00:00:00 --- log: started retro/09.12.19 03:55:23 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 03:55:23 retro10: 3Charles Childers 07master0 SHA1-9a83ce0 03:55:23 add "else", cleanups 03:55:23 http://bit.ly/4QfOrx 03:55:24 --- part: retro-commit left #retro 04:01:12 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 04:01:12 wheke: 3Charles Childers 07master0 SHA1-dfdd713 04:01:12 cleanups to forth94, update image 04:01:12 http://bit.ly/4zvKZX 04:01:13 --- part: retro-commit left #retro 05:27:46 crc: How does that if work on values of true that aren't -1? 09:26:04 --- join: docl (n=luke@74-36-1-108.dr01.hmdl.id.frontiernet.net) joined #retro 10:29:01 --- quit: docl (Read error: 110 (Connection timed out)) 10:44:11 --- quit: virl (Remote closed the connection) 10:53:47 --- quit: crc (Read error: 110 (Connection timed out)) 12:55:51 --- join: docl (n=luke@74-36-1-108.dr01.hmdl.id.frontiernet.net) joined #retro 12:56:23 new changes to rem in my forthlets repo 13:06:38 now backspace works with the dynamic cursor 13:10:58 trying to decide whether custom arrow words (left, right, up, down) are a waste of time. 13:12:27 I think they are. for one thing, it is just as easy to type "l ++" as "right". 13:22:40 --- join: crc (n=charlesc@c-68-80-139-0.hsd1.pa.comcast.net) joined #retro 13:22:40 --- mode: ChanServ set +o crc 13:24:44 SimonRC: what other values are used for true? 13:26:46 anything non-zero? 13:26:57 the output of / or mod? 13:27:21 I have used the quotient from /mod as the condition of an if before 13:27:56 docl: how about up dn lt rt ? 13:28:35 maybe nonzero would be something like : nz ` 0 ` <> ` if ; 13:28:41 immediate 13:28:49 hmm, I could add those. 13:30:03 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 13:30:03 forthlets: 3Charles Childers 07master0 SHA1-bba098b 13:30:03 add bounds checking, navigation between blocks 13:30:03 http://bit.ly/7xhG3n 13:30:04 forthlets: 3Charles Childers 07master0 SHA1-56356ad 13:30:06 dont replace old "v" 13:30:08 http://bit.ly/82V5jo 13:30:10 forthlets: 3Charles Childers 07master0 SHA1-993429b 13:30:12 tweak behaviour of "bounds" 13:30:14 http://bit.ly/6b0tau 13:30:16 forthlets: 3Luke Parrish 07master0 SHA1-76419e2 13:30:18 New versions of p and n that do not invoke v an extra time, which eliminates a console flash within bounds. Also added bounds to p and n hotkeys. 13:30:21 http://bit.ly/7TDuWq 13:30:23 forthlets: 3Luke Parrish 07master0 SHA1-256406d 13:30:25 Accept input dynamically and show cursor as you type it in. 13:30:27 http://bit.ly/6nC4o1 13:30:29 forthlets: 3Luke Parrish 07master0 SHA1-9d18266 13:30:31 added refresh command and clean backspace. 13:30:33 http://bit.ly/4Kua31 13:30:35 forthlets: 3Charles Childers 07master0 SHA1-0c08add 13:30:37 match docls repo 13:30:39 http://bit.ly/6hcs0S 13:30:41 --- part: retro-commit left #retro 13:31:02 SimonRC: makes sense 13:32:49 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 13:32:49 retro10: 3Charles Childers 07master0 SHA1-5132ef6 13:32:49 "if" now executes conditional code if TOS != 0. Old behavior: execute was execute if TOS = -1. 13:32:49 http://bit.ly/6GAVTq 13:32:50 --- part: retro-commit left #retro 13:34:40 docl: I don't think we need to call 'bounds' after 'p' and 'n' handlers 13:34:50 docl: since neither one alters the cursor positions 13:35:38 they need to check the block number though don't they? 13:35:45 at least p does 13:36:04 n doesn't, unless there is a max block size that bounds will check 13:36:17 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 13:36:17 wheke: 3Charles Childers 07master0 SHA1-5c16e42 13:36:17 updated image; now makes forth94 IF standard compliant 13:36:17 http://bit.ly/5G79Zl 13:36:18 --- part: retro-commit left #retro 13:36:19 we could have this as a seperate word though 13:36:26 docl: probably a good idea 13:36:35 though the core editor doesn't check... 13:36:44 a separate word would be best 13:40:56 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 13:40:56 forthlets: 3Charles Childers 07master0 SHA1-0f12e7a 13:40:56 add range checking to blocks 13:40:56 http://bit.ly/4PC8By 13:40:57 --- part: retro-commit left #retro 13:42:36 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 13:42:36 retro10: 3Charles Childers 07master0 SHA1-ed6cbe5 13:42:36 fix display of number of blocks 13:42:36 http://bit.ly/4r1mR9 13:42:37 --- part: retro-commit left #retro 13:44:42 it's so cool to be able to see what you're about to change with git diff. 13:45:50 give me a moment to sync against your latest push 13:46:27 ok 13:54:42 docl: done 13:54:46 docl: pull against mine please 13:54:49 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 13:54:49 forthlets: 3Luke Parrish 07master0 SHA1-39ca860 13:54:49 moved match and bounds upward. bounds is used in backspace code so it does not go into negative column numbers. wr added, which uses bounds before writing to the addr. moved bounds references to end of hotkey checker. 13:54:49 http://bit.ly/5A37rl 13:54:51 forthlets: 3Luke Parrish 07master0 SHA1-5b07683 13:54:53 added up dn lt rt 13:54:55 http://bit.ly/63cx9p 13:54:57 forthlets: 3Charles Childers 07master0 SHA1-9aa38a6 13:54:59 update to match docls repo 13:55:01 http://bit.ly/5VUZfP 13:55:03 forthlets: 3Charles Childers 07master0 SHA1-8e1988b 13:55:05 whitespace issues 13:55:07 http://bit.ly/8wDml6 13:55:09 forthlets: 3Charles Childers 07master0 SHA1-e812f28 13:55:11 minor tweaks 13:55:13 http://bit.ly/6UOwQF 13:55:15 --- part: retro-commit left #retro 13:55:20 * crc has made some tweaks, and remove "enter"; since this is a modal editor, use of esc to exit the entry mode is better IMO 13:56:30 docl: one bug I found: if you reach the end of a line, the cursor moves to the start of the previous line rather than the next line 13:56:45 e,.g.: 13:56:51 2 his is a test kkk 13:56:51 3 t 14:01:55 yeah, I see that. line 43 doesn't take column into account. 14:01:57 ISTR that one cah get odd behaviour in .NET if one mixes languages that only use 2 boolean values with ones that can manage to use any byte as a boolean value 14:02:22 e.g. if the C# compiler assumes that true is always 1 14:02:45 c#'s interpretation doesn't affect retro :) 14:03:07 but you don't want to imitate their mistakes if you can avoid it 14:03:26 true 14:03:41 * crc did check in a change so that any non-zero is now considered true 14:06:56 docl: I think I fixed it 14:06:59 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 14:06:59 forthlets: 3Charles Childers 07master0 SHA1-b5d2add 14:06:59 fix bug in cursor movement 14:06:59 http://bit.ly/5bvRMl 14:07:00 --- part: retro-commit left #retro 14:24:13 yeah I think that worked 14:24:20 there's another bug that I just fixed 14:25:32 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 14:25:32 forthlets: 3Luke Parrish 07master0 SHA1-9a190a9 14:25:32 fix bug with leaving values on stack 14:25:32 http://bit.ly/6Y3f0Y 14:25:33 --- part: retro-commit left #retro 14:28:54 why does it keep joining and leaving 14:29:10 surely staying on all the time would be less spammy 14:30:29 SimonRC: I don't know; it's provided by github 14:30:38 * crc should see if he can improve this a bit... 14:36:41 --- join: retro-commit (n=Karere@c-68-80-139-0.hsd1.pa.comcast.net) joined #retro 14:37:51 should I alias esc to do the same as q, so you could use it to enter or exit the editor? 14:37:58 yes 14:39:00 --- quit: retro-commit (Remote closed the connection) 14:40:21 --- join: retro-commit (n=Karere@c-68-80-139-0.hsd1.pa.comcast.net) joined #retro 14:41:12 [crcx/forthlets] b5d2ad: fix bug in cursor movement 14:41:47 SimonRC: I wrote up a quick bot that'll stay connected rather than come/go 14:47:12 [crcx/forthlets] 89fb61: alias escape to q 15:05:12 [crcx/forthlets] f25993: use repeat/again rather than recursion in q 15:34:12 [docl/forthlets] f25993: use repeat/again rather than recursion in q 15:34:43 you don't have TCO? 15:35:17 TCO is in the optimizer library 15:35:19 I remember we used to 15:35:28 ahh 15:35:45 * SimonRC likes the way ColorForth has TCO as standard 15:36:33 Chuck discovered he liked using TCO in preference to loops; there is no general looping construct in CF 15:36:51 wierd how he has something in common with FP folks 15:37:01 the TCO doesn't work properly if there's no code following a "then" 15:37:39 you mean if then is followed by ; ? 15:37:47 yes 15:38:31 I'll fix that soon 15:46:12 [crcx/wheke] 5c16e4: updated image; now makes forth94 IF standard compl... 15:46:54 I added a simple fix for the TCO bug 16:13:51 --- join: virl (n=virl__@chello062178085149.1.12.vie.surfer.at) joined #retro 16:13:58 hi virl 16:15:13 [docl/forthlets] b75d50: top bot beg end 16:16:43 I wonder why it takes so long between showing commits? 16:17:28 the bot checks for commits once per minute; I'm not sure how quickly github sends the email to it 16:17:53 is there a way to submit buggy code to a seperate branch for review? 16:18:33 I've not worked with branches in a single repo. I fork a separate copy of the main repo for potentially buggy stuff. 16:23:12 [crcx/forthlets] b75d50: top bot beg end 16:38:41 hmm 16:42:25 hi crc 16:49:04 virl: what's new? 16:49:13 [crcx/forthlets] 6a7c71: ENTER goes to next line now 16:49:46 my forth is still in hibernation mode and I'm still writing on a forth demo 16:51:58 docl: http://gist.github.com/260300 - the (early) start on a users guide for REM 16:53:26 looks good 16:54:15 I'm trying to code an arrow-command multiplier, such that if you type a number and then an arrow key it goes in that direction that many spaces. 16:54:33 arrow keys are not portable 16:54:40 I mean ijkl 16:54:44 ok :) 16:55:04 :) 16:57:40 * crc missed the capitalized commands; added them to gist 17:01:31 there we go, got it working 17:02:06 not hjkl? 17:02:12 [docl/forthlets] ac8494: movement repetition. moves in a given direction wh... 17:03:12 [crcx/forthlets] ac8494: movement repetition. moves in a given direction wh... 17:04:03 hjkl strains the fingers a heck of a lot more than ijkl 17:04:11 [docl/forthlets] 21f7a6: factored arr (direction commands) from rep 17:05:13 [crcx/forthlets] 21f7a6: factored arr (direction commands) from rep 17:07:12 [docl/forthlets] 08fa61: placed arr in main hotkey loop 17:08:12 [crcx/forthlets] 210310: kill trailing whitespace 17:11:12 [crcx/forthlets] 95d8d1: indention; bound checks in rep 17:11:24 --- quit: retro-commit (Remote closed the connection) 17:11:35 --- join: retro-commit (n=Karere@c-68-80-139-0.hsd1.pa.comcast.net) joined #retro 17:11:49 docl: I pushed a patch to add bounds checking with the rep code 17:12:22 good, I noticed it needs that 17:13:27 [docl/forthlets] 6a7c71: ENTER goes to next line now 17:17:15 docl: I updated the gist again. it's nicely formatted now 17:21:02 be back later 17:21:13 ok 17:22:27 [crcx/forthlets] 08fa61: placed arr in main hotkey loop 17:38:58 --- quit: docl (Read error: 110 (Connection timed out)) 17:50:29 * SimonRC goes to bed 18:21:11 --- join: docl (n=luke@67-60-48-82.cpe.cableone.net) joined #retro 20:02:12 am thinking the tab key is easier to hit on my keyboard than escape. 20:30:32 tab works too 20:45:59 added some backspace features. 21:16:31 oops, s is taken 22:35:02 added something to drag lines up and down, also can drag letters right and left. 23:59:59 --- log: ended retro/09.12.19