00:00:00 --- log: started retro/12.07.14 12:31:09 --- join: erider (~chatzilla@unaffiliated/erider) joined #retro 12:32:12 hi all 13:11:56 can someone explain to me why you would use malloc(sizeof(foo *)) instead of malloc(sizeof(foo))? I am not sure why you would just want the size of a pointer to a structure instead of the size of the structure object itself if you want to put stuff in it 14:17:03 --- join: Mat2 (5b4085c5@gateway/web/freenode/ip.91.64.133.197) joined #retro 14:20:27 hello 14:31:56 hi Mat2 14:32:09 Hi erider 14:32:14 how are you ? 14:35:02 I am doing fine and yourself 14:42:57 quite fine, a lot of work at oment 14:43:04 sorry, moment 14:45:44 hmm, the chat seems quite silent this month 14:48:04 yeah not a lot going on here 14:49:54 there exist now a colorforth inspired project on hive, so I think to build one 14:50:19 3 propeller mcu's with 24 cores appeal to me 14:51:58 * erider forgot all about colorforth :( 14:58:06 ok, what language you current work with ? 14:58:24 C 14:58:37 something python 14:58:42 some python 14:58:57 python is the rage right now 15:01:59 yes, the first python versions had remind me some later basic systems 15:03:36 and i found genie useful, its the compiled phyton dialect of the gnome project 15:08:31 need some sleep, ciao 15:08:37 --- quit: Mat2 (Quit: Page closed) 17:21:10 hi all 18:47:13 good evening 19:01:29 hi crc 19:17:36 what's new? 19:20:15 * crc has been working on the parable decompiler; it can now generate source for a function and its dependencies 19:45:28 playing with c just trying to get a better understanding of it base subset of the language 19:54:37 crc parable is the replacement of toka right? 20:05:53 yes 20:06:46 crc: are you making the decompiler in the language itself 20:07:11 no 20:07:57 you are doing it in retro 20:08:21 parable is written in coffeescript/javascript 20:09:38 in runs in a browser or at the cli 20:09:57 in a browser currently 20:10:41 ok 20:10:47 you have a link 20:10:57 http://rx-core.org/parable 20:13:14 interesting 20:15:19 --- join: Kumul (~Kumul@67.224.135.212) joined #retro 20:23:55 crc: is it backend or clientside 20:24:52 code is run client side. I'm working on server side stuff for storage of compiled code + docs + version control 20:25:38 * erider is not sure how to use it yet :( 20:26:08 prefixes and quotes 20:26:18 # for numbers 20:26:28 $ for chaaracters (unicode) 20:26:33 " for strings 20:26:43 & for pointers to named functions 20:27:36 [ #1 + ] "1+" define 20:27:49 attaches the name 1+ to the quote 20:28:39 I've done a few blog posts on it now 20:29:45 "base" variable 20:29:57 #10 &base ! 20:30:18 &base @ 20:30:20 so the name is inside " " 20:30:23 yes 20:30:42 that is new 20:30:45 the parser is not available to code, so you can't add parsing words 20:31:06 syntax is far more rigid than in my earlier languages 20:31:59 no access to classes, no access to parser, no compile-time macros (e.g, immediate words) 20:42:17 are those things you are trying to implement or is it how you are designing the language 20:46:05 crc: is parable going to be your focus for awhile? 20:52:46 yes, parable is my main focus currently 20:53:10 I won't be adding classes or parser access 20:53:14 crc: what are you going to use it for once you get it done 20:53:48 prototyping, and cross compilation 20:54:55 hmm interesting I have been reading that a lot of folks have been using javascript for cross compilation lately 20:55:30 it's still a vm internally. I could generate different forms of bytecode (e.g., exporting an app to ngaro, for offline use) 20:56:10 the decompiler is a big part of that. given a function, I can now generate a source listing for all dependncies, and nothing else 20:56:32 * erider compilers have aways interested me 20:57:56 my compilers are becoming increasingly simplisitic 20:58:14 why would you say that 20:59:53 my earliest compilers generated either assembly or raw binary code for x86, and I had several optimization techniques. Retro moved the work into the classes, switchin g to a direct threaded model 21:00:31 parable is more rigid in that the compilation is handled by the vm/parser itself. you can't extend it 21:00:44 I am play with generating assembly to tcc 21:01:11 that only issue is tcc doesn't support all the directives 21:01:55 parable seems to be more traditional 21:03:14 a lot of languages are not capable of modification of the language on the fly 21:46:11 --- quit: erider (Quit: ChatZilla 0.9.88.2 [Firefox 13.0.1/20120615064004]) 22:47:26 --- quit: Kumul (Quit: gone) 23:59:59 --- log: ended retro/12.07.14