00:00:00 --- log: started retro/10.02.01
04:08:44 <crcz> good morning
04:09:11 --- quit: crcz (Quit: crcz)
04:09:41 --- join: crcz (~crc@bespin.org) joined #retro
05:02:25 --- join: sunwukong (~vukung@business-80-99-161-225.business.broadband.hu) joined #retro
05:04:44 <crcz> hi sunwukong 
05:04:52 <sunwukong> hi
05:16:17 <sunwukong> I've stumbled upon retroforth about an hour ago, now I'm reading the source...
05:16:27 <sunwukong> ... where is this 'ngaro' name coming from?
05:32:09 <crcz> its a maori word
05:33:17 <sunwukong> a 'delicacy of the dead' in a Maori legend - according to Wikipedia
05:33:57 <sunwukong> I don't quite see the connection here, though ;)
05:35:31 <crcz> I've been using maori words for my newer applications
05:35:37 <crcz> don't read anything into the names :)
05:35:42 <sunwukong> okay :)
05:36:44 <sunwukong> well it certainly helps name clashing
05:47:20 <crcz> yes it does
07:54:59 --- quit: sunwukong (Quit: ERC Version 5.3 (IRC client for Emacs))
09:24:14 <crcz> irsii is pretty nice. I'll be using it a lot more in the future.
09:24:19 --- nick: crcz -> crcx
12:59:22 <crcx> has anyone found any problems with the current image?
12:59:34 <crcx> if not, I'll be pushing out 10.4 tonight
14:08:55 --- join: sixforty (~sixforty@pdpc/supporter/active/sixforty) joined #retro
14:12:35 --- join: erider (~chatzilla@unaffiliated/erider) joined #retro
15:00:49 --- join: bardamu (~carlos@201.250.148.122) joined #retro
15:00:53 --- part: bardamu left #retro
15:13:21 --- join: Cammy (~Cammy@209.37.96.58.static.exetel.com.au) joined #retro
15:13:47 <Cammy> #retro is Retro Forth, is that the programming language?
15:14:03 <Cammy> I guess I'll check those URLs
15:14:41 <Cammy> Ah, yeah damnit
15:14:44 --- part: Cammy left #retro
15:30:13 --- quit: sixforty (Quit: Leaving.)
15:32:13 --- join: sixforty (~sixforty@pdpc/supporter/active/sixforty) joined #retro
15:41:38 --- quit: SimonRC (Ping timeout: 246 seconds)
15:51:58 --- join: SimonRC (~sc@fof.durge.org) joined #retro
16:05:30 <erider> crcx I tried to re-factor your  within like this      : within ( x u l -- f ) over < >r > r> + -2 = ;
16:07:02 <crc> erider: won't work the same:  ( xul ) over ( xulu ) < ( xuf ) >r ( xu ) > ( f ) r> ( ff ) + ( n ) -2 ( nn ) = ( f )
16:07:31 <erider> I think I got the > < wrong 
16:07:40 <crc> erider: in yours, the first comparison (over <) compares the upper and lower limits, not the value
16:08:14 <erider> I thought over grabbed the three item on the stack 
16:08:19 <crc> no
16:08:33 <crc> stack effect ( x y -- x y x )
16:08:38 <erider> damn it :) 
16:09:00 <erider> lol 
16:09:13 <erider> back to the drawing board 
16:09:18 <erider> lol 
16:10:28 <erider> hmm 
16:11:17 <erider> I guess I can -rot over < >r < r> + -2 = 
16:12:07 <erider> crc that should work 
16:12:38 * erider fires up reva to test it
16:14:25 <crc> yours did give me an idea, which seems to work ok in retro:
16:14:26 <crc> : within ( xul-f )
16:14:27 <crc>   push over pop < push > pop + -2 = ;
16:15:10 <crc> though I seem to have the stack effect as xlu-f
16:15:48 <crc> so, : within ( xlu-f ) push over pop < push > pop + -2 = ;
16:16:07 <crc> or, for reva: : within ( xlu-f ) >r over r> < >r > r> + -2 = ;
16:17:06 <erider> whew re-factoring :) 
16:18:52 <erider> : within -rot over < >r > r> + -2 = ; 
16:19:48 <crc> rot and -rot aren't efficient in retro
16:19:58 <erider> oh 
16:20:07 <erider> but it seems to work in reva 
16:20:35 <erider> less words 
16:20:59 <erider> lol one less word 
16:21:35 <crc> rot takes four words to implement in retro
16:23:17 <erider> crc I was trying to implement with bitwise operations for most of the day with no luck 
16:24:06 <erider> :( 
16:25:12 <erider> crc also I was thinking about trying to break it up into two words as well 
16:26:43 <erider> crc why when you fetch here it returns zero? 
16:27:17 <crc> because nothing is allocated from 'here' to the end of the heap
16:27:28 <erider> I am trying to see once I use comma if I can access that data by backing up here by 4 
16:27:32 <crc> not all forths initialize the heap to 0's though
16:28:48 <crc> retro doesn't reinitialize after metacompilation, so bits of the old memory contents can be accessed
16:29:08 <crc> hence the use of 'zallot' as opposed to 'allot' if you need to ensure the memory is a known state
16:29:55 <erider> crc so my question is when you compile something with comma how do you access it 
16:30:22 <crc> in retro, you can rewind here, or save a pointer to it
16:30:27 <erider> : test             ,  here  4 -  ; 
16:30:35 <crc> reva may be different
16:30:51 <erider> how can use rewind here 
16:30:53 <crc> nope
16:30:59 <crc> erider: your test should work
16:31:08 <crc> 98 , here 4 - @ .
16:31:15 <crc> shows 98 on my mac
16:31:23 <erider> hmm 
16:32:07 <erider> ok because I originally tried with cells 
16:32:46 <crc> 98 , here cell- @
16:32:49 <crc> should work too
16:33:36 <erider> ahhhhhh I was trying cells 4 - 
16:34:23 <erider> I wanted to use cells because of the size issues and portability 
16:34:54 * crc no longer has to worry about cells, chars, etc :)
16:36:02 <erider> so you just use offsets? 
16:36:34 <crc> everything is 32-bits in retro, so 1+ 1-, etc. purely linear offsets
16:36:53 <erider> cool! 
16:38:00 <erider> crc is there another way to parse forward without using parse 
16:38:25 <erider> I have been playing around with tick ' 
16:38:52 <crc> all parsing stuff involves parse (or in retro, "accept")
16:39:06 <erider> I am trying to get a word from interpreter mode then place it behind tick 
16:39:36 <erider> ie : test   ' [ the word consumed ] ; 
16:39:59 <crc> ?
16:40:21 <erider> parsews just gets stuff and then it is like a string 
16:40:23 <crc> you just want the address to be in the definition?
16:40:27 <erider> I want an address 
16:40:35 <erider> yes 
16:40:35 <crc> : test ['] wordname ;
16:40:42 <crc> ['] = compile-time version of '
16:41:02 <erider> so + test 
16:41:21 <erider> test will contain the address of +? 
16:41:25 <crc> no
16:41:29 <crc> : test ['] + ;
16:41:47 <crc> ['] parses when encountered
16:41:51 <erider> I want to be able to pass a word to test 
16:41:58 <crc> ' + test
16:42:18 <erider> lol yeah I didn't want to do it that way 
16:42:25 <crc> : test ' ... .;
16:42:27 <crc> test +
16:42:34 <erider> but I guess there is no way around it 
16:42:36 <crc> those are the two approaches that can be used
16:43:14 <erider> test + can you explain this one 
16:43:31 <crc> if test has ', it'll parse for the address when test is called
16:44:58 <erider> so : test '  , here cell-  >r  1 2 r> @ execute 
16:46:04 <erider> crc does that look right 
16:46:12 <crc> ; after execute
16:46:17 <crc> but yes, it looks ok to me
16:46:28 <erider> after? 
16:46:42 <crc> you need to end the definition
16:46:50 <erider> yeah 
16:47:25 <erider> I just was type it in the way I understand it 
16:47:29 <erider> sorry 
17:00:17 <crc> np
17:04:44 <erider> crc topically what is the length of  hdr+definition 
17:04:53 <crc> that's implementation specific
17:05:12 <crc> reva keeps headers separate from code IIRC
17:05:55 <crc> in retro, you have the length of the name, plus one, plus link to the previous, xt, class. so min would be 5.
17:08:20 <erider> hmm 
17:10:54 <erider> crc do you have a word and a find 
17:11:03 <crc> ?
17:11:10 <crc> retro has 'find'
17:11:13 <erider> word find 
17:11:25 <erider> to search the dictionary 
17:11:33 <crc> takes a string, returns an address and a flag indicating whether or not the search was successful
17:11:54 <erider> I am thinking about how I can implement forget in reva 
17:12:26 <crc> look at the code for 'mark' in lib/util/misc for ideas
17:23:02 <erider> crc you said you have a rewind 
17:29:49 <crc> -allot
17:36:17 <crc> in retro, -allot is used to "rewind" the heap pointer ("here")
17:40:41 <erider> mark is ok for now 
17:41:03 * erider is trying to figure it out and how it works
17:52:13 <erider> crc ping 
17:52:20 <crc> yes?
17:53:20 <erider> hey how do you save a new word into the dictionary 
17:53:57 <crc> create a new header?
17:54:00 <erider> reva has save but I think that make a turnkey 
17:54:39 <crc> save newreva
17:54:42 <crc> ./newreva
17:54:56 <crc> any words, etc you create will be present in the new binary
17:55:03 <erider> I don't want a turnkey 
17:55:48 <crc> reva doesn't allow saving the system in a non-turnkey fashion
17:56:25 <erider> ok 
17:56:40 <erider> I guess I have to use -n to load a lib 
18:11:29 --- quit: virl (Remote host closed the connection)
18:33:40 <crc> no problems found, so 10.4 will be out tomorrow or later tonight
19:03:25 --- topic: set to 'Retro Forth | Logs @  http://tr.im/Kept | http://retroforth.org | Stable Release 10.4 | Wiki Key: d98t~3 | http://www.forthcommunity.com' by crc
19:05:17 <crc> http://s3.retroforth.org/download/10.x/retro-10.4.tar.gz
19:05:47 <sixforty> just when I managed to open the 10.3 tarball
19:05:55 <crc> heh, sorry 
19:06:04 <sixforty> no yer not :P
19:06:11 <crc> not really
19:06:13 <crc> :)
19:10:08 <sixforty> your rst processor from docutils package?
19:10:49 <crc> yes
19:15:44 <sixforty> image/RetroLanguage.rst : Version 10.3
19:16:17 <crc> the main doc file hasn't been updated; the new stuff is all in separate files
19:19:55 <sixforty> How should new/changed information be notated when it's merged with the main file? Thinscript? Superscript + or x over period? And would you want changelog or so added to main doc.
19:20:25 <sixforty> Nevemind. Couldn't possibly keep up with you.
19:20:47 <crc> I'm moving the main documentation into the wiki currently
19:23:42 --- quit: erider (Ping timeout: 272 seconds)
19:34:15 --- join: ontologist (~4771226c@gateway/web/freenode/x-kkrltlkmaegdkguu) joined #retro
19:35:31 <ontologist> I'm looking for a manual or tutorial
19:38:49 <crc> ontologist: http://code.assembla.com/retro/git/nodes/image/doc/RetroLanguage.rst?rev=d1e38002bd73f1db081bee710481b8a452c71f24
19:41:41 <ontologist> excellent. thank you
19:45:06 --- quit: docl (Ping timeout: 252 seconds)
19:50:06 --- quit: ontologist (Quit: Page closed)
20:19:30 --- join: probonono (~User@unaffiliated/probonono) joined #retro
20:33:16 --- quit: sixforty (Quit: Leaving.)
23:59:59 --- log: ended retro/10.02.01