00:00:00 --- log: started retro/09.11.29 00:55:35 andguent (he is in #plan9) started a port of xv6 to vx32, you could be interested 01:15:54 --- quit: docl (Read error: 110 (Connection timed out)) 02:54:25 --- quit: virl (Remote closed the connection) 07:07:45 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 07:07:45 ngaro: 3Charles Childers 07master0 SHA1-62b3321 07:07:45 golang: minor fixes 07:07:45 http://bit.ly/5uQvnV 07:07:45 --- part: retro-commit left #retro 07:09:40 --- join: nigol (n=nigol@63.120.broadband10.iol.cz) joined #retro 07:09:56 hi nigol 07:10:06 crc: Hi 07:10:28 How's going? 07:10:38 pretty good 07:12:07 Great. I'm doing some programming and in spare time chatting on IRC ;) 07:15:23 crc: But I have no time to improve MIDP version of Retro for MIDP... 07:15:54 nigol: that's ok. the current MIDP version still works good enough for me :) 07:18:29 crc: For me too, but I have some plans to extend ports to allow receive GPS coords and implement some kind of canvas drawing. It could be funny for some experiments on the go. 07:18:40 cool 07:19:50 crc: But I have no free time currently :( 07:20:11 I understand fully 07:33:17 crc: I also have some thoughts about link retro with Objective-C, so man can do GUI with Interface Builder and action code write in Forth. But I'm no very good Mac programmer and my knowledge of Forth is limited. 07:34:13 crc: Retro's has big advantage in portability. Port to another platform is quite easy. 07:35:45 crc: So on Linux, it can be easy to create GUI for example with Glade, and the core code of application can be the same - written in Forth. 07:37:44 I've thought about that myself, but haven't had the time to work on anything 07:40:55 crc: Few days ago, I have successfully conected Haskell with ObjC Mac UI. So maybe in near future I take a look for connecting retro with. 07:42:03 crc: But as I said, I have limited knowledge of Forth, that's slowing me down. 08:02:21 crc: Have a nice day, I must disconnect. 08:02:27 --- quit: nigol () 08:06:07 --- join: cavva (i=5895f42b@gateway/web/freenode/x-wwvvrebyapnsqcvx) joined #retro 08:06:16 hi 08:06:50 i'm looking the meta.retro sources 08:07:16 and cannot understand how the conditionals works .... 08:07:31 =if ... then 08:07:37 for example 08:09:18 i know how to use in retro, but i wish to understand the "under the hood" 08:10:01 the conditionals in the metacompiler just lay down code for the conditional jump forms, just like the actual words in retro 08:11:28 --- join: docl (n=luke@67-60-48-82.cpe.cableone.net) joined #retro 08:12:39 hi docl 08:12:42 hi crc 08:14:22 * crc has written server-side image saving code for the javascript port :) 08:14:30 cool 08:14:56 great crc 08:15:45 mmm i need to understand how branch works 08:16:57 i don't understand why, if the conditions is false (in =if) the exectution switch at the "then" point 08:17:29 if true, excecute the code between the jump and the then 08:17:38 if false, jump to the then 08:18:23 ok, but who tell "jump" where "then" is? 08:18:32 then does 08:18:50 the conditionals lay down a jump to address 0, and leave a pointer on the stack 08:19:09 the "then" figures out where it is, and uses that pointer to change the jump from 0 to the proper location 08:19:58 this explain the "0 m, ;" code ... 08:20:10 in the cond word 08:22:13 @crc what you say is an enlightenment for me :) 08:23:57 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 08:23:57 ngaro: 3Charles Childers 07master0 SHA1-fca32eb 08:23:57 reorganize javascript implementations 08:23:57 http://bit.ly/6Fo4p8 08:23:58 ngaro: 3Charles Childers 07master0 SHA1-0f9549d 08:24:00 add javascript port with server-side storage 08:24:02 http://bit.ly/7YnwTF 08:24:04 --- part: retro-commit left #retro 08:24:17 cavva: I'm glad to help 08:35:13 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 08:35:13 ngaro: 3Charles Childers 07master0 SHA1-c1c8423 08:35:13 add index.html for browser with client-side saving 08:35:13 http://bit.ly/8fO72x 08:35:14 ngaro: 3Charles Childers 07master0 SHA1-5978214 08:35:16 add some notes on the javascript ports 08:35:18 http://bit.ly/6gtps5 08:35:20 --- part: retro-commit left #retro 08:46:05 is there a way to set the maximum length of a string for accept? 08:46:13 no 08:46:44 accept can safely get input up to 512 bytes 08:47:15 since that's the size of the TIB 08:47:34 (which can be altered, but only by rebuilding the image) 08:47:41 hmm 08:50:22 the tib pointer is a constant, right? 08:50:35 yes 09:07:13 i'm still thinking on =if 09:09:47 conditionals fetch the target value and set the new "target cell" with 0, right? 09:11:54 next we have the code to call the word in case of "true" condition ... 09:12:09 ah!! 09:13:46 next we have "then", it get the new target cell, cahnge its value (previous 0) with the "then" address and leave a nop, 09:15:32 but is the final nop, required? 09:16:05 just padding 09:16:11 I'll remove it soon :) 09:17:14 removing the padding is for "saving space" only? 09:17:37 yes 09:17:50 it used to be needed, but no longer is due to cleanups elsewhere 09:19:53 --- join: virl (n=virl__@chello062178085149.1.12.vie.surfer.at) joined #retro 09:25:52 about padding 09:26:52 the two nop, in the t: definition cannot be removed right? 09:38:01 no 09:38:13 those are used to allow words to be revectored 09:38:47 ok 09:39:57 docl: http://retroforth.com/paste/?id=1935 09:40:12 varient of 'accept' with a max count 09:40:17 doesn't use tib 09:41:08 crc: I see you're following my toy vm on github :-P 09:41:16 just remember that the buffer has to be at least one byte longer than the max count (for terminator char) 09:41:29 zbrown: yup 09:42:37 crc: cool! 09:49:51 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 09:49:51 retro10: 3Charles Childers 07master0 SHA1-85f06cf 09:49:51 no more unneeded padding in conditionals 09:49:51 http://bit.ly/5cJiOd 09:49:52 --- part: retro-commit left #retro 10:02:16 docl: does this do what you want? 10:06:51 it seems to be filtering the whitespace, so I can't use cr 10:07:03 but I know how to fix that :) 10:07:09 whitespace off 10:07:12 whitespace on 10:07:13 :) 10:33:54 http://retroforth.com/paste/?id=1938 10:36:16 I sometimes insert lines longer than 64 chars 10:51:09 docl: actually, you shouldn't have get-input directly write into the block buffer 10:51:39 docl: since it appends 0 to the end of the input string, it'll leave these values inside the block 10:51:55 docl: which could cause issues when evaluating the block 11:40:19 --- quit: cavva ("Page closed") 11:48:35 * docl has to leave on an 8 hour drive to Salem :( 11:48:54 I'll probably be on later tonight 11:51:29 ok, I'll likely be here until midnight or so, eastern time 11:58:50 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 11:58:50 ngaro: 3Charles Childers 07master0 SHA1-e78d449 11:58:50 golang: ReaderWriter interface instead of I/O channels 11:58:50 http://bit.ly/4BrXxd 11:58:51 --- part: retro-commit left #retro 12:07:44 --- quit: docl (Read error: 110 (Connection timed out)) 14:38:23 --- quit: virl (Remote closed the connection) 17:00:13 --- join: retro-commit (n=retro-co@sh1-ext.rs.github.com) joined #retro 17:00:13 ngaro: 3Charles Childers 07master0 SHA1-95af014 17:00:13 golang: removed pipes, but got i/o to children right 17:00:13 http://bit.ly/5jAxbL 17:00:14 --- part: retro-commit left #retro 20:28:48 --- join: docl (n=luke@97-120-119-247.ptld.qwest.net) joined #retro 23:32:49 --- join: sixforty (n=sixforty@pdpc/supporter/active/sixforty) joined #retro 23:59:59 --- log: ended retro/09.11.29