal.net) joined #forth 01:30:56 --- join: ygrek (debian-tor@gateway/tor-sasl/ygrek) joined #forth 02:04:20 --- quit: gogonkt (Read error: Connection reset by peer) 02:08:22 --- join: gogonkt (~info@183.27.211.85) joined #forth 03:35:55 I think I know what the problem with C-written Forth's is. 03:36:10 In catch-throw aspect. 03:54:05 --- join: Rods_Tiger (~ian_tinda@host86-174-218-221.range86-174.btcentralplus.com) joined #forth 03:54:15 --- part: Rods_Tiger left #forth 04:22:22 --- quit: ygrek (Ping timeout: 245 seconds) 04:44:20 --- join: MayDaniel (~MayDaniel@unaffiliated/maydaniel) joined #forth 05:51:36 --- quit: MayDaniel () 05:52:39 --- join: MayDaniel (~MayDaniel@unaffiliated/maydaniel) joined #forth 06:41:29 --- join: tathi (~josh@dsl-216-227-95-5.fairpoint.net) joined #forth 06:42:00 ASau: what's that? 06:43:37 --- quit: Deformative (Ping timeout: 276 seconds) 06:54:07 --- join: nighty-- (~nighty@x122091.ppp.asahi-net.or.jp) joined #forth 07:01:10 tathi: part of problem is that usually implementors don't have 07:01:11 clear understanding of necessity of non-local exits and how they are done. 07:01:24 At least this is what I see in FICL and pForth. 07:02:42 When you're implementing Forth in C, "catch" is more fundamental than "execute". 07:04:03 At least catch-throw mechanics should be thought of ahead 07:04:22 rather then implemented some time afterwards. 07:05:04 This all happens because C has more strict nesting. 07:06:41 Ah. 07:06:46 Yeah. 07:07:39 Now I'm thinking if I can rework FICL without breaking compatibility. 07:09:24 I think going the completely alternative path and implement another primitives. 07:09:25 E.g. 07:09:30 : execute catch throw ; 07:09:59 Or use more primitive catch-like and throw-like operations. 07:10:53 Ones which don't unwind the data stack? Or going so far as setjmp/longjmp-like operators? 07:13:02 You have to unwind at least one stack, 07:13:25 if you don't want to implement another stack for setjmp-like operations. 07:13:42 Yes, but that can be the Forth return stack, can't it? 07:13:43 Not that the latter is bad idea... 07:13:48 Sure. 07:14:28 Did I tell you that I tried operations that don't unwind data stack? 07:14:58 You said you were thinking about it, but I didn't know you had actually tried it. 07:15:01 How did that go? 07:15:35 I've reimplemented all necessary operations with them. 07:15:46 catch, throw, abort, abort" 07:16:06 I didn't try anything more massive though. 07:16:32 Oh. And abort" needs reimplementing anyway. 07:16:49 In a _non-portable_ way. 07:17:05 The current semantics is evil. 07:17:13 It smells 70ies. 07:17:27 --- join: Monev (~nal@cvx-ppp-66-50-131-14.coqui.net) joined #forth 07:17:36 Heh. 07:17:46 How would you do it? 07:18:06 I'm thinking about introducing global buffer for the message, 07:18:19 and operations to clear it and append to it. 07:19:40 Negative side is that it goes far beyond classical catch-throw mechanism. 07:20:29 The main problem is that you can't distinguish easily between 07:20:57 exceptions thrown with abort" and thus carrying message, 07:22:12 and exceptions that are thrown or rethrown after abort"-ed exceptions were caught. 07:22:16 right 07:22:40 If the mechanism were more elaborate, 07:22:52 e.g. there were restarts like in CL. 07:23:05 sure, where raising a condition passes some sort of object 07:23:27 This problem could be solved, at least had chances. 07:24:32 Alright, I'll do this first, then we'll see what it means. 07:26:37 Yeah, first things first. :) 07:26:46 What makes me feel bad is that 07:26:59 as long as I dive deeper into all this, 07:27:18 my opinion regarding implementation of very non-standard Forths 07:27:22 changes. 07:27:25 is going up? :) 07:27:46 Recent trends are real suckage. 07:27:57 I'll agree with that. 07:28:35 People who don't program numerics and don't deal with FPNs 07:28:35 tell me what application programmers want. 07:29:03 Another people who program numerics and do it ONE TRUE 07:29:18 (read Fortan) way, just continue doing it that one true way. 07:29:22 eh 07:29:23 heh 07:29:55 I don't understand the point of FSL _at_all_. 07:30:09 Each time I look at it, I see Fortran, 07:30:23 only in awkward syntax 07:30:32 as if it were written for unhumans. 07:31:01 I can't comment there...I don't do numerics at all, really. 07:31:51 Have you seen recent waxman's reply to mhx? 07:32:25 Ruby-man points top Forth programmer to factoring! 07:32:39 I don't usually read waxman's messages...they're usually just the obvious Ruby code snippets. 07:33:09 They're usually one-liners, so there's almost no time lost. 07:33:11 Oh, right. I did see that one. :) 07:34:14 Honestly, I can't say that mhx write "Forthy" Forth code. 07:34:18 I don't remember that. 07:35:09 I can't boast that I write one-liners in CM or KC5TJA style myself, 07:35:26 yet it isn't 10-liners usually. 07:36:06 BTW, this is again point against SF/TF series. 07:36:58 How so? 07:37:02 If newcomer isn't idiot, he'll notice the disparity. 07:37:48 If top Forth coders don't follow what SF/TF suggest, 07:37:58 then there's something preventing from doing so. 07:38:10 :) 07:38:27 Either SF/TF recommendations are unrealistic 07:38:27 or top Forth coders are bad coders actually. 07:39:42 I wonder, do they never talk to their colleagues? 07:40:06 How come that these simple arguments are hard to get into their heads? 07:41:16 BTW, I've stopped my work on CL-like packages. 07:42:20 for Forth? 07:42:26 Yes. 07:42:41 There's fundamental difference between what is described in 07:42:41 those packages for FPL and Lisp/Prolog-style languages 07:42:47 and what is used in Forth. 07:43:21 I'm not sure how this can be solved. 07:43:58 The problem is that "wordlist" is _really_ a linked list. 07:44:36 It isn't injection/mapping. 07:44:59 Perhaps I should dig out that article about skip-lists. 07:45:44 How is a wordlist different from an injection/mapping? 07:45:55 Oh, redefinitions? 07:45:59 Right. 07:46:49 You have more generic relation, and if you're going to 07:46:49 support "marker" or "forget", you have to deal with it. 07:47:09 And I'm going to support both. 07:47:27 Otherwise Forth loses most his nice features. 07:47:33 OK 07:48:25 This agains brings up my point that Forth is more complex 07:48:26 than those "bind-and-dominate" languages. :D 07:49:24 In some ways, yes. 07:49:46 It makes perfect sense as an assembly language programmer's answer to "what's the simplest higher-level language I can design." :) 07:50:07 Er, maybe "design" was a poor word choice there. 07:50:08 Well, at the same level of complexity you can write AIM-8 Lisp. 07:50:43 Yes, but if all you have ever known is assembly language and maybe FORTRAN... 07:50:50 Perhaps even Standard Lisp is managable. 07:50:54 I don't get the impression that Chuck was familiar with Lisp at all. 07:51:26 Anyway. I should get back to doing work. Talk to you later. 07:51:40 There're claims that SWI Prolog VM is really simple :) 07:51:49 Alright. See you later. 08:10:04 --- quit: nighty-- (Ping timeout: 276 seconds) 08:19:26 --- join: nighty-- (~nighty@x122091.ppp.asahi-net.or.jp) joined #forth 08:49:55 --- quit: nighty-- (Quit: Disappears in a puff of smoke) 09:22:13 --- quit: gogonkt (Ping timeout: 276 seconds) 09:44:36 --- join: gogonkt (~info@183.27.211.85) joined #forth 09:48:30 --- quit: MayDaniel () 10:01:52 --- quit: martinhex (Remote host closed the connection) 10:02:30 --- join: martinhex (~mjc@93-97-29-243.zone5.bethere.co.uk) joined #forth 10:12:14 --- join: Deformative (~Joseph@205-36.adsl.umnet.umich.edu) joined #forth 10:13:42 --- quit: Judofyr (Remote host closed the connection) 10:20:44 --- join: qFox (~C00K13S@5356B263.cm-6-7c.dynamic.ziggo.nl) joined #forth 10:29:29 --- join: ygrek (debian-tor@gateway/tor-sasl/ygrek) joined #forth 10:45:58 --- join: MayDaniel (~MayDaniel@unaffiliated/maydaniel) joined #forth 11:01:56 --- quit: Deformative (Ping timeout: 240 seconds) 11:37:01 --- join: roarde (~roarde@pdpc/supporter/active/sixforty) joined #forth 11:39:52 --- quit: MayDaniel (Read error: Connection reset by peer) 12:14:06 --- quit: roarde (Ping timeout: 255 seconds) 12:16:44 --- join: roarde (~roarde@pdpc/supporter/active/sixforty) joined #forth 12:50:02 --- join: Deformative (~Joseph@c-68-84-165-215.hsd1.mi.comcast.net) joined #forth 13:13:12 --- quit: Deformative (Ping timeout: 245 seconds) 14:00:10 --- join: tgunr (~tgunr@cust-66-249-166-13.static.o1.com) joined #forth 14:11:32 --- quit: ygrek (Ping timeout: 245 seconds) 14:20:56 --- quit: qFox (Quit: Time for cookies!) 14:31:24 --- quit: tgunr (Quit: Leaving...) 14:41:37 --- join: tgunr (~tgunr@cust-66-249-166-13.static.o1.com) joined #forth 14:49:47 --- join: Deformative (~Joseph@c-68-84-165-215.hsd1.mi.comcast.net) joined #forth 14:55:40 --- quit: roarde (Quit: Leaving.) 15:05:53 --- quit: TreyB (Quit: Leaving.) 15:46:32 --- join: MayDaniel (~MayDaniel@unaffiliated/maydaniel) joined #forth 16:14:52 --- quit: MayDaniel (Ping timeout: 245 seconds) 16:58:40 --- quit: Snoopy_1611 () 17:06:09 * ASau sighs. 17:07:03 This community is plain dead. 17:07:40 If it rejects moves to reuse standardized code from 90s, 17:07:45 it can't progress. 17:09:33 ? 17:11:19 Phil asks if I don't break anything when I replace 17:11:19 (*s == BLANK) || (*s == '\t') || (*s == 17:11:32 \r') || (*s == '\n') 17:11:44 with isspace((unsigned char)*s) 17:12:35 Sure, I break. 17:12:57 I break all those non-standard compilers noone uses already. 17:13:26 I'm not sure that they accept other constructs either. 17:13:54 How do I know what bugs their developers planted back 30 years ago? 17:15:25 Why would that break anything? 17:15:42 Oh, break old C compilers 17:15:48 Sheesh. 17:19:23 Yeah. 17:20:06 Everyone breaks up with c89 and commits to c99 at least. 17:20:26 K&R shit is forgotten long time ago! 17:20:52 Fortran(!) programmers broke with F77. 17:21:15 You still can fetch F77-compliant LAPACK fron NetLib, 17:21:22 but you can't build new LAPACK with F77. 17:21:57 --- quit: Deformative (Ping timeout: 245 seconds) 17:22:31 Ah, but '77 was 12 years longer ago than '89. ;-) 17:23:30 F90-standard code is rare occurence. 17:23:36 You need F95 at least. 17:24:10 Wait, didn't C89 have isspace? 17:24:22 Yes. 17:24:26 That's exactly the point. 17:24:34 Oh. 17:24:42 [censored] C89 mandates ctype functions! 17:24:51 OK, I get it now. 17:28:17 So far all my efforts to break up with that fossilised shit end in trash bin. 17:28:39 * crc would just fork it in that case 17:29:01 crc: This is straight path to trash bin. 17:29:13 I don't have that much time and energy. 17:33:33 Anything I'm trying to feed back into main repository ends either with 17:33:51 "This needs a review, just in case it doesn't break anything" 17:34:09 or "We want portabililty, does every C compiler support it?" 17:34:24 NO! 17:34:54 There's ALWAYS at least one bloody compiler that is broken 17:34:54 enough not to support any feature you use! 17:35:09 So what? 17:35:18 Are you using exactly that compiler? 17:35:32 If not, then why do you ask those bloody questions? 17:36:31 If you want to scare people away, then tell is in plain words, 17:36:54 "I'm the President of Universe, and I don't want you to touch the code." 17:49:42 http://www.online-literature.com/wilde/1289/ 18:01:07 Heh. 18:48:12 --- quit: tgunr (Quit: Leaving...) 20:05:30 --- join: Snoopy_1611 (Snoopy_161@dslb-088-069-133-028.pools.arcor-ip.net) joined #forth 21:53:10 --- join: nighty__ (~nighty@x122091.ppp.asahi-net.or.jp) joined #forth 22:10:23 --- quit: ASau (Remote host closed the connection) 22:12:34 --- join: ASau (~user@95-26-92-70.broadband.corbina.ru) joined #forth 22:22:49 --- quit: tathi (Quit: leaving) 22:24:13 --- quit: nighty^ (Quit: Disappears in a puff of smoke) 22:33:33 --- join: nighty^ (~nighty@x122091.ppp.asahi-net.or.jp) joined #forth 22:40:39 --- quit: Monev (Quit: Monev) 23:26:54 --- quit: ASau (Remote host closed the connection) 23:29:12 --- join: ASau (~user@95-26-92-70.broadband.corbina.ru) joined #forth 23:43:06 --- join: Deformative (~Joseph@c-68-84-165-215.hsd1.mi.comcast.net) joined #forth 23:59:59 --- log: ended forth/10.12.22