Hi, Your dump sounds a lot like the FORTH and Factor return stack. The return stack is visible and can be used for temporary storage: 1 3 7 >r + r> ==> 4 7 ......
Slava Pestov
slava@...
Jul 1, 2004 9:26 am
1902
... In essence, yes. Actual return addresses in the Joy0 and Joy1 interpreters are stored on the C stack; the dump is used because the precise garbage...
... I get around this problem in the new C rewrite of Factor by keeping as little as possible in C! Right now the C core is 1,400 lines. The rest -- control...
Slava Pestov
slava@...
Jul 1, 2004 5:33 pm
1904
... I urge you again, when you add garbage collection, to use the Boehm conservative collector. Writing bulletproof garbage collectors is *hard*. I've...
... From: "John Cowan" <cowan@...> [:] ... my first cut at cK relies on K's stack, which is the C stack. but this imposes a hard limit on the depth of...
stevan apter
sa@...
Jul 1, 2004 10:08 pm
1906
On Thu, 1 Jul 2004, John Cowan wrote: [..] ... Yes, and for most cases local C variables could replace the dumps. But there are a few troublesome cases for...
I think the simplest solution is to write these combinators in Joy itself, with a minimal set in C (ifte and such). Why was this approach not chosen for the...
Slava Pestov
slava@...
Jul 5, 2004 6:37 am
1908
... "The trampoline that doesn't bend your brain is not the true trampoline." This is the best brief explanation of GNU C trampolines I know, from...
Hi everybody, I decided to start a weblog. It will be mainly about Factor. http://www.jroller.com/page/slava/ Slava...
Slava Pestov
slava@...
Jul 6, 2004 7:06 am
1910
chris: i think you may have inadvertently forgotten to include the definition of 'init-amb': init-amb solve => "female" "male" "female" Just for fun, here's a...
sa@...
Jul 8, 2004 7:00 pm
1911
... Yes, it's really a global renaming problem, sorry about that. 'init-amb' should be 'init-choose', and the code for that is included. It sets up the global...
Chris Double
chris.double@...
Jul 11, 2004 12:47 am
1912
Hi everybody, I've done a lot of work on the C port of Factor in the last few days. Notable achievements: - The VM is now 16kb when compiled -- the library...
Slava Pestov
slava@...
Jul 12, 2004 6:41 am
1913
... [.. long quote from the above ..] ... Thanks a lot for all that, John. It was very helpful. To summarise, for a future re-implementation of Joy: The...
On Thu, 8 Jul 2004 sa@... wrote: [..] ... It may be of interest that backtracking can be implemented in Joy without the general primitives "amb" and...
On Mon, 12 Jul 2004 18:22:45 +1100, phimvt@... ... Thanks Manfred, I'll take a look. I'm still getting used to the different ways of doing...
Chris Double
chris.double@...
Jul 12, 2004 9:26 pm
1916
Hi everybody, Factor is now in the SourceForge CVS repository. There is a new download archive: factor.sf.net/Factor.zip. It includes Factor.jar, as well as...
Slava Pestov
slava@...
Jul 16, 2004 7:35 am
1917
hi all this is an interim report on some work i've been doing on a new (toy) language, which i'm calling 'xy'. the code is here: www.nsl.com/k/xy/xy.k xy is...
stevan apter
sa@...
Jul 18, 2004 4:49 pm
1918
Hi Steve, Good to hear about your experiments with concatenative languages again! ... I don't understand this. ... Does this mean code can be inspected and...
Slava Pestov
slava@...
Jul 19, 2004 2:32 am
1919
Hello everybody, I have added error handling using catch/throw to Factor (both backends). You can download the latest code from http://factor.sf.net. The...
i've fixed a few bugs, and further trimmed the set of primitives. here they are: k: ~!@#$%^&*_-+=|<,>?.: f dyad f: monad f. commute dyad k system functions,...
stevan apter
sa@...
Jul 20, 2004 10:26 pm
1922
oops - a few mistakes in the earlier draft. ... xy primitives: k: ~!@#$%^&*_-+=|<,>?.: f dyad f: monad f. commute dyad system functions, n:, &c. ( )...
basic documentation for XY is done (http://www.nsl.com/papers/xy.htm), and most of the bugs have been knocked out of the code (accessible from the paper.) i've...
sa@...
Jul 28, 2004 6:06 pm
1925
i hate my mail program. hate hate hate. find the un-pretzeled version on my website....
sa@...
Jul 28, 2004 6:38 pm
1926
It would be useful for readers if there was some kind of compact comparison of the various concatenative languages that we have been discussing here. I am...
I'd be glad to write about Forth. When you ask for a comparison, though, do you mean something more like a precis, or a summary? I couldn't compare Forth to...
Hi, I think this is a great idea! Do you want each section to be 'free standing', or discuss differences with other concatenative languages?...
Slava Pestov
slava@...
Aug 4, 2004 1:56 am
1929
i'm not sure when i'll be able to get to this - perhaps in a few weeks. meanwhile, i've encountered some serious difficulties with the xy design, which i'm not...
sa
sa@...
Aug 4, 2004 7:51 am
1930
... This makes sense. ... This doesn't make sense to me. Oh, yes, I suspect you can make it work this way, but allowing things to execute inside your stack...
