00:00:00 --- log: started retro/09.12.01 01:07:29 --- join: virl (n=virl__@chello062178085149.1.12.vie.surfer.at) joined #retro 03:24:59 --- join: Mat2 (i=50eff25e@gateway/web/freenode/x-yfxvnufcmekiqgfm) joined #retro 03:25:46 --- quit: Raystm2 (Read error: 54 (Connection reset by peer)) 04:13:28 --- quit: Mat2 (Ping timeout: 180 seconds) 04:59:01 --- join: Mat2 (i=50eff25e@gateway/web/freenode/x-zypoudfqewnkhcqj) joined #retro 05:04:19 --- part: Mat2 left #retro 06:08:05 --- join: Mat2 (i=50eff25e@gateway/web/freenode/x-mkpwphpvukzighmo) joined #retro 06:19:56 hi Mat 06:28:45 hi crc 06:29:14 what's new? 06:29:38 have read your twitter blog and hope your blood sugar decreases (hope, this is the right translation) 06:30:28 new: working on my functional colorforth compiler 06:30:51 my blood sugar is down a bit today. I'm hoping it stays down. 06:30:55 cool 06:31:54 I think ngaro benefit also from this 06:32:17 as I will metacompile a ngaro vm from this 06:32:57 dialect 06:33:48 *call it an experiment* 06:35:34 ok 06:38:39 have you experimented with the new go port ? I found is an elegant language but miss some thinks from C like the register qualifier. 06:39:26 I've tested to make sure retro works with it 06:39:47 and played a little with the go-specific features 06:41:17 I have written a prototype vm with utilize pthreads in C some time ago 06:42:08 cool 06:43:27 sadly one drawback with parallel vm threads is the syncronisation overhead needed 06:44:34 so I think of some kind of fine-grained parallelisation of vm opcodes would be a better approach 06:46:54 *just an idea this time so far* 06:51:19 have a nice day 06:51:25 *now at work again* 06:52:14 you too 06:52:19 * crc will be going to work soon 06:52:28 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 06:52:28 wheke: 3Charles Childers 07master0 SHA1-4afd4c5 06:52:28 add WITHIN to forth94 06:52:28 http://bit.ly/5jnW0Z 06:52:29 wheke: 3Charles Childers 07master0 SHA1-d148835 06:52:31 add ACCEPT to forth94 06:52:33 http://bit.ly/5nCdbR 06:52:35 --- part: retro-commit left #retro 06:54:26 --- quit: Mat2 (Ping timeout: 180 seconds) 07:37:56 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 07:37:56 wheke: 3Charles Childers 07master0 SHA1-f89a42b 07:37:56 mark obsolete words as such 07:37:56 http://bit.ly/5NrRDl 07:37:57 --- part: retro-commit left #retro 07:49:16 --- quit: virl (Remote closed the connection) 08:14:22 http://gist.github.com/246393 09:10:11 --- join: crcx (i=d8012b82@gateway/web/freenode/x-wjimhlgpofipltgz) joined #retro 11:21:19 --- quit: crcx ("Page closed") 13:23:29 --- join: virl (n=virl__@chello062178085149.1.12.vie.surfer.at) joined #retro 15:26:01 --- quit: virl (Remote closed the connection) 15:50:25 good evening 15:56:05 docl: let me see what I can do with that idea 16:08:16 --- quit: yiyus (Read error: 110 (Connection timed out)) 16:13:03 --- join: yiyus (i=12427124@je.je.je) joined #retro 16:13:53 docl: try this varient: 16:13:54 http://gist.github.com/246775 16:41:13 hey that works 16:55:41 good 16:56:53 it even looks good grammatically 16:56:54 : r `` pop dup push '' ; immediate 17:02:52 I wonder, would it be less readable if you used ;i instead of ; immediate 17:03:47 or ;m for macro 17:08:40 wow, that was simpler than I thought... 17:10:36 http://gist.github.com/246817 17:11:33 heh, that paste was unnecessary since it was only one line 17:11:45 : ;m `` immediate '' ; 17:23:48 I used to have macro: 17:24:11 but having special case creators was problematic with extra classes 17:24:49 what kinds of problems did it cause? 17:24:55 : ;m `` ; immediate '' ; immediate 17:25:05 docl: mostly confusion. we had: 17:25:15 : macro: compiler: 17:25:41 in the core language, and classes required creation of new words when they were added 17:28:40 hmm 17:28:55 you could automate the creation of new words... 17:29:42 anything: ( creates as .anything ) 17:30:04 plus most macros should be compile-only, not immediate (immediate is really needed only if you require the word to be state-smart) 17:31:43 there's a lot of ways to handle the creation of words. I just found myself preferring a single : word 17:33:21 what's an example of a state-smart macro? 17:33:29 ." 17:35:58 which either displays a string, or compiles the code to display it, based on the 'compiler' state 17:36:04 I love the commented "see" :) 17:36:36 I can actually read the output pretty well now 17:37:16 good :) 17:46:36 "see" has proven very valuable to me when debugging 17:46:48 I love it! 17:47:43 you know what would also be nice? the ability to use arrow keys to walk through the memory and see the values 17:47:56 that's not possible yet 17:48:14 different systems report different values for the arrow keys :( 17:48:16 lack of portable arrow key scanning? 17:48:25 how about hjkl then? 17:48:25 yes 17:48:42 that would be doable 17:48:49 with escape letting you start and quit the mode 17:48:55 * crc intends to write a debugger tool sometime soon 18:19:35 what's a good name for a word that is blank but will be vectored later? 18:19:46 foo 18:19:47 :) 18:20:06 dwim ("do what I mean") 18:20:23 :P 18:20:29 create-as-word 18:20:45 create-as-word foo create-as-word bar create-as-word baz 18:21:22 as opposed to : foo ; : bar ; : baz ; 18:22:04 _: maybe 18:22:13 _: foo _: bar _: baz 18:25:20 v: 18:25:27 for "vector" 18:25:37 or perhaps "stub:" 18:25:51 stub is nice 18:27:47 yeah; it makes it easy to understand what it is for 18:47:12 http://gist.github.com/246886 18:50:06 oops 19:08:01 docl: commentary on your last gist: http://gist.github.com/246894 19:12:57 cool :) 19:22:20 http://gist.github.com/246905 19:25:30 the definition of cursor leaves something to be desired 19:26:07 and 'char' is unused 19:26:34 yeah it was having issues 19:26:42 : char: ( "-n ) 32 accept tib @ .data ; immediate 19:26:53 dup char: h =if .... 19:26:55 and so on 19:27:01 cool 19:31:10 http://gist.github.com/246907 19:32:18 nice 19:47:16 why does it say word not found the first time a key is pressed? 19:51:44 oh, the 27 must be in the tib at that point 19:55:06 replace it with a 32 and it works: 19:55:07 :: ( -c ) default: key dup 27 =if drop 32 ['] cursor ['] key :is then ; 19:55:07 is scan-for-esc 21:29:22 http://gist.github.com/246972 23:32:07 --- quit: docl ("Lost terminal") 23:59:59 --- log: ended retro/09.12.01