miloignis/kraken
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
Files
2874be33324bb3b6c8e90025f430daae9604f260
kraken/to_compile.kp

163 lines
10 KiB
Plaintext
Raw Normal View History

Fix & and error checking for compiling environments as code & value, add a todo for things deferred
2022-01-05 01:05:16 -05:00
((wrap (vau root_env (quote)
((wrap (vau (let1)
(let1 lambda (vau se (p b1) (wrap (eval (array vau p b1) se)))
Clean up strip, have default memory allocation scale based on constants, added more until the next bug found, map seems not to be partially evaluating properly
2022-01-28 00:26:46 -05:00
(let1 current-env (vau de () de)
Fix & and error checking for compiling environments as code & value, add a todo for things deferred
2022-01-05 01:05:16 -05:00
(let1 cons (lambda (h t) (concat (array h) t))
Clean up strip, have default memory allocation scale based on constants, added more until the next bug found, map seems not to be partially evaluating properly
2022-01-28 00:26:46 -05:00
(let1 Y (lambda (f3)
((lambda (x1) (x1 x1))
(lambda (x2) (f3 (lambda (& y) (lapply (x2 x2) y))))))
Ah, found additional infinate recursion. In the process of fixing the second, trickier one, caused by compile & partial eval together. Should be fixed? but there's a nil coming out somewhere
2022-01-26 01:55:38 -05:00
(let1 vY (lambda (f)
((lambda (x3) (x3 x3))
Removed check_for_symbols, which was a bad idea generally. Simplify into combiner_return_ok, and fix *eval bug* - it didn't check for any sort of combiner_return_ok like thing, even though it's doing the same thing as a function call basically, by changing the env something is evaluated in. Also just re-write eval using the parital_eval wrapper, enforce it taking in total-values. In the future, can allow more partial eval 1) veval, returned from eval and posssibly others, essentailly (unwrap eval), which allows embedding partially-complete evals places without having to do something ugly like (eval (ensure_val partially_evaluated)) 2) Relax paramter is total-value to head-value and check for env_id inside it Immediatly: Debug bad partial evaluation on main test cases & crash with list index out of range
2022-02-03 02:41:14 -05:00
(lambda (x4) (f (vau de1 (& y) (vapply (x4 x4) y de1))))))
(let1 let (vY (lambda (recurse) (vau de2 (vs b) (cond (= (len vs) 0) (eval b de2)
true (vapply let1 (array (idx vs 0) (idx vs 1) (array recurse (slice vs 2 -1) b)) de2)))))
Implement drop_redundent_veval and make veval properly re-partially-eval and update the env it's called with. Y combiner test added and works now, map still seems not to, about to look at it
2022-02-07 00:31:51 -05:00
(let (
Move over much more code, including the tricky destructuring lambda which revealed bug in the need_for_progress system - a call that takes in the dynamic env that failed and had to be re-constructed would set attempted on the call to true, but would not note the dynamic environment as one of the needed-for-progress idxs. Often I think it would be anyway, so this didn't come up too often, and of course finally revealed itself when doing nested let/lambda destructuring stuff. Fixed by having attempted record not just true or false, but in the case where it's a call that takes in the dynamic env, makes it that env's id, which gets added to the for_progress_idxs.
2022-02-19 00:14:36 -05:00
lcompose (lambda (g f) (lambda (& args) (lapply g (array (lapply f args)))))
Implement drop_redundent_veval and make veval properly re-partially-eval and update the env it's called with. Y combiner test added and works now, map still seems not to, about to look at it
2022-02-07 00:31:51 -05:00
rec-lambda (vau se (n p b) (eval (array Y (array lambda (array n) (array lambda p b))) se))
Move over much more code, including the tricky destructuring lambda which revealed bug in the need_for_progress system - a call that takes in the dynamic env that failed and had to be re-constructed would set attempted on the call to true, but would not note the dynamic environment as one of the needed-for-progress idxs. Often I think it would be anyway, so this didn't come up too often, and of course finally revealed itself when doing nested let/lambda destructuring stuff. Fixed by having attempted record not just true or false, but in the case where it's a call that takes in the dynamic env, makes it that env's id, which gets added to the for_progress_idxs.
2022-02-19 00:14:36 -05:00
if (vau de (con than & else) (cond (eval con de) (eval than de)
(> (len else) 0) (eval (idx else 0) de)
true false))
Removed check_for_symbols, which was a bad idea generally. Simplify into combiner_return_ok, and fix *eval bug* - it didn't check for any sort of combiner_return_ok like thing, even though it's doing the same thing as a function call basically, by changing the env something is evaluated in. Also just re-write eval using the parital_eval wrapper, enforce it taking in total-values. In the future, can allow more partial eval 1) veval, returned from eval and posssibly others, essentailly (unwrap eval), which allows embedding partially-complete evals places without having to do something ugly like (eval (ensure_val partially_evaluated)) 2) Relax paramter is total-value to head-value and check for env_id inside it Immediatly: Debug bad partial evaluation on main test cases & crash with list index out of range
2022-02-03 02:41:14 -05:00
Implement drop_redundent_veval and make veval properly re-partially-eval and update the env it's called with. Y combiner test added and works now, map still seems not to, about to look at it
2022-02-07 00:31:51 -05:00
map (lambda (f5 l5)
; now maybe errors on can't find helper?
(let (helper (rec-lambda recurse (f4 l4 n4 i4)
(cond (= i4 (len l4)) n4
Made vapply and lapply primitives for efficiency/partial eval reasons. It means they can exist in emitted code without calling out to eval, which is good. to_compile.kp now both compiles and runs! Now to add more
2022-02-15 23:16:27 -05:00
(<= i4 (- (len l4) 4)) (recurse f4 l4 (concat n4 (array
(f4 (idx l4 (+ i4 0)))
(f4 (idx l4 (+ i4 1)))
(f4 (idx l4 (+ i4 2)))
(f4 (idx l4 (+ i4 3)))
)) (+ i4 4))
Implement drop_redundent_veval and make veval properly re-partially-eval and update the env it's called with. Y combiner test added and works now, map still seems not to, about to look at it
2022-02-07 00:31:51 -05:00
true (recurse f4 l4 (concat n4 (array (f4 (idx l4 i4)))) (+ i4 1)))))
(helper f5 l5 (array) 0)))
Move over much more code, including the tricky destructuring lambda which revealed bug in the need_for_progress system - a call that takes in the dynamic env that failed and had to be re-constructed would set attempted on the call to true, but would not note the dynamic environment as one of the needed-for-progress idxs. Often I think it would be anyway, so this didn't come up too often, and of course finally revealed itself when doing nested let/lambda destructuring stuff. Fixed by having attempted record not just true or false, but in the case where it's a call that takes in the dynamic env, makes it that env's id, which gets added to the for_progress_idxs.
2022-02-19 00:14:36 -05:00
map_i (lambda (f l)
(let (helper (rec-lambda recurse (f l n i)
(cond (= i (len l)) n
(<= i (- (len l) 4)) (recurse f l (concat n (array
(f (+ i 0) (idx l (+ i 0)))
(f (+ i 1) (idx l (+ i 1)))
(f (+ i 2) (idx l (+ i 2)))
(f (+ i 3) (idx l (+ i 3)))
)) (+ i 4))
true (recurse f l (concat n (array (f i (idx l i)))) (+ i 1)))))
(helper f l (array) 0)))
filter_i (lambda (f l)
(let (helper (rec-lambda recurse (f l n i)
(if (= i (len l))
n
(if (f i (idx l i)) (recurse f l (concat n (array (idx l i))) (+ i 1))
(recurse f l n (+ i 1))))))
(helper f l (array) 0)))
filter (lambda (f l) (filter_i (lambda (i x) (f x)) l))
not (lambda (x) (if x false true))
; Huge thanks to Oleg Kiselyov for his fantastic website
; http://okmij.org/ftp/Computation/fixed-point-combinators.html
Y* (lambda (& l)
((lambda (u) (u u))
(lambda (p)
(map (lambda (li) (lambda (& x) (lapply (lapply li (p p)) x))) l))))
vY* (lambda (& l)
((lambda (u) (u u))
(lambda (p)
(map (lambda (li) (vau ide (& x) (vapply (lapply li (p p)) x ide))) l))))
let-rec (vau de (name_func body)
(let (names (filter_i (lambda (i x) (= 0 (% i 2))) name_func)
funcs (filter_i (lambda (i x) (= 1 (% i 2))) name_func)
overwrite_name (idx name_func (- (len name_func) 2)))
(eval (array let (concat (array overwrite_name (concat (array Y*) (map (lambda (f) (array lambda names f)) funcs)))
(lapply concat (map_i (lambda (i n) (array n (array idx overwrite_name i))) names)))
body) de)))
let-vrec (vau de (name_func body)
(let (names (filter_i (lambda (i x) (= 0 (% i 2))) name_func)
funcs (filter_i (lambda (i x) (= 1 (% i 2))) name_func)
overwrite_name (idx name_func (- (len name_func) 2)))
(eval (array let (concat (array overwrite_name (concat (array vY*) (map (lambda (f) (array lambda names f)) funcs)))
(lapply concat (map_i (lambda (i n) (array n (array idx overwrite_name i))) names)))
body) de)))
flat_map (lambda (f l)
(let (helper (rec-lambda recurse (f l n i)
(if (= i (len l))
n
(recurse f l (concat n (f (idx l i))) (+ i 1)))))
(helper f l (array) 0)))
flat_map_i (lambda (f l)
(let (helper (rec-lambda recurse (f l n i)
(if (= i (len l))
n
(recurse f l (concat n (f i (idx l i))) (+ i 1)))))
(helper f l (array) 0)))
; with all this, we make a destrucutring-capable let
let (let (
destructure_helper (rec-lambda recurse (vs i r)
(cond (= (len vs) i) r
(array? (idx vs i)) (let (bad_sym (str-to-symbol (str (idx vs i)))
;new_vs (flat_map_i (lambda (i x) (array x (array idx bad_sym i))) (slice (idx vs i) 1 -1))
new_vs (flat_map_i (lambda (i x) (array x (array idx bad_sym i))) (idx vs i))
)
(recurse (concat new_vs (slice vs (+ i 2) -1)) 0 (concat r (array bad_sym (idx vs (+ i 1))))))
true (recurse vs (+ i 2) (concat r (slice vs i (+ i 2))))
))) (vau de (vs b) (vapply let (array (destructure_helper vs 0 (array)) b) de)))
; and a destructuring-capable lambda!
only_symbols (rec-lambda recurse (a i) (cond (= i (len a)) true
(symbol? (idx a i)) (recurse a (+ i 1))
true false))
lambda (vau se (p b) (if (only_symbols p 0) (vapply lambda (array p b) se)
(let (
sym_params (map (lambda (param) (if (symbol? param) param
(str-to-symbol (str param)))) p)
body (array let (flat_map_i (lambda (i x) (array (idx p i) x)) sym_params) b)
) (wrap (eval (array vau sym_params body) se)))))
; and rec-lambda - yes it's the same definition again
rec-lambda (vau se (n p b) (eval (array Y (array lambda (array n) (array lambda p b))) se))
Fix the bug with de getting seperate dynamic envs, now unified but no opts for now (doing it seperetly was broken because of tyring to access it from inner closures. Also add a lot more runtime logging
2022-02-20 22:12:21 -05:00
;test0 (map (lambda (x) (+ x 1)) (array 1 2))
;test1 (map_i (lambda (i x) (+ x i 1)) (array 1 2))
;test2 (filter_i (lambda (i x) (> i 0)) (array 1 2))
;test2 (filter (lambda ( x) (> x 1)) (array 1 2))
;test3 (not 1)
;test4 (flat_map (lambda (x) (array 1 x 2)) (array 1 2))
;test5 (flat_map_i (lambda (i x) (array i x 2)) (array 1 2))
;test6 (let ( (a b) (array 1 2) c (+ a b) ) c)
;test7 ((rec-lambda recurse (n) (cond (= 0 n) 1
; true (* n (recurse (- n 1))))) 5)
;test8 ((lambda (a b c) (+ a b c)) 1 13 14)
;test9 ((lambda (a (b c)) (+ a b c)) 1 (array 13 14))
Implement drop_redundent_veval and make veval properly re-partially-eval and update the env it's called with. Y combiner test added and works now, map still seems not to, about to look at it
2022-02-07 00:31:51 -05:00
;monad (array 'open 3 "test_self_out" (lambda (fd code)
; (array 'write fd "wabcdefghijklmnopqrstuvwx" (lambda (written code)
; (array 'exit (if (= 0 written) 12 14))))))
;old 4
;test (+ old 4)
;test 4
Move over much more code, including the tricky destructuring lambda which revealed bug in the need_for_progress system - a call that takes in the dynamic env that failed and had to be re-constructed would set attempted on the call to true, but would not note the dynamic environment as one of the needed-for-progress idxs. Often I think it would be anyway, so this didn't come up too often, and of course finally revealed itself when doing nested let/lambda destructuring stuff. Fixed by having attempted record not just true or false, but in the case where it's a call that takes in the dynamic env, makes it that env's id, which gets added to the for_progress_idxs.
2022-02-19 00:14:36 -05:00
;monad (array 'write 1 "test_self_out2" (vau (written code) (map (lambda (x) (+ x 133)) (array written code))))
;monad (array 'write 1 "test_self_out2" (vau (written code) (map_i (lambda (i x) (+ x i 133)) (array written code))))
;monad (array 'write 1 "test_self_out2" (vau (written code) (filter_i (lambda (i x) (> i 0)) (array written code))))
;monad (array 'write 1 "test_self_out2" (vau (written code) (filter (lambda (x) (> x 0)) (array written code))))
;monad (array 'write 1 "test_self_out2" (vau (written code) (not (array written code))))
;monad (array 'write 1 "test_self_out2" (vau (written code) (flat_map (lambda (x) (array 1 x 2)) (array written code))))
;monad (array 'write 1 "test_self_out2" (vau (written code) (flat_map_i (lambda (i x) (array i x 2)) (array written code))))
Fix the bug with de getting seperate dynamic envs, now unified but no opts for now (doing it seperetly was broken because of tyring to access it from inner closures. Also add a lot more runtime logging
2022-02-20 22:12:21 -05:00
;monad (array 'write 1 "test_self_out2" (vau (written code) (let ( (a b) (array written code) c (+ a b test8 test9)) c)))
monad (array 'write 1 "test_self_out2" (vau (written code) ((lambda (a (b c)) (+ a b c)) 1 (array written code))))
Implement drop_redundent_veval and make veval properly re-partially-eval and update the env it's called with. Y combiner test added and works now, map still seems not to, about to look at it
2022-02-07 00:31:51 -05:00
)
monad
)
;(array 'write 1 "test_self_out2" (vau (written code) 7))
Fix & and error checking for compiling environments as code & value, add a todo for things deferred
2022-01-05 01:05:16 -05:00
; end of all lets
Made vapply and lapply primitives for efficiency/partial eval reasons. It means they can exist in emitted code without calling out to eval, which is good. to_compile.kp now both compiles and runs! Now to add more
2022-02-15 23:16:27 -05:00
))))))
Fix & and error checking for compiling environments as code & value, add a todo for things deferred
2022-01-05 01:05:16 -05:00
; impl of let1
Implement veval and combiner_return_ok enhancements, now let4.3 behaves like the proper macro expansion. Compiling our test doesn't work though, it doesn't partial evalaute the vaus away as expected - changing let1 implementations does have an effect. Will need to investigate, as well as add support for compiling away veval
2022-02-05 12:14:13 -05:00
; this would be the macro style version (((
)) (vau de (s v b) (eval (array (array wrap (array vau (array s) b)) v) de)))
;)) (vau de (s v b) (eval (array (array vau (array s) b) (eval v de)) de)))
Fix & and error checking for compiling environments as code & value, add a todo for things deferred
2022-01-05 01:05:16 -05:00
; impl of quote
I belive I fixed the bug revealed by the let definition - it was a similar thing where <comb>s were counted as values even if there were fake envs inside, and then thouse fake envs could be moved inside another env_stack and then later resolve to a wrong env, or at lest I think that's what was happening. The let test takes too long to run now - I killed it at 20 minutes and 48GB of RAM taken. Given that, it's now time to move on towards optimization
2022-01-07 22:08:29 -05:00
)) (vau (x5) x5))
Reference in New Issue Copy Permalink