Datasets:
Tasks:
Text Generation
Modalities:
Text
Sub-tasks:
language-modeling
Languages:
English
Size:
100K - 1M
License:
(* ========================================================================= *) | |
(* Create search database from OCaml / modify search database dynamically. *) | |
(* *) | |
(* This file assigns to "theorems", which is a list of name-theorem pairs. *) | |
(* The core system already has such a database set up. Use this file if you *) | |
(* want to update the database beyond the core, so you can search it. *) | |
(* *) | |
(* The trickery to get at the OCaml environment is due to oleg@pobox.com *) | |
(* (see his message to the caml-list on Tuesday 26th September 2006). *) | |
(* ========================================================================= *) | |
(* !!!!!!! You must set this to point at the source directory in | |
!!!!!!! which OCaml was built. (And don't do "make clean" beforehand.) | |
*) | |
let ocaml_source_dir = | |
Filename.concat (Sys.getenv "HOME") | |
("software/ocaml-"^Sys.ocaml_version);; | |
do_list (fun s -> Topdirs.dir_directory(Filename.concat ocaml_source_dir s)) | |
["parsing"; "typing"; "toplevel"; "utils"];; | |
(* This must be loaded first! It is stateful, and affects Predef *) | |
#load "ident.cmo";; | |
#load "misc.cmo";; | |
#load "path.cmo";; | |
#load "types.cmo";; | |
#load "btype.cmo";; | |
#load "tbl.cmo";; | |
#load "subst.cmo";; | |
#load "predef.cmo";; | |
#load "datarepr.cmo";; | |
#load "config.cmo";; | |
#load "consistbl.cmo";; | |
#load "clflags.cmo";; | |
#load "env.cmo";; | |
#load "ctype.cmo";; | |
#load "printast.cmo";; | |
#load "oprint.cmo";; | |
#load "primitive.cmo";; | |
#load "printtyp.cmo";; | |
(* ------------------------------------------------------------------------- *) | |
(* Get the toplevel environment as raw data. *) | |
(* ------------------------------------------------------------------------- *) | |
let get_value_bindings env = | |
let rec get_val acc = function | |
| Env.Env_empty -> acc | |
| Env.Env_value (next, ident, val_descr) -> | |
get_val ((ident,val_descr)::acc) next | |
| Env.Env_type (next,_,_) -> get_val acc next | |
| Env.Env_exception (next,_,_) -> get_val acc next | |
| Env.Env_module (next,_,_) -> get_val acc next | |
| Env.Env_modtype (next,_,_) -> get_val acc next | |
| Env.Env_class (next,_,_) -> get_val acc next | |
| Env.Env_cltype (next,_,_) -> get_val acc next | |
| Env.Env_open (next,_) -> get_val acc next | |
in get_val [] (Env.summary env);; | |
(* ------------------------------------------------------------------------- *) | |
(* Convert a type to a string, for ease of comparison. *) | |
(* ------------------------------------------------------------------------- *) | |
let type_to_str (x : Types.type_expr) = | |
Printtyp.type_expr Format.str_formatter x; | |
Format.flush_str_formatter ();; | |
(* ------------------------------------------------------------------------- *) | |
(* Put an assignment of a theorem database in the named file. *) | |
(* ------------------------------------------------------------------------- *) | |
let make_database_assignment filename = | |
let all_bnds = get_value_bindings (!Toploop.toplevel_env) in | |
let thm_bnds = filter (fun (ident,val_descr) -> | |
type_to_str val_descr.Types.val_type = "thm") | |
all_bnds in | |
let names = | |
subtract (map (fun (ident,val_descr) -> Ident.name ident) thm_bnds) | |
["it"] in | |
let entries = map (fun n -> "\""^n^"\","^n) (uniq(sort (<) names)) in | |
let text = "theorems :=\n[\n"^ | |
end_itlist (fun a b -> a^";\n"^b) entries^"\n];;\n" in | |
file_of_string filename text;; | |
(* ------------------------------------------------------------------------- *) | |
(* Remove bindings in first list from second assoc list (all ordered). *) | |
(* ------------------------------------------------------------------------- *) | |
let rec demerge s l = | |
match (s,l) with | |
u::t,(x,y as p)::m -> | |
if u = x then demerge t m | |
else if u < x then demerge t l | |
else p::(demerge s m) | |
| _ -> l;; | |
(* ------------------------------------------------------------------------- *) | |
(* Incrementally update database. *) | |
(* ------------------------------------------------------------------------- *) | |
let update_database = | |
let value_bindings_checked = ref 0 | |
and theorem_bindings_existing = ref undefined in | |
let listify l = if l = [] then "[]" | |
else "[\n"^end_itlist (fun a b -> a^";\n"^b) l^"\n]\n" in | |
let purenames = map (fun n -> "\""^n^"\"") | |
and pairnames = map (fun n -> "\""^n^"\","^n) in | |
fun () -> | |
let old_count = !value_bindings_checked | |
and old_ths = !theorem_bindings_existing in | |
let all_bnds = get_value_bindings (!Toploop.toplevel_env) in | |
let new_bnds = funpow old_count tl all_bnds in | |
let new_count = old_count + length new_bnds | |
and new_ths = | |
rev_itlist (fun (ident,val_descr) -> | |
let n = Ident.name ident in | |
if type_to_str val_descr.Types.val_type = "thm" && n <> "it" | |
then (n |-> ()) else undefine n) new_bnds old_ths in | |
value_bindings_checked := new_count; | |
if new_ths = old_ths then () else | |
(print_string "Updating search database\n"; | |
theorem_bindings_existing := new_ths; | |
let all_ths = combine (fun _ _ -> ()) (fun _ -> false) old_ths new_ths in | |
let del_ths = combine (fun _ _ -> ()) (fun _ -> true) all_ths new_ths | |
and add_ths = combine (fun _ _ -> ()) (fun _ -> true) all_ths old_ths in | |
let del_names = mergesort (<) (foldr (fun a _ l -> a::l) del_ths []) | |
and add_names = mergesort (<) (foldr (fun a _ l -> a::l) add_ths []) in | |
let exptext = | |
"theorems :=\n merge (increasing fst) (demerge "^ | |
(listify(purenames del_names))^ | |
" (!theorems)) "^ | |
(listify(pairnames add_names))^ | |
";;\n" in | |
(let filename = Filename.temp_file "database" ".ml" in | |
file_of_string filename exptext; | |
loadt filename; | |
Sys.remove filename));; | |
(* ------------------------------------------------------------------------- *) | |
(* Include a call to this on each search. *) | |
(* ------------------------------------------------------------------------- *) | |
let search = | |
let rec immediatesublist l1 l2 = | |
match (l1,l2) with | |
[],_ -> true | |
| _,[] -> false | |
| (h1::t1,h2::t2) -> h1 = h2 && immediatesublist t1 t2 in | |
let rec sublist l1 l2 = | |
match (l1,l2) with | |
[],_ -> true | |
| _,[] -> false | |
| (h1::t1,h2::t2) -> immediatesublist l1 l2 || sublist l1 t2 in | |
let exists_subterm_satisfying p (n,th) = can (find_term p) (concl th) | |
and name_contains s (n,th) = sublist (explode s) (explode n) in | |
let rec filterpred tm = | |
match tm with | |
Comb(Var("<omit this pattern>",_),t) -> not o filterpred t | |
| Comb(Var("<match theorem name>",_),Var(pat,_)) -> name_contains pat | |
| Comb(Var("<match aconv>",_),pat) -> exists_subterm_satisfying (aconv pat) | |
| pat -> exists_subterm_satisfying (can (term_match [] pat)) in | |
fun pats -> | |
update_database(); | |
let triv,nontriv = partition is_var pats in | |
(if triv <> [] then | |
warn true | |
("Ignoring plain variables in search: "^ | |
end_itlist (fun s t -> s^", "^t) (map (fst o dest_var) triv)) | |
else ()); | |
(if nontriv = [] && triv <> [] then [] | |
else itlist (filter o filterpred) pats (!theorems));; | |
(* ------------------------------------------------------------------------- *) | |
(* Update to bring things back to current state. *) | |
(* ------------------------------------------------------------------------- *) | |
theorems := [];; | |
update_database();; | |