00:08:40 -!- jonrafkind [~jon@crystalis.cs.utah.edu] has quit [Read error: Operation timed out] 00:10:57 -!- fradgers- [~fradgers-@5adafe9d.bb.sky.com] has left #scheme 00:10:59 hadronzoo [~hadronzoo@ppp-70-251-108-227.dsl.rcsntx.swbell.net] has joined #scheme 00:17:19 Toekutr [~toekutr@adsl-69-107-109-7.dsl.pltn13.pacbell.net] has joined #scheme 00:21:37 turbofail [~user@adsl-69-238-246-201.dsl.pltn13.pacbell.net] has joined #scheme 00:25:33 -!- Riastradh [debian-tor@fsf/member/riastradh] has quit [Ping timeout: 245 seconds] 00:25:41 -!- hadronzoo [~hadronzoo@ppp-70-251-108-227.dsl.rcsntx.swbell.net] has quit [Quit: hadronzoo] 00:28:01 -!- mejja [~user@c-14bee555.023-82-73746f38.cust.bredbandsbolaget.se] has quit [Remote host closed the connection] 00:33:19 -!- Adamant [~Adamant@unaffiliated/adamant] has quit [Quit: Adamant] 00:37:27 Riastradh [debian-tor@fsf/member/riastradh] has joined #scheme 00:41:23 -!- dfkjjkfd [~paulh@109-13-ftth.onsnetstudenten.nl] has quit [Quit: Lost terminal] 00:41:57 -!- ray [ray@xkcd-sucks.org] has quit [Remote host closed the connection] 00:43:06 -!- rbarraud [~rbarraud@118-92-9-12.dsl.dyn.ihug.co.nz] has quit [Read error: Operation timed out] 00:43:36 rbarraud [~rbarraud@118-92-9-12.dsl.dyn.ihug.co.nz] has joined #scheme 00:44:02 ray [ray@xkcd-sucks.org] has joined #scheme 00:53:51 -!- hosh_office [~hosh@c-71-204-27-0.hsd1.ga.comcast.net] has quit [Ping timeout: 272 seconds] 01:02:52 hosh_office [~hosh@c-71-204-27-0.hsd1.ga.comcast.net] has joined #scheme 01:13:23 -!- turbofail [~user@adsl-69-238-246-201.dsl.pltn13.pacbell.net] has quit [Remote host closed the connection] 01:23:49 hypercube32 [~hypercube@169-91.202-68.tampabay.res.rr.com] has joined #scheme 01:33:12 -!- dr_saccade is now known as saccade_ 01:45:59 -!- bremner [~bremner@yantan.tethera.net] has quit [Quit: ZNC - http://znc.sourceforge.net] 01:46:09 bremner [~bremner@yantan.tethera.net] has joined #scheme 01:48:00 MrFahrenheit [~RageOfTho@users-55-117.vinet.ba] has joined #scheme 01:48:25 -!- RageOfThou [~RageOfTho@users-55-117.vinet.ba] has quit [Read error: No route to host] 01:48:34 arcus [~ajt@125-236-141-218.jetstream.xtra.co.nz] has joined #scheme 01:49:27 Hello, arcus! I haven't seen you in a while around here. 01:49:38 I haven't been here in a while :] 01:49:41 hi, Riastradh 01:49:55 how are you? 01:50:28 Golly! I'd never have guessed that might be the explanation. Surely you must have a more elaborate one! 01:51:22 I'm doing pretty well. Been in any Jean-Pierre Jeunet films lately? 01:51:41 actually, I changed my name to zzarcus, and hid under zbigniew all this time 01:52:01 no :] 01:52:26 although I was gazing at a copy of delicatessan in the video store yesterday, thinking I must watch that. 01:52:29 is he in that? 01:52:58 Why, yes, of course you are in it. 01:53:04 You played the main character! 01:54:17 ah, how silly of me to forget :] 01:55:53 It's a pretty good film. Not quite _La Cit\'e_ or _Am\'elie_, but pretty good nevertheless. 01:59:08 so what's new in the scheme neck of the woods 01:59:19 I hear there's a new standards effort underway 01:59:32 although you're probably entirely sick of such things 02:00:18 -!- acarrico [~acarrico@pppoe-68-142-49-233.gmavt.net] has quit [Ping timeout: 240 seconds] 02:02:39 I am only dimly aware of the new standards process. 02:03:37 and everyone is going on and on about how fantastically great clojure is 02:04:14 Ha. 02:05:00 Well, language design aside, at the moment I want to stay as far away from the JVM as possible, after Oracle's recent present for Google. 02:05:19 (not that I have ever been particularly attracted to the JVM to begin with!) 02:05:35 I was actually wondering whether that might make it easier to target for scheme compilers 02:05:52 although I never got a straight story of what the modifications actually were 02:05:58 -!- MrFahrenheit [~RageOfTho@users-55-117.vinet.ba] has quit [Ping timeout: 265 seconds] 02:06:03 so it was a pretty idle wonder :] 02:08:27 What might make what easier to target for Scheme compilers? 02:08:28 Modifications? 02:08:55 aren't you talking about the recent modifications to the JVM? 02:09:52 No, I'm talking about the recent lawsuit that Oracle handled Google. 02:09:59 oh. 02:10:10 it involves google's use of it in android 02:10:27 the ceo at oracle has lost his mind 02:10:58 Well, sort of. `JVM' is a slightly red herring; the issue is Java more generally, and its standard libraries, on which there are software patents that Oracle claims Google has violated. 02:11:07 is that still Larry Ellison 02:11:15 Yes, it's Larry Ellison. 02:11:15 ya, JVM has nothing whatever to do with android 02:11:27 No, it does have something to do with the Android. 02:11:57 how? 02:12:00 What it has to do with the Android is that Google carefully avoided implementing it on the Android in order to get around certain legal constraints that Sun had put on the Java system. 02:12:04 oh. 02:12:13 offby1, yes it does actually 02:12:31 if you want to code in android in the sdk, its all java 02:12:34 However, in getting around those legal constraints, what Google ended up with was not an implementation of Java, as specified by Sun. Consequently, Google was not automatically granted a licence to the patents that cover various things in Java. 02:12:40 hypercube32: java, yes. JVM, no. 02:12:56 (It's not Java(TM) either.) 02:13:03 (That was deliberate on Google's part.) 02:13:09 I thought Java had been open-sourced. 02:13:17 -!- askhader [~askhader@taurine.csclub.uwaterloo.ca] has quit [Remote host closed the connection] 02:13:22 offby1, nope 02:13:24 Sun's implementation of Java has been released under the GPL. 02:13:31 However, there's a caveat. 02:13:48 Sun still held patents on various things needed to implement a Java standard library. 02:13:50 oracle is fking with mysql and doing shenanegans as well 02:14:07 and oracle leaked shutting down opensolaris and going back to the broken idea of solaris11 02:14:21 all very bad decisions 02:14:24 Sun granted a royalty-free licence for the use of those patents to anyone who implemented Java(TM) -- i.e., created a specification-compliant implementation. 02:14:37 Since Google didn't do this -- they implemented Java, not Java(TM) --, they didn't get the patent licence. 02:14:41 Now Oracle is suing them. 02:14:49 *offby1* rolls eyes 02:15:08 That's my rough, high-level understanding of the situation, anyway. 02:15:19 Riastradh, thats what i heard as well 02:15:25 atleast roughly 02:17:50 So, anyway, going near the JVM implies going near Java, and now I have even more reasons not to do that. 02:18:14 let's write a programming language that does nothing but web service API calls. 02:18:32 lets change that subject before i vomit in my mouth? :/ 02:19:18 Do you want us to pop the stack back to Clojure, hypercube32? 02:20:37 im not a huge fan of that language but ya i guess its better i dunno 02:20:39 Or would that only further encourage your stomach to rbellion? 02:20:42 Rebellion, even. 02:20:59 it would only alleviate things as per my ignorance of it :) 02:21:28 -!- proq [~user@unaffiliated/proqesi] has quit [Read error: Operation timed out] 02:24:38 how about we pop the stack back to dim sum 02:24:49 -!- bgs100 [~ian@unaffiliated/bgs100] has quit [Quit: Leaving] 02:27:13 -!- Riastradh [debian-tor@fsf/member/riastradh] has quit [Remote host closed the connection] 02:27:25 Spewns [~Spewns@97-92-222-240.dhcp.stls.mo.charter.com] has joined #scheme 02:27:39 askhader [~askhader@taurine.csclub.uwaterloo.ca] has joined #scheme 02:29:30 Riastradh [debian-tor@fsf/member/riastradh] has joined #scheme 02:30:11 ...boink. 02:34:26 -!- Spewns [~Spewns@97-92-222-240.dhcp.stls.mo.charter.com] has quit [Quit: Spewns] 02:36:27 acarrico [~acarrico@pppoe-68-142-49-233.gmavt.net] has joined #scheme 02:40:03 -!- kniu [~kniu@pool-71-106-0-142.lsanca.dsl-w.verizon.net] has quit [Remote host closed the connection] 02:47:10 does the name Hugh Aguilar mean anything to anyone? 02:47:17 I'm wondering why it sounds familiar to me. 02:50:25 why .... that's the fellow who ran off with my first wife! 02:51:13 Not at all to me, arcus. 02:51:38 google it and save yourself some time 02:52:21 google has yet to get to the stage where it tells me where I personally heard a name :] 02:52:27 -!- timj_ [~timj@e176194023.adsl.alicedsl.de] has quit [Read error: Connection reset by peer] 02:52:32 Serious about that Oracle stuff earlier??? 02:52:35 timj_ [~timj@e176194023.adsl.alicedsl.de] has joined #scheme 02:52:52 Ugh yea seems true 02:53:14 Patents and software just don't mix IMO... 02:53:45 he's a character on comp.lang.forth - I was wondering whether he'd ever had anything to do with scheme, as I don't think I've read any posts on comp.lang.forth until two weeks ago. 02:53:50 I doubt that's a controversial opinion here. :) 02:53:57 Lol 02:54:20 I did google it, didn't help. It's not worth giving any further thought to :] 03:03:28 Riastradh: so you're not enthusiastically embracing the new standards effort, you haven't abandoned scheme in favour of clojure, and you're not busying yourself with developing scheme compilers targeting the JVM... 03:04:09 -!- aymeric [~aymeric@CPE0019e33fe10f-CM0017ee42b49e.cpe.net.cable.rogers.com] has quit [Quit: aymeric] 03:04:13 why are you so anti-progress? :] 03:05:17 why abandon scheme for clojure? 03:05:20 Can't you tell? I'm running for a seat in the House. 03:05:21 i dont know that id call anything using the jvm progress lol 03:05:30 Riastradh, ha 03:05:36 isn't it way slower 03:05:53 i'd rather have the nicer language and speed w/ scheme if i can 03:06:33 although java is pretty good lately 03:06:45 on the performance (aside from memory i guess) 03:07:14 -!- hosh_office [~hosh@c-71-204-27-0.hsd1.ga.comcast.net] has quit [Ping timeout: 264 seconds] 03:07:59 the performance so they claim has increased but i still dont see it anything close to c++ 03:08:16 it's not all that far off 03:08:28 but one problem is smaller applications 03:08:40 the language itself is horrible anyway 03:08:45 if you need to start something up quickly and do some short work, the jvm is bad 03:08:52 java i mean 03:09:00 arcus: He's not anti-progress. He goes ``boink'' and everything! 03:09:05 probelm with 'java' is that it means two different things 03:09:17 "Java" the language and "Java" the platform 03:09:35 i worked with sun implementation and i wanted to throw it out the window every day 03:09:42 Lol 03:09:48 chandler: yes, that's the same sound that scientific progress makes, isn't it? 03:10:35 Lately I just do a Scheme and C/C++ combo and I've been pretty happy with it for the most part 03:11:31 But I think if Java (the language) wasn't so verbose, it wouldn't be so bad 03:12:01 i do php lately for money, but scheme and c++ is where ill keep my motivation when im not doing math 03:12:01 It's just that almost anything I've tried to do or seen people do takes a billion lines of code 03:12:24 Then you've got to "refactor" it and shove around big fat piles of code 03:12:47 that billion lines of code shows how aweful the language itself is, you shouldnt have to write a book to say hello world 03:12:53 I know 03:13:15 i got paid to write in it for 6 months, alot of money, ... it wasnt worth it 03:13:16 The language just isn't that flexible IMO 03:13:33 Not much can be done to compress the code down 03:14:03 They went insane on the whole OOP thing to the point where it's just silly 03:14:14 so about scheme, i have used guile and tinyscheme, but these seem to be r5rs? i want to use a 6 version (im new into this language so sorry if i misrepresent what im trying to say) 03:14:26 what implementation is good for 6? 03:14:29 There are a few r6rs 03:14:31 on a linux system 03:14:42 (and free) 03:14:43 ypsilon, ikarus, a couple others 03:14:50 is there a list somewhere? 03:14:52 presumably PLT? 03:14:54 I don't think ikarus works on linux (at least amd64) 03:14:55 thanx btw ill check it out 03:14:59 and wtf is PLT mean? 03:15:00 I know ypsilon works 03:15:07 hypercube32: it's a scheme implementation 03:15:08 Isn't that called Racket now? 03:15:11 is it? 03:15:13 i keep seeing that PLT everywhere 03:15:15 'tis 03:15:16 I'm out of touch :] 03:15:20 ok 03:15:32 They declared it a "new language" 03:15:38 plt? 03:15:44 Although it's really Scheme with lots of extensions 03:15:48 hm 03:15:59 Yea, it was PLT but they renamed it to Racket 03:16:15 Idk I still use r5rs tbh 03:16:16 i want to stick with something that can atleast stick around for quite a while and that has atleast 6 in it (not screwing it up on the standards part) 03:16:34 r6rs might not even stick for all I know 03:16:38 true 03:16:40 Racket is a dialect of Scheme. It's neither R5RS nor R6RS. 03:16:50 ok i want to avoid weird dialects like that 03:17:02 R6RS is widely acknowledged not to have stuck. 03:17:10 Hold on I'll find out what I had workincg 03:17:18 chandler, so you're saying most people still are using 5? 03:17:29 atleast to be safe for now 03:17:31 Racket is a multilingual environment. You can use the R5RS in it if you want. 03:17:32 I use mostly gambit-c and chicken 03:17:47 but i need C/C++ interop 03:17:54 thats why im using guile lately 03:18:07 jyaan, how does guile compare to those two? 03:18:21 Racket isn't "weird"; it's a very well-used dialect with good support. 03:18:22 Pretty good 03:18:32 what is going on??? my capital g doesn't work?? 03:18:46 guile doesn't compile to native 03:18:51 but it has very good interpreter speed lately 03:19:02 it's faster than Lua, for example 03:19:12 jyaan, i can use the api i guess to compile C code, how does that mean its not native? 03:19:23 That's not what I mean 03:19:33 jyaan, or are you talking like straight from scheme to binary? 03:19:40 With gambit or chicken, you compile to C then to binary 03:19:49 with guile you run like a script 03:19:53 true 03:20:00 Guile 1.8.x is a poor implementation. It's slow and its hygiene is flawed. Guile 2.0 addresses these issues. 03:20:02 ofc you can still run scripts w/ the others 03:20:03 and you can embed the scheme into a c file 03:20:07 with guile 03:20:22 chandler, ive been using 1.8 :/ 03:20:33 but im so new at it i dont think ive noticed issues 03:20:41 i have 1.8.7 guile 03:20:47 If you're picking an implementation to get started with, I'd recommend Racket. You'll be able to get the best support here for it. 03:20:55 he's on linux btw 03:21:07 i don't think it really matters all that much tbh 03:21:12 pick one that you like 03:21:26 i basicaly need something with at the least a good r5rs implementation in it 03:21:48 guile, gambit, chicken, gauche, even bigloo are all good 03:21:50 anything beyond that is just bonus, but id like to stick to r5rs as close as possible so i dont get confused between what is, and is not r5rs 03:22:00 Racket also has the best documentation I've seen from an implementation. 03:22:14 chandler, cool ill check that out too 03:22:30 i haven't used racket mainly cause tehre are no linux packages 03:22:41 Er, what? 03:22:43 no src to compile either? 03:22:50 I'm using Racket on Linus. 03:22:53 for 64 03:22:55 er, Linux. 03:23:04 i see source tho 03:23:11 guess i should try it out 03:23:34 ...*what* are you doing to Mr Torvalds, chandler? 03:23:37 hm maybe i was tinking of something else 03:23:54 ya ive been doing c/c++ for a long time, to compile something is easy and get it going, no need for packages, but sometimes its nice to have them 03:24:03 hypercube32: MIT Scheme is another very good implementation, as is Scheme48. 03:24:41 cool ill try to look through a few others and see whats up with them 03:24:58 I installed Racket on amd64 Linux from a distribution package. Your luck may vary depending on how quickly your distribution picks up new versions. 03:25:55 i must have been thinking of another scheme 03:26:08 downloads look OK or they changed recently 03:26:19 like check this out, ive been having issues trying to figure out how to use guile with gtk and a database interface, which one of these implementations has a good straight forward way to use these two things? 03:26:22 or is there one? 03:26:45 for C libraries generally you have to write a binding 03:26:47 or find one 03:26:56 chandler, im using ubuntu for the fun of it, but have used everything from opensolaris, linux, whatever, gentoo, ubuntu, you name it 03:27:03 Of the other implementations mentioned, Chicken is a good choice; its only major flaw is the lack of indefinite-size exact numbers by default. 03:27:04 chicken has a lot of C bindings though 03:27:21 yea, make suer to enable GMP if you need that 03:27:37 Reason it doesn't come by default is that Chicken is BSD license 03:28:30 I don't know about GTK specifically with Scheme. Racket doesn't currently use GTK, but it has a built-in GUI library and available database libraries. 03:28:49 guile definitely has a binding but i haven't tried it 03:29:04 Another reason the GMP may not come by default is that its memory allocation interface is ridiculously broken for use in a garbage-collected environment. 03:29:28 In particular, the GMP wants memory allocation failure to cause the process to terminate. 03:30:21 brb my keyboard is driving me insane 03:30:26 -!- jyaan [~jyaan@c-98-250-102-194.hsd1.mi.comcast.net] has quit [Quit: Leaving.] 03:31:04 So an escapee from the local loonie bin walks into a bar with a keyboard attached to his head, and tries to make a lewd joke about pirates... 03:31:25 leftylink [~lefty@c-24-16-114-188.hsd1.wa.comcast.net] has joined #scheme 03:31:53 -!- leftylink [~lefty@c-24-16-114-188.hsd1.wa.comcast.net] has left #scheme 03:35:12 hypercube32: If you're willing to explore the various implementations, go nuts; there are certainly plenty to choose from. I tend to give one recommendation for new users (Racket) because I've seen people transfer a negative impression resulting from a regrettable implementation choice to Scheme as a whole, and that's unfortunate. 03:36:44 Not all implementations are of good quality, especially those designed with the view that Scheme is easy to implement. (It really isn't!) 03:38:02 chandler, thanks, im very sure ill try a few, but unfortunately for my few future projects id like to try to stick with one long enough to implement something useful, vs keep checking out others 03:38:17 jyaan [~jyaan@c-98-250-102-194.hsd1.mi.comcast.net] has joined #scheme 03:38:22 chandler, im very much a command line kind of person, i dont like IDE's 03:38:35 me too, sort of 03:38:36 and i know scheme and C++ is what i want to go with 03:38:46 I generally just use Emacs 03:38:54 what scheme implementation handles OOM gracefully? 03:38:56 im a vim guy 03:39:03 I got Semantic w/ that so I'm really not missing any IDE-like features 03:39:47 ide's to me just get in the way, and their non-command mode editors make me insane 03:39:52 Oh btw, another plus of Chicken is the nice "eggs" system (package system) 03:39:55 and the auto complete is the first thing id shut off 03:40:09 just sudo chicken-install something to install something 03:40:10 jyaan, ya i just saw that actually, sounds good 03:40:18 It's pretty fast too 03:40:22 jyaan, how is it if i want to mix in c++ though? 03:40:37 jyaan, i mean i know how guile does that, but does chicken allow me to do that somehow? 03:40:41 I think it does C++ through Swig 03:40:49 Maybe you'd have to do extern C on stuff not sure 03:40:54 jyaan, basically i need to do stuff in gnome with a canvas 03:41:06 gambit can be compiled with a c++ compiler though, so you can just mix whatever you like that way 03:41:08 Rakko [~rakko@71-90-73-192.dhcp.ftbg.wi.charter.com] has joined #scheme 03:41:10 so i can do some vector math, visually 03:41:13 even objective-c can be mixed in lol 03:41:20 foof, every one of them! Here's how you handle it: run the GC. 03:41:46 Well GNOME is C-based 03:42:04 yes but there are gtkmm and other mm wrappers for c++ 03:42:08 It's hard to gracefully handle the case when the GC reclaims too little memory, but that's not the issue here. 03:42:26 I think it would be a bad idea to bind to gtkmm instead of plain gtk+ 03:42:43 extra layer for no reason, scheme just needs to call the libs 03:42:59 jyaan, true, but i do need objects in c++ anyway, so either way id like to stay oop with the c/c++ style code 03:43:02 Riastradh: i was curious about that case 03:43:02 hypercube32: I'll reiterate my Racket suggestion; there's already a built in canvas-like surface and drawing primitives. 03:43:13 chandler, ill definately check that out then 03:43:52 sorry if i sound a little confusing, im new at the scheme world :) 03:44:02 Well, what are you trying to get then, just a scripting language on top of your C++? 03:44:04 tired of the imperative languages finally 03:44:08 foof, both MIT Scheme and Scheme48 put some effort into handling that case, by reserving some memory for that case (basically, by running the GC when the memory is not actually full, but nearly full). MIT Scheme tries to abort whatever evaluation it is running and clean anything it can to release more memory. 03:44:42 jyaan, well i was going with the c/c++ to bind in some libs i couldnt understand to get in there with scheme previously, thus my crutch on it 03:44:49 Probably would be a good idea to write apps that don't use all memory if possible.. :) 03:44:49 foof: I believe Racket can recover from an out-of-heap condition, but only by embiggening the heap quota. 03:44:50 hosh_office [~hosh@c-71-204-27-0.hsd1.ga.comcast.net] has joined #scheme 03:45:15 Scheme48 suspends all running threads and pushes a new command level, at which you can try to eliminate useless data and abort. 03:45:21 jyaan, my goal is to do some gui coding for a canvas style vector graphics type work and set theory type stuff on the back end (amoung other) like heavy math based stuff 03:45:37 What I've been doing is writing the "lower layer" of my code in C++ then basically doing everything on top of that Scheme 03:45:50 jyaan, like in the c++ world it would be gtk, gmp, cairo, blas, and some other stuff 03:46:46 scheme ffi is generally pretty easy to use, it's just really tedious for huge libraries 03:46:57 whats ffi? 03:46:59 chibi returns a pre-allocated OOM object 03:47:03 some sort of function interface? 03:47:04 foreign function interface 03:47:07 ;) 03:47:18 yea for calling c/c++/objective-c/whatever 03:47:25 and ya i can see how that would be a bit of a pain 03:47:46 OK folks I'm outa here 03:47:47 ciao 03:47:48 i really only bind my app's one layer 03:48:02 -!- arcus [~ajt@125-236-141-218.jetstream.xtra.co.nz] has left #scheme 03:48:03 jyaan, what would be ideal too is being able to have it compile to c/c++ for say inclusion into php pecl libs (php extensions) (not the gui stuff but other things) 03:48:13 jonrafkind [~jon@c-67-172-254-235.hsd1.ut.comcast.net] has joined #scheme 03:48:38 -!- REPLeffect [~REPLeffec@69.54.115.254] has quit [Ping timeout: 265 seconds] 03:48:39 Well i don't know much about PHP 03:48:56 But I know you can write the Scheme code and make a little C interface to access it 03:49:00 jyaan, as long as you can run c code, you can make a php extention, its pretty easy actually 03:49:12 jyaan, and php can auto bind in external libs 03:49:16 Oh 03:49:29 jyaan, so as long as i have a "c function" i can do it :) 03:49:45 I think there is something like that for Chicken called "easy-ffi" but it's slower than writing your own 03:50:13 jyaan, one of the other things im tasked with at work is writing a php extention to do some custom math stuff... which id normally have to do in c :/ but i wanna do it in scheme instead 03:50:33 Yea Scheme would make it easier 03:50:47 jyaan, so the amount of interfaces would actually be minimal, just enough to call what i need, the guts would be deep though 03:51:06 Yea only write interface to the parts you call 03:51:11 Binding goes both ways btw- 03:51:18 Scheme can call C and C can call Scheme 03:51:32 ya i saw something like that in guile :) very very cool to me 03:51:38 thats what actually attracted me into scheme 03:51:39 guile is a bit difference 03:51:42 different* 03:51:57 well i just mean conceptually when that idea hit me, it was like a bolt of lightning 03:51:58 That's more like Lua where you call the Guile library and stuff like tat 03:51:59 -!- hosh_office [~hosh@c-71-204-27-0.hsd1.ga.comcast.net] has quit [Ping timeout: 276 seconds] 03:52:23 For Schemes like gamit you write some little scheme code for it 03:52:28 gambit* 03:52:43 I'll put a small example, 1 sec 03:52:57 I'm very new to Scheme. How do I invoke a Scheme program so that Scheme terminates when it's done? 03:52:57 especially for a newbie like me thats already a c pro, it makes it easy cause i can do what i dont understand yet in scheme (still in c) and then dig into scheme when where i know how to swim 03:53:48 hosh_office [~hosh@c-71-204-27-0.hsd1.ga.comcast.net] has joined #scheme 03:54:29 -!- Riastradh [debian-tor@fsf/member/riastradh] has quit [Quit: leaving] 03:54:43 jyaan pasted "gambit-c binding" at http://paste.lisp.org/display/113724 03:55:13 well it's nice being able to choose the language 03:55:24 gives a lot of freedom 03:55:34 cool looking at it now 03:55:55 what i do lately is write scheme code that gets loaded as script at first 03:56:14 then i check it out and test it and whatever -- if it's good, then i just have it compiled instead 03:56:35 having a repl is really just awesome 03:57:19 oh dang should have put a c-lambda too 03:57:34 that's actually from the examples 03:57:55 ive actually only read the first chapter all the way through on the book mentioned in the topic (the tspl one) 03:58:06 so the function at the bottom would be available in C as jytek_scriptinterface_eval_string 03:58:09 and i know a bit of lambda calc 03:58:25 but ya i recognize the extern c stufff :) 03:58:55 you can also do static and w/e else 03:58:58 so there are like ways i can make this scheme into a shared C object as such? 03:59:02 like a .so? 03:59:12 there is also c-declare tat you can use to put in arbitrary C code 03:59:16 then link it in later 03:59:22 and c-define-type to handle types 03:59:23 oh cool 03:59:39 well you can do gsc -c code.scm and ten you get a C file 04:00:00 the very idea of these two languages intertwined the way they are makes me thrilled 04:00:35 one neat thing about it too, is that you can (include "somefiles.scm") into the "bindings" file and it will just get built in 04:00:54 oh right on 04:00:56 And you can load in code and runtime, all the neat stuff like that 04:01:01 at* 04:01:07 I love it for testing thing out 04:01:22 cool sounds like my kind of stuff 04:01:46 only problems are with macros 04:02:02 because when you compile a file the macros are expanded 04:02:07 jyaan, one other thing im really looking for is something with sqlite sort of bindings or some sort of connectivity to that specific database 04:02:18 and any macros in that file can't be called cause compile time is over 04:02:22 jyaan, i rarely use macros 04:02:34 then you need to learn more about them. :P 04:02:37 REPLeffect [~REPLeffec@69.54.115.254] has joined #scheme 04:02:42 Lisp macros aren't C macros 04:02:47 oh ok 04:02:53 They're actually a GOOD thing lol 04:02:57 lol 04:03:07 ya you know then how i could see that from the past thats ok :P 04:03:28 They take in code and produce code, it's not like the crazy string-ish stuff from C and C++ 04:03:46 One great thing you can do with macros is compile-time evaluation 04:04:02 teurastaja [~Samuel@modemcable173.144-131-66.mc.videotron.ca] has joined #scheme 04:04:08 oh ok cool 04:04:12 So if you can figure out a result at compile-time, then your code will be faster 04:04:36 There are some gotchas but overall it's good 04:05:05 C++ lately made a big deal about templates and template metaprogramming 04:05:11 jyaan, say if i wanted to find an example of a certain math algorithm in scheme, is there some sort of place id find that on the web? like for instance a lambda or whatever of a cartesian product? 04:05:16 Well, that's basically a very weak version of this idea 04:05:29 ya templates are huge in c++ 04:05:39 Sample code for it? 04:05:44 sure 04:06:14 Mainly the kind of sites that focus on algorithms 04:06:17 -!- acarrico [~acarrico@pppoe-68-142-49-233.gmavt.net] has quit [Ping timeout: 276 seconds] 04:06:31 generally they have tons of languages 04:06:44 Just do a little searching and it shouldn't be hard to find a couple 04:06:49 thats one thing thats been holding me back a bit like i read through an entire chapter of that book and at the end i realized i didnt even know how to find the intersection of two lists 04:06:55 Haha 04:06:59 i want a r5rs or r6rs compliant compiler/interpreter with good networking support that keeps "little features" in mind does that exist? 04:07:46 Mmm 04:07:57 What kind of networking 04:08:20 I'm not really the person to ask though (weakest at networking.. :() 04:08:34 like socket support? 04:08:41 nicktick [~nicktick@unaffiliated/nicktick] has joined #scheme 04:08:46 now that i think about that, that would be useful 04:08:50 well most have at least socket communication 04:08:54 of some kind 04:08:58 cool 04:09:09 I'd want more than just bare sockets though lol 04:09:32 yes socket support and preferably tcp/udp support also 04:09:34 anything after bare sockets are a plus i guess, but ive been down that far so its just like ok fine ill do that hehe 04:09:45 Chicken probably has the most libraries 04:09:58 http://wiki.call-cc.org/chicken-projects/egg-index-4.html#net 04:10:06 jyaan, how does chicken compare to bigloo? or have you used bigloo? 04:10:22 I only really know the C-compiler Schemes though 04:10:36 Bigloo has some nonstandard stuff but it's good too 04:10:48 It has things like the ability to specify types 04:10:50 i cant stand looking at all these scheme implementations that finally look more like common lisp with no standard 04:11:02 And also you can compile to several targets like jvm 04:11:08 lol 04:11:29 i know what you mean 04:11:42 jyaan, thats a selling point for me, being able to do both, c and jvm 04:11:48 plus just considering all the duplication of effort is a bit painful 04:12:00 does .Net also 04:12:02 jyaan, especially now that all the phones do java 04:12:16 jyaan, i work in the mobile industry :P 04:12:18 Not sure how good the .net is tough 04:12:20 though* 04:12:30 my h key isn't working correctly, it's driving me crazy 04:12:43 .net ofcourse is just the 'clr' aka what they'd have comparable to a jvm 04:12:51 Also lshift+g isn't working, has to be rshift+g, also driving me nuts 04:12:57 yea 04:13:12 jyaan, buy yourself another keyboard and stop getting malware from porn sites ;)lol 04:13:19 clr is pretty nice, but it's pretty far behind jvm on performanc 04:13:20 e 04:14:07 i think it's just stupid that they call it cross-platform but they really mean the different windows versions lol 04:14:07 oh wow cairo and opengl bindings in chicken 04:14:17 cross-windows :) 04:14:17 yea they have a lot 04:14:39 well the interesting thing about .net is the mono project, and that MS actually made most of those things open standards 04:14:49 yea i know about mono 04:15:02 but it still doesnt have the ground java does, but we'll see maybe oracle will fk up and kill off people wanting to use java 04:15:05 it's really what actually makes .net x-platform 04:15:11 stupid move for how prevalent android is right now 04:15:23 yea i was kind of irritated 04:15:38 irritated by what? 04:15:50 everyone is pissed off about that and the way they are mishandling mysql stuff, people are jumping ship now that opensolaris is being tanked with mysql too 04:15:52 oracle's patent/copyright thing against google 04:15:59 yep 04:16:01