00:00:00 --- log: started retro/06.08.18
05:49:20 --- join: timlarson_ (n=timlarso@65.116.199.19) joined #retro
07:45:27 --- nick: docl -> lukeparrish
11:02:35 --- join: neceve (n=claudiu@unaffiliated/neceve) joined #retro
11:03:32 --- quit: Cheery (niven.freenode.net irc.freenode.net)
11:03:33 --- quit: lukeparrish (niven.freenode.net irc.freenode.net)
11:03:33 --- quit: neceve (niven.freenode.net irc.freenode.net)
11:03:34 --- quit: virl (niven.freenode.net irc.freenode.net)
11:03:35 --- quit: Shain (niven.freenode.net irc.freenode.net)
11:03:43 --- join: [Shain] (i=steve@c-67-161-56-76.hsd1.ca.comcast.net) joined #retro
11:04:09 --- join: neceve (n=claudiu@unaffiliated/neceve) joined #retro
11:04:18 --- join: Cheery (n=Cheery@81.197.20.242) joined #retro
11:04:21 --- join: virl (n=virl@62.178.85.149) joined #retro
11:08:27 --- quit: [Shain] (Client Quit)
11:08:39 --- join: Shain (i=steve@c-67-161-56-76.hsd1.ca.comcast.net) joined #retro
11:19:03 --- join: lukeparrish (n=docl@74-36-211-202.dr01.hmdl.id.frontiernet.net) joined #retro
11:39:48 <Quartus> crc, new version of the retro ANS layer up: http://retroforth.net/paste/?id=140
11:40:22 <Quartus> I haven't included the conditional-compilation stuff yet, waiting to test against the is: version in your new release.
12:58:26 --- join: snoopy_1711 (i=snoopy_1@dslb-084-058-134-226.pools.arcor-ip.net) joined #retro
13:06:44 --- quit: Snoopy42 (Read error: 145 (Connection timed out))
13:06:44 --- nick: snoopy_1711 -> Snoopy42
13:23:52 --- log: started retro/06.08.18
13:23:52 --- join: clog (n=nef@66.114.33.57) joined #retro
13:23:52 --- topic: 'RetroForth | Pastebin @ http://retroforth.net/paste | The editing key is 'despair''
13:23:52 --- topic: set by crc on [Sat Aug 05 13:13:30 2006]
13:23:52 --- names: list (clog Snoopy42 lukeparrish @ChanServ timlarson timlarson_ neceve virl Shain)
13:24:20 --- join: nighty_ (n=nighty@62.128.181.22) joined #retro
13:24:24 --- join: Quartus (n=trailer@72.136.201.3) joined #retro
13:29:10 --- quit: timlarson_ ("Leaving")
14:03:24 --- quit: nighty_ (Read error: 104 (Connection reset by peer))
14:39:30 <Quartus> crc, some weirdness in rf I've discovered relating to inconsistent behaviour of include and -f.
14:39:36 <Quartus> oops, no crc
14:39:38 <Quartus> :)
14:46:59 --- join: crc (n=crc@pool-70-110-183-96.phil.east.verizon.net) joined #retro
14:47:09 <Quartus> crc, some weirdness in rf I've discovered relating to inconsistent behaviour of include and -f.
14:47:25 <crc> what weirdness?
14:48:06 --- mode: ChanServ set +o crc
14:48:43 <Quartus> under windows.    File a includes file b.  Fails to work if I do  include a   at the prompt, terminates loading at the  include b  silently.  If I do rf-windows -f a  it works.
14:49:33 <crc> I'll take a look at this
14:49:38 <Quartus> File A:  ." hi " include b ." everybody"
14:49:45 <Quartus> File B: ." there"
14:49:56 <Quartus> rf-windows -f A  ->  hi there everybody
14:50:03 <Quartus> rf-windows -> include A -> hi
14:50:19 <crc> http://retroforth.org/bin.tar.gz has the 9.2.9 binaries if you want to test your ANS code against it
14:50:25 <Quartus> ok!
14:53:10 <Quartus> Runs my tests on the version I pasted -- http://retroforth.net/paste/?id=140
14:53:20 <Quartus> Blows up huge on my new conditional compilation words.
14:53:38 <Quartus> I'll work on that.
14:54:37 <crc> ok
14:54:47 * crc is working on the problem with nested includes
14:59:03 <Quartus> Something changed in 9.2.9 that makes my conditional words operate very strangely.
14:59:43 <crc> other than my "is:" word?
14:59:49 <Quartus> Oh, it may be my own astonishing stupidity. Hang on.
15:00:30 * crc only added  is: (is) #!  and made a slight tweak to  ;;
15:01:03 <Quartus> I vote for 'now', also. I need it. :)
15:01:15 <crc> do you have a definition?
15:01:23 <Quartus> : now  1+ dup @ 4 + + ;
15:01:28 * crc will add it
15:01:45 <Quartus> Great!  It pulls the xt out of a vectored word.  ' foo now 
15:03:41 <Quartus> Ok, I've resolved my personal insanity. :)
15:09:23 <Quartus> You know, only three small issues stand between where it is now and a 100%-ANS Forth compliant system.
15:09:33 <crc> >in being one of them
15:09:59 <crc> I may be able to change that in 9.3 (sometime in September)
15:10:02 <Quartus> >IN needs to be zero-indexed against the line currently being parsed,  SOURCE needs to return the addr and len of the current line being parsed, and the interpreter needs to have the option of displaying an OK prompt.
15:10:09 <Quartus> Well, and I have to fix MOVE.
15:10:27 <crc> :: ."  ok" cr ; is ui
15:10:37 <Quartus> One down!
15:10:37 <crc> should do the ok prompt
15:10:54 <crc> >in will require some internal changes, hence the 9.3 timeframe
15:11:02 <Quartus> I figured it would.
15:11:18 <Quartus> That prompt coughs up two 'oks' a lot of the time.
15:11:24 <crc> on windows
15:11:27 <Quartus> Yes.
15:11:30 <crc> the cr/lf crap
15:11:36 * crc will take a look at that as well
15:11:49 <Quartus> I'll leave it out of the ans layer for the time being.
15:11:52 <crc> it's hard for me to test the windows version, since I don't have a windows install on this box
15:11:55 <crc> ok
15:12:09 <Quartus> The differences appear minor.
15:13:08 <Quartus> If you have a build with now in it, I'd be happy to test and give you this newest release with the conditional-compilation goodies.
15:14:36 <crc> http://retroforth.org/bin.tar.gz now has now
15:14:46 <Quartus> That's the filename?
15:14:48 <crc> I think I fixed 'include' to do nesting
15:14:54 <crc> yes
15:16:35 <Quartus> I think your include fix truncates larger includes.
15:16:58 <Quartus> It's halting silently partway through coretest.fs
15:17:01 <crc> it has a limit of 32k per include
15:17:03 <crc> hmm
15:17:50 <Quartus> Coretest is 25K or so.
15:18:02 <crc> hmm
15:18:10 <crc> it shouldn't fail on that
15:18:17 * crc will take another look
15:18:40 <Quartus> It just suddenly stops including anything after about 65% of the way through.
15:18:57 <Quartus> -f works like a champ still, though.
15:19:25 <Quartus> No problems with: rf-windows.exe -f ans.txt -f tester.fs -f coretest.fs
15:19:43 <Quartus> Big problems with: rf-windows.exe -> include tester.fs -> include coretest.fs
15:19:57 <Quartus> er, sorry
15:20:00 <Quartus> Big problems with: rf-windows.exe -f ans.txt -> include tester.fs -> include coretest.fs
15:20:55 <crc> I updated the bin.tar.gz with a fixed version
15:21:13 * crc had it set to 16k max per include :(
15:21:48 <Quartus> Smooth; works fine.
15:22:03 <crc> good
15:25:08 <Quartus> include now behaves oddly in another more esoteric test that I do, where I re-alias and re-class all the Standard words into one wordlist and seal the rest of the system for testing.  I can pursue that later, though, it doesn't affect the ANS layer itself.
15:26:02 <Quartus> Updated ANS layer: http://retroforth.net/paste/?id=141
15:40:00 <Quartus> Earlier I discovered that inside a block comment delimited by 0 [if] and [then], you can't talk about [then]. :)
15:46:35 <crc> makes sense
15:47:21 <Quartus> So I talk about [THEN] instead, as the case-folding search isn't implemented until later.
15:48:30 <Quartus> I've been using retro with this layer to do development work, to give it a real test.
15:51:52 <crc> cool
15:51:57 <crc> 9.2.9 is up
15:52:05 <Quartus> Great.
15:53:56 <Quartus> You spelled my name wrong in your changelog ;)
15:56:33 <crc> sorry :(
15:56:37 <Quartus> No worries.
15:56:45 <Quartus> Happens a lot.
15:56:49 * crc will check more carefully in the future
15:58:21 * crc fixes it
15:58:25 <Quartus> :)
15:58:53 <Quartus> with an 'a' is ye olde Celtic spelling
16:25:56 <virl> buha '\' comments how disgusting, 
16:26:14 <Quartus> Thanks for the constructive input, virl.
16:26:28 <lukeparrish> hehe
16:26:45 <virl> heh docl
16:26:56 <lukeparrish> how ya doing virl?
16:31:30 <virl> fine..
16:31:49 <lukeparrish> cool
16:32:21 <virl> I have a basic xell system with the primitives now I need todo the real stuff.
16:32:55 <lukeparrish> sweet
16:39:26 <virl> ans - it's dead jim
16:44:54 <virl> and what are you doing?
16:45:45 <lukeparrish> nothing much
16:46:06 <virl> oh..
16:54:21 <virl> I can't understand why the ans folks chose uppercase for important words. words are words and there isn't a difference between them.
17:00:54 <virl> no fun with ans, they call it immediate word and I call it a macro, they do big code and I do small one
17:02:46 <virl> how did that become like that?
17:05:19 --- join: jas2o (n=jas2o@wnpp-p-144-134-163-120.prem.tmns.net.au) joined #retro
17:34:14 <virl> | or \ oh my god.
17:47:16 <virl> somehow I got used to crcs | comments, they don't read like msdos directories.
17:48:08 <virl> the cool stuff about ans perhaps is 'see', one name which fits. others don't
17:52:54 --- quit: virl ("Verlassend")
18:36:40 --- quit: jas2o ("Chatzilla 0.9.74 [XULRunner 1.8.0.1/2006012608]")
18:51:38 <Quartus> Be curious to hear opinions on where the ANS layer could use additional explanatory commenting.
21:46:58 --- quit: Quartus (Read error: 104 (Connection reset by peer))
21:48:18 --- join: Quartus (n=trailer@CPE0001023f6e4f-CM013349902843.cpe.net.cable.rogers.com) joined #retro
22:51:47 --- join: nighty (n=nighty@sushi.rural-networks.com) joined #retro
23:41:07 <Quartus> crc, interesting to note that only 8 retro words have names with different meanings than in Standard Forth: # \ exit forth i >number repeat find
23:42:41 <Quartus> I suppose find doesn't count.  And >number doesn't either -- slightly different return results.
23:43:26 <Quartus> Even leaving them in, that's only 8 out of 251 words.
23:59:59 --- log: ended retro/06.08.18