00:00:00 --- log: started retro/11.06.18 05:30:05 --- topic: set to 'Retro Language | http://retroforth.org | Logs @ http://rx-core.org/dev/rancid | 11.0, Beta5 @ http://retroforth.org/beta/retro-11.0_beta5.tar.gz | Fossil Repo @ http://rx-core.org | Corpse Blog @ http://rx-core.org/dev/corpse | Forums @ http://forthcommunity.com | Donations @ https://www.wepay.com/donate/179527' by crc 05:30:13 11.0, beta 5 is out 05:47:31 --- join: docl (~docl@unaffiliated/docl) joined #retro 06:07:14 crc: I'm confused by the behavior of clone. Stack diagram indicates that it takes an address, but what it does with that is call allot... maybe it is supposed to grab the size of the struct from that first instead? 06:49:18 https://gist.github.com/1033126 06:58:10 that definition is buggy 06:59:27 I took out the curry. not sure if that broke it. 07:00:15 (my dfinition of 'clone', to clarify) 07:00:55 ah yes, thought that might be the case 07:03:18 hmm, the order of ` @ .data looks backwards 07:04:24 fixed 07:10:45 I fixed mine: https://gist.github.com/1033126#comments 07:11:11 yours isn't quite right yet: look at 'here' before and after a clone 07:11:29 err, after an allocation, not a clone 07:17:49 hmm, mine is still not working right either 07:22:01 ok, apparently the @ was unneeded 07:31:22 I refreshed mine as well (still using curry) 07:36:26 the curry version will have slightly higher overhead at runtime, but saves a bit of space when creating each struct (since the common code is in a shared quote) 07:41:29 ah, that makes sense 07:47:06 * docl ponders a version of "fields" similar to "elements" but with varying sizes possible by distinguishing numeric values. example: { .id .initials 3 .jobcode } employee 07:49:09 alternately something like: 3 fields .id .initials 3 .jobcode 07:50:03 when it sees "3" it would know to increment the offset and allotment by 3 07:52:55 perhaps a tweak to create that allows it to recognize numbers and execute a quote if something is numeric 08:06:34 --- quit: SimonRC (Ping timeout: 260 seconds) 08:22:59 docl: something like this: 08:23:00 {{ 08:23:00 : new ( nn$- ) over header ]] .data ` + ` ; + ; 08:23:00 ---reveal--- 08:23:00 : fields ( n"- ) 08:23:01 [ getToken dup isNumber? [ toNumber getToken new ] [ 1 swap new ] if ] times ; 08:23:03 }} 08:23:08 { 08:23:14 3 fields .a .b 5 .c 08:23:17 } foo 08:23:35 defaults to size 1 cell, unless override is specified 08:24:00 --- join: SimonRC (~sc@fof.durge.org) joined #retro 08:28:10 or better yet: https://gist.github.com/1033211 08:43:53 --- quit: SimonRC (Ping timeout: 260 seconds) 08:51:02 cool, works 09:02:14 I updated https://gist.github.com/1032096 and pushed the fields word to the repo 09:08:12 very nice! 09:10:04 the modified hash generator is better I think. 09:19:39 interesting 09:49:48 --- quit: docl (Remote host closed the connection) 09:54:24 --- join: SimonRC (~sc@fof.durge.org) joined #retro 10:07:24 --- join: docl (~luke@unaffiliated/docl) joined #retro 19:36:03 * docl is reading the book at http://producingoss.com/ 21:28:59 --- quit: docl (Quit: Lost terminal) 21:32:05 --- join: docl (~luke@unaffiliated/docl) joined #retro 22:31:05 --- join: Raystm2 (Raystm2@c-75-70-81-159.hsd1.co.comcast.net) joined #retro 22:35:59 hi Raystm2 22:36:15 Hey docl! how have you been? 22:36:42 what have you been working on, lately? 23:14:54 G'night. 23:14:58 --- part: Raystm2 left #retro 23:59:59 --- log: ended retro/11.06.18